ReadingTimeMachine/rtm-sgt-ocr-v1

1source,target2" The mean alieumieut angles are 62° for both siuple aud couples dust features in D2 sources. iud GS"" aid 387 for simple and"," The mean alignment angles are $62\degr $ for both simple and complex dust features in B2 sources, and $68\degr $ and $38\degr $ for simple and"3believe that there is good evidence that the absorbers in T1821|613 are truly intervening. based on the proximity of intervening galaxies to the line of sight (Tripp.Lu& 2000).,"believe that there is good evidence that the absorbers in H1821+643 are truly intervening, based on the proximity of intervening galaxies to the line of sight \citep{tripp98,bowen00}."4. Tripp.Lu&Savage(1998) report a strong. bleuded line of Lvo at 2=0.12123.0.12157.," \citet{tripp98} report a strong, blended line of $\alpha$ at $z=0.12123, 0.12157$."5 In addition to detecting the the Ly. line in this svstem. we report the detection of a line at wwhich we have identified as at 2=0.121237 (see Figure 2).," In addition to detecting the the $\beta$ line in this system, we report the detection of a line at which we have identified as at $z=0.12137$ (see Figure 2)."6 The line is clearly seen in both the LiET and LiF2 spectra., The line is clearly seen in both the LiF1 and LiF2 spectra.7 The Line is blended with the Lvà line in the +=0.225 absorber aud could not be measured., The line is blended with the $\delta$ line in the $z=0.225$ absorber and could not be measured.8 This is the sixthintervening aabsorptiou svsteni observed towards T1821)613. aud provides further evideuce that Ίο: sxvstenuis contain a large fraction of the barvous at the present epoch (Tripp.Savage&Jenkins2000:TrippSavage 2000)..," This is the sixth absorption system observed towards H1821+643, and provides further evidence that $z$ systems contain a large fraction of the baryons at the present epoch \citep{tripp00,tripp00_2}."9 These observations are in accord with cosmological sinmlatious by Cen&Ostriker(1999). that predict a substantial fraction of present-day barvons are in a shock-heatedphase at 10° LOK. We have identified an observed line at aas lam90977 at 2=0.1705., These observations are in accord with cosmological simulations by \citet{cen99} that predict a substantial fraction of present-day baryons are in a shock-heatedphase at $10^5-10^7$ K. We have identified an observed line at as 977 at $z=0.1705$.10 This ideutification was made possible by comparison with the wavelength of a Lyra absorber recently observed by Tripp.Savage&Jeulkins(2000) with STIS/IIST.," This identification was made possible by comparison with the wavelength of a $\alpha$ absorber recently observed by \citet{tripp00}11 with STIS/HST."12 The Lv line for this absorber is not conclusively detected in the FUSE. spectrum. although a weak. rather broad feature is at the predicted wavelength.," The $\gamma$ line for this absorber is not conclusively detected in the FUSE spectrum, although a weak, rather broad feature is at the predicted wavelength."13 The Ly. line in this absorber is redshitted iuto the strong Calactic ttriplet at, The $\beta$ line in this absorber is redshifted into the strong Galactic triplet at.14 Tutervening absorption at 2=0.225 has been detected in the Lya aud Ly? lines aud in wwith DST (Savage.Tripp&Lu1998)., Intervening absorption at $z=0.225$ has been detected in the $\alpha$ and $\beta$ lines and in with HST \citep{savage98}.15. Recently. Tripp.Savage&Jeukius(2000) have also clearly detected in this svstem. and the compoucut structure establishes that this is a multiphase absorber.," Recently, \citet{tripp00} have also clearly detected in this system, and the component structure establishes that this is a multiphase absorber."16 However. this absorber is not detected by IIST in low ionization lines such as oor the high ionization lines of aud.," However, this absorber is not detected by HST in low ionization lines such as or the high ionization lines of and."17 Savage.Tripp&Lu(1998) showed that the absorption could occur in low density. extended eas pliotoionized by the UV. background or in hot collisionally ionized gas iu an iutervening galaxy or galaxy eroup.," \citet{savage98} showed that the absorption could occur in low density, extended gas photoionized by the UV background or in hot collisionally ionized gas in an intervening galaxy or galaxy group."18 Evidence for the presence of a galaxy eroup at 2=0.225 has been provided by Schneideretal.(1992) and Tripp.Lu&Savage(1998)., Evidence for the presence of a galaxy group at $z=0.225$ has been provided by \citet{schneider92} and \citet{tripp98}.19. The Lyd. Lye aud Lyà lines at +=0.22191 axe detected in the FUSE spectrum.," The $\delta$ , $\epsilon$ and $\zeta$ lines at $z=0.22491$ are detected in the FUSE spectrum."20 Two of these lines are shown in Figure 2., Two of these lines are shown in Figure 2.21 The lines are clearly resolved iuto 2 compoucuts with velocity separation of ~70., The lines are clearly resolved into 2 components with velocity separation of $\sim 70$.22 We have not identified amy metal lines arising in the =0.225 svstem in the FUSE spectrum., We have not identified any metal lines arising in the $z=0.225$ system in the FUSE spectrum.23 It would be interesting to obtain FUSE short-waveleneth (SiC) spectra covering 000995A©where we have the possibility of detecting the redshitted LL7770.780 doublet.," It would be interesting to obtain FUSE short-wavelength (SiC) spectra covering $900-995$,where we have the possibility of detecting the redshifted 770,780 doublet."24 hhas a higher ionization potential than ((207 eV vs. LLL eV)., has a higher ionization potential than (207 eV vs. 114 eV).25 Tf detected. it would indicate the presence of collisionally ionized gas at a teniperature of T~10 ON. We report the detection of several “associated” (tute2 tem) absorption lines in the spectrum of T18211612.," If detected, it would indicate the presence of collisionally ionized gas at a temperature of $T\sim10^{5.5}$ K. We report the detection of several “associated” $z_{abs} \approx26z_{em}$ ) absorption lines in the spectrum of H1821+643."27 We identify the observed line at L021.ΙΑ Pas absorption by the rest-frame EUV line lam7787.7 (sco Figure 3)., We identify the observed line at $1021.45$ as absorption by the rest-frame EUV line 787.7 (see Figure 3).28 This line has a EWIIM ~150 ©. which places it in the category of nirrow absorption line (NAL) absorbers(Wevinannetal. 2000)...," This line has a FWHM $\sim 150$ , which places it in the category of narrow absorption line (NAL) absorbers\citep{weymann79,hamann00}. ."29 We have also ideutified lami8832.93 with an observed line at Aj4;~1080.05À ©., We have also identified 832.93 with an observed line at $\lambda_{obs} \sim 1080.05$ .30 Finally. we have teutatively identified a weak line at Aups71019:SSA ," Finally, we have tentatively identified a weak line at $\lambda_{obs}31\sim 1019.85$ "3222222). although more conservative interpretations have been provided by other analysis (aseg.by?)..,", although more conservative interpretations have been provided by other analysis \citep[as \eg by][]{Mortonson_Hu_Huterer_2011}."33 The main difficulties in driving sufficiently robust conclusions on a possible tension with the standard ACDM model from such detections arise as a consequence of the observational uncertainties on the mass determination of the clusters and on the estimation of the statistical significance of the cosmological volumes covered by the cluster surveys (22)..," The main difficulties in driving sufficiently robust conclusions on a possible tension with the standard $\Lambda $ CDM model from such detections arise as a consequence of the observational uncertainties on the mass determination of the clusters and on the estimation of the statistical significance of the cosmological volumes covered by the cluster surveys \citep{Sheth_Diaferio_2011,Waizmann_Ettori_Moscardini_2011}."34 Such significance will need to be better claritied before claiming a diserepaney with respect to the predictions of the standard. model., Such significance will need to be better clarified before claiming a discrepancy with respect to the predictions of the standard model.35 Nevertheless. it is in any case interesting to investigate whether such possible discrepancy could be alleviated by alternative cosmological scenarios.," Nevertheless, it is in any case interesting to investigate whether such possible discrepancy could be alleviated by alternative cosmological scenarios."36 Several attempts have been made in this direction in the recent past., Several attempts have been made in this direction in the recent past.37 On one side. a deviation from statistical Gaussianity in the primordial density field has been shown to give rise to a higher number density of massive halos as compared to the standard Gaussian case (seee.g.22222222).," On one side, a deviation from statistical Gaussianity in the primordial density field has been shown to give rise to a higher number density of massive halos as compared to the standard Gaussian case \citep[see \eg][]{Matarrese_Verde_Jimenez_2000, Grossi_etal_2007,Jimenez_Verde_2009,Holz_Perlmutter_2010,Cayon_Gordon_Silk_2010,Sartoris_etal_2010,Hoyle_Jimenez_Verde_2011,LoVerde_Smith_2010}."38 However. in order to significantly increase the probability of detection this approach requires a level of primordial non-Gaussianity that seems to be already ruled out by CMB constraints (2). unless a strongly scale-dependent non-Gaussianity is invoked (asproposedbye.g. ???).," However, in order to significantly increase the probability of detection this approach requires a level of primordial non-Gaussianity that seems to be already ruled out by CMB constraints \citep{wmap7}, unless a strongly scale-dependent non-Gaussianity is invoked \citep[as proposed by \eg][]{LoVerde_etal_2008,Cayon_Gordon_Silk_2010,Hoyle_Jimenez_Verde_2011}."39 Alternatively. ?. have shown that interacting DE models. which are generically characterized by a faster growth of density perturbations with respect to ACDM due to the presence of a mediated by the DE scalar field. also predict a larger number density of massive halos at all redshifts that would result in a higher detection probability.," Alternatively, \citet{Baldi_Pettorino_2011} have shown that interacting DE models, which are generically characterized by a faster growth of density perturbations with respect to $\Lambda $ CDM due to the presence of a fifth-force mediated by the DE scalar field, also predict a larger number density of massive halos at all redshifts that would result in a higher detection probability."40 Both these approaches. however. are faced by the additional problem of overpredicting the abundance of massive clusters at low redshift.," Both these approaches, however, are faced by the additional problem of overpredicting the abundance of massive clusters at low redshift."41 In fact. while the detection of massive clusters at high redshifts seems to imply an anomalous growth of density perturbations as compared to the predictions of the standard model. the observed number counts at low redshifts are found to be in ver-- good agreement with the predictions of \CDM (seee.g.2222).," In fact, while the detection of massive clusters at high redshifts seems to imply an anomalous growth of density perturbations as compared to the predictions of the standard model, the observed number counts at low redshifts are found to be in very good agreement with the predictions of $\Lambda $ CDM \citep[see \eg][]{Reiprich_Boehringer_2002,42Vikhlinin_etal_2009b, Mantz_etal_2010, Rozo_etal_2010}."43 In other words. a non-Gaussian initial density tield or a faster growth of density fluctuations can determine an increased number density of massive halos at high redshifts. thereby alleviating the possible tension with the model arising from the unexpected detection of an exceeding number of such objects. but will then also necessarily imply the further growth of such massive halos from these high redshifts to the present time. resulting in a clear discrepancy with present bounds on the cluster mass function at low redshifts.," In other words, a non-Gaussian initial density field or a faster growth of density fluctuations can determine an increased number density of massive halos at high redshifts, thereby alleviating the possible tension with the model arising from the unexpected detection of an exceeding number of such objects, but will then also necessarily imply the further growth of such massive halos from these high redshifts to the present time, resulting in a clear discrepancy with present bounds on the cluster mass function at low redshifts."44 In the present study we will show how this problem can be naturally solved by an interacting DE model with a suitable self-interaction potential different from the standard exponential or power-law potentials assumed in previous investigations of coupled DE (cDE) cosmologies., In the present study we will show how this problem can be naturally solved by an interacting DE model with a suitable self-interaction potential different from the standard exponential or power-law potentials assumed in previous investigations of coupled DE (cDE) cosmologies.45" In particular. we will show how the existence of a global minimum in the potential. and the consequent ""bounce"" of the DE field on the cosmological constant barrier. is the key feature to address the existence of exceedingly massive clusters at high redshift without necessarily affecting the halo mass function at the present epoch."," In particular, we will show how the existence of a global minimum in the potential, and the consequent “bounce"" of the DE field on the cosmological constant barrier, is the key feature to address the existence of exceedingly massive clusters at high redshift without necessarily affecting the halo mass function at the present epoch."46 Although our proposed scenario does not benetit in general of the same scaling properties typical of the exponential and power-law potentials. it naturally provides a mechanism to enhance the expected number of massive halos up to some high redshift and to subsequently reduce it again to the standard \CDM prediction at the present time. thereby accounting at the same time for the existence of massive clusters at early epochs and for the observed halo abundance at low redshift.," Although our proposed scenario does not benefit in general of the same scaling properties typical of the exponential and power-law potentials, it naturally provides a mechanism to enhance the expected number of massive halos up to some high redshift and to subsequently reduce it again to the standard $\Lambda $ CDM prediction at the present time, thereby accounting at the same time for the existence of massive clusters at early epochs and for the observed halo abundance at low redshift."47 We therefore regard the specific model presented in this work as a toy example of how the dynamical nature of a DE scalar field interacting with cold dark matter (CDM) particles can account for the presence of a high redshift peak in the deviation of the expected cluster number counts from the ACDM prediction., We therefore regard the specific model presented in this work as a toy example of how the dynamical nature of a DE scalar field interacting with cold dark matter (CDM) particles can account for the presence of a high redshift peak in the deviation of the expected cluster number counts from the $\Lambda $ CDM prediction.48" This feature could not arise in standard gravity models or in non-Gaussian cosmological scenarios. and would therefore represent a ""smoking gun"" for the dynamical nature of DE."," This feature could not arise in standard gravity models or in non-Gaussian cosmological scenarios, and would therefore represent a “smoking gun"" for the dynamical nature of DE."49 In the present work we will illustrate in detail this peculiar behavior. also by means of large N-body simulations of structure formation.," In the present work we will illustrate in detail this peculiar behavior, also by means of large N-body simulations of structure formation."50 The paper is organized as follows., The paper is organized as follows.51 In Section 2. we briefly review the main features of standard cDE models for he background evolution of the universe. and we highlight how bouncing CDE models deviate from this standard and well-known behavior.," In Section \ref{sec:cDE} we briefly review the main features of standard cDE models for the background evolution of the universe, and we highlight how bouncing cDE models deviate from this standard and well-known behavior."52 In Section 3 we investigate the evolution of linear density yerturbations in the context of both standard and bouncing cDE scenarios. and we show how the latter can provide a possible explanation for a larger amplitude of density perturbations at vigh redshift without affecting the normalization of the linear matter power spectrum at the present time.," In Section \ref{sec:linear} we investigate the evolution of linear density perturbations in the context of both standard and bouncing cDE scenarios, and we show how the latter can provide a possible explanation for a larger amplitude of density perturbations at high redshift without affecting the normalization of the linear matter power spectrum at the present time."53 This effect is further investigated for the nonlinear regime of structure formation in Section 4.. where we show by means of large N-body simulations how the real halo mass function evolves in time in each of the models considered. and how bouncing ceDE models can predict an enhanced number of massive clusters at high redshift without changing the halo abundance at the present epoch.," This effect is further investigated for the nonlinear regime of structure formation in Section \ref{sec:sims}, where we show by means of large N-body simulations how the real halo mass function evolves in time in each of the models considered, and how bouncing cDE models can predict an enhanced number of massive clusters at high redshift without changing the halo abundance at the present epoch."54 Finally. in Section 5 we draw our conclusions.," Finally, in Section \ref{sec:concl} we draw our conclusions."55 Interacting DE models involving a direct exchange of energy-momentum between a DE scalar field © and other matter components of the universe have been widely investigated in the past. in particular for the case of a coupling with the CDM fluid (2?2??) or with massive neutrinos (?2)..," Interacting DE models involving a direct exchange of energy-momentum between a DE scalar field $\phi $ and other matter components of the universe have been widely investigated in the past, in particular for the case of a coupling with the CDM fluid \citep{Wetterich_1995,Amendola_2000,Amendola_2004,Pettorino_Baccigalupi_2008,Baldi_2011a}56 or with massive neutrinos \citep{Amendola_Baldi_Wetterich_2008,Baldi_etal_2011a}."57 The effects of these interactions on observable quantities such as the CMB angular power spectrum (222).. the Lyman-a forest (2).. the large scale structure of the universe (2).. or the structural properties of highly nonlinear objects (2222) have been extensively studied by means of both analytic and numerical techniques.," The effects of these interactions on observable quantities such as the CMB angular power spectrum \citep{Bean_etal_2008,Xia_2009,LaVacca_etal_2009}, the $\alpha $ forest \citep{Baldi_Viel_2010}, the large scale structure of the universe \citep{Baldi_Pettorino_2011}, or the structural properties of highly nonlinear objects \citep{Baldi_etal_2010,Li_Barrow_2011,Baldi_2011b,Baldi_Lee_Maccio_2011} have been extensively studied by means of both analytic and numerical techniques."58 Most of the previous works on interacting DE models assume a monotonic function for the self-interaction potential of the DE scalar field. as an exponential (2). or an inverse power law (?) potential. which determine the existence of attractor solutions that are almost independent on the initial conditions of the scalar field.," Most of the previous works on interacting DE models assume a monotonic function for the self-interaction potential of the DE scalar field, as an exponential \citep{Wetterich_1988} or an inverse power law \citep{Ratra_Peebles_1988} potential, which determine the existence of attractor solutions that are almost independent on the initial conditions of the scalar field."59 These potentials do not possess local minima. and therefore define a unique direction for the time evolution of the scalar field ©.," These potentials do not possess local minima, and therefore define a unique direction for the time evolution of the scalar field $\phi $."60 However. other possible functions. characterized in some cases by the presence of local or global minima. have been considered in the literature for the case of non-interacting DE scenarios. as the SUGRA potential (2) arising naturally in supersymmetrie theories of gravity.," However, other possible functions, characterized in some cases by the presence of local or global minima, have been considered in the literature for the case of non-interacting DE scenarios, as the SUGRA potential \citep{Brax_Martin_1999} arising naturally in supersymmetric theories of gravity."61 The existence of a local or a global minimum in the scalar field self-interaction potential allows the field to oscillate and invert its motion during the expansion history of the Universe. as recently suggested by ?..," The existence of a local or a global minimum in the scalar field self-interaction potential allows the field to oscillate and invert its motion during the expansion history of the Universe, as recently suggested by \citet{Tarrant_etal_2011}."62 Such feature could have very significant consequences if a coupling to CDM is also present. as we will show in the present work.," Such feature could have very significant consequences if a coupling to CDM is also present, as we will show in the present work."63 We will study the case of an interacting DE model with a SUGRA self-interaction potential and compare it with the standard ACDM cosmology and with coupled DE models with, We will study the case of an interacting DE model with a SUGRA self-interaction potential and compare it with the standard $\Lambda $ CDM cosmology and with coupled DE models with64depends ou both position. aud redshift.,depends on both position and redshift.65 As a result of lensing. some regions of the source plane are observed to ercater depths than others. even if the scusitivity across the observed mapthe mage planeis uniformi.," As a result of lensing, some regions of the source plane are observed to greater depths than others, even if the sensitivity across the observed map—the image plane—is uniform."66 With a sufficiently detailed: miass model for the leus. corrections can be applied to model the effects of leusiug accurately. and thus the cuhanced seusitivitv to faint backerouud sources can be exploited.," With a sufficiently detailed mass model for the lens, corrections can be applied to model the effects of lensing accurately, and thus the enhanced sensitivity to faint background sources can be exploited."67 Moreover. lensine provides a further advantage: because sub-nnua wave telescopes have coarse angular resolution as compared with optical telescopes. source confusion can contribute noise in faint inages (Blain. Ivison. Sul 1998: ITughes et 11998).," Moreover, lensing provides a further advantage: because sub-mm wave telescopes have coarse angular resolution as compared with optical telescopes, source confusion can contribute noise in faint images (Blain, Ivison, Smail 1998; Hughes et 1998)."68 The flux deusitv of all backeround sources is iucreased by a lensing cluster. while their mean separation on the skv is also increased.," The flux density of all background sources is increased by a lensing cluster, while their mean separation on the sky is also increased."69 Both of these features reduce the problem of confusion., Both of these features reduce the problem of confusion.70 The lensing effect of the rich clusters exploited in the survey is modeled accurately by using multiple-componcut lass distributious that describe the extended poteutial well of the clusters aud their more massive individual 1iember galaxies IKI&neib et 11996)., The lensing effect of the rich clusters exploited in the survey is modeled accurately by using multiple-component mass distributions that describe the extended potential well of the clusters and their more massive individual member galaxies Kneib et 1996).71 These models are clerived using the positions of multiply imaged features identified iu high-resolution optical nuages and are very well constrained using the spectroscopic redshifts of these multiple images., These models are derived using the positions of multiply imaged features identified in high-resolution optical images and are very well constrained using the spectroscopic redshifts of these multiple images.72 Details of the mass models emploved cal be found in €100021]|16 (Smail et 11996). A310. (παν et 11993: Bézzecourt et 11999). | 02 (Cola et 11998). CTO0939| LF (Seitz et 11996). 11835 (Edee et 11999). 22390. (I&neib et 11999). and 11 02(J.-P. I&neib 1998. private cohuuunication).," Details of the mass models employed can be found in $+$ 16 (Smail et 1996), 370 (Kneib et 1993; Bézzecourt et 1999), $+$ 02 (Gioia et 1998), $+$ 47 (Seitz et 1996), 1835 (Edge et 1999), 2390 (Kneib et 1999), and $-$ 02 (J.-P. Kneib 1998, private communication)."73 The uncertaintiesiu the maenification of backeround galaxies derived frou these models are and so are comparable with the wncertainty in the absolute calibration of the SCUBA maps., The uncertainties in the magnification of background galaxies derived from these models are and so are comparable with the uncertainty in the absolute calibration of the SCUBA maps.74" The Ίος amplification depends ou both the redshift of the leus (2) aud source (2.). although if +.29+, this effect is minor."," The lens amplification depends on both the redshift of the lens $z_l$ ) and source $z_s$ ), although if $z_s \gg z_l$ this effect is minor."75" The lensing clusters are at redshifts between O.19 and 0.11. aud so the effect is small if 7,21."," The lensing clusters are at redshifts between 0.19 and 0.41, and so the effect is small if $z_s \gs 1$."76 A complete redshift distribution of SCUBA-sclected galaxies is eradually being determined (Barger ct 11998. 1999: IIughes et 11998: Lilly et 11999: Simail et 119985).," A complete redshift distribution of SCUBA-selected galaxies is gradually being determined (Barger et 1998, 1999; Hughes et 1998; Lilly et 1999; Smail et 1998)."77 Dased on identifications made in high-quality optical nuages and extremely deep radio maps. aud follow-up optical spectroscopy (Dargor ct 11999). we are able to distinguish which SCUBA ealaxics are within the clusters. and in the foreground and backerouud.," Based on identifications made in high-quality optical images and extremely deep radio maps, and follow-up optical spectroscopy (Barger et 1999), we are able to distinguish which SCUBA galaxies are within the clusters, and in the foreground and background."78 The nunibers of ealaxies assigned to cach category are listed in 11., The numbers of galaxies assigned to each category are listed in 1.79 Tt is likely that more than of thebackground galaxies are at tyx (Suiall ct 11998). and no backerouud candidateshave been spectroscopically confiiied at <0.9 (Bareer et 11999).," It is likely that more than of the galaxies are at $z_s \ls 5$ (Smail et 1998), and no background candidateshave been spectroscopically confirmed at $z_s < 0.9$ (Barger et 1999)."80 lence. :;21 for the backerouud galaxies. aud so is iudeed expected to be much ercater than +).," Hence, $z_s \gs 1$ for the background galaxies, and so is indeed expected to be much greater than $z_l$."81 Uncertaiuties iu the redshift distribution of the SCUBA ealaxics should not affect the derived counts sienificautlv., Uncertainties in the redshift distribution of the SCUBA galaxies should not affect the derived counts significantly.82 The full sample of spectroscopic identifications of foreground/cluster/backeround. ealaxics span the redshift range 0.2 to 2.8: Ll out of the 17 detected galaxies have plausible identifications (Darger ct 11999)., The full sample of spectroscopic identifications of foreground/cluster/background galaxies span the redshift range 0.2 to 2.8; 14 out of the 17 detected galaxies have plausible identifications (Barger et 1999).83 The potential svstematic errors duc to these uncertainties are discussed iu Section33.1., The potential systematic errors due to these uncertainties are discussed in 3.1.84 At about the level. they comprise a small contribution to the error budget in the count calculations.," At about the level, they comprise a small contribution to the error budget in the count calculations."85 Two approaches were taken to derive counts from our sub-nun naps and catalogs: a direct inversion of the observed source catalog. and a Monte Carlo method to coustrain a parametric model for the background source counts.," Two approaches were taken to derive counts from our sub-mm maps and catalogs: a direct inversion of the observed source catalog, and a Monte Carlo method to constrain a parametric model for the background source counts."86 A comparison of the two techuiques allows us to verify the reliability of the results., A comparison of the two techniques allows us to verify the reliability of the results.87 The detailed mass models available for the seven clusters observed in the Simail et ((1998) snple were used to reconstruct the counts of backeround ealaxics w correcting for the effects of lensing., The detailed mass models available for the seven clusters observed in the Smail et (1998) sample were used to reconstruct the counts of background galaxies by correcting for the effects of lensing.88 Due to the difficulties of incorporating mceonipleteuess corrections mn his method. we conceutrated ou the (greater than Lo) sample discussed by σπα] et ((1997. 1998).," Due to the difficulties of incorporating incompleteness corrections in this method, we concentrated on the (greater than $\sigma$ ) sample discussed by Smail et (1997, 1998)."89 We aραQ conducted the analysis using the more liberal (greater han 30) suuple. in order to both compare the results and assess the systematic error expected from the procedure.," We also conducted the analysis using the more liberal (greater than $\sigma$ ) sample, in order to both compare the results and assess the systematic error expected from the procedure."90 SCUBA detections identified with ealaxics within the eusng clusters were excluded from: our analysis., SCUBA detections identified with galaxies within the lensing clusters were excluded from our analysis.91 The mnber of galaxies identified with cluster sources aud oreeround galaxies are listec in 11., The number of galaxies identified with cluster sources and foreground galaxies are listed in 1.92 At the faintest flix ceusity. the aud samples coutain ciglt aud ll noncluster galaxies respectively.," At the faintest flux density, the and samples contain eight and 14 noncluster galaxies respectively."93 Usiug the appropriate LENSTOOL models. the detected sources identified with background galaxies were mapped frou their observed. positious back onto the source planes at four values of ὃς=1. 2. 3. aud. LI.," Using the appropriate LENSTOOL models, the detected sources identified with background galaxies were mapped from their observed positions back onto the source planes at four values of $z_s = 1$, 2, 3, and 4."94 The flux doeusities of the galaxies were also corrected individually for leus auplification. leading to true flux deusities in the source plane that are less than the observed values.," The flux densities of the galaxies were also corrected individually for lens amplification, leading to true flux densities in the source plane that are less than the observed values."95 Again. we stress that this correction is oulv weakly depeucent ou redshift for our sources at τν21.," Again, we stress that this correction is only weakly dependent on redshift for our sources at $z_s \gs 1$."96 None of the backeround SCUBA sources can be explained casily as iiultiple images of the same galaxy: however. ππο umltiple inages could be detected in deeper integrations.," None of the background SCUBA sources can be explained easily as multiple images of the same galaxy; however, such multiple images could be detected in deeper integrations."97 The umber of galaxies in the catalog that are brighter than a flux density 5. Aga58.1) is calculated at cach redshift ly simply counting the nuniber of sources brighter than S lensing.," The number of galaxies in the catalog that are brighter than a flux density $S$, $N_{\rm raw}(>S, z)$ is calculated at each redshift by simply counting the number of sources brighter than $S$ ."98 A siuple Poisson uncertaintv is attached to this value., A simple Poisson uncertainty is attached to this value.99 ΜΗ, 2mm100Quasars and other tvpes of AGN activity are believed to be powered by supermassive black holes.,Quasars and other types of AGN activity are believed to be powered by supermassive black holes.101 The AGN activity may produce highly collimated outflows from the immediate vicinitv of the black hole (e.g. Rees 1954: Blancllore 1990). which can power a large-scale radio source.," The AGN activity may produce highly collimated outflows from the immediate vicinity of the black hole (e.g. Rees 1984; Blandford 1990), which can power a large-scale radio source."102 Two defining properties of astrophyvsical black holes that can be measured in principle are the black hole mass and spin., Two defining properties of astrophysical black holes that can be measured in principle are the black hole mass and spin.103 A significant amount of progress has been made, A significant amount of progress has been made104we used the masseradius relations of Llamaca(1961).,we used the mass-radius relations of \citet{ham1961}.105. We also confirmed the high mass for the hot white dwarl in the binary LER 8210., We also confirmed the high mass for the hot white dwarf in the binary HR 8210.106 A high mass was initially implied by the binary mass function (Landsmanetal.1993)., A high mass was initially implied by the binary mass function \citep{lan1993}.107. We fitted the Lyman line spectrum of the white dwarf and constrained the mass to be Af=LOS.L24M. in agreement. in the upper mass range. with the binary parameters.," We fitted the Lyman line spectrum of the white dwarf and constrained the mass to be $M= 1.08 - 1.24\ M_\odot$ in agreement, in the upper mass range, with the binary parameters."108 Finally. we show that the EUV-selected. population of white cwarl stars is composed. of &104 objects with masses in excess of LlAZ..," Finally, we show that the EUV-selected population of white dwarf stars is composed of $\approx 10$ objects with masses in excess of $1.1\ M_\odot$."109 A similar vield. was obtained bv Liebertetal.(2005). based. on the PCr survey and. by Ixepleretal.(2007) based on SDSScorrections were applied. due to the magnitude-Iimited. nature of the samples collected., A similar yield was obtained by \citet{lie2005} based on the PG survey and by \citet{kep2007} based on SDSS were applied due to the magnitude-limited nature of the samples collected.110 La particular. it should be noted that only seven objects out. of 347 from the Palomar-Creen sample. or à fraction of24.. met the criterion.," In particular, it should be noted that only seven objects out of 347 from the Palomar-Green sample, or a fraction of, met the criterion."111 By applving V/Vses corrections due to incompleteness at fainter magnitudes. the estimated. fraction. was re-evaluated at in agreement with the vield directly measured in the EUV selection.," By applying $V/V_{\rm max}$ corrections due to incompleteness at fainter magnitudes, the estimated fraction was re-evaluated at in agreement with the yield directly measured in the EUV selection."112 The origin of ultra-massive white cwarls remains unc, The origin of ultra-massive white dwarfs remains uncertain.113 lnitial-mass to final-mass relations (Catalánοἱ indicate that main sequence stars with masses in excess of &6AL. generate white clwarls with masses in excess |Αι. a situation best illustrated by the massive white dwarf (LB 1497) member of the Pleaides.," Initial-mass to final-mass relations \citep{cat2008a} indicate that main sequence stars with masses in excess of $\approx 6\ M_\odot$ generate white dwarfs with masses in excess $1\ M_\odot$, a situation best illustrated by the massive white dwarf (LB 1497) member of the Pleaides."114 Dv. re-evaluating available cluster cata Catalanctal.(2008a) revised the final masses upward. ancl managed to reproduce the high-mass peak in both SDSS and PC empirical mass distributions.," By re-evaluating available cluster data \citet{cat2008a} revised the final masses upward, and managed to reproduce the high-mass peak in both SDSS and PG empirical mass distributions."115 It is therefore possible that white cdwarls with masses in excess of 1.1AZ. are the products of single star evolution and that the binary merger scenario may only apply to à minority of peculiar. objects such as the fast rotating magnetic white dwarl \WD 857 (seeVennesetal.2003.andreferences thercin).., It is therefore possible that white dwarfs with masses in excess of $1.1\ M_\odot$ are the products of single star evolution and that the binary merger scenario may only apply to a minority of peculiar objects such as the fast rotating magnetic white dwarf WD $-$ 857 \citep[see][and references therein]{ven2003}.116 The existence. of a substantial population of ultra-massive white dwarfs supports the concept of a steeper initial-piass to final-mass relations linking 6AL. progenitors with LIA. white dwarfs as proposed by Catalánctal. (2008a)., The existence of a substantial population of ultra-massive white dwarfs supports the concept of a steeper initial-mass to final-mass relations linking $6\ M_\odot$ progenitors with $\ga 1.1 \ M_\odot$ white dwarfs as proposed by \citet{cat2008a}.117. Ultra-massive white dwarfs in close binarics are also likely Tvpe la supernova progenitors (Parthasarathyal.2007), Ultra-massive white dwarfs in close binaries are also likely Type Ia supernova progenitors \citep{par2007}.118 ον V. is grateful for the hospitality and. support of the Astronomical Institute. at Oncdrejov Observatory., S. V. is grateful for the hospitality and support of the Astronomical Institute at Ondrejov Observatory.119 Ads. acknowledges support [rom the Centre for Theoretical Astrophysics (LCOGOI4)., A.K. acknowledges support from the Centre for Theoretical Astrophysics (LC06014).120 We are grateful to Jay Farihi for pointing out recent developments in the study of CD 362., We are grateful to Jay Farihi for pointing out recent developments in the study of GD 362.121stagnation point is for the 0.005yam erains at 61AU. while the furthest is at 227AU for 0.056jou erains.,"stagnation point is for the $0.005 ~\micron$ grains at $64 ~{\rm AU}$, while the furthest is at $227 ~{\rm AU}$ for $0.056 ~\micron$ grains."122 The temperature coded mage in Figure 19 shows the surface brightuess temperature of the bow shock tthe temperature of a black body. that would eive the same surface brightness in the MIPS wavelengths as observed).," The temperature coded image in Figure \ref{fig:color} shows the surface brightness temperature of the bow shock the temperature of a black body, that would give the same surface brightness in the MIPS wavelengths as observed)."123 Table 1) shows good agreement between the mocel aud the measuredvalues., Table \ref{tab:par} shows good agreement between the model and the measuredvalues.124 The original observed images at 21jun aud 70pau were compared to the model., The original observed images at $24 ~\micron$ and $70 ~\micron $ were compared to the model.125 We ecuerated model images with high resolution that included the bow shock aud he central star with its photospherie brightucss value at he central pixel., We generated model images with high resolution that included the bow shock and the central star with its photospheric brightness value at the central pixel.126 We couvolved these mages with a 1.5 iative pixel boxcar siioothed STiTiu PSF2007)., We convolved these images with a $1.8$ native pixel boxcar smoothed STinyTim PSF.127. These mages were subtracted from he observed ones (Figure 113)., These images were subtracted from the observed ones (Figure \ref{fig:24-70model}) ).128 The residuals are sniall and ecnerally consistent with the expected noise., The residuals are small and generally consistent with the expected noise.129 Finally. he best fitting SED of the svstem (rap=1366 AU) is uotted in Figure 12..," Finally, the best fitting SED of the system $r_{\rm ext}=1366 ~{\rm AU}$ ) is plotted in Figure \ref{fig:sed}. ."130 The total mass of the dust inside, The total mass of the dust inside131universe. al least as abundant as stars.,"universe, at least as abundant as stars."132 Even if we accept such speculation. there are still large uncertainties in the value of » because the distribution of wormholes is not specified.," Even if we accept such speculation, there are still large uncertainties in the value of $n$ because the distribution of wormholes is not specified."133 llere. we introduce (wo possibilities.," Here, we introduce two possibilities."134 One is that wormholes are bound (ο the Galaxy. aud (he number density is approximately equal to the local stellar density., One is that wormholes are bound to the Galaxy and the number density is approximately equal to the local stellar density.135 The other possibility is (hat wormholes are not bound to the Galaxy and are approximately uniformly distributed throughout the universe., The other possibility is that wormholes are not bound to the Galaxy and are approximately uniformly distributed throughout the universe.136" For the bound hypothesis. we use n=pr./CMau)0.14Tpe where pp, is the local stellar density in the solar neighborhood. pp.=0.044.pe.oE. and (Mag) is the average mass of stars."," For the bound hypothesis, we use $n = \rho_{Ls}/\langle M_{star} \rangle = 0.147 pc^{-3}$ , where $\rho_{Ls}$ is the local stellar density in the solar neighborhood, $\rho_{Ls} = 0.044 M_\odot pc^{-3}$, and $\langle M_{star} \rangle$ is the average mass of stars."137 We use (Mu)=0.3.M.. a typical mass of an M ναΕςie. (he dominant stellar component in the Galaxy.," We use $\langle M_{star} \rangle = 0.3 M_\odot$, a typical mass of an M dwarf;, the dominant stellar component in the Galaxy."138 For the unbound hypothesis. we asstuned that the munber density of the wormholes is the same as the average stellar density ol the universe.," For the unbound hypothesis, we assumed that the number density of the wormholes is the same as the average stellar density of the universe."139 The stellar densitv of the universe is estimated assuming that the fraction of barvonic matter accounted [or by star is the same as that of the solar neighborhood., The stellar density of the universe is estimated assuming that the fraction of baryonic matter accounted for by star is the same as that of the solar neighborhood.140" Then we obtain n=p.Qypi/(ppsMaa))4.97x10?pe7. where p.=LASxLOTAL.pe? is the critical density. Ον=0.042 is the barvon densitv of the universe divided by the critical density. and pp.=0.04.M..pe and pj;=OLRM.pe*oE ave the local star and local barvon densities. respectively,"," Then we obtain $n = \rho_c \Omega_b \rho_{Ls} / (\rho_{Lb} \langle M_{star} \rangle) = 4.97 \times 10^{-9} pc^{-3}$, where $\rho_c = 1.48 \times 10^{-7} M_\odot pc^{-3}$ is the critical density, $\Omega_b = 0.042$ is the baryon density of the universe divided by the critical density, and $\rho_{Ls} = 0.044 M_\odot pc^{-3}$ and $\rho_{Lb} = 0.18 M_\odot pc^{-3}$ are the local star and local baryon densities, respectively."141 Using these values. we caleulated (he optical depths and event rates for bulge and LMC lensings.," Using these values, we calculated the optical depths and event rates for bulge and LMC lensings."142 Table 3. presents the results for the bulge lensines., Table \ref{tbl-3} presents the results for the bulge lensings.143 In an ordinary Schwarzschild microlensing survey. observations are made of more than 10 million stus.," In an ordinary Schwarzschild microlensing survey, observations are made of more than 10 million stars."144 Thus. we can expect approximately LO'T events in a vear.," Thus, we can expect approximately $10^7 \Gamma$ events in a year."145 However. the situation is different in a wormbhole search.," However, the situation is different in a wormhole search."146 As mentioned previously. the magnification of wormhole lensing is less than that of Schwarzschild lensing. and a remarkable feature of wormhole lensing is the decreasing brightness around the Einstein radius crossing times.," As mentioned previously, the magnification of wormhole lensing is less than that of Schwarzschild lensing, and a remarkable feature of wormhole lensing is the decreasing brightness around the Einstein radius crossing times."147 Past microlensing surveys have mainly searched [or stars that increase in brightness., Past microlensing surveys have mainly searched for stars that increase in brightness.148 The stars monitored are those with magnitudes down to the limiting magnitude or less., The stars monitored are those with magnitudes down to the limiting magnitude or less.149 However. we need to find stars that decrease in brightness in thewormhole search.," However, we need to find stars that decrease in brightness in thewormhole search."150 To do so. we need to watel brighter," To do so, we need to watch brighter"151in a solar abundance plasma (1.2). the solar abundance of oxvgen (8.5x10.+. (1989))). and an estimate of the fraction of oxvgen atoms in the ilonizatlion level.,"in a solar abundance plasma (1.2), the solar abundance of oxygen $8.5 \times 10^{-4}$ , \citet{grevesse_anders}) ), and an estimate of the fraction of oxygen atoms in the ionization level."152 During the plasmas extraordinary cooling history. it has experienced some degree of recombination. as evidenced by the strong lines ofIIT. IIL. and iin Breitschwerdt (2001)s predictions.," During the plasma's extraordinary cooling history, it has experienced some degree of recombination, as evidenced by the strong lines of, , and in \citet{breitschwerdt}' 's predictions."153 Here. we take the lower limit on recombinations of oxvgen to the sstate as that of a 5xI0? IN collisional ionizational equilibrium plasma.," Here, we take the lower limit on recombinations of oxygen to the state as that of a $5 \times 10^5$ K collisional ionizational equilibrium plasma."154 In this case. the fraction of oxveen in the O VI state is 0.037 (Mazzottaetal.1998:Schimutzler& 1993)..," In this case, the fraction of oxygen in the O VI state is 0.037 \citep{mazzotta_etal,schmutzler_tscharnuter}."155 As the plasma recombines further. the VI//oxvgen fraction increases. and. due to recombinations ofVIL. does not decrease until the most of the oxvgen has recombined to andIL.," As the plasma recombines further, the /oxygen fraction increases, and, due to recombinations of, does not decrease until the most of the oxygen has recombined to and."156 An upper limit on the extent of the recombinations is taken from the isochoric cooling predictions of Schmutzler&Tscharuuter(1993) at the model temperature. again vielding an (lo oxvgen fraction of 0.037.," An upper limit on the extent of the recombinations is taken from the isochoric cooling predictions of \citet{schmutzler_tscharnuter} at the model temperature, again yielding an to oxygen fraction of 0.037."157 We take 0.037 as our estimate. noting that the VI//oxvgen ratio and (he intensity prediction may be higher.," We take 0.037 as our estimate, noting that the /oxygen ratio and the intensity prediction may be higher."158 The resulting intensity estimate. ~1900 > ↕↽≻∐∪∩≻∐⋟∖⊽≺∢∐↓−⋟∖⊽↓⋟∖⊽↕⋅↓⋅↕⋟∖⊽∐↓∪↕⋅≼↲⊔⋯↴∐↥∖∖⇁↕≺∢≼↲⊔∐↲∃⋟∖⊽↕≸↽↔↴∐↓≀↕↴∏↕↽≻↕↽≻≼↲↕⋅∐∐∐↥≼↲⋟∖⊽↥≀↧↴∣↽≻∐⋟∖⊽∐≼↲≼⊔∐⊔∐⋟∖⊽↕↽≻≀↧↴↕↽≻≼↲↕⋅⋅ ∙ ∙ ∙ ∙ ∙ ∙ ," The resulting intensity estimate, $\sim1900$ photons $^{-2}$ $^{-1}$ $^{-1}$, is more than twice the 2 sigma upper limit established in this paper."159∙ For this caleulation. the lions are assumed to be mixed throughout the remnant.," For this calculation, the ions are assumed to be mixed throughout the remnant."160 If the assumed abundance of metal atoms were reduced from (he solar value. then (he soft X-ray scaling relationships used in fincling the model parameters would need to be adjusted to a similar degree.," If the assumed abundance of metal atoms were reduced from the solar value, then the soft X-ray scaling relationships used in finding the model parameters would need to be adjusted to a similar degree."161 As a result. the model parameters would need to be adjusted. probably leading to a bigger. denser. aud/or hotter bubble.," As a result, the model parameters would need to be adjusted, probably leading to a bigger, denser, and/or hotter bubble."162 As far as the predicted intensity is concerned. these adjustments would probably offset the hypothetical abundance adjustment.," As far as the predicted intensity is concerned, these adjustments would probably offset the hypothetical abundance adjustment."163 Not only is there a significant discrepancy between our null result and the intensity calculated from the model. there is a factor of 20 difference between the observed column density. (~1.6xLOM em7. Shelton&Cox(1994).. Oegerleetal. (2002).. see below) and the column density calculated from the model (2.7x1011 7). and a significant discrepancy between (he 2 sigma upper limit on the observed 90TT intensity (7300 photons 7s J| sr !) and the published pprediction (>8500 photons 7? ! !. Figure 2 Dreitschwerdt. (2001))).," Not only is there a significant discrepancy between our null result and the intensity calculated from the model, there is a factor of 20 difference between the observed column density $\sim1.6 \times 10^{13}$ $^{-2}$, \citet{shelton_cox}, \citet{oegerle_etal}, see below) and the column density calculated from the model $2.7 \times 10^{14}$ $^{-2}$ ), and a significant discrepancy between the 2 sigma upper limit on the observed 977 intensity $7300$ photons $^{-2}$ $^{-1}$ $^{-1}$ ) and the published prediction $\geq 8500$ photons $^{-2}$ $^{-1}$ $^{-1}$, Figure 2 \citet{breitschwerdt}) )."164 The number and magnitude ofthese discrepancies practically eliminate thisclass of models., The number and magnitude ofthese discrepancies practically eliminate thisclass of models.165age-metallicity relation in the solar vicinity are determined.,age-metallicity relation in the solar vicinity are determined.166 Auch elfort has been devoted. to the determination. of reliable individual isochrone ages for our sample of stars., Much effort has been devoted to the determination of reliable individual isochrone ages for our sample of stars.167 For this task. the isochrones of Berebusch VandenBere (2001) and. VandenDerg (2003) as transformed to the weby system by the empirically constrained: colour-temperature relations of Clem et al. (," For this task, the isochrones of Bergbusch VandenBerg (2001) and VandenBerg (2003) as transformed to the $uvby$ system by the empirically constrained colour-temperature relations of Clem et al. ("1682004). are adopted.,"2004), are adopted."169 The sample of Nidever et ((2002) contains SSO stars from the Doppler planet search project at. the Weck and Lick Observatories (Butler ct al., The sample of Nidever et (2002) contains 889 stars from the Doppler planet search project at the Keck and Lick Observatories (Butler et al.170 2000. Vogt et al.," 2000, Vogt et al."171 2000)., 2000).172 This sample contains mostly. main-sequence and. sub-giant stars from F7 to M5. within 50 pe.," This sample contains mostly main-sequence and sub-giant stars from F7 to M5, within 50 pc."173" Chromospherically active stars have been removed. from. this list as well as stars with known stellar companions within 2 "" (including known spectroscopic binaries).", Chromospherically active stars have been removed from this list as well as stars with known stellar companions within 2 $\arcsec$ (including known spectroscopic binaries).174 This list provides several advantages for our study of the age-metallicity relation: (a) it includes many Cooler chwarl stars and so avolds one of the selection alleets discussed by Feltzing et al. (, This list provides several advantages for our study of the age-metallicity relation: (a) it includes many cooler dwarf stars and so avoids one of the selection affects discussed by Feltzing et al. (175(2001): when these cool dsvarfs are left out of a sample. the old. metal-rich stars are preferentiallv exeluded. depleting the upper right-hand corner of an age-metallicity diagram. (,"2001); when these cool dwarfs are left out of a sample, the old, metal-rich stars are preferentially excluded depleting the upper right-hand corner of an age-metallicity diagram. ("176"b) the racial velocities are especially. accurate (systematic. errors .£0.53 aq.kms 1 7) and precise. (random errors 0.1 km "" 7) (Nidever"" et 22002) providing us with especially good kinematic data or these stars allowing us to separate more cleanly the stellar populations within the sample. such as the separation »etween the thin- and thick-disk stars. (","b) the radial velocities are especially accurate (systematic errors $\la \pm 0.3$ km $^{-1}$ ) and precise (random errors $\la \pm0.1$ km $^{-1}$ ) (Nidever et 2002) providing us with especially good kinematic data for these stars allowing us to separate more cleanly the stellar populations within the sample, such as the separation between the thin- and thick-disk stars. ("177ο) As suggested by CGiménnez (2000) the iebg photometry can be used to refine re survey Lists of planetary-search. projects and. to study 1e metallicities and ages of stars found to have planets.,c) As suggested by Giménnez (2000) the $uvby$ photometry can be used to refine the survey lists of planetary-search projects and to study the metallicities and ages of stars found to have planets.178 This paper is organized as follows: Section 2 describes je data: in Section 3 the calculation. procedure for. the space velocities is given: the derivation of astrophysical »wameters:: reddening. metallicity. absolute magnitucles. and ages is given in Section 4. plus age-metallicity ciagranis and a discussion. of these: ancl finally the conelusions in Section 5.," This paper is organized as follows: Section 2 describes the data; in Section 3 the calculation procedure for the space velocities is given; the derivation of astrophysical parameters: reddening, metallicity, absolute magnitudes, and ages is given in Section 4, plus age-metallicity diagrams and a discussion of these; and finally the conclusions in Section 5."179 For the SSO stars in the catalogue of Nidever et ((2002). ucby photometry of 437 stars and {41 photometry for 214 have been collected. from the web site of the General Catalogue of Photometric of Hauck. Alermilliod (1998).," For the 889 stars in the catalogue of Nidever et (2002), $uvby$ photometry of 437 stars and $H\beta$ photometry for 214 have been collected from the web site of the General Catalogue of Photometric of Hauck Mermilliod (1998)."180 This veby1:3 photometry (5y. ana. ey. 44 :1) of these 437/214 stars are given in Cols.," This $uvby-H\beta$ photometry $b-y$, $m_{\rm 1}$, $c_{\rm 1}$, $H\beta$ ) of these 437/214 stars are given in Cols."181 25 of Table 1. respectively.," 2–5 of Table 1, respectively."182" In Table 1 this veby photometry of the 437 E and Cr main-sequence. turn-oll and. sub-giant stars covers the ranges O.0NT<=m,<O4907. O.116xο0.492. 0.313x(b.gy)<0.580."," In Table 1 this $uvby$ photometry of the 437 F and G main-sequence, turn-off and sub-giant stars covers the ranges $0.087 \leq m_{\rm 1} \leq 0.497$, $0.116 \leq c_{\rm 1} \leq 0.492$, $0.313 \leq (b-y) \leq 0.580$."183 The 214 stars with {1.1 values fall in the range. 2.53044:3<2.680.," The 214 stars with $H\beta$ values fall in the range, $2.530 \leq H\beta \leq 2.680$."184 Visual magnitudes for our sample stars. all brighter than VY=10 mag. and (2Y) colours. given in Cols.," Visual magnitudes for our sample stars, all brighter than $V=10$ mag, and $(B-V)$ colours, given in Cols."185 78 of Table 1. respectively. were taken from the Appearcos and Tycho Reference Catalogues 11991).," 7–8 of Table 1, respectively, were taken from the $Hipparcos$ and $Tycho$ Reference Catalogues 1997)."186 The barycentrie racial velocities. (V4). with a typical accuracy of 20.3 km s+ were taken from Nidever οἱ ((2002).," The barycentric radial velocities $(V_{\rm rad})$, with a typical accuracy of $\la 0.3$ km $^{-1}$, were taken from Nidever et (2002)."187" Parallaxes (x). the proper motion coniponents (pi,0089. ges) and their associated uncertainties were taken mainty from the Lffpparcos and Tycho catalogues 1997) and Z'ycho 2 catalogue (Hog et al."," Parallaxes $\pi$ ), the proper motion components $\mu_{\alpha}cos\delta$, $\mu_{\delta}$ ) and their associated uncertainties were taken mainly from the $Hipparcos$ and $Tycho$ catalogues 1997) and $Tycho$ –2 catalogue (Hog et al."188 2000)., 2000).189 Parallaxes and their uncertainties for three stars (ID 185295. LID 92020 and LED 1854) not found in the Zpparcos and Lycho catalogues were taken from the catalogue of IXharchenko (2001).," Parallaxes and their uncertainties for three stars (HD 185295, HD 192020 and HD 1854) not found in the $Hipparcos$ and $Tycho$ catalogues were taken from the catalogue of Kharchenko (2001)."190 The radial velocities (Via). parallaxes (3). proper motion components (fncosó. pn) ancl their associated uncertainties are listed in Table 2.," The radial velocities $(V_{\rm rad})$, parallaxes $\pi$ ), proper motion components $\mu_{\alpha}cos\delta$, $\mu_{\delta}$ ) and their associated uncertainties are listed in Table 2."191 Most of the ZLippearcos xwallaxes in our sample have relative errors that are much ess than 20 per cent., Most of the $Hipparcos$ parallaxes in our sample have relative errors that are much less than 20 per cent.192 In Tables 1.3 those stars with extra-solar planetary systems have a ΠΟ after their HD. number in Col., In Tables 1–3 those stars with extra-solar planetary systems have a “P” after their HD number in Col.193 1: these stars are discussed further in Section 4.4 and especially in Fig., 1; these stars are discussed further in Section 4.4 and especially in Fig.194 9c., 9c.195 Our sample was selected based on the existence of accurate radial velocities for SSO FCINAL main-sequence. ancl sub-giant stars of Nidever et ((2002)., Our sample was selected based on the existence of accurate radial velocities for 889 FGKM main-sequence and sub-giant stars of Nidever et (2002).196 Phe sample. used in this work is reduced to only 437 stars. lor which nebygI3 photometry is available. to be able to determine a photometric metallicity for cach star.," The sample used in this work is reduced to only 437 stars, for which $uvby-H\beta$ photometry is available, to be able to determine a photometric metallicity for each star."197 Our sample is [ree from binary contamination according to Nidever ct (000:), Our sample is free from binary contamination according to Nidever et (2002).198 The accurate radial velocities of Nidever οἱ (000:) are suitable for the detection of extra-solar. planets. and so the high accuracy of their racial velocities ollers us an opportunity to derive. very. goocl kinematic parameters and thus the age-metallicity relations for sub-classes of E and Ci stars.," The accurate radial velocities of Nidever et (2002) are suitable for the detection of extra-solar planets, and so the high accuracy of their radial velocities offers us an opportunity to derive very good kinematic parameters and thus the age-metallicity relations for sub-classes of F and G stars."199 Compared with the samples of Edvardsson et al. (, Compared with the samples of Edvardsson et al. (2001993) and. Chen et al... (,1993) and Chen et al. (2012003). the size of our sample is large. whereas compared to the samples anc metallicity distributions of Feltzing οἱ al. (,"2003), the size of our sample is large, whereas compared to the samples and metallicity distributions of Feltzing et al. ("202(2001) and Nordstrómum et al. (,2001) and Nordströmm et al. (2032004). the size of our sample is small but covering a wide metallicity range. 2.0< ΜΗ]|0.5 dex. with a very high percentage of the stars having L.0« κ|0.5 dex.,"2004), the size of our sample is small but covering a wide metallicity range, $-2.0 <$ $ <+0.5$ dex, with a very high percentage of the stars having $-1.0 <$ $ < +0.5$ dex."204 Our sample is not alfected by one of the main biases discussed by Feltzing et al. (, Our sample is not affected by one of the main biases discussed by Feltzing et al. (205(2001) in the sense that old. metal-rich stars are amply represented (see Fig.,"2001) in the sense that old, metal-rich stars are amply represented (see Fig."206 9 below): the presence of these old. metal-rich stars produces. an age-metallicity. relation very similar to other modern metallicity relations. such as those mentioned above.," 9 below); the presence of these old, metal-rich stars produces an age-metallicity relation very similar to other modern age-metallicity relations, such as those mentioned above."207" One advantage of having accurate radial velocities [ου our sample together with the £0.120.14 dex uncertainty in MM] (Schuster Nissen 1989. and see Section 4.3 below) is that it allows us to determine well the NX cut-olls of the stellar populations from Vy. and M/L]. where Y is the stellar-population criterion. defined in Schuster. Parrao Contreras-\lartinnez (1993). a linear combination of V, and MILI]."," One advantage of having accurate radial velocities for our sample together with the $\pm0.12-0.14$ dex uncertainty in [M/H] (Schuster Nissen 1989, and see Section 4.3 below) is that it allows us to determine well the $X$ cut-offs of the stellar populations from $V_{\rm rot}$ and [M/H], where $X$ is the stellar-population criterion defined in Schuster, Parrao nez (1993), a linear combination of $V_{rot}$ and [M/H]."208 Galactic space. velocity components (0. V. MW)," Galactic space velocity components $U$ , $V$ , $W$ )"209in pre-main sequence stars with periods as short as 18 min (Amado et al 2004). approximately 90 percent of 6 Set stars have a pulsation period in the range 40 min to 5.3 hrs (cf Table 1. Rodrfeeucz ct al 2000).,"in pre-main sequence stars with periods as short as 18 min (Amado et al 2004), approximately 90 percent of $\delta$ Sct stars have a pulsation period in the range 40 min to 5.3 hrs (cf Table 1, Rodrígguez et al 2000)."210 SX Phe and 9 Scu stars have a well defined period-Iuminosity relationship and can therefore be used as distance indicators and hence map Galactic structure (eg Nemec. Linnell Nemec Lutz 1993).," SX Phe and $\delta$ Scu stars have a well defined period-luminosity relationship and can therefore be used as distance indicators and hence map Galactic structure (eg Nemec, Linnell Nemec Lutz 1993)."211 Since we took our first set. of data in 2003. we have obtained a significant amount of further data (Barclay et al VI).," Since we took our first set of data in 2003, we have obtained a significant amount of further data (Barclay et al 2011)."212 We therefore have made a svstematic search for blue. gadort period. pulsating stars.," We therefore have made a systematic search for blue, short period, pulsating stars."213 Our light curves are typically 2.5 hrs in duration. so the longest period we can determine =vith confidence is less than 2 hrs.," Our light curves are typically 2–2.5 hrs in duration, so the longest period we can determine with confidence is less than 2 hrs."214 For stars with periods garorter than 40 mins. it becomes increasingly difficult to letermine the nature of the source based. on colour and period information (cf Table 1 and 2. Barclay οἱ al 2011).," For stars with periods shorter than 40 mins, it becomes increasingly difficult to determine the nature of the source based on colour and period information (cf Table 1 and 2, Barclay et al 2011)."215 In js paper. we therefore have decided to restrict our search for blue pulsating variables in the range of 40 min to 2 hrs.," In this paper, we therefore have decided to restrict our search for blue pulsating variables in the range of 40 min to 2 hrs."216 The RATS observing strategy is to take à series of 30 sec exposures of a given field using the wicle-fielcl cameras on the Isaac Newton Telescope in La Palma and the MPG/IESO 2.2m on La Silla for a curation of ~2 hrs., The RATS observing strategy is to take a series of 30 sec exposures of a given field using the wide-field cameras on the Isaac Newton Telescope in La Palma and the MPG/ESO 2.2m on La Silla for a duration of $\sim$ 2 hrs.217 To date our survey has discovered around 107. variable stars (see Barclay et al 2011 for a full description of our reduction. process)., To date our survey has discovered around $\times10^{5}$ variable stars (see Barclay et al 2011 for a full description of our reduction process).218 Based on their photometric properties. a small sub-sample has been selected for followup spectroscopic observations το determine their nature.," Based on their photometric properties, a small sub-sample has been selected for followup spectroscopic observations to determine their nature."219 To narrow our search for blue pulsators in our RATS data. we restricted our search to a range in both magnitude and colour.," To narrow our search for blue pulsators in our RATS data, we restricted our search to a range in both magnitude and colour."220" “Phe intrinsic colour of SX Phe stars is typically (BVo ~0.1 to 0.35 (eg Poretti ct al 2008). which corresponds to (9gr),= 0.12 to 0.14."," The intrinsic colour of SX Phe stars is typically $(B-V)_{o}\sim$ 0.1 to 0.35 (eg Poretti et al 2008), which corresponds to $(g-r)_{o}=$ –0.12 to 0.14."221 This is virtually identical to the colours for 6 Set stars (eg Rodrigguez οἱ al 2000)., This is virtually identical to the colours for $\delta$ Sct stars (eg Rodrígguez et al 2000).222 Alany of our fields lie at low Galactic latitude and hence have high extinction., Many of our fields lie at low Galactic latitude and hence have high extinction.223 To reduce the contamination with other types of sources at low Galactic. latitudes we allow a maximum extinction. corresponding to. 21 —0.40 (Ig.v 0.13 for R=3.1)., To reduce the contamination with other types of sources at low Galactic latitudes we allow a maximum extinction corresponding to $A_{V}$ =0.40 $E_{B-V}$ =0.13 for R=3.1).224" For blue stars Jg. =0.13 equates to fe,0.13.", For blue stars $E_{B-V}$ =0.13 equates to $E_{g-r}$ =0.13.225 HE we add à conservative uncertainty of 0.13 mag in our observed colours (Barclay et al 2011) our search region therefore covers 0.12<(gor)O40. while the brightness of stars were in the range 15<q23.," If we add a conservative uncertainty of 0.13 mag in our observed colours (Barclay et al 2011) our search region therefore covers $-0.12<(g-r)<0.40$, while the brightness of stars were in the range $15<g<23$."226 llere we note that the shape of the light curves and the colour of our sources is similar to some cataclvsmic variables (CVs: eg Szkody et al 2002 who present the first sample of CVs discovered. using SDSS data)., Here we note that the shape of the light curves and the colour of our sources is similar to some cataclysmic variables (CVs; eg Szkody et al 2002 who present the first sample of CVs discovered using SDSS data).227 “Phe hydrogen accreting CVs have a minimum orbital period of 80 mins (Günnsicke et al 20090). implving a possible over-lap with our arget selection. C," The hydrogen accreting CVs have a minimum orbital period of $\sim$ 80 mins (Gännsicke et al 2009), implying a possible over-lap with our target selection. ("228Ehe helium dominated. CVs with orbital »riods in the range 4070 mins do not show a photometric modulation. implving we are not overlapping with these objects).,The helium dominated CVs with orbital periods in the range 40–70 mins do not show a photometric modulation implying we are not overlapping with these objects).229 Llowever. CVs show optical spectra dominated by ine emission although we obtained a spectrum for only 7 out of the 31 new sources none show evidence for emission ines (ef 85)).," However, CVs show optical spectra dominated by line emission – although we obtained a spectrum for only 7 out of the 31 new sources – none show evidence for emission lines (cf \ref{spectra}) )."230 Moreover. although pulsating blue stars are not X-ray sources. CVs are weak to moderately strong X-ray sources (Verbunt et al 1997).," Moreover, although pulsating blue stars are not X-ray sources, CVs are weak to moderately strong X-ray sources (Verbunt et al 1997)."231 We therefore eross-correlated he position of our sources with that of catalogues derived rom the All-Sky. X-ray Survey (RASS)., We therefore cross-correlated the position of our sources with that of catalogues derived from the All-Sky X-ray Survey (RASS).232 None of our sources has an X-ray counterpart within 20 aresec of the optical position., None of our sources has an X-ray counterpart within 20 arcsec of the optical position.233 In contrast 30 of the 48 CVs with orbital »eriods in the range 70.120 mins in the 2009 catalogue of utter Ixolb (2003) were detected in the RASS., In contrast 30 of the 48 CVs with orbital periods in the range 70–120 mins in the 2009 catalogue of Ritter Kolb (2003) were detected in the RASS.234 Although at this stage we cannot preclude that none of our 31 sources are CVs we consider this rather unlikely., Although at this stage we cannot preclude that none of our 31 sources are CVs we consider this rather unlikely.235 For sources which fell within our search range we manually inspected each light curve to verily that the light curve was consistent with that of a stellar. pulsator (cf Rodrigguez et al 2007 for a recent example of a light curve for an SX Phe star) and to exclude light. curves of low quality., For sources which fell within our search range we manually inspected each light curve to verify that the light curve was consistent with that of a stellar pulsator (cf Rodrígguez et al 2007 for a recent example of a light curve for an SX Phe star) and to exclude light curves of low quality.236 We found. 31 sources which showed a modulation in their lighteurve on à period between 51.83 mins and had a mean brightness of g=15.920.8., We found 31 sources which showed a modulation in their lightcurve on a period between 51–83 mins and had a mean brightness of $g=15.9-20.8$.237 The full-amplit of the modulation is in the range 0.050.65., The full-amplitude of the modulation is in the range 0.05–0.65.238 We show theiruce photometrie properties in Table 1. and the light curves in Figure 3. and 4.., We show their photometric properties in Table \ref{candidates} and the light curves in Figure \ref{curves1} and \ref{curves2}.239 There is à clear relationship between Ady ancl pulsation period which is applicable to SX Phe. 6 Set and RR Lye stars (eg AleNamara L997).," There is a clear relationship between $M_{V}$ and pulsation period which is applicable to SX Phe, $\delta$ Sct and RR Lyr stars (eg McNamara 1997)."240 Vhis Periocd-Luminosity (PL) relationship is consistent with a study of dillerent. types of short period pulsators in the Fornax cdwarl spheroidal galaxy (Poretti et al 2008)., This Period-Luminosity (PL) relationship is consistent with a study of different types of short period pulsators in the Fornax dwarf spheroidal galaxy (Poretti et al 2008).241 Since the PL relationship of MeNamara (1997) is calibrated with respect to Ady. we applied a small correction to transform our g band magnitudes to that of the Y band (Jester et al 2005).," Since the PL relationship of McNamara (1997) is calibrated with respect to $M_{V}$, we applied a small correction to transform our $g$ band magnitudes to that of the $V$ band (Jester et al 2005)."242 Further. we used a NASA/LPAC which uses the maps of Schlegel. Finkbeiner Davis (1998) to determine extinction to the edge of the Galaxy.," Further, we used a NASA/IPAC which uses the maps of Schlegel, Finkbeiner Davis (1998) to determine extinction to the edge of the Galaxy."243 The PL relationship assumes that the period. is the undamental radial pulsation mode rather than the first over-tone which can also be observed in these stars., The PL relationship assumes that the period is the fundamental radial pulsation mode rather than the first over-tone which can also be observed in these stars.244 Given, Given245coolest stars giving the lowest (Cu/Fe].,coolest stars giving the lowest [Cu/Fe].246 This is likely due to errors in Che continuum fitting., This is likely due to errors in the continuum fitting.247 In the coolest eiants. molecular band heads of TiO are particularly strong; (see 1994.. their Fig.," In the coolest giants, molecular band heads of TiO are particularly strong (see \citealt{Sneden1994}, their Fig."248 D). and a TiO band head does exist near the 5782 feature (at 5758Α.. degraded to the red: Pearse&Gaydon 1963)).," 1), and a TiO band head does exist near the 5782 feature (at 5758, degraded to the red; \citealt{Pearse1963}) )."249 HF these band heads are not properly taken into account. the tendency is to incorrectly identify the continuum so that the Cu line becomes shallower than it ought to be and gives a falselv low Cu abundance.," If these band heads are not properly taken into account, the tendency is to incorrectly identify the continuum so that the Cu line becomes shallower than it ought to be and gives a falsely low Cu abundance."250 Copper lies at the end of the Fe-peak and may or may not actually be formed in the same processes (hat create the lighter elements of the Fe-peak., 	Copper lies at the end of the Fe-peak and may or may not actually be formed in the same processes that create the lighter elements of the Fe-peak.251 However. like the Fe-peak elements. it shows no real star to star variation in these globular clusters.," However, like the Fe-peak elements, it shows no real star to star variation in these globular clusters."252 The standard deviation of the mean [Cu/Fe] value for every cluster in the sample is quite small and well within the observational errors., The standard deviation of the mean [Cu/Fe] value for every cluster in the sample is quite small and well within the observational errors.253" In Table 2 we list for each cluster the values ο and the standard deviation (0). and in Figure 4.. we plot [(Cu/Fel,,, vs. [Fe/I],,4 for the clusters in our sample."," In Table \ref{tbl-2} we list for each cluster the values $_{avg}$, $_{avg}$, and the standard deviation $\sigma$ ), and in Figure \ref{trend}, we plot $_{avg}$ vs. $_{avg}$ for the clusters in our sample."254 As a supplement (to Figure 4.. we offer in Figure 5. box plots in whieh the clusters are arranged in order of increasing metallicity. and. which illustrate both the median [Cu/Fe| and the range in [Cu/Fe| for each cluster.," 	As a supplement to Figure \ref{trend}, we offer in Figure \ref{box} box plots in which the clusters are arranged in order of increasing metallicity, and which illustrate both the median [Cu/Fe] and the range in [Cu/Fe] for each cluster."255 Figure 5 appears to show a rather large range in (Cu/Fe] in MIO. when the outliers are considered.," Figure \ref{box} appears to show a rather large range in [Cu/Fe] in M10, when the outliers are considered."256 However. an examination of the spread in [Ni/Fe]| (which ought not to vary. from star to star) in MIO as measured by indicates that the spread in [Cu/Fe| is not real: an investigation of each of the five Ni lines analyzed by INraftetal.(1995). reveals that &([Ni/Fe]) ranges from 0.10 dex to 0.15 dex.," However, an examination of the spread in [Ni/Fe] (which ought not to vary from star to star) in M10 as measured by \citet{Kraft1995} indicates that the spread in [Cu/Fe] is not real: an investigation of each of the five Ni lines analyzed by \citet{Kraft1995} reveals that $\sigma$ ([Ni/Fe]) ranges from 0.10 dex to 0.15 dex."257 o([Cu/Fe|)—0.10 is therefore not significant., $\sigma$ ([Cu/Fe])=0.10 is therefore not significant.258 The total range of [Ni/Fe| in each individual Ni line is as much as 0.3J) dex in this cluster., The total range of [Ni/Fe] in each individual Ni line is as much as 0.3 dex in this cluster.259 It should also be noted that the single low outlying [Cu/Fe| value. M10:G. is the warmest star in our MIO sample. and has," It should also be noted that the single low outlying [Cu/Fe] value, M10:G, is the warmest star in our M10 sample, and has"260drift-Alfvénn case has been discussed in detail in VranjesDPoedts (2010).,drift-Alfvénn case has been discussed in detail in \citet{v6}.261. For the purpose of the present study. in order to quantitatively cleseribe some details of the heating and acceleration of plasma particles. any of the mentioned instabilities of the drift wave will do.," For the purpose of the present study, in order to quantitatively describe some details of the heating and acceleration of plasma particles, any of the mentioned instabilities of the drift wave will do."262 Hence. we shall use the kinetic. density eradient driven instability from. Vranjes&Poects (2009a.b).," Hence, we shall use the kinetic, density gradient driven instability from \citet{v1,v2}. ."263. From standard textbooks (Ichinaru198)Weilanel2000).. the frequeney and the growth rate of the kinetic drift wave are given by and," From standard textbooks \citep{ichi,weil}, the frequency and the growth rate of the kinetic drift wave are given by and."264 llere. the geometry is chosen as follows.," + Here, the geometry is chosen as follows."265" The constant magnetic field is given by By=Duc. the equilibrium density. gradient in the electrically neutral plasma with No=eoΠο is given bv Ving=CanQmoGedfee. and theperturbations are assumed. to be of the form fOryexp(Aetη|des). with [Asκlh, and an uredependent wave amplitude f."," The constant magnetic field is given by $\vec B_0=B_0 \vec e_z$, the equilibrium density gradient in the electrically neutral plasma with $n_{i0}=n_{e0}=n_0$ is given by $\nabp n_0= - \vec266e_x n_0'\equiv -\vec e_x d n_0/dx$, and theperturbations are assumed to be of the form $f(x) \exp(- i \omega t + i k_y267y + i k_z z)$, with $|k_z|\ll |k_y|$ and an $x$ -dependent wave amplitude $f$."268 The local approximation will be used., The local approximation will be used.269 Llence. it will be assumed that ‘ας|hy," Hence, it will be assumed that $|d/dx|\ll |k_y|$."270 In the cerivations of Eqs. (1)), In the derivations of Eqs. \ref{k1}) )271 and2 . the condition ol a strongly magnetized plasma is used n&O;. together with the smallness of the acoustic response of ions in the direction parallel to the magnetic field vector. implving that wel cs where ὃς is the ion sound speed.," and \ref{k2}) ), the condition of a strongly magnetized plasma is used, $|\omega|\ll \Omega_i$, together with the smallness of the acoustic response of ions in the direction parallel to the magnetic field vector, implying that $\omega_r/k_z\gg c_s$ , where $c_s$ is the ion sound speed."272 The notation in Eqs. (12) , The notation in Eqs. \ref{k1}) )273and (2)) is standard. (Weiland2000).. ic. where {ο is the modified Bessel function of the first kind and of the order 0.," and \ref{k2}) ) is standard \citep{weil}, i.e., where $I_0$ is the modified Bessel function of the first kind and of the order 0."274 From Eq. (2)), From Eq. \ref{k2}) )275 it ds seen that within the present instability. the sien of 5 can be changed only by the electron. term if [|[ws].," it is seen that within the present instability, the sign of $\gamma$ can be changed only by the electron term if $|\omega_r|<|\omega_{*e}|$."276 This is. however. only a necessary condition.," This is, however, only a necessary condition."277 A sullicient condition is obtained when the contribution of the rest. of the expression. (the ion damping terms) is taken into account., A sufficient condition is obtained when the contribution of the rest of the expression (the ion damping terms) is taken into account.278 As demonstrated by various graphs in Vranjes&Poedts(2009a.b).. in the case of the solar corona it is more clillicult to find a regime where the mode is camped than a regime where it is growing.," As demonstrated by various graphs in \citet{v1,v2}, in the case of the solar corona it is more difficult to find a regime where the mode is damped than a regime where it is growing."279" In these studies itis shown that the mode grows more strongly for shorter perpendicular wave lengths A, (meter or sub-meter size) and for longer parallel wavelengths that may exceed the perpendicular ones for 4 to orders of magnitude.", In these studies it is shown that the mode grows more strongly for shorter perpendicular wave lengths $\lambda_y$ (meter or sub-meter size) and for longer parallel wavelengths $\lambda_z$ that may exceed the perpendicular ones for 4 to 5 orders of magnitude.280 Notethat a similar ratio of the two components5 A; holds for tokamak plasmas as well., Notethat a similar ratio of the two components $\lambda_j$ holds for tokamak plasmas as well.281 The same conclusions can also be drawn in the case of the reactive (the so-called 5;-instability) discussed in. Vranjes&Pocdts(2009d)., The same conclusions can also be drawn in the case of the reactive (the so-called $\eta_i$ -instability) discussed in \citet{v4}.282. ‘To provide some details of the particle acceleration and heating by the growing drift wave in the context of the solar corona. we need specific data.," To provide some details of the particle acceleration and heating by the growing drift wave in the context of the solar corona, we need specific data."283" Hence. as an example we here use à set of parameters from Vranjes&Pocdts(2009a.b) that are known to vield the instability: By=107? T. nj=rim ο, ἐν-(dingfdr)fino]+s.100m. A,=0.5 m. and À.=s-101 m- ""These are just representative parameters with no particular significanee."," Hence, as an example we here use a set of parameters from \citet{v1,v2} that are known to yield the instability: $B_0=10^{-2}\;$ T, $n_0=10^{15}\;$ $^{-3}$, $L_n= [(dn_0/dx)/n_0]^{-1}=s\cdot 100\;$ m, $\lambda_y=0.5\;$ m, and $\lambda_z=s \cdot 10^4\;$ m. These are just representative parameters with no particular significance."284 As discussed in Vranjes&Poedts(2009b).. the instability will in fact develop. even when the density is varied by several orders of maguituce around the given value. and the same holds for the magnetic field and the temperature.," As discussed in \citet{v2}, the instability will in fact develop even when the density is varied by several orders of magnitude around the given value, and the same holds for the magnetic field and the temperature."285" The parameter s is introduced in Vranjes&Poedts(2009a.b). for convenience only. because it was realized that the ratio 5fw, remains exactly the same in Cus Horatia PL, is fixed."," The parameter $s$ is introduced in \citet{v1,v2} for convenience only, because it was realized that the ratio $\gamma/\omega_r$ remains exactly the same in case the ratio $\lambda_z/L_n$ is fixed."286" Here. £,, is the characteristic 1scale-lenei πι|ETAT hie inhomogencous equilibrium density."," Here, $L_n$ is the characteristic scale-length for the inhomogeneous equilibrium density."287" Phis further impliesthat we may go to very dillerent scales. c.g.bv taking s in the range 0.11073. As long as A, is kept constant. theratio 5fw, will remain the same. although the actual values for these two quantities will certainly change."," This further impliesthat we may go to very different scales, e.g.,by taking $s$ in the range $0.1 - 10^4$ As long as $\lambda_y$ is kept constant, theratio $\gamma/\omega_r$ will remain the same, although the actual values for these two quantities will certainly change."288 Lence. for the given parameters. from Eqs. (1))," Hence, for the given parameters, from Eqs. \ref{k1}) )"289" ancl (2)) .one finds +""welt", and \ref{k2}) ) one finds $\gamma/\omega_r=0.26$.290" In the case s=L1. this in [act το.""uiplies uw,— n] uxl 5=66 Lz."," In the case $s=1$, this in fact implies $\omega_r=254\;$ Hz and $\gamma=66\;$ Hz."291 n in the case s—10°. this would eive a mode with o=254 Hz ands LIz.," While in the case $s=10^3$, this would give a mode with $\omega_r=0.254\;$ Hz and $\gamma=0.66\;$ Hz."292" Going to such small values of L,, (tens of meters) in act makes sense because the perpendicular dilfusion. tha naturally clevclops in such a plasma with density eraclicnts. is indeed. very small. with the cdilfusion coellicient 24. of he order of 0.01. m?/s and the corresponding illusion velocity οιτηη of just a few millimeters per. secon (Vranjes&Pocets2008)."," Going to such small values of $L_n$ (tens of meters) in fact makes sense because the perpendicular diffusion, that naturally develops in such a plasma with density gradients, is indeed very small, with the diffusion coefficient $D_\bot$ of the order of $0.01\;$ $^2$ /s and the corresponding diffusion velocity $D_\bot \nabla_\bot n/n$ of just a few millimeters per second \citep{vx}."293" Short values of L, in the same ime imply very high frequency drift waves and fast growing instabilities. so that the expected heating will develop a ime scales that are orders of magnitude shorter than the xdasma diffusion."," Short values of $L_n$ in the same time imply very high frequency drift waves and fast growing instabilities, so that the expected heating will develop at time scales that are orders of magnitude shorter than the plasma diffusion."294" Note that the parameters A. and L, can of course also be varied independently of cach other.", Note that the parameters $\lambda_z$ and $L_n$ can of course also be varied independently of each other.295 The physics of the mentioned stochastic heating is described in detail in various sources., The physics of the mentioned stochastic heating is described in detail in various sources.296 Here. we give some short sketches obit by following Bellan (2006).. and usingthe stanclard drift wave theory.," Here, we give some short sketches of it by following \citet{bel}, , and usingthe standard drift wave theory."297 Withinthe tsvo-Iuid theory. the perpendicular velocity of the ion unit. volume is given by the recurrent formula(Vranjes&Poedts2006.200015] Oe eB Mable.The four terms here correspond to. respectively. the ££ L-drift that is usually the leading order one. the diamagneticdrift ρε the stress tensor draft ὅτι and the polarization dift c5.," Withinthe two-fluid theory, the perpendicular velocity of the ion unit volume is given by the recurrent formula \citep{v5, v2}298 + e_z + .The four terms here correspond to, respectively, the $\vec E\times \vec B$ -drift that is usually the leading order one, the diamagneticdrift $\vec v_{*i}$ , the stress tensor drift $ \vec v_{\pi i}$ , and the polarization drift $\vec299v_{pi}$ ."300 Lhe velocity can be calculated up to small ternis, The velocity can be calculated up to small terms301The spectral analvsis of the CS (CirciumStellar) envelopes of AGB stars has been an active field of observational astronomy for decades. (see. for instance. Bujarrabal et al. (1994)..,"The spectral analysis of the CS (CircumStellar) envelopes of AGB stars has been an active field of observational astronomy for decades (see, for instance, Bujarrabal et al. \cite{buj},"302 Woods οἱ al. (2003)))., Woods et al. \cite{woo}) ).303 As a result. the list of detected atoms. ions. raclicals ancl molecules is continuously enriched ancl their characterization improves in extent and equality as instruments and telescopes are constantlv upgraded (see van Dishoek (2004) [or references).," As a result, the list of detected atoms, ions, radicals and molecules is continuously enriched and their characterization improves in extent and quality as instruments and telescopes are constantly upgraded (see van Dishoek \cite{vdi} for references)."304 One of the most abundant molecules in both O-rich and C-rich AGDs is carbon monoxide. CO. so much so that the dominant isotopic varietv. CO. is optically thick in its characteristic inlrared bands. precluding direct determination of it column density. towards the star.," One of the most abundant molecules in both O-rich and C-rich AGBs is carbon monoxide, CO, so much so that the dominant isotopic variety, $^{12}$ CO, is optically thick in its characteristic infrared bands, precluding direct determination of its column density towards the star."305" One wav of cireumventing this diffieulty is to measure instead (he density of the rarer isotope. SCO, from which that of ""CO is deduced. assuming their ratio is somewhere between 1/50 and 1/100. because of Candetermined) fractionation processes which locally alter (he relative cosnic abundance of 1/39 (see Durgh at al. (2007).."," One way of circumventing this difficulty is to measure instead the density of the rarer isotope, $^{13}$ CO, from which that of $^{12}$ CO is deduced, assuming their ratio is somewhere between 1/50 and 1/100, because of (undetermined) fractionation processes which locally alter the relative cosmic abundance of 1/89 (see Burgh at al. \cite{bur},"306 Liszt (2001)))., Liszt \cite{lis}) ).307 In spite of these achievements and because of observational uncertainties (such as the CO isotopic ratio) and the relative scarcity of stars amenable (o in-depth analvsis. it is still οΠας to draw a complete and accurate observational picture of the variations of CO abundance with the (vpe of AGB star.," In spite of these achievements and because of observational uncertainties (such as the CO isotopic ratio) and the relative scarcity of stars amenable to in-depth analysis, it is still difficult to draw a complete and accurate observational picture of the variations of CO abundance with the type of AGB star."308 For want ofa better guess. one may assume an arbitrary but reasonable relative abundance. such as ΟΗΕ)=10* (e.g. Woods (2003)))).," For want of a better guess, one may assume an arbitrary but reasonable relative abundance, such as $[CO]/[H_{2}]=10^{-3}$ (e.g. Woods \cite{woo}) ))."309 Another simple rule of thumb has often been used: all (he oxygen (respectively carbon) expelled by C-rich (respectively O-richi) stus is locked into gaseous CO., Another simple rule of thumb has often been used: all the oxygen (respectively carbon) expelled by C-rich (respectively O-rich) stars is locked into gaseous CO.310 This rule is strongly suggested bv Che hieh reciprocal chemical allinities of the two elements. and the strength of the CO bond.," This rule is strongly suggested by the high reciprocal chemical affinities of the two elements, and the strength of the CO bond."311 Precisely because of the latter. the carbon and oxygen atoms locked in CO are irreversibly lost for carbonacious and silicate grams which ultimately condense in the envelope.," Precisely because of the latter, the carbon and oxygen atoms locked in CO are irreversibly lost for carbonacious and silicate grains which ultimately condense in the envelope."312 I strictly applied. the rule above implies (hat only stars with C/O1 can contribute to carbon grains.," If strictly applied, the rule above implies that only stars with $>$ 1 can contribute to carbon grains."313 Now. for many. carbon grain models. the C budget is already (uite üght. so the available C input deserves careful scrutiny.," Now, for many carbon grain models, the C budget is already quite tight, so the available C input deserves careful scrutiny."314 In particular. the carrier of the extinction bump at 2175 4 (see. for instance. Bless aud Savage (1972))) requires large quantities of pure carbon (see snow and Witt (1995) for a discussion).," In particular, the carrier of the extinction bump at 2175 $\AA{\ }$ (see, for instance, Bless and Savage \cite{ble}) ) requires large quantities of pure carbon (see Snow and Witt \cite{sno} for a discussion)."315 Since C'S shells. together with supernovae aud novae. Where erain condensation probably occurs in similar wavs. are the main. if not sole.," Since CS shells, together with supernovae and novae, where grain condensation probably occurs in similar ways, are the main, if not sole,"316bv them. it is diffieult to discuss 5-rav emission of the considered objects in 5general.,"by them, it is difficult to discuss $\gamma$ -ray emission of the considered objects in general."317 Instead. a detailed analvsis can be performed for each individual source.," Instead, a detailed analysis can be performed for each individual source."318 In this section. we discuss the large scale jet y-ray radiative outputs of Centaurus A and M 37.," In this section, we discuss the large scale jet $\gamma$ -ray radiative outputs of Centaurus A and M 87."319 One may note at this point. that both sources are quite unique.," One may note at this point, that both sources are quite unique."320 Thev are relatively nearby. what allowed in the past to study (heir jets at different Irequencies and scales. with hieh spatial and spectral resolutions.," They are relatively nearby, what allowed in the past to study their jets at different frequencies and scales, with high spatial and spectral resolutions."321 The laree scale jet in M 87 is the first one discovered al optical Irequencies., The large scale jet in M 87 is the first one discovered at optical frequencies.322 Both sources were among a few racio-loud AGNs known to posses the large scale X-ray jets before. was lannchecl., Both sources were among a few radio-loud AGNs known to posses the large scale X-ray jets before was launched.323 During the last decades. M 87 ancl Centaurus A were also being frequently observed αἱ 5-ravs. [rom sub-MeV (o TeV photon energies. what resulted in positive detections or constraints on the flix upper limits.," During the last decades, M 87 and Centaurus A were also being frequently observed at $\gamma$ -rays, from sub-MeV to TeV photon energies, what resulted in positive detections or constraints on the flux upper limits."324 Centaurus A galaxy is a giant elliptical. which most probably mergered in the past with an other small and gas rich galaxy.," Centaurus A galaxy is a giant elliptical, which most probably mergered in the past with an other small and gas rich galaxy."325 With the distance D=3.5 AIpe Centaurus A hosts the closest AGN. which is however dillicult to be observed at IR and optical frequencies because of the obscuring dust lane.," With the distance $D = 3.5$ Mpc Centaurus A hosts the closest AGN, which is however difficult to be observed at IR and optical frequencies because of the obscuring dust lane."326" VLA studies (e.g..Burnsetal.1983: show a double-lobed morphology of the Centaurus A radio source. with the jet and the counterjet of the FR I (vpe and a kinetic power e2-10"" erg/s. VLBI observations followed the jet and the counterjet to the pc-scales. suggesting the jet viewing angle ϐ~50""— 80""."," VLA studies \citep[e.g.,][]{bur83,cla92} show a double-lobed morphology of the Centaurus A radio source, with the jet and the counterjet of the FR I type and a kinetic power $\sim 2 \cdot 10^{43}$ erg/s. VLBI observations \citep[e.g.,][]{jon96}327 followed the jet and the counterjet to the pc-scales, suggesting the jet viewing angle $\theta \sim 50^0 - 80^0$ ."328 Optical and IR. observations of the Centaurus A nucleus (e.g. revealed the unresolved (<1 pe) and variable central source with a small (40 pc) nuclear disc ancl a possible extremely compact nuclear torus.," Optical and IR observations of the Centaurus A nucleus \citep[e.g.,][]{bai86,haw93,pac96,sch96,sch98,mar00,cap00} revealed the unresolved $< 1$ pc) and variable central source with a small $\sim 40$ pc) nuclear disc and a possible extremely compact nuclear torus."329 No IR/optical emission connected directly with the jet non-thermal radiation was found (butseeJoyοἱal.1991:Sehreier1996.L998:Marconiet2000).," No IR/optical emission connected directly with the jet non-thermal radiation was found \citep[but see][]{joy91,sch96,sch98,mar00}."330". At the X-ravs. the jet was observed in addition to the variable (at the time scales [rom minutes up (o vears) nuclear component (Schreierοἱal.1979:Morini1950,respectively)."," At the X-rays, the jet was observed in addition to the variable (at the time scales from minutes up to years) nuclear component \citep[respectively]{sch79,mor89}."331. Recently.Chandra (telescope revealed details of the jet N-rav emission ancl presence of the X-ray counterjet (Ixralftetal.2002:ELarcdeastle2003).. as well as of the whole Centaurus A svstem (Ixarovskaetal.2002).," Recently, telescope revealed details of the jet X-ray emission and presence of the X-ray counterjet \citep{kra02,hrd03}, as well as of the whole Centaurus A system \citep{kar02}."332.. At higher energies. Centaurus A was observed byCGRO satellite. what allowed to detectthe maximum of its nuclear >-rav emission near ~0.1 MeV (Steinleetal.," At higher energies, Centaurus A was observed by satellite, what allowed to detectthe maximum of its nuclear $\gamma$ -ray emission near $\sim 0.1$ MeV \citep{ste98}."3331998).. For ο>0.1 MeV the 5-rax [lux decreases. although al 100 MeV photon energies it is still detectable (Sreekumarοἱal.1999).," For $\varepsilon_{\gamma} > 0.1$ MeV the $\gamma$ -ray flux decreases, although at $\sim 100$ MeV photon energies it is still detectable \citep{sre99}."334". observations of ihe extended (~14 kpe) region around Centaurus A nucleus put the upper limit on the emission al the VILE range. Sls,>1.5TeV)<128-10|phem?s! (Rowelletal. 1999).."," observations of the extended $\sim 14$ kpc) region around Centaurus A nucleus put the upper limit on the emission at the VHE range, $S(\varepsilon_{\gamma} \geq 1.5 \, {\rm335 TeV}) < 1.28 \cdot 10^{-11} \, {\rm ph \, cm^{-2} \, s^{-1}}$ \citep{row99}. ."336 llowever. the discussed object was detected in the past at TeV energies. with theobserved," However, the discussed object was detected in the past at TeV energies, with theobserved"337"free path goes to infinity, for qo€kr.","free path goes to infinity, for $q_0\le k_F$."338 Thus the lowest energy for which A is finite corresponds to the lowest momentum allowed by Pauli blocking of the final state., Thus the lowest energy for which $\lambda$ is finite corresponds to the lowest momentum allowed by Pauli blocking of the final state.339" Table 1, together with Eq. ("," Table 1, together with Eq. ("340"5), should facilitate the interpretation of the mean free path behaviour observed in Fig.","5), should facilitate the interpretation of the mean free path behaviour observed in Fig."341 1., 1.342 Reconnecting with the previous discussion which followed Eq. (, Reconnecting with the previous discussion which followed Eq. (343"3), we also show, see Fig.","3), we also show, see Fig."344" 2, the mean free path calculated without considerations of Pauli blocking of the final states."," 2, the mean free path calculated without considerations of Pauli blocking of the final states."345" In this case, \ becomes very small at low energy, due to the large values of the cross section in that region."," In this case, $\lambda$ becomes very small at low energy, due to the large values of the cross section in that region."346 Back to Fig., Back to Fig.347" 1, and focusing on the higher density first (solid line), we see the sharp drop from infinity at low energy, after which the mean free path slowly decreases with energy, due to the fact that the in-medium cross sections actually start to go up with energy at high densities, see Table 1."," 1, and focusing on the higher density first (solid line), we see the sharp drop from infinity at low energy, after which the mean free path slowly decreases with energy, due to the fact that the in-medium cross sections actually start to go up with energy at high densities, see Table 1."348" This feature, which may apppear counterintuitive (being opposite to what is seen in free space), has been reported in other works as well [3,0].."," This feature, which may apppear counterintuitive (being opposite to what is seen in free space), has been reported in other works as well \cite{Jiang07,Fuchs}."349" A similar behavior also sets in at the lower density (dashed line), but in that case the mean free path, after the sharp drop from infinity, rises with energy at first (corresponding to a reduction of the in-medium cross section)."," A similar behavior also sets in at the lower density (dashed line), but in that case the mean free path, after the sharp drop from infinity, rises with energy at first (corresponding to a reduction of the in-medium cross section)."350" Notice that the tendency to rise with energy in dense matter appears more pronounced for scattering of identical nucleons, a behaviour which was traced to in-medium enhancement of some isospin-1 partial waves [1].."," Notice that the tendency to rise with energy in dense matter appears more pronounced for scattering of identical nucleons, a behaviour which was traced to in-medium enhancement of some isospin-1 partial waves \cite{sk05}."351 We now move to mean free path considerations in asymmetric matter., We now move to mean free path considerations in asymmetric matter.352" For scattering of like nucleons, the cross section is set equal to zero when qo<ki, (i=n, p), whereas for np scattering it is set to zero for qy<kp, theaverage Fermi momentum."," For scattering of like nucleons, the cross section is set equal to zero when $q_0\le k_F^i$ $i=n,p$ ), whereas for $np$ scattering it is set to zero for $q_0\le k_F$, the Fermi momentum."353 The corresponding behavior of the mean free path is shown in Fig., The corresponding behavior of the mean free path is shown in Fig.354 3 for a=0.5 and in Fig., 3 for $\alpha$ =0.5 and in Fig.355" 4 for a greater degree of asymmetry, a=0.8."," 4 for a greater degree of asymmetry, $\alpha$ =0.8."356" The large differences between the mean free path for protons and neutrons at the lowest energies are to be expected from what we stated above, namely the suppression of pp and nn cross sections is controlled by the (unequal) proton and neutron Fermi momenta."," The large differences between the mean free path for protons and neutrons at the lowest energies are to be expected from what we stated above, namely the suppression of $pp$ and $nn$ cross sections is controlled by the (unequal) proton and neutron Fermi momenta."357" Proceeding from the lowest to the highest energies, the proton mean free path is infinity when both pp and np cross sections are Pauli blocked, followed by the small rise around 25 MeV, and then again the sharp drop when the np cross section starts to contribute."," Proceeding from the lowest to the highest energies, the proton mean free path is infinity when both $pp$ and $np$ cross sections are Pauli blocked, followed by the small rise around 25 MeV, and then again the sharp drop when the $np$ cross section starts to contribute."358 Similar considerations, Similar considerations359Iu this final section we coustruct examples of four-folds with trivial canonical »uuncdles which. are fibred1 over I752 by Jacobiaus: of 2Meeuus two curves.,In this final section we construct examples of four-folds with trivial canonical bundles which are fibred over $\P^2$ by Jacobians of genus two curves.360 Moreover. we Classify all such four-folds.," Moreover, we classify all such four-folds."361 The results in this section owe a lot to Markusheviclis vapers [LL15]: indeed. the examples we describe are the Deaunville-Mukai system (a holomorphic sviuplectie fou-fold) aud Example 1 from ΤΗ (which we prove is à Calabi-Yau four-fold).," The results in this section owe a lot to Markushevich's papers \cite{markushevich95,362 markushevich96}; indeed, the examples we describe are the Beauville-Mukai system (a holomorphic symplectic four-fold) and Example 1 from \cite{markushevich95} (which we prove is a Calabi-Yau four-fold)."363 \larkushevich [15] classified all irreducible nolmorphlic sviuplectie four-folds fibred by Jacobiaus: our classification result is an extension of lis. aud uses similar arguments.," Markushevich \cite{markushevich96} classified all irreducible holomorphic symplectic four-folds fibred by Jacobians; our classification result is an extension of his, and uses similar arguments."364 We start by definingg exactly what we mean by a fibration by Jacobiaus., We start by defining exactly what we mean by a fibration by Jacobians.365 The ⋡∪↕∪↖↖↽↕∐∶↴⋁↕↴∖↴↸∖↴∖↴↴∖↴↸∖∐↑↕⋜↧∐⋅↖⇁↕≻↸∖∱∎⊔↕↑↕∪∐↓↕≯↥⋅∪⋯∏, The following isessentially Definition 4 from \cite{markushevich95}. .366while the effect of the pressure gradient on the equilibrium rotation velocity leaves us a margin with respect to the Rayleigh criterion κ;>04).,while the effect of the pressure gradient on the equilibrium rotation velocity leaves us a margin with respect to the Rayleigh criterion $\kappa^{2}>0$.367". The bump is centered at rg=3rj, far enough from the inner edge of the disk to avoid a strong effect of the boundary condition there."," The bump is centered at $r_{B}=3 r_i$, far enough from the inner edge of the disk to avoid a strong effect of the boundary condition there."368" The vertical density profile is initially at hydrostatic equilibrium, giving an aspect ratio of the order of 10-1."," The vertical density profile is initially at hydrostatic equilibrium, giving an aspect ratio of the order of $10^{-1}$."369 with10-?r20;(r;)?pl;., with.370". Finally, we use initially a density pj,=10-?p,, in order to avoid getting too low densities in the corona."," Finally, we use initially a density $\rho_{min}=10^{-2}\rho_m$ in order to avoid getting too low densities in the corona."371 Figure 2 shows the resulting isodensity contours in a vertical cut of the disk., Figure \ref{FigCI} shows the resulting isodensity contours in a vertical cut of the disk.372" The initial velocity field is purely toroidal, v/v""=0 and v?"" has been chosen for the disk to be in radial equilibrium in a Newtonian potential: The numerical simulations presented here use the Versatile Advection Code (VAC) developped by ?.."," The initial velocity field is purely toroidal, $v_r^{ini}=v_z^{ini}=0$ and $v_{\phi}^{ini}$ has been chosen for the disk to be in radial equilibrium in a Newtonian potential: The numerical simulations presented here use the Versatile Advection Code (VAC) developped by \citet{TOT96}."373 In the version we use the code solves the 3D hydrodynamics equations for an isentropic flow., In the version we use the code solves the 3D hydrodynamics equations for an isentropic flow.374" We use VAC with the total variation diminishing monotonic upstream-centered scheme for conservation laws (TVD-MUSCL), a Roe Riemann solver, a Hancock predictor step and a Woodward limiter (?).."," We use VAC with the total variation diminishing monotonic upstream-centered scheme for conservation laws (TVD-MUSCL), a Roe Riemann solver, a Hancock predictor step and a Woodward limiter \citep{COL84}."375" The TVD-MUSCL scheme detailed in ? is one of the less dissipative schemes included in the VAC code, which is useful in order to observe the full development of the instability and properly characterize its saturation."," The TVD-MUSCL scheme detailed in \citet{TOO96} is one of the less dissipative schemes included in the VAC code, which is useful in order to observe the full development of the instability and properly characterize its saturation."376"and 16 um channels of the Spitzer IRS, in which regions the modelled radiance is dominated by CO» absorption, as explained below.","and 16 $\mu$ m channels of the $Spitzer$ IRS, in which regions the modelled radiance is dominated by $_{2}$ absorption, as explained below."377" The H2O functional derivatives all peak in the ~100-500 mbar region, showing that the measurements can only constrain the H2O abundance in this altitude range."," The $_{2}$ O functional derivatives all peak in the $\sim$ 100–500 mbar region, showing that the measurements can only constrain the $_{2}$ O abundance in this altitude range."378" In contrast, the CO2 functional derivatives are divided into two separate pressure levels (peaks between ~0.1 bar and 0.1—1 mbar) as are clearly shown in the second row of Fig. 4.."," In contrast, the $_{2}$ functional derivatives are divided into two separate pressure levels (peaks between $\sim$ 0.1 bar and 0.1–1 mbar) as are clearly shown in the second row of Fig. \ref{f4}."379 The H ST/NICMOS channels between 1.584-1.698 and 1.929-2.216 um are only sensitive to the CO2 abundance at altitudes below 100 mbar whereas the Spitzer broadband photometry and IRS spectroscopy are sensitive to the CO» in the 0.1-1 mbar region as well., The $HST$ /NICMOS channels between 1.584–1.698 and 1.929–2.216 $\mu$ m are only sensitive to the $_{2}$ abundance at altitudes below 100 mbar whereas the $Spitzer$ broadband photometry and IRS spectroscopy are sensitive to the $_{2}$ in the 0.1–1 mbar region as well.380" Despite this sensitivity to a range of altitudes for CO», a combination of these channels, however, may not determine CO» abundance at both levels due to the small sensitivity in the upper atmosphere (cf."," Despite this sensitivity to a range of altitudes for $_{2}$, a combination of these channels, however, may not determine $_{2}$ abundance at both levels due to the small sensitivity in the upper atmosphere (cf."381 Section 5.4)., Section 5.4).382" CO can only be detected in three HST/NICMOS channels in the range 2.33-2.45 um at 100 mbar, and these channels have been extensively used to constrain the CO mixing ratio (S09 and MS09)."," CO can only be detected in three $HST$ /NICMOS channels in the range 2.33–2.45 $\mu$ m at 100 mbar, and these channels have been extensively used to constrain the CO mixing ratio (S09 and MS09)."383" The sensitivities for CH4 are so marginal, as seen in the fourth row, that constraining its abundance may not be possible from the given measurements."," The sensitivities for $_{4}$ are so marginal, as seen in the fourth row, that constraining its abundance may not be possible from the given measurements."384" In summary, the functional derivatives indicate the altitudes and channels showing sensitivity to each particular molecule."," In summary, the functional derivatives indicate the altitudes and channels showing sensitivity to each particular molecule."385" The sensitivities to the molecules are mostly clustered in the lower atmosphere (74100 mbar), and, in particular, CO2 shows an additional peak at ~1 mbar."," The sensitivities to the molecules are mostly clustered in the lower atmosphere $\sim$ 100 mbar), and, in particular, $_{2}$ shows an additional peak at $\sim$ 1 mbar."386" Therefore the molecular abundances can only be constrained in the deep atmosphere (troposphere), with less sensitivity to the upper atmosphere."," Therefore the molecular abundances can only be constrained in the deep atmosphere (troposphere), with less sensitivity to the upper atmosphere."387" Finally, these functional derivatives show that the spectrum is sensitive to the HzO abundance over a broad range in most of the measured channels, implying an inherent degeneracy between temperatures and H3O in these observations."," Finally, these functional derivatives show that the spectrum is sensitive to the $_{2}$ O abundance over a broad range in most of the measured channels, implying an inherent degeneracy between temperatures and $_{2}$ O in these observations."388 Previously P-T' profiles of HD 189733b have been estimated based on theoretical models or parametric retrievals to generate many thousands of model spectra to compare with observations., Previously $P$ $T$ profiles of HD 189733b have been estimated based on theoretical models or parametric retrievals to generate many thousands of model spectra to compare with observations.389" However, this forward modelling approach does not explicitly solve the inverse problem during the constraining process and thus it is unclear if the solutions are biased more towards theoretical expectations than being driven by the measurements themselves."," However, this forward modelling approach does not explicitly solve the inverse problem during the constraining process and thus it is unclear if the solutions are biased more towards theoretical expectations than being driven by the measurements themselves."390" For this reason, we derive temperatures using several different a priori profiles to show that the retrieved temperature converges to a reproducible profile in the altitude range covered by the contribution functions."," For this reason, we derive temperatures using several different $a$ $priori$ profiles to show that the retrieved temperature converges to a reproducible profile in the altitude range covered by the contribution functions."391" Because a biased P-T' profile would vary with the a priori assumptions, we look for an appropriate a priori error and its vertical shape in order that the same P-T' profile can be retrieve irrespective of the shape of the a priori profile."," Because a biased $P$ $T$ profile would vary with the $a$ $priori$ assumptions, we look for an appropriate $a$ $priori$ error and its vertical shape in order that the same $P$ $T$ profile can be retrieve irrespective of the shape of the $a$ $priori$ profile."392 Fig. 5((, Fig. \ref{f5}( (393a) presents the retrieved P-T' profile and its error for a range of selected temperature a priori.,a) presents the retrieved $P$ $T$ profile and its error for a range of selected temperature $a$ $priori$.394" For all cases, the P-T' profiles share a common shape between 0.1 mbar and 1 bar, demonstrating the validity of the temperature retrieval over this range."," For all cases, the $P$ $T$ profiles share a common shape between 0.1 mbar and 1 bar, demonstrating the validity of the temperature retrieval over this range."395 It is shown that even with the simplest possible assumption such as an isothermal temperature throughout the atmosphere [blue line in Fig. 5((, It is shown that even with the simplest possible assumption such as an isothermal temperature throughout the atmosphere [blue line in Fig. \ref{f5}( (396"a)], the measurements still produce a similar thermal profile as the other retrievals.","a)], the measurements still produce a similar thermal profile as the other retrievals."397" As a further test, we take an a priori structure from the retrieved profile in the previous step, offset it by a temperature 400 K, and repeat the temperature retrieval again."," As a further test, we take an $a$ $priori$ structure from the retrieved profile in the previous step, offset it by a temperature 400 K, and repeat the temperature retrieval again."398 Fig. 5((, Fig. \ref{f5}( (399"b) shows that the P-T' profile is still sufficiently constrained by the measurements, even if there are large shifts at levels not covered by the contribution functions, where the solutions relax back to their different a prioris. We conclude here that the available dayside emission spectra can constrain an unbiased atmospheric structure over the levels probed by the contribution functions.","b) shows that the $P$ $T$ profile is still sufficiently constrained by the measurements, even if there are large shifts at levels not covered by the contribution functions, where the solutions relax back to their different $a$ $priori$ s. We conclude here that the available dayside emission spectra can constrain an unbiased atmospheric structure over the levels probed by the contribution functions."400" As a result, we find that the temperature decreases adiabatically from 1900 K at 600 mbar to ~1200 K at 100 mbar, then becomes isothermal up to the upper atmosphere (1 mbar)."," As a result, we find that the temperature decreases adiabatically from 1900 K at 600 mbar to $\sim$ 1200 K at 100 mbar, then becomes isothermal up to the upper atmosphere $\sim$ 1 mbar)."401" These adiabatic and isothermal layers in the thermal structure are dominant features of heat transfer by convection and radiative cooling, respectively."," These adiabatic and isothermal layers in the thermal structure are dominant features of heat transfer by convection and radiative cooling, respectively."402" In comparison S09 claimed a decreasing temperature layer between 0.01 and 1 bar to model the HST/NICMOS measurements, making an adiabatic layer ~10 times thicker than our estimation, and theoretical models also considered adiabatic layers in the troposphere (Fortneyetal.2006;Burrowsetal.2008;Showman 2008)."," In comparison S09 claimed a decreasing temperature layer between 0.01 and 1 bar to model the $HST$ /NICMOS measurements, making an adiabatic layer $\sim$ 10 times thicker than our estimation, and theoretical models also considered adiabatic layers in the troposphere \citep{for06,bur08,sho08}."403. The isothermal temperature (~1100 K) of the dayside hemisphere at pressure levels above the troposphere (100 mbar) is possibly maintained by efficient energy re-distribution throughout the whole planet system as suggested by the observations in the Spitzer IRAC channels (Knutsonetal.2007;Charbonneau2008).," The isothermal temperature $\sim$ 1100 K) of the dayside hemisphere at pressure levels above the troposphere (100 mbar) is possibly maintained by efficient energy re-distribution throughout the whole planet system as suggested by the observations in the $Spitzer$ IRAC channels \citep{knu07,cha08}."404. Various circulation models (Showmanetal.2008;Dobbs-Dixon&Lin2008) predict that the heat transports between the dayside and nightside lead to the isothermal structure at mid and low pressure (<100 mbar) in the dayside atmosphere.," Various circulation models \citep{sho08,dob08} predict that the heat transports between the dayside and nightside lead to the isothermal structure at mid and low pressure $<$ 100 mbar) in the dayside atmosphere."405" In summary, the broad wavelength range of available measurements has provided a strong constraint on the vertical P- T profile (an adiabatic troposphere and isothermal stratosphere) without excessive sensitivity to the a priori assumptions, and without reliance on theoretical modelling or parameterised profiles."," In summary, the broad wavelength range of available measurements has provided a strong constraint on the vertical $P$ $T$ profile (an adiabatic troposphere and isothermal stratosphere) without excessive sensitivity to the $a$ $priori$ assumptions, and without reliance on theoretical modelling or parameterised profiles."406" In general, the best-fitting P-T profile from the retrieval is a"," In general, the best-fitting $P$ $T$ profile from the retrieval is a"407"We may evaluate907,7, [rom ils veclor-invariant form (??)). = —2n(l + + ).","We may evaluate$\delta\sigma_{\zb\zb}$ from its vector-invariant form \ref{sigvec}) ), = -2 ( + + ) ."408(23) Usingυ=ROd and equation (??)). we find + |.," Using$\bb{v}=R\Omega\bb{\hat \phi}$ and equation \ref{b}) ), we find + ]."409(24) The linearized equations are written in a coordinate svstem (hat is shearing with the unperturbed flow., The linearized equations are written in a coordinate system that is shearing with the unperturbed flow.410 The only effects this has on the equations of motion are that the time derivative must be Lagrangian. of ≀↧↴∐≼⇂⊔∐↲↕⋅⋯∐≀↧↴↥↕↽≻≀↧↴↕⋅∐≀↧↴↥≺⇂≼↲↕⋅↕∖↽≀↧↴∐∖↽≼↲↕⊳∖⇁↕⋅≼↲↕↽≻↥≀↧↴≺∢≼↲≼⊓↽≻∡∖↽⊔∐↲∣⋅⊲∫≖≽⋜⋝∣⇄⋝∖∖⊽≀↧↴∖↽≼↲∐⋯∐∣↽≻≼↲↕⋅↕∐≼," The only effects this has on the equations of motion are that the time derivative must be Lagrangian, +, and the radial partial derivative is replaced by the $k_R(t)$ wavenumber in equation \ref{kR}) ) \citep{bh92}."411"↲≺⇂∏≀↧↴∐∪∐≼∕∙↗↕↗∣⋝↕⋝ ≼⊥⊰≀↧↴∐↽≻∏⊳∖⊽≪↽∖↽∐≀↧↴∖∖↽↥≼↲∡∖⇁⊥≤∍≤⊔⇄⋝⋅⋅↴∏∐↲≼⇂⋡∖↽∐≀↧↴∐∐≺∢≀↧↴↥≼↲≺⇂∏≀↧↴∐∪∐⊳∖⇁≀⋯↲ = 0. + ((op = yg, EN XyóUp μαERN PI =0."," The dynamical equations are = 0, + ( P - )= 0, + v_R -) ] = 0,"412"↲≺⇂∏≀↧↴∐∪∐≼∕∙↗↕↗∣⋝↕⋝ ≼⊥⊰≀↧↴∐↽≻∏⊳∖⊽≪↽∖↽∐≀↧↴∖∖↽↥≼↲∡∖⇁⊥≤∍≤⊔⇄⋝⋅⋅↴∏∐↲≼⇂⋡∖↽∐≀↧↴∐∐≺∢≀↧↴↥≼↲≺⇂∏≀↧↴∐∪∐⊳∖⇁≀⋯↲ = 0. + ((op = yg, EN XyóUp μαERN PI =0.("," The dynamical equations are = 0, + ( P - )= 0, + v_R -) ] = 0,"413"↲≺⇂∏≀↧↴∐∪∐≼∕∙↗↕↗∣⋝↕⋝ ≼⊥⊰≀↧↴∐↽≻∏⊳∖⊽≪↽∖↽∐≀↧↴∖∖↽↥≼↲∡∖⇁⊥≤∍≤⊔⇄⋝⋅⋅↴∏∐↲≼⇂⋡∖↽∐≀↧↴∐∐≺∢≀↧↴↥≼↲≺⇂∏≀↧↴∐∪∐⊳∖⇁≀⋯↲ = 0. + ((op = yg, EN XyóUp μαERN PI =0.(2"," The dynamical equations are = 0, + ( P - )= 0, + v_R -) ] = 0,"414"↲≺⇂∏≀↧↴∐∪∐≼∕∙↗↕↗∣⋝↕⋝ ≼⊥⊰≀↧↴∐↽≻∏⊳∖⊽≪↽∖↽∐≀↧↴∖∖↽↥≼↲∡∖⇁⊥≤∍≤⊔⇄⋝⋅⋅↴∏∐↲≼⇂⋡∖↽∐≀↧↴∐∐≺∢≀↧↴↥≼↲≺⇂∏≀↧↴∐∪∐⊳∖⇁≀⋯↲ = 0. + ((op = yg, EN XyóUp μαERN PI =0.(28"," The dynamical equations are = 0, + ( P - )= 0, + v_R -) ] = 0,"415"↲≺⇂∏≀↧↴∐∪∐≼∕∙↗↕↗∣⋝↕⋝ ≼⊥⊰≀↧↴∐↽≻∏⊳∖⊽≪↽∖↽∐≀↧↴∖∖↽↥≼↲∡∖⇁⊥≤∍≤⊔⇄⋝⋅⋅↴∏∐↲≼⇂⋡∖↽∐≀↧↴∐∐≺∢≀↧↴↥≼↲≺⇂∏≀↧↴∐∪∐⊳∖⇁≀⋯↲ = 0. + ((op = yg, EN XyóUp μαERN PI =0.(28)"," The dynamical equations are = 0, + ( P - )= 0, + v_R -) ] = 0,"416The iain advantage of SIAL over the VIRGO/SPM channels is that a ιο. more conyplete sampling of wavelonetls is avaiable.,The main advantage of SIM over the VIRGO/SPM channels is that a much more complete sampling of wavelengths is available.417 Fig., Fig.418 δ shows the RAIS variahiitv between Alay anc August 2001 over the whole analysed SORCE/SIM wavecheth rauge., \ref{fig:var_all} shows the RMS variability between May and August 2004 over the whole analysed SORCE/SIM wavelength range.419 The normalised saudard deviation of the SORCE/SIAL data is indicated bv the solid black lines iu ]xuels a to c. though note that t16 grev- area on panel a indicates the range in variability hat is obtained depending on whether a linear YO 15] noved from the data or not.," The normalised standard deviation of the SORCE/SIM data is indicated by the solid black lines in panels a to c, though note that the grey-shaded area on panel a indicates the range in variability that is obtained depending on whether a linear trend is removed from the data or not."420 The horizoital bars du Figs., The horizontal bars in Figs.421 δ » aud c duicate the wavelengtli regioIn Ww.rere the nwsured variabili vods dominated by iustrunieutal noise (back dotted lines)., \ref{fig:var_all} b and c indicate the wavelength regions where the measured variability is dominated by instrumental noise (black dotted lines).422 The modelled variability Is ludicaed w the due lines, The modelled variability is indicated by the blue lines.423 T16 bottom plots (d. ο and f) show he conributions of the spots (dottc4( blacs lues) aud aculae (dashed purpe lues) tot1ο overall variabiliv.," The bottom plots (d, e and f) show the contributions of the spots (dotted black lines) and faculae (dashed purple lines) to the overall variability."424 To calculate these. we repaced the facular (resp.," To calculate these, we replaced the facular (resp."425 sunspot) contribution bv a cuiΤΟ conrilttion., sunspot) contribution by a quiet-Sun contribution.426" The «""ves show very clearly tha tt16 waveleeth 6epeudenuce O| spot variability is spectrallylv uncle siuether than the facular variability.", The curves show very clearly that the wavelength dependence of spot variability is spectrally much smoother than the facular variability.427 This has τοςo with the act hat the darkeinue due to spots is dominated by the «clrop iu contiumm intensity., This has to do with the fact that the darkening due to spots is dominated by the drop in continuum intensity.428 Clanges in spectral lines produced. by the lower tenrperature iu spot umbrae aixl peuunibrae av a secondarv role., Changes in spectral lines produced by the lower temperature in spot umbrae and penumbrae play a secondary role.429 For faculac. the absolute teniperature differeuce dis less pronounced (oesveclally in the kWer atinosphere). aud it is the different teniperature eraclicut that produces chauges iu the coutinuun as well as iu the lues.," For faculae, the absolute temperature difference is less pronounced (especially in the lower atmosphere), and it is the different temperature gradient that produces changes in the continuum as well as in the lines."430 Especially at shorter wavleugths. individual and eroups of lues provide the doniuait coutributiou (Mitchell&Livineston1991:αιetal.2000)..," Especially at shorter wavelengths, individual and groups of lines provide the dominant contribution \citep{mitchell91,unruh2000issi}."431 Deow about 280 nn. the variability due to faculae eenerallv exceeds that «ue to spots: then ollows a region up to LOO) num where they are inostlv of comparable magnitude.," Below about 280 nm, the variability due to faculae generally exceeds that due to spots; then follows a region up to 400 nm where they are mostly of comparable magnitude."432" Above 100 in. the modelled spot variability is always lareer than the facular variaημίν, aud the combined variability. folows the spot variability. closcly."," Above 400 nm, the modelled spot variability is always larger than the facular variability, and the combined variability follows the spot variability closely."433 Note hat there is a further cross-over ix»nd 1.1 au where the modelled variability of tl| spots aud. faculae drops olov the total variability., Note that there is a further cross-over around 1.4 $\mu$ m where the modelled variability of the spots and faculae drops below the total variability.434 This 1uarks the transition where the facular model becomes dar.s averaged over the solar disk aud thus no longer acts to counterbalance the spots., This marks the transition where the facular model becomes dark averaged over the solar disk and thus no longer acts to counterbalance the spots.435 The fact that the model shows : vstualler variability than SIM at these wavelengths caurot be due to this property of the model: dark faculae euvance the darkeni18o due to spots. and thus increase the staidard deviatiou (see also Sec. 5 yy.," The fact that the model shows a smaller variability than SIM at these wavelengths cannot be due to this property of the model; dark faculae enhance the darkening due to spots, and thus increase the standard deviation (see also Sec. \ref{sec:conclusion}) )."436 In other words. if the model faculae were bright at these waveleneths the disepauev between t SATIRE iodel and SIM data would )o larger.," In other words, if the model faculae were bright at these wavelengths the discrepancy between the SATIRE model and SIM data would be larger."437 The shift iun importance away froii faculae to spots at waveleugths as low as about 300 1 is expected when considering variability on the order of a couple of mouth. Le. on the rotation time scale when tl1¢ influence of spots tends to dominate the TSI.," The shift in importance away from faculae to spots at wavelengths as low as about 300 nm is expected when considering variability on the order of a couple of month, i.e., on the rotation time scale when the influence of spots tends to dominate the TSI."438 Ou longer tine scales such as that of the solar cevcle. however. bright sinall-scale magnetic features dominate the TSI variations. aud are thus also expected to dominate variability iu both the UV aud visible.," On longer time scales such as that of the solar cycle, however, bright small-scale magnetic features dominate the TSI variations, and are thus also expected to dominate variability in both the UV and visible."439 Iu this section. we compare the variability iu à nuniber of wavelonet1 bands in more detail," In this section, we compare the variability in a number of wavelength bands in more detail."440 The wavelength bands have been picked so as to show he change in behaviour going from the UV around 220 um. up to the IR at 1.5 san. Thev are tvpically also chosen at wavecheths where t relative variabilivods hnieh aud tje data quality is good.," The wavelength bands have been picked so as to show the change in behaviour going from the UV around 220 nm, up to the IR at 1.5 $\mu$ m. They are typically also chosen at wavelengths where the relative variability is high and the data quality is good."441 The wavecheth bands are ste in Tab. l..," The wavelength bands are listed in Tab. \ref{tab:bands},"442 along wi heir bane widtli and the ΠΟ of daa points include in the iutegratkli O| the SIM data., along with their band widths and the number of data points included in the integration of the SIM data.443 So as to improve t S/N level of tlicποπ πιο fine series. he UV bands. in urtieubu. have bec1 chosen to iuclude a large nunb of wavelength ])oluts.," So as to improve the S/N level of the resulting time series, the UV bands, in particular, have been chosen to include a large number of wavelength points."444 Note tha the wise level of t nodelled tiue series ds larecly governed bv the noise 1 he maenetoeramsoOC» and cannot be decreased by increasi1g he number of wavecneth poiits considered: the absolte flux level. however. is influeuced by the coarseness of t1e wavelonet1 exid.," Note that the noise level of the modelled time series is largely governed by the noise in the magnetograms and cannot be decreased by increasing the number of wavelength points considered; the absolute flux level, however, is influenced by the coarseness of the wavelength grid."445 The measure (uad modeled timescries are shown iu Έπος 9 and 10 aud the correlation cocfiicients as well as the sloaes between the model and the data are listexl in Tab. L., The measured and modelled timeseries are shown in Figs \ref{fig:timeseries1} and \ref{fig:timeseries2} and the correlation coefficients as well as the slopes between the model and the data are listed in Tab. \ref{tab:bands}.446 The| upper panes show the measured auc modclled solar variability while the bottom panels show the 1oeled contribitions of the faculae and spots to he nvracdianuce varialolis., The upper panels show the measured and modelled solar variability while the bottom panels show the modelled contributions of the faculae and spots to the irradiance variations.447 The figures illustrate again the rapid decrease in the varbilitv towards longer waveleustlis., The figures illustrate again the rapid decrease in the variability towards longer wavelengths.448 Additionally.la trey show the chauge iu he lighteurve aspects as one noves from the UV to the visibe.," Additionally, they show the change in the lightcurve aspects as one moves from the UV to the visible."449 Iu he UV. as ilhstratevd by the 230 ix| 280 111 bauds (Fig. 9) .," In the UV, as illustrated by the 230 and 280 nm bands (Fig. \ref{fig:timeseries1}) ),"450 tjo Influcuce of the sdots IS SOSΠα that t1011 ¢larkeniis ejfect is more tWL conpensatec for bv the faculae. eve‘ll on solar rotatioial tine scales.," the influence of the spots is so small that their darkening effect is more than compensated for by the faculae, even on solar rotational time scales."451" Furthermore. the facular contrast Increase towards the iub is not sufücieut at tjese waveleusgls to counteract the ]xojectioi effects,"," Furthermore, the facular contrast increase towards the limb is not sufficient at these wavelengths to counteract the projection effects."452 C'onsequeib. tjo Su appears brightest when the main ni)ot grouds are nearly at disk centre.," Consequently, the Sun appears brightest when the main spot groups are nearly at disk centre."453 At our resoltiou. this behaviour is no louger observed at longer wavelengths. such as in the Ca aud the € bans.," At our resolution, this behaviour is no longer observed at longer wavelengths, such as in the Ca and the G bands."454 T16 sprMo darkeune is now sufficient to offset the facular brigiteniug. at least when he active regioIs are near disk ceutre.," The spot darkening is now sufficient to offset the facular brightening, at least when the active regions are near disk centre."455" The overall lighteuves thus appear somewhat confising. with rapid sequoaces of peaks arxd dips. and very fow stretches of ""quieSun behavior1."," The overall lightcurves thus appear somewhat confusing, with rapid sequences of peaks and dips, and very few stretches of `quiet'-Sun behaviour."456 Further high-resolution calculations wotul be required Q check whether tus behaviour also holds for the individual liue cores. or whether we are cuirenlv secing a nrIX of line and continuum behaviour.," Further high-resolution calculations would be required to check whether this behaviour also holds for the individual line cores, or whether we are currently seeing a mix of line and continuum behaviour."457 The| low waveleughi resolution of tlic! jnodel aud the failure o calculate exact lue profiles partly explains the relativev huge difference, The low wavelength resolution of the model and the failure to calculate exact line profiles partly explains the relatively large difference458emit f.=250 photons per atom over ἐς = 3 Myr leading to a photon Lux A oL. where AZ is the halo mass.,"emit $f_{\gamma} = 250$ photons per atom over $t_s$ = 3 Myr leading to a photon flux $\dot{N}_{\gamma}$ of, where $M$ is the halo mass."459 The svstem was then evolved for 0.4 Myr., The system was then evolved for 0.4 Myr.460 In order for SPHBAYto. complete this test dH was necessary to convert the density. field. data [rom a eric based. representation into an SPILL density. field., In order for to complete this test it was necessary to convert the density field data from a grid based representation into an SPH density field.461 Εις was accomplished using the following procedure: we start from a smooth elass distribution representing a constant density Ποιά equal to the peak density pias of the grid. data.," This was accomplished using the following procedure: we start from a smooth glass distribution representing a constant density field equal to the peak density $\rho_{\rm462max}$ of the grid data."463" Particles are then selected. for removal according to a comparison of the density p, at the particle position with", Particles are then selected for removal according to a comparison of the density $\rho_p$ at the particle position with464termination shock. the wind electrons can be accelerated to multi-TeV energies.,"termination shock, the wind electrons can be accelerated to multi-TeV energies."465 The radiation of these electrons results in a phenomenon called pulsar wind nebula., The radiation of these electrons results in a phenomenon called pulsar wind nebula.466 Since (he wind is cold. i.e. particles remain at rest in the wind co-moving system. no svnchrotron emission is expected from the wind before its termination.," Since the wind is cold, i.e. particles remain at rest in the wind co-moving system, no synchrotron emission is expected from the wind before its termination."467 On the other hand. the comptonization of the ultrarelativise wind by external radiation fields. through which the wind propagates. can lead to detectable eanmuna ray emission.," On the other hand, the comptonization of the ultrarelativistic wind by external radiation fields, through which the wind propagates, can lead to detectable gamma ray emission."468 This effect is relatively weak in isolated pulsars. and can achieve a reasonable elliciency only in powerful pulsars. provided that the particle dominated wind is formed close to the light evlinder (Bogovaloy&Aharonian2000).," This effect is relatively weak in isolated pulsars, and can achieve a reasonable efficiency only in powerful pulsars, provided that the particle dominated wind is formed close to the light cylinder \citep{bogovalov00}."469. In binary systems. (he process operates wilh an enhanced efficiency. thanks to the presence of (he dense radiation field of (he optical companion (Ball&Ixirk2000:BallDodd2001).," In binary systems, the process operates with an enhanced efficiency thanks to the presence of the dense radiation field of the optical companion \citep{ball00,ball01}."470. The interaction rate in this channel depends on different parameters characterizing (lie svstem: (1) Iuminosity and temperature of the optical star: Gi) orbital separation and inclination: ii) distance to (he svstem: (iv) size of the region occupied bv (the pulsar wind: (v) pulsar wind bulk Lorentz factor., The interaction rate in this channel depends on different parameters characterizing the system: (i) luminosity and temperature of the optical star; (ii) orbital separation and inclination; (iii) distance to the system; (iv) size of the region occupied by the pulsar wind; (v) pulsar wind bulk Lorentz factor.471 Remarkably. the recent optical observations of have significantly revised the parameters (1). Gi) and (1i) in favor of higher temperature and luminosity. smaller inclination angle and further svstem location.," Remarkably, the recent optical observations of have significantly revised the parameters (i), (ii) and (iii) in favor of higher temperature and luminosity, smaller inclination angle and further system location."472 Regarding the wind bulk Lorentz factor. it remains a hiehlv uncertain parameter.," Regarding the wind bulk Lorentz factor, it remains a highly uncertain parameter."473 The size of the region. where (he pulsar wind can propagate depends on the ratio of the wind ram pressures. 77) (Dogovalovetal.2003).," The size of the region, where the pulsar wind can propagate depends on the ratio of the wind ram pressures, $\eta$ \citep{bogovalov08}."474". Since the pulsar spindown luminosity is known. one can estimate (he ram pressure of (he pulsar wind where L, and + are the spindown Iuminositv of the pulsar aud distance to (he pulsar. respectively."," Since the pulsar spindown luminosity is known, one can estimate the ram pressure of the pulsar wind where $L_{\rm sd}$ and $r$ are the spindown luminosity of the pulsar and distance to the pulsar, respectively."475 It should be noted that (his relation ienores (he possible effects related to the anisotropy of the pulsar wind., It should be noted that this relation ignores the possible effects related to the anisotropy of the pulsar wind.476 Although. the level of the anisotropy maw be quite high ab large distances lrom (he pulsar. e.g. in the case of the Crab-like pulsars 2002).. in this paper we limit our consideration by the case of an isotropic wind.," Although, the level of the anisotropy may be quite high at large distances from the pulsar, e.g. in the case of the Crab-like pulsars \citep{bogovalov02}, in this paper we limit our consideration by the case of an isotropic wind."477 To obtain the ram pressure of the stellar wind. one needs detailed information about the properties of the optical star. including the mass-loss rate and wind velocity profile. which are currently not firmly established.," To obtain the ram pressure of the stellar wind, one needs detailed information about the properties of the optical star, including the mass-loss rate and wind velocity profile, which are currently not firmly established."478 For the given optical star luminosity. the mass-loss rate can be estimated at the level of AJ—6x10M.vr.| (Vinketal.2000).," For the given optical star luminosity, the mass-loss rate can be estimated at the level of $\dot{M}=6\times10^{-8} M_\sun \rm yr^{-1}$ \citep{vink00}."479".. Accounting for the wind velocity at interaction point V,<V4=13504200kms! (MeCollum1993). 1 is possible to estimate the expected value of the ratio of the momentum [flux densities."," Accounting for the wind velocity at interaction point $V_{\rm w}<V_\infty=1350\pm200\,\rm km\, s^{-1}$ \citep{mccollum93}, it is possible to estimate the expected value of the ratio of the momentum flux densities,"480polynomials and rational functions (see e.g.. Press et 11992).,"polynomials and rational functions (see e.g., Press et 1992)."481 The fact that the minimax condition is satistied for our MGE fits shows that our algorithm is able to converge to the minimax MGE approximation of a power-law., The fact that the minimax condition is satisfied for our MGE fits shows that our algorithm is able to converge to the minimax MGE approximation of a power-law.482 The same minimax condition can be veritied in the case of MGE tits to more realistic profiles composed of multiple power-law regimes (Pig. 2).," The same minimax condition can be verified in the case of MGE fits to more realistic profiles composed of multiple power-law regimes (Fig. \ref{fig:mge_fit_1d_m32}) ),"483 and essentially for any 'uncetion that appears smooth in a log-log plot., and essentially for any function that appears smooth in a log-log plot.484 Here the Gaussians will not be precisely logarithmically spaced in the a). but will be more closely spaced where the profile is steeper (Fig. 39).," Here the Gaussians will not be precisely logarithmically spaced in the $\sigma_j$, but will be more closely spaced where the profile is steeper (Fig. \ref{fig:mge_fit_1d}) )."485 Also in hese cases however the least-squares solution essentially provides he minimax MGE approximation of the given function., Also in these cases however the least-squares solution essentially provides the minimax MGE approximation of the given function.486 We emphasize the fact that the logarithmic sampling of the orofiles is a necessary condition for the least-squares solution to coincide. within the numerical approximations. with the minimax solution for the power-law.," We emphasize the fact that the logarithmic sampling of the profiles is a necessary condition for the least-squares solution to coincide, within the numerical approximations, with the minimax solution for the power-law."487 A linear sampling would have produced an MGE fit with smaller errors at large radii and larger errors owards the centre., A linear sampling would have produced an MGE fit with smaller errors at large radii and larger errors towards the centre.488 With real photometrie data the above condition. for the minimax solution cannot be always verified. due to the noise and o the fact that the observed profile is known at a finite number of points.," With real photometric data the above condition for the minimax solution cannot be always verified, due to the noise and to the fact that the observed profile is known at a finite number of points."489 The convergence to a global solution however can still be recognized. by simple visual inspection of the result. from the ‘act that every Gaussian has to give a significant contribution to a specific part of the profile: a Gaussian that remains always lower han any other Gaussian is generally an indication that the solution is not the global minimum (unless the profile becomes very flat).," The convergence to a global solution however can still be recognized, by simple visual inspection of the result, from the fact that every Gaussian has to give a significant contribution to a specific part of the profile: a Gaussian that remains always lower than any other Gaussian is generally an indication that the solution is not the global minimum (unless the profile becomes very flat)."490 To determine the fits of Fig. l-, To determine the fits of Fig. \ref{fig:mge_fit_1d_powerlaw}-491—3. we made use of the important fact that equation (1)) 1s in the Gaussian total luminosity £;., \ref{fig:mge_fit_1d} we made use of the important fact that equation \ref{eq:surf_twist}) ) is in the Gaussian total luminosity $L_j$.492 Since a global solution for the linear variables £; can be found easily for any given combination of the nonlinear ones oj. it is crucial to separate the treatment of these two sets of variables.," Since a global solution for the linear variables $L_j$ can be found easily for any given combination of the nonlinear ones $\sigma_j$, it is crucial to separate the treatment of these two sets of variables."493 In practice our ID MGE fitting algorithm consists of he following steps: In case a solution with negative Gaussians is needed (e.g. to reproduce a profile that is monotonically decreasing with radius) the solution of the linear system with NNLS. in step (iii) above. must be replaced by a solution using the Singular Value Decomposition (SVD) to provide a zero order mminimum," In practice our 1D MGE fitting algorithm consists of the following steps: In case a solution with negative Gaussians is needed (e.g., to reproduce a profile that is monotonically decreasing with radius) the solution of the linear system with NNLS, in step (iii) above, must be replaced by a solution using the Singular Value Decomposition (SVD) to provide a zero order minimum"494of these GCs contaimed the most Iuniuous N-rav sources (?)..,of these GCs contained the most luminous X-ray sources \citep{silk75}. .495 7?— later confirmed that GCs hosting ταν sources 21076 ore &| were siguificauth deuser., \citet{bellazzini95} later confirmed that GCs hosting X-ray sources $>$ $^{36}$ erg $^{-1}$ were significantly denser.496" They found that IRohnuogorov-Siürnov testing eave a probability that the XN-rav bright Calactic GCs were draw from a differeut deusitv distribution than those GCs without N-ravs: sinularly, the probability that the M31. X-ray GCs were significantly: deuser that the noun-N-rav GCs was98."," They found that Kolmogorov-Smirnov testing gave a probability that the X-ray bright Galactic GCs were drawn from a different density distribution than those GCs without X-rays; similarly, the probability that the M31 X-ray GCs were significantly denser that the non-X-ray GCs was."4978%... Since the N-rav. bright Galactic CCS are mostly in the dise. it was necessary fo demonstrate that the metalicitv was a driver of the XN-rav huuinositv. rather than a bvproduct of their locations.," Since the X-ray bright Galactic GCs are mostly in the disc, it was necessary to demonstrate that the metalicity was a driver of the X-ray luminosity, rather than a byproduct of their locations."498 ? found a simular metalicity Chhancement for N-rav bright CCs in AI31. over nou-N-rav GCs iu the same regiou of the skx. hence thev showed that inetalicitv. rather than location. was imuportaut.," \citet{bellazzini95} found a similar metalicity enhancement for X-ray bright GCs in M31, over non-X-ray GCs in the same region of the sky, hence they showed that metalicity, rather than location, was important."499 Receutly 7.— conducted a mmchli-expauded survey of ND GCs. associating N-rav sources with Ll out of [16 old clusters.," Recently, \citet{peacock10b} conducted a much-expanded survey of M31 GCs, associating X-ray sources with 41 out of 416 old clusters."500 Their K-S tests showed a probability that GCs hosting NBs are redder than the eeucral population. and a probability that they are more etal rich than the ecuecral population.," Their K-S tests showed a probability that GCs hosting XBs are redder than the general population, and a probability that they are more metal rich than the general population."501 ? note that the metallicity uncertaitics are large. and likely weaken aux. genuine relationships between the clusters.," \citet{peacock10b} note that the metallicity uncertainties are large, and likely weaken any genuine relationships between the clusters."502 When ?— surveved the N-rayv sources in the eiut elliptical galaxv NGC £172. they found that  of the bright N-ray sources were associated with GCs. a significauth hieher fraction than or the Milkv. Was.," When \citet{kundu02} surveyed the X-ray sources in the giant elliptical galaxy NGC 4472, they found that $\sim$ of the bright X-ray sources were associated with GCs, a significantly higher fraction than for the Milky Way."503 Tlowever. the fraction of X-rav bright GCs was ~ remarkably similar o the fractious observed iu the Milkv. Wax. aud other sealaxies studied: they therefore concluded hat the formation of X-ray binaries in CCS iust ve Influenced more by the GC properties than he galaxy.," However, the fraction of X-ray bright GCs was $\sim$, remarkably similar to the fractions observed in the Milky Way, and other galaxies studied; they therefore concluded that the formation of X-ray binaries in GCs must be influenced more by the GC properties than the galaxy."504 Thev also used the strongly bimodal uetalicity distribution to show that X-ray binaries were 3 ties more miuuerous in the metal-rich CC yopulation than the metal-poor GC., They also used the strongly bimodal metalicity distribution to show that X-ray binaries were 3 times more numerous in the metal-rich GC population than the metal-poor GCs.505 The most secure method of ideuntifius black hole binaries requires the mass function to be derived from. the radial velocity. curve of the conrpauion star., The most secure method of identifying black hole binaries requires the mass function to be derived from the radial velocity curve of the companion star.506 Towever. this method is uufeasible for N-rav binaries in elobular clusters: these systems are iuost likely to be located iu the cluster center. making ideutifviug the counterpart nupossible with current instrumentation.," However, this method is unfeasible for X-ray binaries in globular clusters; these systems are most likely to be located in the cluster center, making identifying the counterpart impossible with current instrumentation."507 IIence alternative indicators are needed to identify CC X-ray sources containing black holes., Hence alternative indicators are needed to identify GC X-ray sources containing black holes.508 Iu sunnarx. our method for ideutifving black holes iu extragalactic X-ray binaries involves searching for high Iuninosity Coniptouized emission. conumion to neutron star and black hole binarics at low accretion rates (2): auv thermal component is too faint to observe in extragalactic N-rav binarics.," In summary, our method for identifying black holes in extragalactic X-ray binaries involves searching for high luminosity Comptonized emission, common to neutron star and black hole binaries at low accretion rates \citep{vdk94}; any thermal component is too faint to observe in extragalactic X-ray binaries."509 At higher accretion rates. black hole biuaries can exhibit a high/soft state (290% thermal Cluission). or a two-component steep power law state cousisting of thermal aud Coniptouized colmpoucuts (soee.g.7.andreferenceswithin).," At higher accretion rates, black hole binaries can exhibit a high/soft state $\ga$ thermal emission), or a two-component steep power law state consisting of thermal and Comptonized components \citep[see e.g.][ and references within]{mr06}."510 The various neutron star binary behaviors at hieher huuinosities all include a strong thermal contribution to the emission (seeeg.22?)," The various neutron star binary behaviors at higher luminosities all include a strong thermal contribution to the emission \citep[see e.g.][]{white86,hasinger89,vdk94}."511 A colupreheusive survey of Galactic neutron star rav binaries by? revealed that no neutron star rav binary exhibited low state behaviour at 0.011000 keV bhuninosities 210% of the Eddington luüt (Lg): for à 2 M. neutron star. 0.1 Lg M24 107 ores +.," A comprehensive survey of Galactic neutron star X-ray binaries by \citet{glad07} revealed that no neutron star X-ray binary exhibited low state behaviour at 0.01--1000 keV luminosities $\ga$ of the Eddington limit $L_{\rm Edd}$ ); for a 2 $_{\odot}$ neutron star, 0.1 $L_{\rm Edd}$ $\sim$ $\times 10^{37}$ erg $^{-1}$."512" THeuce. if we see Comptonized cnussion (represeuted by a power law with slope ο2) and no significant thermal emission at luninosities siguificautlv hieher than 10°"" erg st. then the aceretor is a candidate black hole."," Hence, if we see Comptonized emission (represented by a power law with slope $\sim$ 1.4–2) and no significant thermal emission at luminosities significantly higher than $\times$ $^{37}$ erg $^{-1}$ , then the accretor is a candidate black hole."513 We note that our criterion is similar iu principle to that of ?.. who found that oulv black hole binarics can produce 20200 keV. huuinosities exceeding ~L5<10% eres !.," We note that our criterion is similar in principle to that of \citet{barret00}, who found that only black hole binaries can produce 20–200 keV luminosities exceeding $\sim$ $\times10^{37}$ erg $^{-1}$ ."514 7? found the first such black hole candidate iu he M31 GC Do 15. named following the Revised Bologna Catalogue Vil (2???.hereafterRBC)..," \citet{barnard08} found the first such black hole candidate in the M31 GC Bo 45, named following the Revised Bologna Catalogue V4 \citep[ hereafter RBC]{galleti04,galleti06,galleti07,galleti09}."515" The X-ray source associated with Bo I5 exhibited ow state behaviour at a 0.310 keV Lbhuuinositv of ~2«107 ere 1, 6 times brighter than the ow state buuinositv threshold for neutron stars: it appears to have been porsisteutlv bright for he last ~30 vears. and is cousistent with the xedietious of ? for a binary formed by tidal capture."," The X-ray source associated with Bo 45 exhibited low state behaviour at a 0.3–10 keV luminosity of $\sim 2\times 10^{38}$ erg $^{-1}$, 6 times brighter than the low state luminosity threshold for neutron stars; it appears to have been persistently bright for the last $\sim$ 30 years, and is consistent with the predictions of \citet{kalogera04} for a binary formed by tidal capture."516 7.— found a second black hole cancidate associated with the M31 GC Bo τις this also appears to be persistently bright., \citet{barnard09} found a second black hole candidate associated with the M31 GC Bo 144; this also appears to be persistently bright.517It haslong been kuown thatthe Iuuiuosities of state transitions vary considerably between X- binarics.and even between outburst of a single system furthermore. transicuts exhibit lavsteresis,"It haslong been known thatthe luminosities of state transitions vary considerably between X-ray binaries,and even between outburst of a single system; furthermore, transients exhibit hysteresis"518In all cases. P increases with mi.,"In all cases, $P$ increases with $\dot m$."519 This is because the electron. number density anc hence. scattering. probability increases with ri.," This is because the electron number density and hence, scattering probability increases with $\dot m$."520 Indeed. in all the low m cases. the average number of scatterings is l. whereas photons scatter on average more than once in all the high m cases (see Fig. 7)).," Indeed, in all the low $\dot m$ cases, the average number of scatterings is $\leq 1$, whereas photons scatter on average more than once in all the high $\dot m$ cases (see Fig. \ref{scatnum}) )."521 The scattering probability is also slightly. higher for larger Hír.. producing higher { for these column. geometries.," The scattering probability is also slightly higher for larger $H/r_{\rm c}$ , producing higher $P$ for these column geometries."522 Figs., Figs.523 2 and 3 also reveal that /? increases with 7., \ref{M1} and \ref{M05} also reveal that $P$ increases with $i$.524 This trend arises because photons in the initial seed distribution at the column base escape quickly with none or very Low scatterings across the radius of the column (7z90 irection). leaving behind a photon distribution that becomes. increasingly more anisotropic (see Fig. 4)).," This trend arises because photons in the initial seed distribution at the column base escape quickly with none or very few scatterings across the radius of the column $i\approx90\degree$ direction), leaving behind a photon distribution that becomes increasingly more anisotropic (see Fig. \ref{photonprop}) )."525 Photons that now escape in directions 7zz90 must do so through large-angle scatterings which result in a high net 2., Photons that now escape in directions $i \approx 90\degree$ must do so through large-angle scatterings which result in a high net $P$.526 Conversely. photons do not need to undergo large-angle scatterings in order to escape in directions /20. hence 2? is lowest for these viewing angles.," Conversely, photons do not need to undergo large-angle scatterings in order to escape in directions $i \approx 0$, hence $P$ is lowest for these viewing angles."527 The steep rise in 2 towards @z907. reaching values up to in the high rh cases. dilfers considerably from the results of Matt.(2004)... which show a gradual rise in Z with { towards a maximum. of at ¢z907.," The steep rise in $P$ towards $i\approx90\degree$, reaching values up to in the high $\dot m$ cases, differs considerably from the results of \citet{Matt04}, which show a gradual rise in $P$ with $i$ towards a maximum of at $i\approx 90\degree$."528 This cdillerence can be attributed to the assumption of a uniform column in Alatt’s model., This difference can be attributed to the assumption of a uniform column in Matt's model.529 To demostrate this dillerence. in Fig.," To demostrate this difference, in Fig."530 5. we plot P vs. £ for a uniform column with the same parameters as those used for Fig., \ref{uniform} we plot $P$ vs. $i$ for a uniform column with the same parameters as those used for Fig.531 3bb. In this case. the photons are emitted uniformly throughout the column.," \ref{M05}b b. In this case, the photons are emitted uniformly throughout the column."532 The resulting 7? at ie907 ds substantially lower than that. predicted for the nonuniform column (where the photons are emitted at the base of the column)., The resulting $P$ at $i\approx 90\degree$ is substantially lower than that predicted for the nonuniform column (where the photons are emitted at the base of the column).533 Εις is because the photon distribution remains quasi-isotropic throughout the uniform column and thus. fewer photons undergo large angle scatterings to escape in the direction ὁzz90.," This is because the photon distribution remains quasi-isotropic throughout the uniform column and thus, fewer photons undergo large angle scatterings to escape in the direction $i\approx 90\degree$."534 Fig., Fig.535 7. shows the average number of scatterings each photon undergoes before. escaping the column at a particular ἐς, \ref{scatnum} shows the average number of scatterings each photon undergoes before escaping the column at a particular $i$.536 In the case of a uniform. column the average number of scatterings is lower than the uniform case for the the same m., In the case of a uniform column the average number of scatterings is lower than the non-uniform case for the the same $\dot m$.537 We also investigate the clleet of evelotron cooling dominated acerction Lows on polarization., We also investigate the effect of cyclotron cooling dominated accretion flows on polarization.538 Fig., Fig.539" 6 compares the polarization degrees for an mCV. with ApΞ05M. m=lWeem7s| and fr=5 when evelotron cooling at the shock is negligible (ο=0. dotted curve) and when it dominates (c,=10. solid curve)."," \ref{cyclotron} compares the polarization degrees for an mCV with $\Mwd = 0.5 \, \Msun$, $\dot m = 10 \, \gcms$ and $H/r_{\rm c} = 5$ when cyclotron cooling at the shock is negligible $\epsilon_{\rm s} = 0$, dotted curve) and when it dominates $\epsilon_{\rm s} = 10$, solid curve)."540 Interestingly. here is not much cillerence between these two cases.," Interestingly, there is not much difference between these two cases."541 This can be understood. as follows., This can be understood as follows.542 Ehe presence of evelotron cooling makes the accretion column. more compact. and he density enhancement increases the average number of Whoton scatterings (Fig. 7))., The presence of cyclotron cooling makes the accretion column more compact and the density enhancement increases the average number of photon scatterings (Fig. \ref{scatnum}) ).543 However. for a sullicientlv igh accretion rate. the column will be optically thick regardless of whether bremsstrahlung cooling or evclotron cooling dominates at the shock.," However, for a sufficiently high accretion rate, the column will be optically thick regardless of whether bremsstrahlung cooling or cyclotron cooling dominates at the shock."544 For multiple scatterings. he resulting polarization degree is. determined. largely ov the final scattering.," For multiple scatterings, the resulting polarization degree is determined largely by the final scattering."545 Generally in evelotron dominated lows. the last scattering surface will be in the cense emsstrahlung cooling zone close to the base of the column. where the physical conditions are very similar to the xemsstrahlung cooling dominated case.," Generally in cyclotron dominated flows, the last scattering surface will be in the dense bremsstrahlung cooling zone close to the base of the column, where the physical conditions are very similar to the bremsstrahlung cooling dominated case."546 In both cases. the scattering electrons are relatively cool with low velocities. and. as a result aberration is negligible and Compton recoil will dominate.," In both cases, the scattering electrons are relatively cool with low velocities, and as a result aberration is negligible and Compton recoil will dominate."547 With three or four scatterings beforehand. incident photon directions for a subsequent final scattering will be approximately randomized and. hence unpolarized.," With three or four scatterings beforehand, incident photon directions for a subsequent final scattering will be approximately randomized and hence unpolarized."548 Thus. the average polarization 2 of photons emerging from the last scattering surface depends only on the scattering angle (and hence the viewing inclination) and is insensitive to evelotron cooling.," Thus, the average polarization $P$ of photons emerging from the last scattering surface depends only on the scattering angle (and hence the viewing inclination) and is insensitive to cyclotron cooling."549 The situation may be clifferent for lower. accretion rates., The situation may be different for lower accretion rates.550 If the post-shock region becomes semi-transparent to scattering. photons can emerge from the entire structure of the post-shock Dow.," If the post-shock region becomes semi-transparent to scattering, photons can emerge from the entire structure of the post-shock flow."551 In that case. evcelotron cooling. and hy implication the white-dwarf magnetic field which. mocifies the flow. can allect the polarization.," In that case, cyclotron cooling, and by implication the white-dwarf magnetic field which modifies the flow, can affect the polarization."552 We may thus conclude that polarization is a robust cliagnostic of geometry and orientation of svstems if they have sulliciently high accretion rates., We may thus conclude that polarization is a robust diagnostic of geometry and orientation of systems if they have sufficiently high accretion rates.553 Finally. we show an application to an mC'V. with high accretion rate and low magnetic field.," Finally, we show an application to an mCV with high accretion rate and low magnetic field."554 We construct a model accretion column using mC'V. parameters similar to that of the intermediate polar Gly Per (Fig. 8):, We construct a model accretion column using mCV parameters similar to that of the intermediate polar GK Per (Fig. \ref{GKPer}) ):555 Mwp=0.63AL: and m=I0gcm7s+ (MoralesRuedactal.2002:Vriel-mannetal. 2005).," $\Mwd = 0.63 \, \Msun$ and $\dot m = 10 \, \gcms$ \citep{Morales02, Vrielmann05}."556. The predicted. fractional polarization increases towards d$=0.33 where P peaks at zz0.075., The predicted fractional polarization increases towards $\Phi = 0.33$ where $P$ peaks at $\approx 0.075$.557 Whether the predicted 2 and its phase-dependent variation could be observed would depend upon. among other things. the extent of dilution by background. emission.," Whether the predicted $P$ and its phase-dependent variation could be observed would depend upon, among other things, the extent of dilution by background emission."558 Note that the predicted. P for Glx Per type mCVs would be higher as the accretion [low resembles a curtain rather than a cvlinder., Note that the predicted $P$ for GK Per type mCVs would be higher as the accretion flow resembles a curtain rather than a cylinder.559 Also. X-ravs reflected. by dense material further upstream in the accretion disk and the disk magnetosphere coupling region will contribute to the observed. polarization.," Also, X-rays reflected by dense material further upstream in the accretion disk and the disk magnetosphere coupling region will contribute to the observed polarization."560 Note that for very strong field svstems with low accretion rates (e.g. polars in an intermediate accretion state). an anisotropic temperature distribution due to thermal decoupling between charged particles at the shock (seeSax-ton2005) may complicate the situation by imprinting an intrinsic polarization on the seed bremsstrahlung radiation. (sceWarner1995.forareviewofpolarsand.intermediate polars)..," Note that for very strong field systems with low accretion rates (e.g. polars in an intermediate accretion state), an anisotropic temperature distribution due to thermal decoupling between charged particles at the shock \citep[see][]{Saxton05} may complicate the situation by imprinting an intrinsic polarization on the seed bremsstrahlung radiation. \citep[see][for a review of polars and intermediate polars]{Warner95}. ."561 Also. an additional polarized Compton component resulting from rellection olf the white dwarf surface. can contribute to the overall spectrum. above a few keV. and could. introduce an energy. dependence on the polarization (Matt2004).," Also, an additional polarized Compton component resulting from reflection off the white dwarf surface can contribute to the overall spectrum above a few keV and could introduce an energy dependence on the polarization \citep{Matt04}."562.. We have investigated the properties of Compton polarized X-rays in the accretion column of mCVs using Monte Carlo simulations., We have investigated the properties of Compton polarized X-rays in the accretion column of mCVs using Monte Carlo simulations.563 The calculations take into account a post-shock region stratified in temperature. density and. velocity.," The calculations take into account a post-shock region stratified in temperature, density and velocity."564 The degree of linear. polarization for scattered X-rays was calculated For a range of different column geometries. white cwarl masses and accretion rates.," The degree of linear polarization for scattered X-rays was calculated for a range of different column geometries, white dwarf masses and accretion rates."565 We have found that the resulting polarization is sensitive to the density structure in addition to the viewing geometry., We have found that the resulting polarization is sensitive to the density structure in addition to the viewing geometry.566 The non-uniform density structure in the post-shock column has a significant elfect on the photon distribution and average number of scatterings., The non-uniform density structure in the post-shock column has a significant effect on the photon distribution and average number of scatterings.567 We have demonstrated that enhanced. emissivity near the base of the column results in photon distributions that become increasingly more anisotropic throughout the column length so that N-ravs escaping at largeangles with respect to the column axis can clo so only through increasinglylarger angle scatterings., We have demonstrated that enhanced emissivity near the base of the column results in photon distributions that become increasingly more anisotropic throughout the column length so that X-rays escaping at largeangles with respect to the column axis can do so only through increasinglylarger angle scatterings.568 This produces more strongly. polarized. N-ravs, This produces more strongly polarized X-rays569where the transition frou high to low state occurs at a lower Iuninositv than the transition from low to high state (2?)..,"where the transition from high to low state occurs at a lower luminosity than the transition from low to high state \citep{miyamoto95, maccarone03}. ."570 7. recently conducted a survey of state transitions in Galactic X-ray binaries by colmparing their fluxes iu the the RNTE/ÀASM aud Switt/BAT: a BAT to ASM fiux ratio 21.0 indicated a low/hard state. while a ratio πι» indicated a high/soft state.," \citet{tang10} recently conducted a survey of state transitions in Galactic X-ray binaries by comparing their fluxes in the the RXTE/ASM and Swift/BAT; a BAT to ASM flux ratio $\ga$ 1.0 indicated a low/hard state, while a ratio $\la$ 0.2 indicated a high/soft state."571 They identified 1258 hard to soft transitions in 28 svsteimis. consisting of 20 neutron star low mass N-rav binaries ΕΕ. T black hole LAENDs aud 1 high mass Nav binary.," They identified 128 hard to soft transitions in 28 systems, consisting of 20 neutron star low mass X-ray binaries (LMXBs), 7 black hole LMXBs and 1 high mass X-ray binary."572 They found that the transition luuinosity was stronely correlated with the peak huninmositv of he soft state. sugeesting that the buuinositv of rausition from lard to soft state is governed bv ion-stationary parameters. such as the accretion rate historv. rather than the accretion rate itself: rowever all transitions occurred at 1550 keV uninosities —0.0010.1 τα (1.0 eve + in the 1550 keV baud = 1.3 ere + in the 0.3.10 keV vad assunmdue power law enüsson with photon index 1.7).," They found that the transition luminosity was strongly correlated with the peak luminosity of the soft state, suggesting that the luminosity of transition from hard to soft state is governed by non-stationary parameters, such as the accretion rate history, rather than the accretion rate itself; however all transitions occurred at 15–50 keV luminosities $\sim$ 0.001–0.1 $L_{\rm Edd}$ (1.0 erg $^{-1}$ in the 15–50 keV band = 1.3 erg $^{-1}$ in the 0.3–10 keV band assuming power law emission with photon index 1.7)."573 We have couducted a variability study for LO X-ray sources associated with objects in the RBC. using 123 Chandra observatious (Barnard et al..," We have conducted a variability study for 40 X-ray sources associated with objects in the RBC, using 123 Chandra observations (Barnard et al.,"574 in xep): these include 31 confirmed globular clusters (GCs). E candidate GCs. 2 stars. 2 galaxies aud a III region.," in prep); these include 31 confirmed globular clusters (GCs), 4 candidate GCs, 2 stars, 2 galaxies and a HII region."575 We produced long term leltcurves Toni these data. and used them to identity yosstble DIICs (BIICs). looking for hard. non-hermal spectra at high luminosities. as in 7. and 7...," We produced long term lightcurves from these data, and used them to identify possible BHCs (BHCs), looking for hard, non-thermal spectra at high luminosities, as in \citet{barnard08} and \citet{barnard09}."576 We used archival NMNI-Newton observatious o obtain hieh quality spectra of these sources where possible., We used archival XMM-Newton observations to obtain high quality spectra of these sources where possible.577 We have identified | new DIIC* associated with the NI elobular clusters Bo 82 (XBos2). Bo 153 (XB153). Bo 163 (XD163) aud Bo 155 (NBIsS5).," We have identified 4 new BHCs associated with the M31 globular clusters Bo 82 (XB082), Bo 153 (XB153), Bo 163 (XB163) and Bo 185 (XB185)."578 The N-rav source associated with the M31 GC Bo 163 (hereafter referred to as XND163) is a c-iown recurrent transient., The X-ray source associated with the M31 GC Bo 163 (hereafter referred to as XB163) is a known recurrent transient.579 ?— found it bright in one Chaudra. and several ROSAT observations. but found uo detection iu anv other Chaudra observation. nor iu auv of the four NMNE-Neowtou observations available at the time.," \citet{trudolyubov04} found it bright in one Chandra, and several ROSAT observations, but found no detection in any other Chandra observation, nor in any of the four XMM-Newton observations available at the time."580" Asstuineg au absorbed power law emission model. with photou dex 1.7 and absorption equivalent to 4107?"" IT atom 7. they obtained Iuminosities 107"" LO cre st,"," Assuming an absorbed power law emission model, with photon index 1.7 and absorption equivalent to $\times 10^{20}$ H atom $^{-2}$, they obtained luminosities $\sim$ $^{35}$ $^{38}$ erg $^{-1}$."581 Tn Section 2. we discuss the observations aud data analysis: next. we present our results iu Section 3.. imcludiug a long-term lighteurves aud analysis of the best available spectra: fnallv. we discuss our findings in Section [..," In Section \ref{obs} we discuss the observations and data analysis; next, we present our results in Section \ref{res}, including a long-term lightcurves and analysis of the best available spectra; finally, we discuss our findings in Section \ref{dis}."582 The central region of M31 has been observed with Chandra on a ~ inouthlv basis for the last —11 wears in order to monitor frausieuts., The central region of M31 has been observed with Chandra on a $\sim$ monthly basis for the last $\sim$ 11 years in order to monitor transients.583 We have analyzed 78 ACTS observations and 15 IIBC observations. in order to discern the variability of N-rav sources associated with CC's in this region.," We have analyzed 78 ACIS observations and 45 HRC observations, in order to discern the variability of X-ray sources associated with GCs in this region."584 We determined the position of cach source from a inerged ACTS image. using the oolIMCENTROLD: locations of the L tarects arc xovided iu our full survey (Barnard et al..," We determined the position of each source from a merged ACIS image, using the tool; locations of the 4 targets are provided in our full survey (Barnard et al.,"585 iu rep)., in prep).586 For cach Chaudra observation. we obtained source and background lighteurves aud spectra iu he 0.5-7.0 keV. band from circular regions: the vackerouud region was the same size as the source reeion. and at a simular offaxis angle.," For each Chandra observation, we obtained source and background lightcurves and spectra in the 0.3-7.0 keV band from circular regions; the background region was the same size as the source region, and at a similar off-axis angle."587" The PSF at cach source was somewhat spread. due to their off-axis angeles: we used extraction radii of 10"" for XDB153 aud XDBls5. 15"" for ND1623 aud 20"" for NBos?2."," The PSF at each source was somewhat spread, due to their off-axis angles; we used extraction radii of $''$ for XB153 and XB185, $''$ for XB163 and $''$ for XB082."588 For ACTS observations. response matrix and ancillary respouse fles were mace.," For ACIS observations, response matrix and ancillary response files were made."589" We initially estimated the conversion from flux to Iuuinositv by asstuuine a power law ciission spectrum with photon iudex⋅ 1.7. ↽−⋅with My- = ---«1079 atom cu1. then determining the unabsorbed 0.3 keV luninosity equivalent to 1 count |,"," We initially estimated the conversion from flux to luminosity by assuming a power law emission spectrum with photon index 1.7, with $N_{\rm H}$ = $\times 10^{20}$ atom $^{-1}$, then determining the unabsorbed 0.3--10 keV luminosity equivalent to 1 count $^{-1}$."590 After correcting for the exposure. vignetting and backeround. imultiphàus the source iuteusitv by this conversion factor gave the source Iuninosity.," After correcting for the exposure, vignetting and background, multiplying the source intensity by this conversion factor gave the source luminosity."591 Source spectra with 2200 uet counts were frecly fitted., Source spectra with $>$ 200 net counts were freely fitted.592 For IIRC observations. we used only eveuts with PI 15293 to reduce the iustimucntal backerouncl.," For HRC observations, we used only events with PI 48–293 to reduce the instrumental background."593 We used the WebPIAINIS tool to find the unabsorbed lunimositv equivalent to 1 count assunine the same cussion model as for the faint ACTS observations.," We used the WebPIMMS tool to find the unabsorbed luminosity equivalent to 1 count $^{-1}$, assuming the same emission model as for the faint ACIS observations."594 We created a 1 keV exposure map for cach observation. aud. compared the exposure within the source region with that of au identical on-axis region. in order to estimate the necessary exposure correction.," We created a 1 keV exposure map for each observation, and compared the exposure within the source region with that of an identical on-axis region, in order to estimate the necessary exposure correction."595 We multiplied the backerouud subtracted. corrected source iuteusitv by theconversion to ect the 0.3.10 keV. luminosity.," We multiplied the background subtracted, corrected source intensity by theconversion to get the 0.3–10 keV luminosity."5960.2F(100—300keV) (=4 per cent of Εποι).,"$300\,{\rm keV})$ $\simeq 4$ per cent of $F_{\rm bol}$ )."597" These estimates are valid for the hard state, in which the spectra fall off exponentially at high energies."," These estimates are valid for the hard state, in which the spectra fall off exponentially at high energies."598" In soft states, the spectra continue at high energies as power laws (see, e.g., 113 in Z02), so the correction factors would be higher."," In soft states, the spectra continue at high energies as power laws (see, e.g., 13 in Z02), so the correction factors would be higher."599" However, most of the contribution in the soft state is from soft X-rays, so that the underestimate is negligible."," However, most of the contribution in the soft state is from soft X-rays, so that the underestimate is negligible."600 We note that in order to calculate Fyj we could have alternatively fitted the ASM and either BAT or BATSE data with some models including absorption., We note that in order to calculate $F_{\rm bol}$ we could have alternatively fitted the ASM and either BAT or BATSE data with some models including absorption.601 This would be more accurate but would have presented a number of additional problems., This would be more accurate but would have presented a number of additional problems.602" Whereas we can use the relatively simple thermal Comptonization in the hard state, a non-thermal electron component is required in the intermediate and soft states (e.g., Gierlinskietal. 1999)), which would add a few free parameters."," Whereas we can use the relatively simple thermal Comptonization in the hard state, a non-thermal electron component is required in the intermediate and soft states (e.g., \citealt{gierlinski99}) ), which would add a few free parameters."603 The fitting would have to be performed for as many as the ~3000 daily spectra., The fitting would have to be performed for as many as the $\sim$ 3000 daily spectra.604" Taking these issues into account, we have opted for the method based on individual-channel contributions described above."," Taking these issues into account, we have opted for the method based on individual-channel contributions described above."605" We fit the X-ray/X-ray and radio/X-ray data by a power law using the method symmetric in two fitted variables, both subject to uncertainties (Pressetal.1992),, Here Fx is either the flux in one of the X-ray energy intervals or a broad-band X-ray flux, Επροι is either a radio flux or the bolometric flux, and (Fx) is the geometric average of the X-ray flux, which is chosen as the normalization in order to minimize the fit uncertainties."," We fit the X-ray/X-ray and radio/X-ray data by a power law using the method symmetric in two fitted variables, both subject to uncertainties \citep{press92}, Here $F_{\rm X}$ is either the flux in one of the X-ray energy intervals or a broad-band X-ray flux, $F_{\rm R,bol}$ is either a radio flux or the X-ray bolometric flux, and $\langle F_{\rm X}\rangle$ is the geometric average of the X-ray flux, which is chosen as the normalization in order to minimize the fit uncertainties."606" The parameters obtained are Fgbo1,9, which is either the radio or bolometric X-ray flux at (Fx), and the index, p."," The parameters obtained are $F_{\rm R,bol,0}$, which is either the radio or bolometric X-ray flux at $\langle F_{\rm X}\rangle$, and the power-law index, $p$."607 Thelogarithms of Fg»; and Fx are fitted.," Thelogarithms of $F_{\rm R,bol}$ and $F_{\rm X}$ are fitted."608" In the case of radio/X-ray correlations, we fit the data only with Fg>Fmuis; below which the correlations found here break down, probably due to large errors on the radio measurements."," In the case of radio/X-ray correlations, we fit the data only with $F_{\rm R}>F_{\rm R,min}$, below which the correlations found here break down, probably due to large errors on the radio measurements."609 The fit parameters are given in Table 1.., The fit parameters are given in Table \ref{t:fit}.610" We encounter here an important issue of the uncertainties on the fitted parameters, Fgsao and p."," We encounter here an important issue of the uncertainties on the fitted parameters, $F_{\rm R,bol,0}$ and $p$."611" The probability calculated by this method assumes that the spread of the points is solely due to measurement errors whereas it is, as usually the case in astrophysics, due to both the intrinsic variability and the measurement errors."," The probability calculated by this method assumes that the spread of the points is solely due to measurement errors whereas it is, as usually the case in astrophysics, due to both the intrinsic variability and the measurement errors."612" If only the errors are included, the fit probabilities are very close to null, which renders invalid the estimates on the fit uncertainties."," If only the errors are included, the fit probabilities are very close to null, which renders invalid the estimates on the fit uncertainties."613" Therefore, we assume fractional uncertainties, 6,, dy, on all the measurements such that the fit probability becomes «0.5, or equivalently, χ~1."," Therefore, we assume fractional uncertainties, $\delta_{\rm x}$, $\delta_{\rm y}$, on all the measurements such that the fit probability becomes $\simeq$ 0.5, or equivalently, $\chi^2_\nu\simeq 1$."614 This takes into account the actual spread of the data points due to intrinsic variability., This takes into account the actual spread of the data points due to intrinsic variability.615" The ratio between the errors in the x and y directions is taken to be the ratio of the observed spread of the points, 6y/5x=Oy/ox, where oxy is the standard deviation (in the log space) of the x or y points, respectively."," The ratio between the errors in the $x$ and $y$ directions is taken to be the ratio of the observed spread of the points, $\delta_{\rm y}/\delta_{\rm x}=\sigma_{\rm y}/\sigma_{\rm x}$, where $\sigma_{\rm x,y}$ is the standard deviation (in the log space) of the $x$ or $y$ points, respectively."616 Since the measurement errors are typically much smaller than the spread of the points we neglect them and include only the errors as above. (, Since the measurement errors are typically much smaller than the spread of the points we neglect them and include only the errors as above. (617"This method is similar to that used in Merlonietal.2003 except that they assumed ó,= óy.)",This method is similar to that used in \citealt{mhd03} except that they assumed $\delta_{\rm x}=\delta_{\rm y}$ .)618" Under our assumptions, the best fit index corresponds to the slope of the covariance ellipse, p~σγ/σκ."," Under our assumptions, the best fit index corresponds to the slope of the covariance ellipse, $p\simeq \sigma_{\rm y}/\sigma_{\rm x}$."619 The best-fit parameter values depend only weakly on ó as long as their ratio is fixed as above., The best-fit parameter values depend only weakly on $\delta$ as long as their ratio is fixed as above.620" The fractional errors required for y?~1 are in general «0.3 except for fits involving the 150-195 keV energy band, where they are required to be slightly larger."," The fractional errors required for $\chi^2_\nu\simeq 1$ are in general $<0.3$ except for fits involving the 150–195 keV energy band, where they are required to be slightly larger."621" For tight correlations, such as between the X-ray flux in a given energy band, the fitted values of p weakly depend on the ó,/óy ratio, and they are almost the same if, e.g., 6,=dy were assumed."," For tight correlations, such as between the X-ray flux in a given energy band, the fitted values of $p$ weakly depend on the $\delta_{\rm x}/\delta_{\rm y}$ ratio, and they are almost the same if, e.g., $\delta_{\rm x}=\delta_{\rm y}$ were assumed."622" On the other hand, if the spread is large, as in the radio/X- correlations, the fitted values depend sensitively on the adopted ó,/ó,."," On the other hand, if the spread is large, as in the radio/X-ray correlations, the fitted values depend sensitively on the adopted $\delta_{\rm x}/\delta_{\rm y}$ ."623" However, our method above does reproduce the actual slope of the dependence, under the assumption that it is linear in the log space and that all points have equal relative weight."," However, our method above does reproduce the actual slope of the dependence, under the assumption that it is linear in the log space and that all points have equal relative weight."624the cooling phase.,the cooling phase.625 Finally we have adopted a set of a=3.0 and ο=0.05 during the wind phase to reproduce the orbital modulation observed by Matsiunotoetal...(2001).. α—OS and 2=03012 the steady hydrogen shell-buruiug phase. aud a=0.8 and ο=0.20 in the cooling phase.," Finally we have adopted a set of $\alpha=3.0$ and $\beta=0.05$ during the wind phase to reproduce the orbital modulation observed by \citet{mat01}, $\alpha=0.8$ and $\beta=0.30$ in the steady hydrogen shell-burning phase, and $\alpha=0.8$ and $\beta=0.20$ in the cooling phase."626 Tere. we adopt svuuuetric configurations of the accretion disk throughout the 2000 outburst.," Here, we adopt symmetric configurations of the accretion disk throughout the 2000 outburst."627 The decline of the carly phase (f~0 70 days) hardly depends on the lvdrogen couteut. IX. of the white dwarf euvelope. but the cuds of the wind phase aud of the hydrogen shell-buruius phase depend seusitively ou the hydrogen couteut as tabulated in Table 5..," The decline of the early phase $t \sim 0$ —70 days) hardly depends on the hydrogen content, $X$, of the white dwarf envelope, but the ends of the wind phase and of the hydrogen shell-burning phase depend sensitively on the hydrogen content as tabulated in Table \ref{decline_rates}. ."628 Therefore. we have finally determined the hydrogen couteut of N=0.35 and shown calculated light curves in Figure Ls for three different bands of B. V. aud J...," Therefore, we have finally determined the hydrogen content of $X=0.35$ and shown calculated light curves in Figure \ref{vmag_irradmix_ciaql00_m15} for three different bands of $B$ , $V$ , and $I_c$."629 Tere. we have used the apparent distance modulus of GnAL=1L06 for the V. baud.," Here, we have used the apparent distance modulus of $(m-M)_V= 14.06$ for the $V$ band."630" Adopting the color excess of E(BVW)=1.0. we obtain the apparent distauce moduli of GaAL),=1506 and GnM);=12.16 for B aud £L. bands. respectively."," Adopting the color excess of $E(B-V)= 1.0$, we obtain the apparent distance moduli of $(m-M)_B= 15.06$ and $(m-M)_I= 12.46$ for $B$ and $I_c$ bands, respectively."631 Tere. we adopted the relation of E(/..V)/E(DB.V)1.60 (e.g.Rieke&Lebotsky 1985).," Here, we adopted the relation of $E(I-V) / E(B-V)= -1.60$ \citep[e.g.,][]{rie85}."632. Both the B aud IL. light curves are in good agreement with the observational points as shown iu Figure 18.., Both the $B$ and $I_c$ light curves are in good agreement with the observational points as shown in Figure \ref{vmag_irradmix_ciaql00_m15}.633 This indicates the cousisteucy of our estimated values;, This indicates the consistency of our estimated values.634 Various physical values of the 2000 outburst are sunuuuized in Table 6.. Matstuotoetal.(200, Various physical values of the 2000 outburst are summarized in Table \ref{envelope_mass2000}.635"1) reported a sharp ~1 mag drop of R,anagnuitude ou 23 November 2000 and INivota (2001. VSNET archives. http://vsucet.kusastro.kvoto-iLacjp/vsuet/) also observed a ~1.5 mag drop of L-magnitude around the same dav as shown in Figure Ἰδ."," \citet{mat01} reported a sharp $\sim 1$ mag drop of $R_c$ -magnitude on 23 November 2000 and Kiyota (2001, VSNET archives, http://vsnet.kusastro.kyoto-u.ac.jp/vsnet/) also observed a $\sim 1.5$ mag drop of $I_c$ -magnitude around the same day as shown in Figure \ref{vmag_irradmix_ciaql00_m15}."636", If we attribute this drop to the cud of wind plase. the hydrogen content of IX=0.35 is consistent with the drop."," If we attribute this drop to the end of wind phase, the hydrogen content of $X=0.35$ is consistent with the drop."637 We have re-estinated the huuinous supersoft X-rav source phase., We have re-estimated the luminous supersoft X-ray source phase.638 No Iuminous supersoft N-rayvs are expected in the early decline phase because the photospheric radius of the white dwarf is rather large aud the photospheric temperature is relatively low., No luminous supersoft X-rays are expected in the early decline phase because the photospheric radius of the white dwarf is rather large and the photospheric temperature is relatively low.639 After the recurrent nova enters the plateau phase. the photospheric temperature of the white dwarf iucreases enough to enit supersoft N-ravs (kato1999).," After the recurrent nova enters the plateau phase, the photospheric temperature of the white dwarf increases enough to emit supersoft X-rays \citep{kat99}."640. During the massive wiud phase. however. we do not expect supersoft N-ravs because they are solt-absorbed bv the wind itself.," During the massive wind phase, however, we do not expect supersoft X-rays because they are self-absorbed by the wind itself."641 U Sco was observed as a huninous supersoft N-rav source in the mid plateau plase of the 1999 outburst just after the massive wind stopped (ISaliabkactal.1999:Tlachisuet2000a).," U Sco was observed as a luminous supersoft X-ray source in the mid plateau phase of the 1999 outburst just after the massive wind stopped \citep{kah99,hkkm00}."642. Our analvsis sugeests that the massive wind stopped before 25 November 2000., Our analysis suggests that the massive wind stopped before 23 November 2000.643" This 110218 that a Iuninuous supoersoft N-rav phase started from November of 2000 aud continued ""util March of 2001 as shown in Figure 18..", This means that a luminous supersoft X-ray phase started from November of 2000 and continued until March of 2001 as shown in Figure \ref{vmag_irradmix_ciaql00_m15}. .644 The rather low N-ray fluxes in June auc August of 2001 (πιο&DiS-tefano2002) are also very consistent with our analysis.," The rather low X-ray fluxes in June and August of 2001 \citep{gre02}645 are also very consistent with our analysis."646 The apparent distance modulus is obtained to be (1)ντ=LL06 bv fitting for the late phase of the 2000 outburst., The apparent distance modulus is obtained to be $(m-M)_V= 14.06$ by fitting for the late phase of the 2000 outburst.647 The color excess is also estimated to he E(BVV)=1.0 from the difference between the observed color and the caleulated color. and the absorption of Ay=3.1 froma relation of Ay=3.1ECBV) (e.g.Ricke&Lehot 150).," The color excess is also estimated to be $E(B-V)=1.0$ from the difference between the observed color and the calculated color, and the absorption of $A_V=3.1$ from a relation of $A_V = 3.1~E(B-V)$ \citep[e.g.,][]{rie85}."648. Therefore. the distance to CT Aq! is derived to be 1.55 kpc.," Therefore, the distance to CI Aql is derived to be 1.55 kpc."649 The distance iiodulus of (00.AL)p=15.06 ix derived from the distance modulus of GaM):=1106 and the color excess of E(BV)=1.0. which are also consistent with the 5 light curve of the 1917 outburst.," The distance modulus of $(m-M)_B= 15.06$ is derived from the distance modulus of $(m-M)_V= 14.06$ and the color excess of $E(B-V)=1.0$, which are also consistent with the $B$ light curve of the 1917 outburst."650 This set of GnAL)=τιοὐ and LEBV)=1.0 are also in very good aerecineut with the V. B. aud { light curves of the 2000 outburst as shown in Figure 185..," This set of $(m-M)_V= 14.06$ and $E(B-V)=1.0$ are also in very good agreement with the $V$, $B$, and $I_c$ light curves of the 2000 outburst as shown in Figure \ref{vmag_irradmix_ciaql00_m15}."651 The orbital elt curve in August of 2001 shows a sienificant asviunietry., The orbital light curve in August of 2001 shows a significant asymmetry.652 This indicates that the shape of the accretion disk is changing in time at least in the final decay phase of the 2000 outbiurst., This indicates that the shape of the accretion disk is changing in time at least in the final decay phase of the 2000 outburst.653 Assuniug the asvuumetry of Glow=0.7 or 0.75. we have calculated a V. helt curve aud added it to Figures 11-13) (solid liue for August of 2001).," Assuming the asymmetry of $\zeta_{\rm low}= 0.7$ or $0.75$, we have calculated a $V$ light curve and added it to Figures \ref{vmag_quiescence_late_m10}- \ref{vmag_quiescence_late_m20}654 (solid line for August of 2001)."655 Ou the other haud. the disk shape is almost sviuinetric for May of 2001 aud in quiescence.," On the other hand, the disk shape is almost symmetric for May of 2001 and in quiescence."656 It should be noted that the model light curve of August 2001 does uot reproduce the observational lack of any secondary eclipses even if we introduce asviunietry of the accretion disk shape., It should be noted that the model light curve of August 2001 does not reproduce the observational lack of any secondary eclipses even if we introduce asymmetry of the accretion disk shape.657 It may stem from large time variations of the edee of the accretion clisk., It may stem from large time variations of the edge of the accretion disk.658" The envelope mass at the visual ασ ἐς estimated to be AM=8.0«10937. for the lydrogen couteut of X=0.35. indicating the average mass transfer rate of Mac—4.0«10TAL 1 betwoeu the 1917 aud 2000 ο or Mad=13410AF, | between the 1911 and 2000 outbursts."," The envelope mass at the visual maximum is estimated to be $\Delta M_{\rm max}= 8.0 \times 10^{-6} M_\sun$ for the hydrogen content of $X=0.35$, indicating the average mass transfer rate of $\dot M_{\rm acc}= 1.0 \times 10^{-7} 659M_\sun$ $^{-1}$ between the 1917 and 2000 outbursts, or $\dot M_{\rm acc}= 1.3 \times 10^{-7} 660M_\sun$ $^{-1}$ between the 1941 and 2000 outbursts."661" This mass accretion rate is a bit higher but almost consistent with our CI Aql model in quicsccuce (sce Figs, L1--13)).", This mass accretion rate is a bit higher but almost consistent with our CI Aql model in quiescence (see Figs. \ref{vmag_quiescence_late_m10}- \ref{vmag_quiescence_late_m20}) ).662 Ou the other hand. if the mean recurrence period is 20 vrs as sugeested by Schacter(2001c¢).. the average nass rausfer rate increasesto HM~qsd10AM 1d.," On the other hand, if the mean recurrence period is 20 yrs as suggested by \citet{sch01c}, , the average mass transfer rate increasesto $\dot M_{\rm acc} \sim 4 \times 10^{-7} M_\sun$ $^{-1}$."663 This value is high. enough to maintain steady hydrogen shell-nunnueg for hydrogen coutent of X.=0.7 on a 1.2M. WD when the mass transfer is steady;, This value is high enough to maintain steady hydrogen shell-burning for hydrogen content of $X=0.7$ on a $1.2~M_\sun$ WD when the mass transfer is steady.664 In other words. it 15 oo high to be compatible withour thermonuclear runaway uodel. unless the mass transter rate itself rapidly iereased just before the outburst.," In other words, it is too high to be compatible with our thermonuclear runaway model, unless the mass transfer rate itself rapidly increased just before the outburst."665 When the hydrogen content of the ransterred matter is as low as (X.~0.35. the lower init to steady hydrogen shel-burning becomes 8&LOTAL. 4 (κουe.g.eq.(7)inTachisu&Kato20015)..," When the hydrogen content of the transferred matter is as low as $X \sim 0.35$, the lower limit to steady hydrogen shell-burning becomes $8 \times 10^{-7} M_\sun$ $^{-1}$ \citep[see e.g. eq.(7) in][]{hac01kb}."666 Then. the nass accretion rate of AL~{x10SAL. vr3 can still cause shell-Bashes on a 1.2AZ. WD. although the orbital modulation of the light curve in quiesceuce is not consistent with such a lieh mass transfer rate as shown iu Figures 11-12..," Then, the mass accretion rate of $\dot M_{\rm acc} \sim 4 \times 10^{-7} M_\sun$ $^{-1}$ can still cause shell-flashes on a $1.2~M_\sun$ WD, although the orbital modulation of the light curve in quiescence is not consistent with such a high mass transfer rate as shown in Figures \ref{vmag_quiescence_late_m10}- \ref{vmag_quiescence_late_m20}."667" About of the euvelope mass is lost by the wind (Δρα=66s10 SAL.) while the residual (AAR,=Lls10 SAL.) is left and added to the helium laver."," About of the envelope mass is lost by the wind $\Delta M_{\rm wind}= 6.6 \times 10^{-6} M_\sun$ ), while the residual $\Delta M_{\rm He}= 1.4 \times 10^{-6} M_\sun$ ) is left and added to the helium layer."668" The net mass increasing rate of the white dwarf is about Mg,=L3<10SAL. | between the 1917 aud 2000 outburst or Ay.=2.1«10SALt between the 1911aud2000 outburst.", The net mass increasing rate of the white dwarf is about $\dot M_{\rm He}= 1.7 \times 10^{-8} M_\sun$ $^{-1}$ between the 1917 and 2000 outburst or $\dot M_{\rm He}= 2.4 \times 10^{-8} M_\sun$$^{-1}$ between the 1941and2000 outburst.669 These satisty the conditions of Type Ia supernova explosion if the white dwarf consists of carbon aud oxveen (Nomoto&Ixoudo 1991)., These satisfy the conditions of Type Ia supernova explosion if the white dwarf consists of carbon and oxygen \citep{nom91}.670.Evolutionary paths to Type Ia superuovae via recurrent novae have been diseussed ia iore detail iu Tachisuetal. C1999a.b).," Evolutionary paths to Type Ia supernovae via recurrent novae have been discussed in more detail in \citet{hkn99, hknu99}. ."671. We thank the VSNET aembers who observed CT Aql and R. EK. IHoneveutt for his providing us the unpublished data of CT Aql in quiescence., We thank the VSNET members who observed CI Aql and R. K. Honeycutt for his providing us the unpublished data of CI Aql in quiescence.672 This researchhas been supported in part by the Cwant-in-Aid for Scicutific Research (11610226)of the Japan Society for the Promotion of Science., This researchhas been supported in part by the Grant-in-Aid for Scientific Research (11640226)of the Japan Society for the Promotion of Science.673a given value of // and &.,a given value of $h$ and $k$ .674 The [p.q] space analysis is essentially the same. except that he data from Orbit 96 are excluded from the fit for tje reasons described above.," The $[p,q]$ space analysis is essentially the same, except that the data from Orbit 96 are excluded from the fit for the reasons described above."675 The conto in Fig., The contours in Fig.676 6 illustrate how X7 varies with [/.A] aud μα.," \ref{charmpar} illustrate how $\chi^2$ varies with $[h,k]$ and $[p,q]$."677 For 1ie [Ak] plot. the best-fit solution is :jah. ak]=[-18.1.9] kin.," For the $[h,k]$ plot, the best-fit solution is at $[ah,ak] = [ -18,1.9]$ km."678 This woid üunply tha the periapse leads the aui-solar direcion by 6°., This would imply that the periapse leads the anti-solar direction by $^\circ$.679 However. the best-fit solutiοι assuming the pejceuter is exactly aiti-aligued with the Sun is uot obviotsly worse tan the overall best fit.," However, the best-fit solution assuming the pericenter is exactly anti-aligned with the Sun is not obviously worse than the overall best fit."680 Note that even Lo‘these best-fi(ing models. the 4? [it is still quilep ου) (TEfor Leegrees of freedom).," Note that even for these best-fitting models, the $\chi^2$ fit is still quite poor (74 for 4 degrees of freedom)."681 This is consistent. with a visual inspectio1 0f the data.wl ich scatter around tle circle by more tha1 their error bars.," This is consistent with a visual inspection of the data, which scatter around the circle by more than their error bars."682 This excess salter could occur fcla number of reasons., This excess scatter could occur for a number of reasons.683 The data used he'e cole [rom a rauge of pliase augles aid are sensilve to different. parts of the size distribtion. which may lead to clillerences |1 the apparent shape of the rineet.," The data used here come from a range of phase angles and are sensitive to different parts of the size distribution, which may lead to differences in the apparent shape of the ringlet."684 Also. our background sibtractiou aleorithin and olier procedures used to derive the raclial positions of tle riuglet 1iiy have introduced systematic errors between dilfereut. scans.," Also, our background subtraction algorithm and other procedures used to derive the radial positions of the ringlet may have introduced systematic errors between different scans."685 For the [p.q] plot. the best-fittiig mode has [ap.ag)=[0.1.2|| lan.," For the $[p,q]$ plot, the best-fitting model has $[ap,aq]=[0,1.2]$ km."686" He'e the 47> value is"" eood (3.1 for 3 degrees of freedor1).", Here the $\chi^2$ value is good (3.1 for 3 degrees of freedom).687 However. the cliffe'ence in the cLality «X the fit between his and [p.q]=[0.0] is only ma'eually significant (assuninug no fη’ced inclination. the 4? is 9.9 [or 5 degrees of freedom).," However, the difference in the quality of the fit between this and $[p,q]=[0,0]$ is only marginally significant (assuming no forced inclination, the $\chi^2$ is 9.9 for 5 degrees of freedom)."688 Furthermore. siuce the Sun is iu te soulieri. hemisphere. and B. is negative. we expect tl€al he best-fit gy should be negative. uot positive.," Furthermore, since the Sun is in the southern hemisphere, and $B_\Sun$ is negative, we expect that the best-fit $q$ should be negative, not positive."689 Thus the est-fittiug model is a bit of a strprise., Thus the best-fitting model is a bit of a surprise.690 Since B. changes significaiy Ove: the time period covered by these observatious. a LO'e complete model would include a time-variable forcecl inclination.," Since $B_\Sun$ changes significantly over the time period covered by these observations, a more complete model would include a time-variable forced inclination."691 However. given the weak evideuce for any forced inclinajon at all. we close Lol Oo consider sucli complications a UCis time.," However, given the weak evidence for any forced inclination at all, we chose not to consider such complications at this time."692 Despite these uncertainties. we can now explore whether the teriporal evolutio1 of the shape parameters are consistent wil ithe above model. which suggests that the par:unelters should drift around the circles at lealy constant rates determiued by he moctifiecl aud uodal-regressiou rates.," Despite these uncertainties, we can now explore whether the temporal evolution of the shape parameters are consistent with the above model, which suggests that the parameters should drift around the circles at nearly constant rates determined by the modified pericenter-precession and nodal-regression rates."693 Given tlie sparseness of the daa. we caliot estabish easily whether auy givei solution is unique.," Given the sparseness of the data, we cannot establish easily whether any given solution is unique."694 However. oreliminary examination of ala showed that they were aj»proxiuiately consistent with tie expected drift rates /day).," However, preliminary examination of the data showed that they were approximately consistent with the expected drift rates $\dot{\varpi}'_o\simeq-\dot{\Omega}'_o\simeq4.7^\circ$ /day)."695" Therefore. for each posssible solutiou lor ey. ej. 2j aud £j. we determined t ase of the shape for each longituclinalscan observation. 1uwrappecd he plase assumi rift rates close to those expected. aud fitted the resultiug phases versus observation tline toa line to obtain estimates of the rates zi aud Qi. as well as the loieitudes at epoct x, aud ο)i (the epocl time beiug taken as the time of the first image 1 ithe Orbit 12 sequence 2007-099122:]|9:10. see Table 1))."," Therefore, for each posssible solution for $e_f$, $e_l$, $i_f$ and $i_l$, we determined the phase of the shape for each longitudinal-scan observation, unwrapped the phase assuming drift rates close to those expected, and fitted the resulting phases versus observation time to a line to obtain estimates of the rates $\dot{\varpi}'_{l}$ and $\dot{\Omega}'_{l}$ , as well as the longitudes at epoch $\varpi'_{l}$ and $\Omega'_{l}$ (the epoch time being taken as the time of the first image in the Orbit 42 sequence 2007-099T22:19:10, see Table \ref{obstab1}) )."696" Regaless of whether we accepted the best-lit solution (iriangles/dashed circles in Fig 6)) or a simplified solution asstuning =180"" and ip=0 (cliamouds/solid circles in Fig 6)). we obtain roughly the same rates."," Regardless of whether we accepted the best-fit solution (triangles/dashed circles in Fig \ref{charmpar}) ) or a simplified solution assuming $\varpi'_f=180^\circ$ and $i_f=0$ (diamonds/solid circles in Fig \ref{charmpar}) ), we obtain roughly the same rates."697 c;=L66° /cday and Q0;=—LT» /day., $\dot{\varpi}'_l=4.66^\circ$ /day and $\dot{\Omega}'_l=-4.75^\circ$ /day.698 Recall that these are the modilied precession rates in a reference frame tied to the Su1., Recall that these are the modified precession rates in a reference frame tied to the Sun.699 The precession rates ina1 Inert]al coordinate system niust account for themovement of Sun A.= 0.03/day., The precession rates in an inertial coordinate system must account for themovement of Sun $\dot{\lambda}_\Sun=0.03^\circ$ /day.700" Thus he precession rates are actually: c= L697/day aud ο, —L72°/day.", Thus the precession rates are actually: $\dot{\varpi}_l=4.69^\circ$ /day and $\dot{\Omega}_l=-4.72^\circ$ /day.701" The expected rates at 119910 kin are ""fday and -L637 αν. respectively. so these numbers are closeto theo'etical expectations."," The expected rates at 119940 km are $^\circ$ /day and $^\circ$ /day, respectively, so these numbers are closeto theoretical expectations."702 This model therefore, This model therefore703the clown-llowing area.,the down-flowing area.704 The right panel in Figure 2. illustrates (he structure of uw. on a vertical eut of the convection zone., The right panel in Figure \ref{initrope} illustrates the structure of $u_{z}$ on a vertical cut of the convection zone.705 9mall-scale down-flowing plasma merges and forms the large-scale persistent downdralts in the deep convection zone., Small-scale down-flowing plasma merges and forms the large-scale persistent downdrafts in the deep convection zone.706 The large-scale cowndralts are verv important in the formation of pores during the building up of the active region. discussed in Section 3.1..," The large-scale downdrafts are very important in the formation of pores during the building up of the active region, discussed in Section \ref{dipoles}."707 The stratification of the atmosphere in the simulation domain is maintained sell-consistently with the implementation of the thermodynamic processes in the model., The stratification of the atmosphere in the simulation domain is maintained self-consistently with the implementation of the thermodynamic processes in the model.708 The superachabatic stratilication provides an unstable background. atmosphere for turbulent convective motion., The superadiabatic stratification provides an unstable background atmosphere for turbulent convective motion.709 As shown bv ?.. convective motion plavs an important role in the ormation of the sunspols ancl active region.," As shown by \cite{cheung2010}, convective motion plays an important role in the formation of the sunspots and active region."710 However. the role of the turbulent convection in (he emergence of the magnetic flux has not been clearly understood. vet. particularly ad arger scales.," However, the role of the turbulent convection in the emergence of the magnetic flux has not been clearly understood yet, particularly at larger scales."711 The aim of our simulation here is thus to study the transfer of the energy and nagnetic [lux of the rope curing its rise from (he deep convection zone. aud its interaction with the surrounding turbulent medium.," The aim of our simulation here is thus to study the transfer of the energy and magnetic flux of the rope during its rise from the deep convection zone, and its interaction with the surrounding turbulent medium."712 So alter the generation of the convection zone with a hot corona. we linearly superinpose a [lux rope upon the ambient magnetic field in (he deep convection zone. at 2=—10 Mm. indicated by the dot-dashed line in the left panel of Figure 1..," So after the generation of the convection zone with a hot corona, we linearly superimpose a flux rope upon the ambient magnetic field in the deep convection zone, at $z = -10$ Mm, indicated by the dot-dashed line in the left panel of Figure \ref{initatm}."713 The initial flux rope is centrally buovant ancl twisted along the c— axis. as described in ? ancl? by the following ecquations:," The initial flux rope is centrally buoyant and twisted along the $x -$ axis, as described in \cite{fan2001} and \cite{manchester2004} by the following equations:"714wind powered cussion from either a neutron star or white dwart.,wind powered emission from either a neutron star or white dwarf.715 However. the measured temperature of ~12 keV is much hotter than found in OB stars. which usuallv have temperatures in the rauge ~0.5 2.5 keV. Evidence that he compact companionin the 5 Cas system is actually a white dwiuf. rather than a neutron star. is strongly suggested by the observation of strong Fe line cinission at 6.8 keV of equivalent width. EW. 280 eV (Alwakaimi ct al.," However, the measured temperature of $\sim$ 12 keV is much hotter than found in OB stars, which usually have temperatures in the range $\sim$ 0.5–2.5 keV. Evidence that the compact companionin the $\gamma$ –Cas system is actually a white dwarf, rather than a neutron star, is strongly suggested by the observation of strong Fe line emission at 6.8 keV of equivalent width, EW, 280 eV (Murakami et al."716 1986)., 1986).717 This προς a liehly ionized plasma of temperature. AL —10 co which is conunonly seen in white dwarf svstems (Mulaü Sliokawa 1993).," This implies a highly ionized plasma of temperature, $kT$ $\sim$ 10 keV which is commonly seen in white dwarf systems (Mukai Shiokawa 1993)."718 Tigh mass neutron star accreting systems. on the other hand. usually show Fe cussion at 6.1. keV. with occasionally weak (EW<LOO eV) cluission at 6.7 keV. Ans.lL ks ROSAT PSPC observation of 3 Cas (Ilaberl 1995) revealed evidence for a soft component which can be modeled as black body radiatiou of KT-—200 eV. It is believe to eninnate from the heated surface of a white dwarf near the magnetic pole.," High mass neutron star accreting systems, on the other hand, usually show Fe emission at 6.4 keV, with occasionally weak $<100$ eV) emission at 6.7 keV. An 8.4 ks ROSAT PSPC observation of $\gamma$ –Cas (Haberl 1995) revealed evidence for a soft component which can be modeled as black body radiation of $\sim$ 200 eV. It is believed to emanate from the heated surface of a white dwarf near the magnetic pole."719 This conrponeut appears to be modulated with a period of 135.3 nüus which may be the spin period of the white dwarf., This component appears to be modulated with a period of 135.3 mins which may be the spin period of the white dwarf.720 The modulation may then arise frou geometrical self occultation by the white dwar (Ning Shaviv 1981). or by photoclectric absorption. as iu the accretion curtain model of Rosen e al. (," The modulation may then arise from geometrical self occultation by the white dwarf (King Shaviv 1984), or by photoelectric absorption, as in the accretion curtain model of Rosen et al. ("7211988).,1988).722 Based on a possible stellar modulation iu the chaotic N-rav. cuussion and a sinularity (spectrally and temporally) to late-type flaring stars (such as RS CVus). Suüth et al. (," Based on a possible stellar modulation in the chaotic X-ray emission and a similarity (spectrally and temporally) to late-type flaring stars (such as RS CVns), Smith et al. ("7231998) proposed a mechaisi iu which N-ravs are produced in magnetically eeucrated hot spots on the surface of + Cas itself.,1998) proposed a mechanism in which X-rays are produced in magnetically generated hot spots on the surface of $\gamma$ –Cas itself.724 Such an origin would be cousisteut with the apparent lack of orbital motion aud the active Be nature of the star., Such an origin would be consistent with the apparent lack of orbital motion and the active Be nature of the star.725 Iu this paper we report ou siunultaneous optical and N-ray ineasurements with the BeppoSAX satellite aud present the first 0.150 keV. broad-band cuerey spectruui of +Cas., In this paper we report on simultaneous optical and X-ray measurements with the BeppoSAX satellite and present the first 0.1–50 keV broad-band energy spectrum of $\gamma$ –Cas.726 As ~ Cas was one of the first stars observed to display the characteristics nowadays categorized as De star phenomenon. the collection of optical observations covers avery lone time baseline (c.e.. AFOEV 1998).," As $\gamma$ –Cas was one of the first stars observed to display the characteristics nowadays categorized as Be star phenomenon, the collection of optical observations covers a very long time baseline (e.g., AFOEV 1998)."727 Based ou photometric and spectroscopic peculiarities in conunuon with other Be stars. Doazan ct al (," Based on photometric and spectroscopic peculiarities in common with other Be stars, Doazan et al. ("728"1983) describe a three phased ""evcle iu the long term behavior of 3 Cas. without inplviue auv periodicity.","1983) describe a three phased “cycle” in the long term behavior of $\gamma$ --Cas, without implying any periodicity."729 Such a cycle starts with the builliug-up of Baluer cussion., Such a cycle starts with the building-up of Balmer emission.730 This phase is characterized by moderate mregular variability iu the Bahuer line iuteusities as well as in the visual iiagnitudo., This phase is characterized by moderate irregular variability in the Balmer line intensities as well as in the visual magnitude.731 The line intensities and magnitudes show a slow merease over several decades (Be phase)., The line intensities and magnitudes show a slow increase over several decades (Be phase).732 This behavior culminates ina second phase of Ligh variability iu both characteristics (Be-shell phase) lasting approximately LO wears. which was last observed in 5 Cas) from 1932 to 1912.," This behavior culminates in a second phase of high variability in both characteristics (Be-shell phase) lasting approximately 10 years, which was last observed in $\gamma$ –Cas from 1932 to 1942."733 The evele is terminates by a third phase with no detectable De characteristics (B-normal phase) lasting about 5 vears., The cycle is terminated by a third phase with no detectable Be characteristics (B-normal phase) lasting about 5 years.734" The visual magnitude of + Cas has more or less continuously risen frou its early Be pliase value of about 246in 1950 to a present value of —272,", The visual magnitude of $\gamma$ –Cas has more or less continuously risen from its early Be phase value of about 6in 1950 to a present value of $\sim$ 2.735 Coutemporaneous to the BeppoSANX observations. optical monitoring of 5 Cas was carried out with Universitatts-Steruwarte Münucheus sOcin Telescope located at the Wencdelstein Observatory.," Contemporaneous to the BeppoSAX observations, optical monitoring of $\gamma$ –Cas was carried out with Universitätts-Sternwarte Münnchen's $\,$ cm Telescope located at the Wendelstein Observatory."736 Measurements were made using the instrament in Johusou D. V and R arotulnc 41:50 UT on 1998. July 21.," Measurements were made using the instrument in Johnson B, V and R around 01:50 UT on 1998, July 21."737 The measured magnitudes were: D-—2.18-:0.06.. V=2.23+0.02 and R=2.56£0.03. which confixiis the “normal” behavior of  Cas at this time.," The measured magnitudes were: $\pm$ 0.06, $\pm$ 0.02 and $\pm$ 0.03, which confirms the “normal” behavior of $\gamma$ –Cas at this time."738 Moreover. the resulting (QUOS is dmn perfect aerecineut with a correlation between visual magnitude aud B-V color iudex described by Horaguchi et al. (," Moreover, the resulting $-$ 05 is in perfect agreement with a correlation between visual magnitude and B-V color index described by Horaguchi et al. ("739199D. which shows that. the xiehter the V maguitude of + Cas. the redder it ποιος in D.V. This trend is conunonly interpreted as απο due to the erowth of a circtuustellar euvelope with time.,"1994), which shows that, the brighter the V magnitude of $\gamma$ –Cas, the redder it becomes in B–V. This trend is commonly interpreted as being due to the growth of a circumstellar envelope with time."740" Ta sumus. ~ Cas, as indicated by its visual observables. is currently in a late but rather normal Be phase. with no obvious signs of a transition to the spectacular De-shell phase."," In summary, $\gamma$ –Cas, as indicated by its visual observables, is currently in a late but rather normal Be phase, with no obvious signs of a transition to the spectacular Be-shell phase."741 The Nav observation was carried out using the BeppoSAX astronomy satellite (Boclla et al., The X-ray observation was carried out using the BeppoSAX astronomy satellite (Boella et al.742 1L997a)., 1997a).743 The platform coutains four coaligued Narrow Field Tustrmuents (NFI) providing broad-banud coverage over the enerev range 1l to 300 keV. The NEI are: the Low Encrev and Medimm Encrey Concentrator Spectrometers (LECS and MECS). the Igel Pressure Gas Scintillation Proportional Counter (IIPCGSPC) aud the Phoswich Detection System (PDS).," The platform contains four coaligned Narrow Field Instruments (NFI) providing broad-band coverage over the energy range 0.1 to 300 keV. The NFI are: the Low Energy and Medium Energy Concentrator Spectrometers (LECS and MECS), the High Pressure Gas Scintillation Proportional Counter (HPGSPC) and the Phoswich Detection System (PDS)."744 The LECS covers the energy range 0.110 keV with an enerev resolution of (Parmar et al., The LECS covers the energy range 0.1–10 keV with an energy resolution of (Parmar et al.745 1997)., 1997).746 The MECS consists of three detectors simular to the LEC'S. with ideutical ΟΠΟΙΟΥ: resolution but with thicker eutrance windows. spannuiig 1.310 keV (Boclla et al.," The MECS consists of three detectors similar to the LECS, with identical energy resolution but with thicker entrance windows, spanning 1.3–10 keV (Boella et al."747 1997b)., 1997b).748 The IIPCISPC is sensitive over the energv rauge L120 keV with au cucrey resolution of ranee 13300 keV with au energy resolutiou of ot al., The HPGSPC is sensitive over the energy range 4–120 keV with an energy resolution of range 13–300 keV with an energy resolution of et al.749 1997)., 1997).750 ~ Cas was observed from 1998 July 20 11:21 UTC to July 21 22:15 UTC. vielding total on-source exposure times of 10.2. 57.1. 20.0 and 18.6 ks in the LECS. MECS. IIPGSPC aud PDS. respectively.," $\gamma$ –Cas was observed from 1998 July 20 11:21 UTC to July 21 22:15 UTC, yielding total on-source exposure times of 10.2, 37.4, 20.0 and 18.6 ks in the LECS, MECS, HPGSPC and PDS, respectively."751 Data were processed with the SANDAS data analvsis svsteni using standard procedures., Data were processed with the SAXDAS data analysis system using standard procedures.752 For the LECS aud AIECS. source events were extracted from regious of radii Saud [/e respectively. centered ou the source position.," For the LECS and MECS, source events were extracted from regions of radii 8 and $'$ , respectively, centered on the source position."753 Spectral fitting was carried out using (version 10.1).The LECS data were fit over the energy rauge 0.1, Spectral fitting was carried out using (version 10.1).The LECS data were fit over the energy range 0.1754L136bb. Ἱς one of the few known Neptune-niass extrasolar λαοί».,b is one of the few known Neptune-mass extrasolar planets.755 T was iscoveredL by radial-velocitv nieasurenüents (2) as a planet with a period of 2.6 days and a ΠΠ mass of 21 M., It was discovered by radial-velocity measurements \citep{Butler:2004dq} as a planet with a period of 2.6 days and a minimum mass of 21 $_\oplus$.756 Follow-up Doppler observations of [136 refined the planetary mass aud the orbital parameters. inchiding au eccentricity of 0.164 (7.hereafterMOT)..," Follow-up Doppler observations of 436 refined the planetary mass and the orbital parameters, including an eccentricity of $0.16\pm0.02$ \citep[hereafter M07]{Maness:2007la}."757 Our team (7.hereafterCOTA) discovered he transiting nature of bb. euabliug us to incasure a planetary radius  1 Bí.," Our team \citep[hereafter G07a]{Gillon:2007b} discovered the transiting nature of b, enabling us to measure a planetary radius $\sim$ 4 $_\oplus$."758 This discovery and the corresponding measurements of the planetary radius and mass inicated a planet composed mostly of ice. probably surrounded by a small ΠΠΠο cuvelope.," This discovery and the corresponding measurements of the planetary radius and mass indicated a planet composed mostly of ice, probably surrounded by a small H/He envelope."759 Because of the simall size of the parent star (2 ~ 0.1 Rjaud the short orbital period of bb. the planet-to-star luuinosity ratio iu the infrared is comparable to that of many known hot Jupiters. despite the planets much simaller radius.," Because of the small size of the parent star $R$ $\sim$ 0.4 $R_\odot$ ) and the short orbital period of b, the planet-to-star luminosity ratio in the infrared is comparable to that of many known hot Jupiters, despite the planet's much smaller radius."760 Furthermore. the M dwarf L136 is rather bright in he infrared (Ik. ~ 6).," Furthermore, the M dwarf 436 is rather bright in the infrared (K $\sim$ 6)."761 Detection of the thermal emission frou lis small plane had thus been expected to be within the reach of theTelescope., Detection of the thermal emission from this small planet had thus been expected to be within the reach of the.762 Following our transit discoverv. we subnütted a Discretionary Director Time (DDT) Spitzer xoposal to better characterize this interesting planet.," Following our transit discovery, we submitted a Discretionary Director Time (DDT) $Spitzer$ proposal to better characterize this interesting planet."763 We applied for photometric observations of the primary transit usine the 8-nu band of the InfraRed Array Camera TRAC (7) in order to get a very accurate radius measurement aud coustrai the bulk conrposition of the planet., We applied for photometric observations of the primary transit using the $\mu$ m band of the InfraRed Array Camera IRAC \citep{Fazio:2004fy} in order to get a very accurate radius measurement and constrain the bulk composition of the planet.764 We also applied. for plotometric observations of the secondary eclipse in the four bands of IRAC (3.6. 1.5. 5.8 and 8 jan). iu the 16-;1u baud of the TutraRed Spectrograph IRS (7) and the 21-40. baud. of the Multibaud huaeimg Photometer MIPS (7). to assess the atinosphieric eniperature. albedo. leat distribution cficiency aud composition.," We also applied for photometric observations of the secondary eclipse in the four bands of IRAC (3.6, 4.5, 5.8 and 8 $\mu$ m), in the $\mu$ m band of the InfraRed Spectrograph IRS \citep{Houck:2004fv} and the $\mu$ m band of the Multiband Imaging Photometer MIPS \citep{Rieke:2004dz} to assess the atmospheric temperature, albedo, heat distribution efficiency and composition."765 Ilowewer. the observations were actually rigecred aud performed as part of an existing Tarect of Opportunity (ToO) program (ID 30129. PI J. IEuviustonu) which has a total priority for the observations of transiting alanets.," However, the observations were actually triggered and performed as part of an existing Target of Opportunity (ToO) program (ID 30129, PI J. Harrington) which has a total priority for the observations of transiting planets."766 The iain goal of this ToO is to deliver to he community without auv proprictary period optimal Spitter observations of transiting plaucts., The main goal of this ToO is to deliver to the community without any proprietary period optimal $Spitzer$ observations of transiting planets.767"Spitzer observed the transit aud the secondary eclipse of L186 in the 8-120 TRAC band ou June 29 and 30 respectively,", observed the transit and the secondary eclipse of 436 in the $\mu$ m IRAC band on June 29 and 30 respectively.768 The data of the primary transit were made publicly available on July 13th 2007., The data of the primary transit were made publicly available on July 13th 2007.769 Our direct analysis of these data allowed us to determine a very accurate radius for L136bb (Rp=L2Ry.?.hereafterGOT) and to confirm the presence of au II/IIe envelope.," Our direct analysis of these data allowed us to determine a very accurate radius for b \citep[Rp = 4.2 R$_\oplus$,][hereafter G07b]{Gillon:2007a} and to confirm the presence of an H/He envelope."770Spitzer data of the secondary eclipse were uot released to the community until July 17th 2007. due to au oversight that occurred at the Spitzer Scieuce Center.," data of the secondary eclipse were not released to the community until July 17th 2007, due to an oversight that occurred at the $Spitzer$ Science Center."771 This explains wliv we separatec or analysis ancl present here our results reearding the secondary eclipse cata., This explains why we separated our analysis and present here our results regarding the secondary eclipse data.772 During the writing of this study. a paper by ? reporting primary and secondary eclipses analyses has been subiuitted to ApJ and put on astro-ph.," During the writing of this study, a paper by \citet{Deming:2007pr} reporting primary and secondary eclipses analyses has been submitted to ApJ and put on astro-ph."773 The preseut analysis has been conducted independently from their, The present analysis has been conducted independently from their774"obscuring torus with an elliptically-shaped cross-section (1), tilted, conical outflows in the polar directions (2), and a ring-like scattering region in the equatorial plane, located between the accretion disk and the inner surfaces of the torus (3).","obscuring torus with an elliptically-shaped cross-section (1), tilted, conical outflows in the polar directions (2), and a ring-like scattering region in the equatorial plane, located between the accretion disk and the inner surfaces of the torus (3)."775 The equatorial torus is filled with neutral matter and centred on the much smaller reprocessing disc., The equatorial torus is filled with neutral matter and centred on the much smaller reprocessing disc.776 The dimensions of all emission and scattering geometries are summarised in Table 1 and a sketch is provided in Fig. 1.., The dimensions of all emission and scattering geometries are summarised in Table \ref{tab:models} and a sketch is provided in Fig. \ref{fig:sketch}.777" The symmetry axes of the torus, the equatorial wedge and the disc are identical and the torus funnel has a half-opening angle @tor=60°, as taken from Rabanetal.(2009)."," The symmetry axes of the torus, the equatorial wedge and the disc are identical and the torus funnel has a half-opening angle $\theta_{\rm tor} = 60\degr$, as taken from \citet{raban2009}."778". Thus, the direct view of the central components is obscured at viewing directions, 7, with i>θιος."," Thus, the direct view of the central components is obscured at viewing directions, $i$, with $i > \theta_{\rm tor}$."779 Both angles are measured with respect to the symmetry axis., Both angles are measured with respect to the symmetry axis.780 We first only consider reprocessing off the irradiated accretion disc alone and then add the torus., We first only consider reprocessing off the irradiated accretion disc alone and then add the torus.781 The following scenarios include polar outflows that we realise by a tilted double-cone., The following scenarios include polar outflows that we realise by a tilted double-cone.782 The centre of the cone is identical with the ones of the reprocessing disc and the torus., The centre of the cone is identical with the ones of the reprocessing disc and the torus.783 The double-cone half-opening angle is measured by θεοπο=40° and its tilting angle with respect to the torus symmetry axis by θωι=18° (seeRabanetal.2009)., The double-cone half-opening angle is measured by $\theta_{\rm cone} = 40\degr$ and its tilting angle with respect to the torus symmetry axis by $\theta_{\rm tilt} = 18\degr$ \citep[see][]{raban2009}.784. We assume that the matter inside the double-cone is ionised and dominated by electron scattering., We assume that the matter inside the double-cone is ionised and dominated by electron scattering.785" The Thomson optical depth, Tcone, is measured between the inner and outer radii of the cone."," The Thomson optical depth, $\tau_{\rm cone}$, is measured between the inner and outer radii of the cone."786 The misalignment between the torus and the ionisation cones is inferred from the mid-IR interferometry measurements of Rabanetal.(2009) in combination with the maser observations of Gallimoreetal.(1996) as well as with the HST-spectroscopy modelling conducted by Dasetal. (2006)., The misalignment between the torus and the ionisation cones is inferred from the mid-IR interferometry measurements of \citet{raban2009} in combination with the maser observations of \citet{gallimore1996} as well as with the -spectroscopy modelling conducted by \citet{das2006}.787. For further details on the geometrical considerations we refer to the very instructive discussion given in Rabanetal. (2009)., For further details on the geometrical considerations we refer to the very instructive discussion given in \citet{raban2009}.788" There is another, potentially important scattering region that is required to explain the dichotomy of the optical/UV polarisation angle."," There is another, potentially important scattering region that is required to explain the dichotomy of the optical/UV polarisation angle."789 The equatorial scattering ring as described by Antonucci(1984) and Smithetal.(2004) efficiently polarises with a projected polarisation vector being parallel to the symmetry axis., The equatorial scattering ring as described by \citet{antonucci1984} and \citet{smith2004} efficiently polarises with a projected polarisation vector being parallel to the symmetry axis.790 The ring should also contribute in the X-ray range and we model it as an equatorial scattering wedge between 0.03 and 0.05 pc., The ring should also contribute in the X-ray range and we model it as an equatorial scattering wedge between 0.03 and 0.05 pc.791 The symmetry axis of the wedge is aligned with the torus axis and its half-opening angle with respect to the equatorial, The symmetry axis of the wedge is aligned with the torus axis and its half-opening angle with respect to the equatorial792"scaltering rate 2, by Py=42,/27 (Field1953).. which is given by £2,={dvcn,oír). where n, is the number density of photons per unit frequency. aud σ(ν} is the cross section for scattering (Aladau.Meiksin.&Rees1997.. hereafter MMB).","scattering rate $P_{\alpha}$ by $P_{10}=4 P_{\alpha}/27$ \citep{F58}, which is given by $P_{\alpha}=\int d\nu c\, n_{\nu}\,\sigma(\nu)$, where $n_{\nu}$ is the number density of photons per unit frequency, and $\sigma(\nu)$ is the cross section for scattering \citealp{MMR97}, hereafter MMR)."793 Besicles coupling the spin temperature and kinetic temperature of (he gas. the scattering of photons may also heat up Che gas.," Besides coupling the spin temperature and kinetic temperature of the gas, the scattering of photons may also heat up the gas."794 In this paper. we present the scattering rate and heating rate due (ο photons in a uniform atomic intergalactic medium as a function of the two parameters they depend on: the gas temperature and the scaltering optical depth of the medium (or Gunu-Pelerson optical depth).," In this paper, we present the scattering rate and heating rate due to photons in a uniform atomic intergalactic medium as a function of the two parameters they depend on: the gas temperature and the scattering optical depth of the medium (or Gunn-Peterson optical depth)."795 There are two different wavs in which photons propagating through the atomic medium mav reach the line. and start being repeatedly scattered by hydrogen: (a) Photous that are emitted between the and Lys wavelengths by sources such as massive stars will be redshifted as they penetrate (he atomic medium until thev reach the wavelength.," There are two different ways in which photons propagating through the atomic medium may reach the line, and start being repeatedly scattered by hydrogen: (a) Photons that are emitted between the and $\beta$ wavelengths by sources such as massive stars will be redshifted as they penetrate the atomic medium until they reach the wavelength."796 These photons enter the line from its blue wing. and will be calledcontinuum photons in this paper. (," These photons enter the line from its blue wing, and will be called photons in this paper. ("797b) Photons that are emitted at wavelengths shorter (han Ly. but still longer (han the Lyman limit so that thev can penetrate deep into the atomic mecium. will redshift until they reach the or a higher Lyman series line. and will then be converted into a Lvo photon when the hydrogen atom {μον excite decavs first to the 2p state before reaching the ground state.,"b) Photons that are emitted at wavelengths shorter than $\beta$, but still longer than the Lyman limit so that they can penetrate deep into the atomic medium, will redshift until they reach the $\beta$ or a higher Lyman series line, and will then be converted into a $\alpha$ photon when the hydrogen atom they excite decays first to the 2p state before reaching the ground state."798 These photons will be injected near the line center with a distribution determined by (the Voiet ΠΟΙΟ. ancl will be called.injected photons.," These photons will be injected near the line center with a distribution determined by the Voigt function, and will be called photons."799 These two tvpes of photons result in different scattering and heating rates. so results will be presented separately for both of them.," These two types of photons result in different scattering and heating rates, so results will be presented separately for both of them."800" MAIR estimated the heating rate due to photons by assuming that the average relative change in a photon energy in each scattering is the same as when (he photon is scattered by an atom at rest: CGAE/E)=—hv,/mgc?.", MMR estimated the heating rate due to photons by assuming that the average relative change in a photon energy in each scattering is the same as when the photon is scattered by an atom at rest: $\langle \Delta E/E \rangle = -h\nu_{\alpha}/m_H c^2$.801" Hence. they assumed that the total energv (ransler rale is At the thermalization rate 2,ο is the rate 27, required to bring the spin temperature close to the kinetic temperature). the heating rate assumed by MMB is E/lgc220KGyr11z)/7. which would heat up the gas kinetic temperature above the CAMB temperature in a fraction of a IIubble (ime at z~6."," Hence, they assumed that the total energy transfer rate is At the thermalization rate $P_{th} =27 A_{10} T_{CMB}/4T_*$ (which is the rate $P_{\alpha}$ required to bring the spin temperature close to the kinetic temperature), the heating rate assumed by MMR is $\dot{E}/k_B \simeq 220 {\rm K\, Gyr}^{-1} (1+z)/7 $, which would heat up the gas kinetic temperature above the CMB temperature in a fraction of a Hubble time at $z\sim 6$."802 In the present investigation we calculate the heating rate of the continuum and injected photons bv computing the background spectrum near the line using the, In the present investigation we calculate the heating rate of the continuum and injected photons by computing the background spectrum near the line using the803"This classification has also been used in other works (Heiles2001;Heiles&Troland2003;AuditHennebelle2005) and is warranted by the positions of the isothermal lines atK,, which — of course only very roughly — describe the region of TI in the intermediate and high-density range.","This classification has also been used in other works \citep{Heiles01,Heiles03,Audit05} and is warranted by the positions of the isothermal lines at, which – of course only very roughly – describe the region of TI in the intermediate and high-density range."804" In the following, the terms cold, unstable, and warm phase always refer to the above definitions."," In the following, the terms cold, unstable, and warm phase always refer to the above definitions."805 The chosen initial conditions of our simulations are located in the thermally unstable region (Fig. 1))., The chosen initial conditions of our simulations are located in the thermally unstable region (Fig. \ref{fig:1}) ).806" In consequence, one can expect that a two-phase-medium will evolve."," In consequence, one can expect that a two-phase-medium will evolve."807" As explained in Sect. 2,,"," As explained in Sect. \ref{sc:numerics},"808" we start the simulations with the gas at rest, and turbulence is generated by stochastic forcing."," we start the simulations with the gas at rest, and turbulence is generated by stochastic forcing."809 The forcing strength is regulated by the characteristic velocity V., The forcing strength is regulated by the characteristic velocity $V$.810" Normalising V by the initial soundspeed cy= (ykaTo/umg)!?, we obtain a characteristic Mach number, Ma=V/co, which determines the magnitude of the stochastic force."," Normalising $V$ by the initial soundspeed $c_0 = (\gamma k_B \mathcal{T}_0/\mu m_H)^{1/2}$ , we obtain a characteristic Mach number, $\mathrm{Ma} = V/c_0$, which determines the magnitude of the stochastic force."811 We performed simulations with Ma between 0.2 and 3.0., We performed simulations with Ma between 0.2 and 3.0.812" However, Ma is not an indicator for the average Mach number of the flow, which depends on the phase separation of the gas."," However, Ma is not an indicator for the average Mach number of the flow, which depends on the phase separation of the gas."813" In one set of simulations, we applied purely compressive forcing (6 = 0) for different initial densities."," In one set of simulations, we applied purely compressive forcing $\zeta$ = 0) for different initial densities."814" In addition, simulations with were also carried out for the solenoidal case (¢= 1)."," In addition, simulations with were also carried out for the solenoidal case $\zeta = 1$ )."815 The parameters of all simulations are listed in Table 1.., The parameters of all simulations are listed in Table \ref{table:sim}.816 Each simulation ran until an approximate statistically steady state was found for at least one dynamical timescale T., Each simulation ran until an approximate statistically steady state was found for at least one dynamical timescale $T$.817" Although it is not obvious that a well-defined statistically steady state can be reached for non-isothermal turbulence, all simulations presented here show a convergence of statistical quantities related to the flow properties and the thermodynamic properties of the gas as time proceeds."," Although it is not obvious that a well-defined statistically steady state can be reached for non-isothermal turbulence, all simulations presented here show a convergence of statistical quantities related to the flow properties and the thermodynamic properties of the gas as time proceeds."818" Thus, the statistics described in the next sections are averaged over the last dynamical timescale."," Thus, the statistics described in the next sections are averaged over the last dynamical timescale."819" Since the computational cost of performing simulations of turbulence in non-isothermal gas is significantly higher than for isothermal gas, all runs were performed on a static grid of N=256? cells."," Since the computational cost of performing simulations of turbulence in non-isothermal gas is significantly higher than for isothermal gas, all runs were performed on a static grid of $N = 256^3$ cells."820" Theonly exceptions are the runs D1Ma1.0_1128 with a resolution of 128? and D1Ma1.0.5512 with 512°, to test the resolution dependency of the results."," Theonly exceptions are the runs 128 with a resolution of $^3$ and 512 with $^3$, to test the resolution dependency of the results."821" Because of the moderate resolution of the simulations, we only investigated global statistics and probability density functions, but not two-point statistics or power spectra."," Because of the moderate resolution of the simulations, we only investigated global statistics and probability density functions, but not two-point statistics or power spectra."822 A crucial question for the following analysis is to what extent statistical properties are affected by the numerical resolution., A crucial question for the following analysis is to what extent statistical properties are affected by the numerical resolution.823" To investigate the dependence on resolution, we consider the simulations D1Ma1.0_1128, DIMal.0 and D1Mal.0.5512 with compressive forcing, a mean number density πο= 1l.0cm?, and Μα-1.0."," To investigate the dependence on resolution, we consider the simulations 128, D1Ma1.0 and 512 with compressive forcing, a mean number density $n_{0}=1.0\,\mathrm{cm^{-3}}$ , and $\mathrm{Ma=1.0}$."824" The resolution varies from 128, over 256? to 512?."," The resolution varies from $128^{3}$, over $256^{3}$ to $512^{3}$."825 Figure 2. shows the time evolution of the temperature field and the Mach number for these three simulations., Figure \ref{fig:2} shows the time evolution of the temperature field and the Mach number for these three simulations.826" For them, we see a similar evolution in time both for the temperature field and the velocity field (represented here by the Mach number)."," For them, we see a similar evolution in time both for the temperature field and the velocity field (represented here by the Mach number)."827" The onset of the TI becomes apparent by a rapidly spreading temperature distribution within the first dynamical timescale, where the temperatures finally range from roughly to several For the lower resolutions, we didnot apply the cutoff at K.."," The onset of the TI becomes apparent by a rapidly spreading temperature distribution within the first dynamical timescale, where the temperatures finally range from roughly to several For the lower resolutions, we didnot apply the cutoff at ."828" As one can see, the temperature rises locally to as much as"," As one can see, the temperature rises locally to as much as"829of the observed peak flux. aad the Euclidean value of7.,of the observed peak flux and the Euclidean value of.830.. The correlation was discovered for GRBs in the GUSBAD catalog!., The correlation was discovered for GRBs in the GUSBAD catalog.831.. Since the Euclidean value of lis a geomelric Cosmological distance indicator (Schmidt2001).. the peak luminosity of a subgroup of GRBs with given ccan be derived without knowing any redshifts.," Since the Euclidean value of is a geometric cosmological distance indicator \citep{sch01}, the peak luminosity of a subgroup of GRBs with given can be derived without knowing any redshifts."832 Using the ecorrelation is not subject to anv of the problems mentioned above for the Amati relation., Using the correlation is not subject to any of the problems mentioned above for the Amati relation.833 The approach in our paper is the following., The approach in our paper is the following.834 We use GUSDAD fluxes in the [our BATSE DISCLA channels to divide them into live spectral classes based on their vvalues., We use GUSBAD fluxes in the four BATSE DISCLA channels to divide them into five spectral classes based on their values.835 We assume that the oof each spectral group is a gaussian of logL with a value of large enough so that the overall sum of the five luminosity [functions is reasonably smooth., We assume that the of each spectral group is a gaussian of $\log L$ with a value of large enough so that the overall sum of the five luminosity functions is reasonably smooth.836 In addition. we assume that the overall rate density of GRBs varies with redshilt as).," In addition, we assume that the overall rate density of GRBs varies with redshift as."837. The central Iuminositwv logL. of each spectral gaussian is derived by a process of iteration. until (he corresponding value of {for the spectral class agrees with that observed.," The central luminosity $\log L_c$ of each spectral gaussian is derived by a process of iteration, until the corresponding value of for the spectral class agrees with that observed."838 The procedure allows deriving the overall GRB luminosity function from the GUSBAD catalog without using anv redshifts., The procedure allows deriving the overall GRB luminosity function from the GUSBAD catalog without using any redshifts.839 The main free parameter is the rate Iunction)., The main free parameter is the rate function.840. To evaluate the predicted Iuminosity. aud redshift distributions for various lorms of).. we do need a sample with redshifts.," To evaluate the predicted luminosity and redshift distributions for various forms of, we do need a sample with redshifts."841 We evaluate and use ddetected GRBs with observed redshifts., We evaluate and use detected GRBs with observed redshifts.842 For the successful Iuminositv functions. the process vields a rest frame vvalue for each of the spectral classes.," For the successful luminosity functions, the process yields a rest frame value for each of the spectral classes."843 This leads to aand ccorrelations entirely based on data in (his paper., This leads to and correlations entirely based on data in this paper.844 The use of aand the derivation of [from GRBs in the GUSDAD catalog are discussed in Section 2. as well as the evidence for the ecorrelation.," The use of and the derivation of from GRBs in the GUSBAD catalog are discussed in Section 2, as well as the evidence for the correlation."845 The framework for deriving the iis discussed in Section 3., The framework for deriving the is discussed in Section 3.846 The status and completeness of oobservalions is covered in Section 4., The status and completeness of observations is covered in Section 4.847 In Section 5 we present two luminosity functions with differing redshilt dependence aad test them on, In Section 5 we present two luminosity functions with differing redshift dependence and test them on848"velocity distribution, with peaks at 106 and 211 km s7!.","velocity distribution, with peaks at 106 and 211 km $^{-1}$."849" While the high velocity component is at approximately the same radial velocity in all eastern fields, and in the Southern field as well, the velocity separation with the lower velocity appears to increase significantly outwards from the SMC componentcentre."," While the high velocity component is at approximately the same radial velocity in all eastern fields, and in the Southern field as well, the velocity separation with the lower velocity component appears to increase significantly outwards from the SMC centre."850" On the other hand, the SMC contribution is very small to non-existent in the most distant field to the East."," On the other hand, the SMC contribution is very small to non-existent in the most distant field to the East."851 This suggests that the edge of the SMC in this direction is close to this field., This suggests that the edge of the SMC in this direction is close to this field.852" To estimate this, we first calculated the number of SMC stars in each of our Eastern fields based on the derived completeness and fraction of stars we attribute to the SMC in Table 1."," To estimate this, we first calculated the number of SMC stars in each of our Eastern fields based on the derived completeness and fraction of stars we attribute to the SMC in Table 1."853" We then integrated a 10 Gyr old, [Fe/H]|2—1.0 luminosity function from Marigoetal.(2008) and derived the surface brightness of SMC stars in our fields, after correcting for incompleteness and sampling."," We then integrated a 10 Gyr old, $-1.0$ luminosity function from \cite{marigo08} and derived the surface brightness of SMC stars in our fields, after correcting for incompleteness and sampling."854" We then fit this to a Hernquist profile and extrapolated to a surfacebrightness of 26 mag arcsec? in K, which we take,"," We then fit this to a Hernquist profile and extrapolated to a surfacebrightness of 26 mag $^{-2}$ in $K$ , which we take,"855The Local Group galaxy. IC! 1613 is a highly-resolved dwarlol (wpe ID(s)m (de Vaucouleurs et al.,"The Local Group galaxy, IC 1613 is a highly-resolved dwarf of type IB(s)m (de Vaucouleurs et al."856 1991)., 1991).857 IC 1613 is rich in egas and is actively forminge stars: as a consequence it also contains many Classical (Population I) Cepheid variables., IC 1613 is rich in gas and is actively forming stars; as a consequence it also contains many Classical (Population I) Cepheid variables.858 These stars have been (he subject, These stars have been the subject859The focus of this dynamical study has been on fitting the model to measured positions and velocities of the 27 galaxies at distances less than 1.5 Mpc.,The focus of this dynamical study has been on fitting the model to measured positions and velocities of the 27 galaxies at distances less than 1.5 Mpc.860 We emphasize again that the measures of the external actors in Table 3 are chosen for a phenomenological description of the effect of external mass on the motions of the LG galaxies implied by the pattern of LG velocities and positions; the parameters of the external actors should not be taken as meaningful estimates of the properties of the named systems., We emphasize again that the measures of the external actors in Table 3 are chosen for a phenomenological description of the effect of external mass on the motions of the LG galaxies implied by the pattern of LG velocities and positions; the parameters of the external actors should not be taken as meaningful estimates of the properties of the named systems.861" We do take it to be an interesting project to attempt to interpret the large distance offsets normal to the line of sight for the actors named Maffei and Centaurus, and their curiously small masses, as an indication of the nature of the external mass distribution, but that is not attempted here."," We do take it to be an interesting project to attempt to interpret the large distance offsets normal to the line of sight for the actors named Maffei and Centaurus, and their curiously small masses, as an indication of the nature of the external mass distribution, but that is not attempted here."862 Since the LG galaxy masses are left nearly free to aid the fit we count as free parameters the ten LG galaxy masses that seem to be large enough to matter., Since the LG galaxy masses are left nearly free to aid the fit we count as free parameters the ten LG galaxy masses that seem to be large enough to matter.863" The redshifts and distances of the external actors are close to their assigned values, meaning they do not seem to matter,"," The redshifts and distances of the external actors are close to their assigned values, meaning they do not seem to matter,"864Eucregy resolved time analvsis of the bursts was performed to study the spectral evolution of the observed events.,Energy resolved time analysis of the bursts was performed to study the spectral evolution of the observed events.865 Due to the above ieutioued nmissiug data in the Aueust 22 burst observation. oulv the August 21 aud September 16 bursts were analyzed this wav (see Fig.," Due to the above mentioned missing data in the August 22 burst observation, only the August 24 and September 16 bursts were analyzed this way (see Fig."866 3)., 3).867 The time histories of the bursts are constructed bv accunulatiug oulv the detector counts associated with the shadoweram obtained for the sky position of the analyzed source. thus improving the signal to noise ratio of the profile.," The time histories of the bursts are constructed by accumulating only the detector counts associated with the shadowgram obtained for the sky position of the analyzed source, thus improving the signal to noise ratio of the profile."868 The backerowud is he sum of (part of) the diffuse A-rav backeround. the particles backeround and the contamination of other sources in the field of view.," The background is the sum of (part of) the diffuse X-ray background, the particles background and the contamination of other sources in the field of view."869 Source contaninatioun is the domunating background component for crowded sky fields like the Galactic Bulge., Source contamination is the dominating background component for crowded sky fields like the Galactic Bulge.870 Nevertheless. the probability of source confusion during a short time-scale eveut (LO100 8) like an N-rav burst is ucelieible.," Nevertheless, the probability of source confusion during a short time-scale event (10–100 s) like an X-ray burst is negligible."871 The burst spectra of GRS 2853 are cousistcut with absorbed blackbody radiation with average color temperatures of ~2 keV. Α suuniuw of the spectral paranieters of the three bursts is given in Table 1., The burst spectra of GRS $-$ 2853 are consistent with absorbed blackbody radiation with average color temperatures of $\sim 2$ keV. A summary of the spectral parameters of the three bursts is given in Table 1.872 The value of the Nyp parameter obtained for the Aneust 2[ πανε is higher with respect to the August 22 and September 16 ones., The value of the $N_{\rm H}$ parameter obtained for the August 24 burst is higher with respect to the August 22 and September 16 ones.873 For burst 2. freezing he NY value to the average value of burst 1 aud 3 (10.3&τοσο 7) leads to higher values of the reduced u (1.30 for 27 d.o.f.).," For burst 2, freezing the $N_{\rm H}$ value to the average value of burst 1 and 3 $10.3\times10^{22} {\rm cm}^{-2}$ ) leads to higher values of the reduced $\chi^{2}$ (1.30 for 27 d.o.f.),"874 to an higher blackbody temperature (2.61+0.15 keV) and to a lower blackbody radius (3.9+0.1 km at 10 kpe)., to an higher blackbody temperature $2.64\pm 0.15$ keV) and to a lower blackbody radius $3.9\pm 0.4$ km at 10 kpc).875 Conversely. if we asstune that all the three bursts had in average the same characteristics (coor temperature and radius of the enütting sphere). this iuplies a 1-dav scale Nyp variability of a factor of ~3.," Conversely, if we assume that all the three bursts had in average the same characteristics (color temperature and radius of the emitting sphere), this implies a 1-day time-scale $N_{\rm H}$ variability of a factor of $\sim 3$."876 Time-resolved spectra were acctunulated for burst 2 and 3. in order to study the time evolution of their spectral parameters.," Time-resolved spectra were accumulated for burst 2 and 3, in order to study the time evolution of their spectral parameters."877 To better coustrain the fits. the Nyp parameter was kept fixed. according to the values obtained for the total bursts. ic. 36.0«107767 and 10.3&107?«3 for burst 2 and burst 3 respectively.," To better constrain the fits, the $N_{\rm H}$ parameter was kept fixed, according to the values obtained for the total bursts, i.e. $36.0\times 10^{22}{\rm cm}^{-2}$ and $10.3\times 10^{22}{\rm cm}^{-2}$ for burst 2 and burst 3 respectively."878" Blackhocy spectra allow to determine the relatiouship between the average radius of the emitting sphere Rig, (1i units of kii) aud the source distance digkpe Gn units of 10 kpc).", Blackbody spectra allow to determine the relationship between the average radius of the emitting sphere $R_{\rm km}$ (in units of km) and the source distance $d_{\rm 10~kpc}$ (in units of 10 kpc).879 Iu Fig., In Fig.880 2 and in Table 2 the tine histories of the measured Rian dtu1 ratios are shown. assuning isotropic enussion and nof correcting for gravitational redshift and conversion to true blackbody temperature from color clupcrature (see Lewin.vanParadijs.&Taain1993 for details).," 3 and in Table 2 the time histories of the measured $R_{\rm km}$ $d_{\rm 10~kpc}$ ratios are shown, assuming isotropic emission and not correcting for gravitational redshift and conversion to true blackbody temperature from color temperature (see \cite{Lewi93} for details)."881 A radius expansion of a factor of 2 is observed iu the September 16 burst., A radius expansion of a factor of $\sim 2$ is observed in the September 16 burst.882 Ou the basis of their spectral aud timing properties. we interpret the three bursts detected from CRS 2853 as tvpelIl Aeray bursts. typically associated. to low-mass binary (LAINB) svstenis (Seo Lewin.vanParadijs.&Taain1995 for a review).," On the basis of their spectral and timing properties, we interpret the three bursts detected from GRS $-$ 2853 as type-I X-ray bursts, typically associated to low-mass binary (LMXB) systems (see \cite{Lewi95} for a review)."883 The dackbody cussion aud the measured color temperatures of ~2 keV are consistent with this hypothesis., The blackbody emission and the measured color temperatures of $\sim 2$ keV are consistent with this hypothesis.884 Spectral softening is observed in the time resolved spectra of the musts (Table 2)., Spectral softening is observed in the time resolved spectra of the bursts (Table 2).885 Moreover. the bursts time profiles can ο fitted with expoucutial decays whose characteristic nues are enerev dependent. beiug shorter at higher energies (soe Fig.," Moreover, the bursts time profiles can be fitted with exponential decays whose characteristic times are energy dependent, being shorter at higher energies (see Fig."886 3)., 3).887 Twpe-I bursts strongly sugecst a ιο(χο star rature for the binary svstem., Type-I bursts strongly suggest a neutron star nature for the binary system.888 This indicates GRS 2853 is a transicut neutrou-star LAINB., This indicates GRS $-$ 2853 is a transient neutron-star LMXB.889 The photospheric radius expansion cerived frou he time resolved spectral analysis of the brightest »ust (burst 23) can be interpreted as adiabatic expansion diving an high bhuumositv (super-Eddiuston) vpe- burst., The photospheric radius expansion derived from the time resolved spectral analysis of the brightest burst (burst 3) can be interpreted as adiabatic expansion during an high luminosity (super-Eddington) type-I burst.890" Actually, the 728 keV time history of the September 16 bust (Fie."," Actually, the 7–28 keV time history of the September 16 burst (Fig."891 23. melt panel) shows a top-flattened and perhaps double-peaked motile which is typical of super-Eddiuston events (c.g. Lewin.vauParadijs.&Taam 1905).," 3, right panel) shows a top-flattened and perhaps double-peaked profile which is typical of super-Eddington events (e.g. \cite{Lewi95}) )."892 Eddinetou-Maunosity N-rav bursts can lead to an estimate of the source distance., Eddington-luminosity X-ray bursts can lead to an estimate of the source distance.893 Asaunug a 2<10σον! Eddiugton olometric bhDunünositv for a 1.1M. neutron star. and aliug iuto account the observed peak flux of burst 3 which extrapolates to an unabsorbed bolometric Dhuuimositv of 527+12 mCrab (3.26+0.26«10Serecus 2st). we obtain d=7.2£0.6 kpc.," Assuming a $2\times 10^{38} {\rm erg~s}^{-1}$ Eddington bolometric luminosity for a $1.4~{\rm M}_{\odot}$ neutron star, and taking into account the observed peak flux of burst 3 which extrapolates to an unabsorbed bolometric luminosity of $527\pm 42$ mCrab $3.26\pm 0.26 \times 10^{-8} 894{\rm erg cm}^{-2}{\rm s}^{-1}$ ), we obtain $d=7.2\pm 0.6$ kpc."895 If we adopt the average huninosity of super-Eddineton bursts proposed by Lewin. van Paradijs. Taam (1995) (3.040.6ς10eres. 1) the distance value becomes d=8.841.2 kpc. indicating GRS 2853 to be very close to the Galactic Centre.," If we adopt the average luminosity of super-Eddington bursts proposed by Lewin, van Paradijs, Taam (1995) $3.0\pm 0.6\times 10^{38}{\rm erg s}^{-1}$ ) the distance value becomes $d=8.8\pm 1.2$ kpc, indicating GRS $-$ 2853 to be very close to the Galactic Centre."896"(pa= plasma at ~80 K, and Kreckeletal. 0.25)reported pa=0.479+0.02 in a warmer (~200 K) (2010)normal hydrogen and argon Finally, in work outside our laboratory, Kreckeletal.(2007) have reported pa=0.4 in a low-temperature ion trap.","$p_2=0.25$ ) plasma at $\sim80$ K, and \cite{kreckel2010} reported $p_3 = 0.479 \pm 0.02$ in a warmer $\sim200$ K) normal hydrogen and argon Finally, in work outside our laboratory, \cite{kreckel2007} have reported $p_3 = 0.4$ in a low-temperature ion trap."897" All of these measurements suggest that it is in fact possible to achieve pa«0.5, and lend some evidence to support the calculated results."," All of these measurements suggest that it is in fact possible to achieve $p_3 < 0.5$, and lend some evidence to support the calculated results."898 We now consider whether there are enough reactive collisions within the lifetime of an average Hj in diffuse molecular clouds to bring the spin modifications into BRE., We now consider whether there are enough reactive collisions within the lifetime of an average $_3^+$ in diffuse molecular clouds to bring the spin modifications into BRE.899" The destruction of in such clouds is dominated by dissociative recombinationHi (DR) with electrons, and the lifetime is simply tafe=(kpmn(e-)) (the reciprocal of the destruction where kpg |is the dissociative recombination raterate),coefficient."," The destruction of $_3^+$ in such clouds is dominated by dissociative recombination (DR) with electrons, and the lifetime is simply $\tau_{life} = (k_{DR} n(e^-))^{-1}$ (the reciprocal of the destruction rate), where $k_{DR}$ is the dissociative recombination ratecoefficient."900" The average time between reactive collisions can be expressed as Tran=(Kyn(Ha))1, where kj, is the reactive collision rate for Hi + Ha."," The average time between reactive collisions can be expressed as $\tau_{rxn} = (k_{rc} n(\rm{H}_2))^{-1}$, where $k_{rc}$ is the reactive collision rate for $_3^+$ + $_2$ ."901 The average number of collisions an Hi will experience is then Neon=Tüfe/Trsn , The average number of collisions an $_3^+$ will experience is then ${\cal N}_{rxn} = \tau_{life}/\tau_{rxn} = [k_{rc}/k_{DR}] [n(\textrm{\hh})/n(e^-)]$ .902"Assuming for the /n(e-moment)]. that kpr is independent of the nuclear spin modification, we adopt a typical value (for T~70 K) of kpg—2x107"" cm? s! (McCalletal. 2004)."," Assuming for the moment that $k_{DR}$ is independent of the nuclear spin modification, we adopt a typical value (for $T\sim70$ K) of $k_{DR}=2 \times 10^{-7}$ $^3$ $^{-1}$ \citep{mccall2004}."903". The ratio can be rewritten as f/2x., where zx, is the electron n(H2)/n(e7fraction,) typically 1.5x1074 assuming charge neutrality and that C* is the dominant ionic species (Cardellietal.1996;Sofia2004)."," The ratio $n(\textrm{\hh})/n(e^-)$ can be rewritten as $f/2x_e$, where $x_e$ is the electron fraction, typically $1.5 \times 10^{-4}$ assuming charge neutrality and that $^+$ is the dominant ionic species \citep{cardelli1996,sofia2004}."904". If we adopt f=0.9, and take k,, to be the full collision rate of + H» (1.5x 107?em? s7!;1987), we Hj.find that Neen~20."," If we adopt $f=0.9$, and take $k_{rc}$ to be the full collision rate of $_3^+$ + $_2$ \citep[$1.5 \times 10^{-9}$ $^3$ $^{-1}$, we find that ${\cal N}_{rxn} \sim 20$."905" However, if we instead adopt the smaller reactive rate coefficient ~3x10-19 cm? s~! of Gerlichetal. we find that Nyon~5."," However, if we instead adopt the smaller reactive rate coefficient $\sim 3 \times 10^{-10}$ $^3$ $^{-1}$ of \cite{gerlich2002}, , we find that ${\cal N}_{rxn} \sim 5$."906" With such a small number of (2002),,collisions in the lifetime ofHi,, ps may not reach the value predicted by equation B4.."," With such a small number of collisions in the lifetime of, $p_3$ may not reach the value predicted by equation \ref{eqp3}. ."907" In the appendix we derive a more complete steady state expression (equation C7)) including the effects of both the Hj + Ha reaction and nuclear-spin-dependent DR rates (Κερ and ke, for aand We"," In the appendix we derive a more complete steady state expression (equation \ref{ssp3}) ) including the effects of both the $_3^+$ + $_2$ reaction and nuclear-spin-dependent DR rates $k_{e,p}$ and $k_{e,o}$ for and )."908" o-Hj)).call this model simply the “steady state"" model, and we adopt the values f—0.9 and z,—1.5x1074 as before."," We call this model simply the “steady state” model, and we adopt the values $f=0.9$ and $x_e = 1.5 \times 10^{-4}$ as before."909 Figure 6 shows the results of this steady state model if we assume that the DR rate coefficient is the same for both nuclear spin modifications (weCalletal. 2004)., Figure \ref{p3p2_4} shows the results of this steady state model if we assume that the DR rate coefficient is the same for both nuclear spin modifications \citep[we have adopted the temperature-dependent value of][]{mccall2004}.910". In this case, the values of p3 depend quite sensitively on 4, as this represents the fraction of ++ ccollisions that are nonreactive during the relatively short lifetime ofH}."," In this case, the values of $p_3$ depend quite sensitively on $S^{id}$, as this represents the fraction of + collisions that are nonreactive during the relatively short lifetime of."911". Consequently with higher values of S4 circles in Figure 6)), the ffraction in (largersteady state is closer to the p-Hjnascent fraction."," Consequently with higher values of $S^{id}$ (larger circles in Figure \ref{p3p2_4}) ), the fraction in steady state is closer to the nascent fraction."912" For S*¢=0.9, which corresponds to a reactive rate coefficient of γε=1.9x10-19 cm? s-!, the calculated pa are in reasonable agreement with most of the observations."," For $S^{id}=0.9$, which corresponds to a reactive rate coefficient of $k_{rc}=1.9\times10^{-10}$ $^{3}$ $^{-1}$, the calculated $p_3$ are in reasonable agreement with most of the observations."913 The range of the X Per uncertainty is consistent with S'4upper=0.7., The upper range of the X Per uncertainty is consistent with $S^{id}=0.7$.914" In Figure 7,, we instead consider the calculated DR rate coefficients for aandpara-Hj presented in dosSantosetal.(2007)."," In Figure \ref{p3p2_5}, we instead consider the calculated DR rate coefficients for and$_3^+$ presented in \cite{santos2007}."915". Their prediction is that iis destroyed considerably p-HZfaster by electrons at low temperatures than o-Hj;; consequently, even for large values of S4, the steady state padoes not approach either the nascent or astronomically-observed values."," Their prediction is that is destroyed considerably faster by electrons at low temperatures than ; consequently, even for large values of $S^{id}$ , the steady state $p_3$does not approach either the nascent or astronomically-observed values."916" In the absence of the Hj + H» reaction (S7= 1), ps wouldbe governed by a steady state determined by the competition between the formation and the spin- DR processes, and this is shown inFigure 8.."," In the absence of the $_3^+$ + $_2$ reaction $S^{id}=1$ ), $p_3$ wouldbe governed by a steady state determined by the competition between the formation and the spin-dependent DR processes, and this is shown inFigure \ref{p3_noh3h2}. ."917" If the calculated rate coefficients of dosSantosetal. are correct, it is difficult to explain the observed"," If the calculated rate coefficients of \cite{santos2007} are correct, it is difficult to explain the observed"918"for T1-3 spectral types, the distance estimates to ULAS J0128—0041, ULAS J0226+0051, and ULAS J0321+0051 are 50 + 10 pc, 70 + 20 pc, and 110 + 30 pc, respectively.","for T1–3 spectral types, the distance estimates to ULAS $-$ 0041, ULAS $+$ 0051, and ULAS $+$ 0051 are 50 $\pm$ 10 pc, 70 $\pm$ 20 pc, and 110 $\pm$ 30 pc, respectively."919 The lowerlimit of the estimate for ULAS J0226+0051 is consistent with 49 + 9 pc presented by Liuetal., The lowerlimit of the estimate for ULAS $+$ 0051 is consistent with 49 $\pm$ 9 pc presented by \citet{liu02}.920" As compiled by Kakazuetal.(2010),, there are only a (2002)..handful of spectroscopically confirmed T dwarfs known today at distances beyond 60 pc, and only a few beyond 100 pc."," As compiled by \citet{kakazu10}, there are only a handful of spectroscopically confirmed T dwarfs known today at distances beyond 60 pc, and only a few beyond 100 pc."921 Therefore the objects presented here are among the most distant T dwarfs., Therefore the objects presented here are among the most distant T dwarfs.922" The three dwarfs are found in the SDSS southern equatorial stripe, where the intensive repeat scans were carried out in the SDSS-II Supernova Survey."," The three dwarfs are found in the SDSS southern equatorial stripe, where the intensive repeat scans were carried out in the SDSS-II Supernova Survey."923 The dense temporal coverage over several years allows us a simple time-series analysis of the objects., The dense temporal coverage over several years allows us a simple time-series analysis of the objects.924 We retrieve all the z- images of the target fields from the SDSS DR 7 archive (Abazajianetal. and create the stacked images of each year., We retrieve all the $z$ images of the target fields from the SDSS DR 7 archive \citep{abazajian09} and create the stacked images of each year.925 The stacking2009) is not performed when less than five images are available in a year., The stacking is not performed when less than five images are available in a year.926" It results in the deep images of the years 2002, 2005, 2006, 2007 for ULAS J0128—0041, years 2002, 2003, 2005, 2006, 2007 for ULAS J0226+0051, and years 2002, 2005, 2006, 2007 for ULAS J0321+0051."," It results in the deep images of the years 2002, 2005, 2006, 2007 for ULAS $-$ 0041, years 2002, 2003, 2005, 2006, 2007 for ULAS $+$ 0051, and years 2002, 2005, 2006, 2007 for ULAS $+$ 0051."927" We measure the coordinates and magnitudes of the objects with theExtractor, version 2.5 (Bertin&Arnouts if the targets are detected on the stacked images 1996),(more than four adjacent pixels above 1.50 of the local background are defined as detections)."," We measure the coordinates and magnitudes of the objects with the, version 2.5 \citep{bertin96}, , if the targets are detected on the stacked images (more than four adjacent pixels above $\sigma$ of the local background are defined as detections)."928 The results for the detected sources are summarized in Table 4.., The results for the detected sources are summarized in Table \ref{tab:timeseries}. .929" Unfortunately, ULAS J0321+0051 is toofaint to be detected on any"," Unfortunately, ULAS $+$ 0051 is toofaint to be detected on any"930 , 931typical weather conditions and resulting map cleXhs of each SILABC-LL field are given in Table 1..,typical weather conditions and resulting map depths of each SHARC-II field are given in Table \ref{tab:sharc}.932" ὃν design. our strategy was to reach a map RAIS of 3a=30mJv. at 350jim. sullicient to detect or dace useful limits on an Sx,~δαν SMCG with Z4 4019 at -JDS"," By design, our strategy was to reach a map RMS of $3\sigma\simeq 30\,\mathrm{mJy}$ at $350\,\mathrm{\mu m}$, sufficient to detect or place useful limits on an $S_{850}\sim8\,\mathrm{mJy}$ SMG with $T_\mathrm{d}\simeq40$ K at $z\lesssim3$."933 Good weather is scarce. and it is import:ui to be as ellicien as possible by only integrating down to 1e planned noise Ievel.," Good weather is scarce, and it is important to be as efficient as possible by only integrating down to the planned noise level."934" We tracked the ellective integrating time on cach source by compensating cach LO minute file for zeith angle and atmospheric opacity at 350μα. as inferre from the TALUQun and mec, dippers."," We tracked the effective integrating time on each source by compensating each 10 minute file for zenith angle and atmospheric opacity at $350\,\mathrm{\mu m}$, as inferred from the $\tau_{\mathrm{350\,\mu m}}$ and $\tau_{\mathrm{225\,GHz}}$ dippers."935 See Appendix A or details., See Appendix \ref{effective} for details.936 ‘TypicaIv. 34 hours of observations were requires [or a map Saremin:l.5arcmin in size.," Typically, 3–4 hours of observations were required for a map $3\,\mathrm{arcmin}\times1.5\,\mathrm{arcmin}$ in size."937 Pointing. focus checks. ancl calibration were performed hourly on standard sources in close proximity to he science targets.," Pointing, focus checks, and calibration were performed hourly on standard sources in close proximity to the science targets."938 The same scan pattern was used as for he science targets. but with integration times of only 120.160 seconds (as typical Uux densities are 2v).," The same scan pattern was used as for the science targets, but with integration times of only 120–160 seconds (as typical flux densities are $\gtrsim2\,\mathrm{Jy}$ )."939 The typical pointing RAIS of 3aaresec a the CSO is a non-neeligible fraction. of the Qaaresee beanisize. and much care is taken to minimise pointing errors.," The typical pointing RMS of arcsec at the CSO is a non-negligible fraction of the arcsec beamsize, and much care is taken to minimise pointing errors."940 Observations of roint-like galaxies. quasars. protostcllar sources. LILLE regions. and evolved stars are used. by the CSO stall for constructing a pointing model.," Observations of point-like galaxies, quasars, protostellar sources, HII regions, and evolved stars are used by the CSO staff for constructing a pointing model."941 The model predictions for the calibrators observed. before ancl ater science observations are compared to the actual pointing measurements: olfsets are calculated and applied to the mocel pointing predictions for he science observations during he map coaccdition stage olt1e reduction., The model predictions for the calibrators observed before and after science observations are compared to the actual pointing measurements; offsets are calculated and applied to the model pointing predictions for the science observations during the map coaddition stage of the reduction.942 This procedure vields a pointing accuracy RAIS of ~2aresec (CD. Dowell. private communication)," This procedure yields a pointing accuracy RMS of $\sim2\,\mathrm{arcsec}$ (D. Dowell, private communication)."943 Flux calibration is performed w comparing the known and measured [lux densities and beamsizes obtained [or standard calibration sources., Flux calibration is performed by comparing the known and measured flux densities and beamsizes obtained for standard calibration sources.944 For sources. in the SADE. oCeti (a Mira. variable) ane occasionally OII231 (a proto-planetary. nebula) were used. as pointing and calibration sources.," For sources in the SXDF, oCeti (a Mira variable) and occasionally OH231 (a proto-planetary nebula) were used as pointing and calibration sources."945 For sources in the LII. CET6 (an evolved star) was used. and the nearby asteroids Pallas and Egeria when CETG was unavailable.," For sources in the LH, CIT6 (an evolved star) was used, and the nearby asteroids Pallas and Egeria when CIT6 was unavailable."946 The standards have well-tabulated 350jun Hux densities. which are available from the calibration webpagel.," The standards have well-tabulated $350\,\mathrm{\mu m}$ flux densities, which are available from the calibration web."947. The final calibraticn ds expected to be better than 15 per cent. with systematic ellects being negligible at that level (CSO stall. private communication)," The final calibration is expected to be better than 15 per cent, with systematic effects being negligible at that level (CSO staff, private communication)."948 We make maps from the raw data ancl then extract 350pm Iluxes from the maps.," We make maps from the raw data and then extract $350\,\mathrm{\mu m}$ fluxes from the maps."949 The map-making cata reduction package is (Comprehensive Reduction Utility for ΗΛΙΟ]. a Java-basecl tool developed. by. Ἱνονασς(2006)...," The map-making data reduction package is (Comprehensive Reduction Utility for SHARC-II), a Java-based tool developed by \citet{Kovacsthesis}."950 The software iterates a least-squares algorithm to solve. for celestial emission along with instrumental and atmospheric contributions to the total signal., The software iterates a least-squares algorithm to solve for celestial emission along with instrumental and atmospheric contributions to the total power signal.951 accesses the polvnomialpower fits to obtain a low noise 350pam sky opacitv-basedT350j;au estimate of the atmospheric signal.," accesses the $\tau_{\mathrm{350\,\mu m}}$ polynomial fits to obtain a low noise $350\,\mathrm{\mu m}$ sky opacity-based estimate of the atmospheric signal."952 Using the μεν. as is recommenced for sources fainter than ‘deep?100mv. the maps of cach field. are coaddecd on a grid. of L.G2aresee pixels.," Using the `deep' utility, as is recommended for sources fainter than $100\,\mathrm{mJy}$, the maps of each field are coadded on a grid of $1.62\,\mathrm{arcsec}$ square pixels."953 The outer four rows of pixels are removed automaticallysquare: bv since they are not sampled sulficientIv well to converge lo useful nieastwements., The outer four rows of pixels are removed automatically by since they are not sampled sufficiently well to converge to useful measurements.954 The data are fitted By single.Ciaussian xam. profile with a ENVIILNl of 9aarcsec., The data are fitted with a singleGaussian beam profile with a FWHM of arcsec.955 Thumbnails of the reduced maps centred. on the SILADISZS S50pm positions are shown in Lig. 1.," Thumbnails of the reduced maps centred on the SHADES $850\,\mathrm{\mu m}$ positions are shown in Fig. \ref{fig:thumb}."956 The bulk of the structure in these images is detector noise., The bulk of the structure in these images is detector noise.957 Poaks are identified in. maps of the signal-to-noise ratio (S/N) using an algorithm which only aCCODls SOULCCS separated by at least 3neu=11.5 aresec.," Peaks are identified in maps of the signal-to-noise ratio (S/N) using an algorithm which only accepts sources separated by at least $3\,\sigma_{\mathrm{beam}}\!=\!11.5\,\mathrm{arcsec}$ ."958 The maps are, The maps are9595mmag) can harborLyrn/Lp ratios even larger than found in the WR region.,mag) can harbor$L_{MIR}/L_B$ ratios even larger than found in the WR region.960" As an example we included in reffig:lir,b-ompthecaseofthe“hypernebula” starburstenshroudzalixübgcoreo f 55253( Alonso—Herre"," As an example we included in \\ref{fig:lir_lb_comp} the case of the “hypernebula"" starburst enshrouded in the core of 5253 \citep{Alonso04b} as well as the massive star clusters identified in 1365 by \citet{Galliano08}."961roetal.2004)aswe," These complexes of star formation share strong similarities with the WR region in the 980425 host in terms of SFR, age and luminosities."962llast! to—infraredcolors.," Yet, the optical depth toward these sources is much more significant, explaining their extreme optical-to-infrared colors."963Assumingthattheobscurationandthel R/optlndledqithesqbyopeibibecrdsserwe," Assuming that the obscuration and the IR/optical luminosity ratio decrease with the starburst age, what we are seeing in the WR region could thus be one or several star clusters that have already evolved and partly escaped the original molecular cloud where they were born."964dhthebhis NGC benvitwganedtatweare see: formingregionviatheiremissionoflong − lastingsupersonicwinds.," Alternatively, and rather than an “aging"" effect, the difference of visual extinction between the WR region of the 980425 host and the enshrouded clusters of 5253 and 1365 could also be a direct consequence of the presence of the WR stars themselves, which may have accelerated the process of clearing out the local ISM of the star-forming region via their emission of long-lasting supersonic winds."965Dependingongeometrye ffectssuchasnemaondyotoaheh dho&+Radsl whileastill Rwavelengthscouldarise fromamorphouscar bongrainsheat distances ipuis bRyspectaescopst aesdumemnatih gthespectr: le," Depending on geometry effects such a scenario could lead to small silicate absorption as observed, while a still important emission of very hot dust at mid-IR wavelengths could arise from amorphous carbon grains heated to very high temperatures by the massive stars dominating the spectrum of the WR region."966"m As demonstrated in 44, the most striking result of our analysis resides in the bolometric luminosity of the 9980425 host, half of which originates from a single and compact star-forming complex located at more than 2kkpc from the galaxy center."," 1cm As demonstrated in 4, the most striking result of our analysis resides in the bolometric luminosity of the 980425 host, half of which originates from a single and compact star-forming complex located at more than kpc from the galaxy center."967" T'his large contribution of extra-nuclear star formation has also been observed in à few other spiral galaxies of the local Universe (e.g.,NGC5257,6670,Arp256,Haanetal.2011),, although these cases are rare and they have much higher mass and SFR than the host of 9980425."," This large contribution of extra-nuclear star formation has also been observed in a few other spiral galaxies of the local Universe \citep[e.g., NGC\,5257, NGC\,6670, Arp\,256,][]{Haan11}, although these cases are rare and they have much higher mass and SFR than the host of 980425."968" Most often, the star-forming activity of morphologically-evolved systems in the luminosity range of LGRB hosts appears to be preferentially distributed among a large number of HII regions or centrally concentrated toward ring-like structures of multiple knots and star clusters."," Most often, the star-forming activity of morphologically-evolved systems in the luminosity range of LGRB hosts appears to be preferentially distributed among a large number of HII regions or centrally concentrated toward ring-like structures of multiple knots and star clusters."969 'The fact that this atypical configuration is found in a galaxy initially selected as the host of a rare event such as a GRB raises therefore the question of a possible link between the physical properties of the WR region and the trigger of the long Gamma-Ray Bursts in star-forming l, The fact that this atypical configuration is found in a galaxy initially selected as the host of a rare event such as a GRB raises therefore the question of a possible link between the physical properties of the WR region and the trigger of the long Gamma-Ray Bursts in star-forming galaxies.970ook similar to a number of specific characteristics iostateddmsiesptitowasntek," Indeed, the properties observed in this environment look similar to a number of specific characteristics commonly found in GRB host galaxies at cosmological distances."971"egieal pheanice the SSFR of the WR region both confirm the presence of extremely young star formation (ie., « MMyr) in this area."," First, the mid-IR spectroscopic features and the SSFR of the WR region both confirm the presence of extremely young star formation (i.e., $<$ Myr) in this area."972" Assuming a constant star formation rate, the SSFR that we derived in refsec:ir,frindicatesthatthiscomplexo fmassivestar formationwoul fts"," Assuming a constant star formation rate, the SSFR that we derived in \\ref{sec:ir_sfr} indicates that this complex of massive star formation would double its stellar mass in only Myr."973(Christensenet al.2004;Thor forminggalaxiesatsimilarredshi so," This is extremely short compared to the typical time scale of galaxy evolution, and although it is not as extreme in other GRB host galaxies it is consistent with the higher SSFR displayed by this population when compared to field star-forming galaxies at similar redshifts \citep{Christensen04,Thoene08b}."974"larmetallicity, onlymodestextinctionbydustandthere foreveryblu [NIT/Ho,"," Second, the WR region has sub-solar metallicity, only modest extinction by dust and therefore very blue optical colors \citep{Fynbo00,Hammer06,Christensen08,Michalowski09} similar to what is typically observed for GRB hosts at higher redshifts \citep{Fruchter99,LeFloch03,Han10}."975 [SII]/Ha and ," Among the numerous HII environments detected in the 980425 host it also harbors the largest surface brightness region of the galaxy, which is consistent with the link that was established by \citet{Fruchter06} between the location of GRBs and the very brightest regions of the hosts."976[NII]/[OIT| suggest that the ionizing conditions and the stellar populations in the WR region are very similar to that observed in more distant and unresolved GRB host galaxies (Christensenetal.2008).," Finally, optical emission line ratios like $\beta$, $\alpha$, $\alpha$ and [NII]/[OII] suggest that the ionizing conditions and the stellar populations in the WR region are very similar to that observed in more distant and unresolved GRB host galaxies \citep{Christensen08}."977". Without claiming a direct link between the birth of the GRB9980425 progenitor and the WR region, it would seem therefore difficult from all of these observations to argue a kind of causal link between the two."," Without claiming a direct link between the birth of the 980425 progenitor and the WR region, it would seem therefore difficult from all of these observations to argue a kind of causal link between the two."978" Besides, and contrary to the characteristics of the WR region, it is worth noting that the integrated properties of the 9980425 host galaxy differ from those of other GRB hosts at high redshift."," Besides, and contrary to the characteristics of the WR region, it is worth noting that the integrated properties of the 980425 host galaxy differ from those of other GRB hosts at high redshift."979" For example, the host of 9980425 is an already-evolved galaxy that is still forming stars, but at a very low rate with respect to the stellar mass that it has already built etal.2009;Michalowski2009)."," For example, the host of 980425 is an already-evolved galaxy that is still forming stars, but at a very low rate with respect to the stellar mass that it has already built \citep{Savaglio09,Michalowski09}."980. Furthermore(Savaglio the line ratio measured in the other HII regions[OIII]/H8 of the 9980425 host are substantially smaller than what is typically observed in more distant LGRB host galaxies (Christensenetal. 2008)., Furthermore the $\beta$ line ratio measured in the other HII regions of the 980425 host are substantially smaller than what is typically observed in more distant LGRB host galaxies \citep{Christensen08}.981. These differences further emphasize the connection that must exist between the episode of star formation currently taking place in the WR region and the trigger of the 9980425 in this galaxy., These differences further emphasize the connection that must exist between the episode of star formation currently taking place in the WR region and the trigger of the 980425 in this galaxy.982" Nonetheless, the most unsettling fact is that 9980425 occur within the WR region but at a projected distance of ~9900 ppc away."," Nonetheless, the most unsettling fact is that 980425 occur within the WR region but at a projected distance of $\sim$ pc away."983 What can then be the physical connection between the?, What can then be the physical connection between the?984? Long GRBs represent extremely rare phenomena which are more likely to occur from large populations of massive stars., Long GRBs represent extremely rare phenomena which are more likely to occur from large populations of massive stars.985" Given the very low density of the latter in the area where 9980425 was found, and considering the presence of numerous Wolf-Rayet and OB stars in the WR region, Hammeretal.(2006) suggested a “run-away scenario"" in which the 9980425 progenitor may have come from the WR region itself."," Given the very low density of the latter in the area where 980425 was found, and considering the presence of numerous Wolf-Rayet and OB stars in the WR region, \citet{Hammer06} suggested a “run-away scenario” in which the 980425 progenitor may have come from the WR region itself."986" This progenitor would have been dynamically ejected after a kick or close stellar encounters, and traveled at velocities — ss~! across the ISM of the galaxy before ending its life as a GRB."," This progenitor would have been dynamically ejected after a kick or close stellar encounters, and traveled at velocities $\sim$ $^{-1}$ across the ISM of the galaxy before ending its life as a GRB."987" Very massive runaway stars with such velocities are actually expected from numerical simulations of dynamical encounters in stellar clusters (e.g.,Gvaramadze&Guala"," Very massive runaway stars with such velocities are actually expected from numerical simulations of dynamical encounters in stellar clusters \citep[e.g.,][]{Gvaramadze11}. ."988"ndris In the mid- the detection of very high-excitation2011). transition lines such as [OIV] at can indicate the presence of very hot Wolf-Rayet stars (Schaerer&Stasitiska or recent supernovae, which in the case of the WR region1999) could further strengthen this possible connection with"," In the mid-infrared, the detection of very high-excitation transition lines such as [OIV] at can indicate the presence of very hot Wolf-Rayet stars \citep{Schaerer99b} or recent supernovae, which in the case of the WR region could further strengthen this possible connection with"989mass-ratios.,mass-ratios.990 This is due to the fact that the calculated era of rapid mass transfer is very short so that the binary rushes through the states with large mass-ratios., This is due to the fact that the calculated era of rapid mass transfer is very short so that the binary rushes through the states with large mass-ratios.991 Binary evolutionary calculations yielding eras of rapid mass transfer lasting for a longer time with somewhat lower peak values of the Nass transfer rate would produce more Algols with large mass-ratios but have never been published., Binary evolutionary calculations yielding eras of rapid mass transfer lasting for a longer time with somewhat lower peak values of the mass transfer rate would produce more Algols with large mass-ratios but have never been published.992 We have compared the Nass transfer rates as obtained by our binary evolutionary code in the conservative mode with mass transfer rates for conservative evolution as produced by previous authors., We have compared the mass transfer rates as obtained by our binary evolutionary code in the conservative mode with mass transfer rates for conservative evolution as produced by previous authors.993 According to Kippenhahn et al. (1967a.. b))," According to Kippenhahn et al. \cite{Kippenhahn1}, , \cite{Kippenhahn2}) )"994 a 9 M. donor transfers more than 5 M. to his initial 5 M. gainer in 6 x 10? years during hydrogen core burning of the donor and almost 7 Ms in 4 x I0* years when RLOF starts after exhaustion of hydrogen in the core of the donor., a 9 $M_{\odot}$ donor transfers more than 5 $M_{\odot}$ to his initial 5 $M_{\odot}$ gainer in 6 $\times$ $10^4$ years during hydrogen core burning of the donor and almost 7 $M_{\odot}$ in 4 $\times$ $10^4$ years when RLOF starts after exhaustion of hydrogen in the core of the donor.995 A2 M donor transfers 0.45 M. to his 1 M. companion in 3.1 x 10° years in the rapid phase of mass transfer during hydrogen core burning of the donor., A2 $M_{\odot}$ donor transfers 0.45 $M_{\odot}$ to his 1 $M_{\odot}$ companion in 3.1 $\times$ $10^5$ years in the rapid phase of mass transfer during hydrogen core burning of the donor.996 Paezyfisski et al. (1967a.. b))," Paczyńsski et al. \cite{Paczynski1}, \cite{Paczynski2}) )"997 calculated. the conservative evolution of a binary with a 16 M. future donor at birth and a 10.67 M. companion., calculated the conservative evolution of a binary with a 16 $M_{\odot}$ future donor at birth and a 10.67 $M_{\odot}$ companion.998 With an initial orbital period leading to case A RLOF he finds that almost 8 M. are transferred to the gainer in 4 x 10 years., With an initial orbital period leading to case A RLOF he finds that almost 8 $M_{\odot}$ are transferred to the gainer in 4 $\times$ $10^4$ years.999 When RLOF starts after exhaustion of hydrogen in the core of the donor a short era of mass of mass transfer is found with a peak value as high. as 3.44 x 1077_ 2.AM, When RLOF starts after exhaustion of hydrogen in the core of the donor a short era of mass of mass transfer is found with a peak value as high as 3.4 $\times$ $10^{-4}$ $M_{\odot}\over year$.1000 Our calculated durations. of rapid mass transfer are very similar to those mentioned above whereas our peak values are somewhat lower., Our calculated durations of rapid mass transfer are very similar to those mentioned above whereas our peak values are somewhat lower.1001 This is due to the fact that our stellar models are calculated with Rogers- Iglesias opacities (1992)) which were not available previously., This is due to the fact that our stellar models are calculated with Rogers- Iglesias opacities \cite{Rogers}) ) which were not available previously.1002 Therefore. our calculated durations and peak values of mass transfer rates agree very well with those as published by Nelson and Eggleton (2001)) for a representative set of interacting binaries.," Therefore, our calculated durations and peak values of mass transfer rates agree very well with those as published by Nelson and Eggleton \cite{Nelson}) ) for a representative set of interacting binaries."1003 The occurrence of many observed Algols with large mass-ratios thus remains unexplained., The occurrence of many observed Algols with large mass-ratios thus remains unexplained.1004 Future investigations should explore other interactions between the gravitational RLOF and the internal thermal structure driving the evolution of the radius of the donor., Future investigations should explore other interactions between the gravitational RLOF and the internal thermal structure driving the evolution of the radius of the donor.1005 Podsiadlowki et al. (1992)), Podsiadlowki et al. \cite{Podsiadlowski}) )1006 pointed out that liberal theoretical caleulations depend very much on the amount of mass lost from the system (characterized by the parameter P) and the amount of angular momentum taken away by this matter (characterized by the parameter a)., pointed out that liberal theoretical calculations depend very much on the amount of mass lost from the system (characterized by the parameter $\beta$ ) and the amount of angular momentum taken away by this matter (characterized by the parameter $\alpha$ ).1007 Our liberal code calculates f(t) self-consistently within the model and assumes that matter is lost from the hot spot on the gamer (or edge of its aceretion disk) so that the escaping matter takes only the angular momentum of the gainer’s orbit., Our liberal code calculates $\beta$ (t) self-consistently within the model and assumes that matter is lost from the hot spot on the gainer (or edge of its accretion disk) so that the escaping matter takes only the angular momentum of the gainer's orbit.1008 [tis clear that if matter would escape at another location (another choice of the parameter a. e.g. characteristic for L5» as the position of mass loss from the system) the calculated populationof Algols could be different.," It is clear that if matter would escape at another location (another choice of the parameter $\alpha$ , e.g. characteristic for $L_{2}$ as the position of mass loss from the system) the calculated populationof Algols could be different."1009"Applying the correction as explained in Sect. 4,","Applying the correction as explained in Sect. \ref{method},"1010 the 8.5 hours of MAGIC data of PG11553+113 from July 2006 were analyzed., the 8.5 hours of MAGIC data of 1553+113 from July 2006 were analyzed.1011" Fitting a power law to the differential spectrum, a flux of (1.4+0.3)-1079phTeV-7!s!m? at 200GGeV and a spectral index of —4.1+0.3 were determined."," Fitting a power law to the differential spectrum, a flux of $(1.4\pm0.3)\cdot10^{-6}~{\rm ph\,TeV^{-1}s^{-1}m^{-2}}$ at GeV and a spectral index of $-4.1\pm0.3$ were determined."1012 A daily light curve shows a flux consistent with a constant flux within the errors., A daily light curve shows a flux consistent with a constant flux within the errors.1013 The results were shown in detail in ? and were used for MWL studies by ?, The results were shown in detail in \citet{magic1553-2} and were used for MWL studies by \citet{mwl}.1014" Correcting the flux for the effect of the SAL, the systematic uncertainty in the method of 0.2.10-6phTeV7!s-!m-? in the absolute flux."," Correcting the flux for the effect of the SAL, the systematic uncertainty in the method of $0.2\cdot10^{-6}~{\rm ph\,TeV^{-1}s^{-1}m^{-2}}$ in the absolute flux."1015Unike in t1e case of most other non-thermal radio sources. the polarization of the radiation frou pulsars plaved an earvy and fundamental role in attempts to understand and mode the operative enission mechanism,"Unlike in the case of most other non-thermal radio sources, the polarization of the radiation from pulsars played an early and fundamental role in attempts to understand and model the operative emission mechanism."1016 The high percentage of linear polarization. well over the maxima theoretica lit for svuchrotrou radiation. together with a special vpe of systematic sweep of the PLA. observed in the Vela Pulsar led to the “maguetic pole model (Racdhakzishwn Cooke 1969).," The high percentage of linear polarization, well over the maximum theoretical limit for synchrotron radiation, together with a special type of systematic sweep of the P.A. observed in the Vela Pulsar led to the “magnetic pole model"" (Radhakrishnan Cooke 1969)."1017" The sweep of the PLA. across the pulse was interpreted in terius of the line of sight anecutially encounterne different field lines cose to the naenuetic poe as the pulsar rotated: aud the parameters of the so caled S. curve of the P.A, SWOCD have ever since been iuerpreted iu terms of «vy and o. 1c angles uade by the magnetic axis to the τεtational axis aud to he liic of sigelt (a minium mapact angle} res)ectivelv."," The sweep of the P.A. across the pulse was interpreted in terms of the line of sight tangentially encountering different field lines close to the magnetic pole as the pulsar rotated; and the parameters of the so called 'S' curve of the P.A. sweep have ever since been interpreted in terms of $\alpha$ and $\beta$, the angles made by the magnetic axis to the rotational axis and to the line of sight (at minimum impact angle) respectively."1018" An importan poi ids that while the geomerv of the uo Clrve ds intimately related. through «a aux  to the OCIS of the sight line. the actual augle between 16 plane of poarization and the operative nuvenctic fie liue cau ive auv valiο, as long as it remaiis fixed."," An important point is that while the geometry of the 'S' curve is intimately related, through $\alpha$ and $\beta$, to the locus of the sight line, the actual angle between the plane of polarization and the operative magnetic field line can have any value, as long as it remains fixed."1019 Iu he case of sviclirotroji radiation. the most widespreacL eniission nuechanisui invoked for non-thenua ποιαος ICore the discovery of pulsars. the electric vec‘tor of the racliaticna would be perpendicular to the projcEe. mae1οic field. as the acc‘cleration of the charged articles was due to 1Lely evratevration around| the field 1lines.," In the case of synchrotron radiation, the most widespread emission mechanism invoked for non-thermal sources before the discovery of pulsars, the electric vector of the radiation would be perpendicular to the projected magnetic field, as the acceleration of the charged particles was due to their gyration around the field lines."1020 lu tlhe| case of pulsars. the svsteimaties of t1ο ooludzation sweep. and its indepeXxence of observi18o Yequenicv. indicated clearly that the radiation enanated Toni closc| fo the volar cap iu a region that had 10 iuterual Faraday τςstation.," In the case of pulsars, the systematics of the polarization sweep, and its independence of observing frequency, indicated clearly that the radiation emanated from close to the polar cap in a region that had no internal Faraday rotation."1021" The strοicth of the fields associatec with these reelous was so high tlat any Pallsversc omentum aud CLOPSV WOud be raciated away ""justautlv and the charged particles would be in tlaur owest Laida levels aud. constrained {ο move aone t1ο uaenetic feld nes. like beads ou a stiie."," The strength of the fields associated with these regions was so high that any transverse momentum and energy would be radiated away “instantly"", and the charged particles would be in their lowest Landau levels and constrained to move along the magnetic field lines, like beads on a string."1022 An appreciaticna of this constraint led to the suggestionOO (Radhakrishnan 1969) tha the radiation could be due to the acceleration in the plane of the curved field lines. and has been known since then as “curvature radiation.," An appreciation of this constraint led to the suggestion (Radhakrishnan 1969) that the radiation could be due to the acceleration in the plane of the curved field lines, and has been known since then as “curvature radiation""."1023" As the motion of the particOs, whether electrous or positrons. could be oulv alonge the fold lines. the polarizaion of the emitted radiatio1 should have the electric vector paralel to the projected field limes."," As the motion of the particles, whether electrons or positrons, could be only along the field lines, the polarization of the emitted radiation should have the electric vector parallel to the projected field lines."1024 Α οςnsequence of this was t1C identification of the intrinsic plane of polarization at tie centre of the nse (Or Dore coirectlv the inflexion poiut of the S ciive). with the projection of the rotation axis of he pulsar on the sky.," A consequence of this was the identification of the intrinsic plane of polarization at the centre of the pulse (or more correctly the inflexion point of the S curve), with the projection of the rotation axis of the pulsar on the sky."1025 This has had important iuplicatioIs or a varicty of studies over the vears relating to the space velocitics of pulsars., This has had important implications for a variety of studies over the years relating to the space velocities of pulsars.1026 According to the above picture. the PA of the yolarization cal ave one and only one value at any pulse ongitude since the anele of the projected field line is fixed.," According to the above picture, the PA of the polarization can have one and only one value at any pulse longitude since the angle of the projected field line is fixed."1027 But as early as 1975 (Maichester et al., But as early as 1975 (Manchester et al.1028 1975: Backer ct al., 1975; Backer et al.1029 1976) it was cliscoverec that the PA could have mor han one value at a given longitude!, 1976) it was discovered that the PA could have more than one value at a given longitude!1030 Closer investigation revealed that the PA switcred ])otwoeen two modes. talàug any one of two valies which werο orthogonal to cach other (Backer Raulsin 1980).," Closer investigation revealed that the PA switched between two modes, taking any one of two values which were orthogonal to each other (Backer Rankin 1980)."1031 The 4»obluization sweep patter- iu any one mode appearec identical to that iu the other. barring the 90° shi tin PA. Tne has been uo shortage of attempted models or the radiation inechauisa. but in the absence of auv other that could be meannuetully compared with observations. the simple Dicture of the magnetic pole model. with its rules for cderiviug o and 2. has survived for over three decades. despite the blatant sweeping uudoer the rug of the observed freedo not the polarization vector to take oue of two orthogonal νιdues. neither of which was ever shown to have a definite orieitation with respect to the field directiou!!," The polarization sweep pattern in any one mode appeared identical to that in the other, barring the $^o$ shift in P.A. There has been no shortage of attempted models for the radiation mechanism, but in the absence of any other that could be meaningfully compared with observations, the simple picture of the magnetic pole model, with its rules for deriving $\alpha$ and $\beta$, has survived for over three decades, despite the blatant sweeping under the rug of the observed freedom of the polarization vector to take one of two orthogonal values, neither of which was ever shown to have a definite orientation with respect to the field direction!!"1032 We turn now to a cliscussion of sole observations which appear to offer for the first time the possibility of establishing a clear relatiouship between the directions of polarization aud the magnetic field of the pulsar., We turn now to a discussion of some observations which appear to offer for the first time the possibility of establishing a clear relationship between the directions of polarization and the magnetic field of the pulsar.1033 Recent observations of tlie Vela pulsar. and its iunediate surrounudiues. with the Chandra Nouv Observatory show a tfwo-side jet at a position angle," Recent observations of the Vela pulsar, and its immediate surroundings, with the Chandra X-ray Observatory show a two-sided jet at a position angle"1034reflected light) of aat the diagonally opposite corner (near W N) of its real position. but the contamination is confined to a small region where the 3.29 IIEF emission ts faint or absent.,"reflected light) of at the diagonally opposite corner (near W N) of its real position, but the contamination is confined to a small region where the 3.29 IEF emission is faint or absent."1035 Imaging at L’ (broadband 3.8 jpim)) was also attempted during the same night and at the same site. but failed due to the saturation of the array by the high sky background at the 072295 sscale.," Imaging at $L'$ (broadband 3.8 ) was also attempted during the same night and at the same site, but failed due to the saturation of the array by the high sky background at the 295 scale."1036 Images of 1-0 S(1) lline emission at 2.12 aand of the -2.1 ccontinuum emission were derived from three NIR narrowband (1%)) filter images at 2.09jmm.. 2.12jm.. and 2.14jm.," Images of 1–0 S(1) line emission at 2.12 and of the $\sim$ 2.1 continuum emission were derived from three NIR narrowband ) filter images at 2.09, 2.12, and 2.14."1037. These images were obtained on the Perkins 1.8 m Telescope at Lowell Observatory. Flagstatf. Arizona.," These images were obtained on the Perkins 1.8 m Telescope at Lowell Observatory, Flagstaff, Arizona."1038" After standard reduction procedures (sky subtraction and flat field correction). the image of the underlying continuum emission at -2.1 wwas constructed by combining the 2.09 aand 2.14 images, wavelengths which are free of eemission lines (Martinietal.1997.1999)."," After standard reduction procedures (sky subtraction and flat field correction), the image of the underlying continuum emission at $\sim$ 2.1 was constructed by combining the 2.09 and 2.14 images, wavelengths which are free of emission lines \citep{MSH97,MSD99}."1039. To obtain an image of the 1-0 S(1) eemission line. the -2.1 ccontinuum image was subtracted from the 2.12 image after the continuum had been properly scaled to have on average the same counts for ten unsaturated stars in the field. which are believed not to have eemission.," To obtain an image of the 1–0 S(1) emission line, the $\sim$ 2.1 continuum image was subtracted from the 2.12 image after the continuum had been properly scaled to have on average the same counts for ten unsaturated stars in the field, which are believed not to have emission."1040 In addition. Lemaireetal.(1996) kindly provided us with their high-resolution reduced narrowband (1%)) filter images of aat 2.12 aand 2.18pm.," In addition, \citet{LFG96} kindly provided us with their high-resolution reduced narrowband ) filter images of at 2.12 and 2.18."1041. Their images cover all of the field of our 3.29 image of7023., Their images cover all of the field of our 3.29 image of.1042. The images exhibit some instrumental artifacts. notably ghost images of including one near «οὐ N of the star. and a row of bad pixels running ~34” N of200775.," The images exhibit some instrumental artifacts, notably ghost images of including one near $\sim$ N of the star, and a row of bad pixels running $\sim$ N of."1043. We subtracted the 2.18 image from the 2.12 image to create an image of the 1-0 S(1) eemission line., We subtracted the 2.18 image from the 2.12 image to create an image of the 1–0 S(1) emission line.1044 We obtained images of aat K’ on the University of Hawai (UH) 2.2 m telescope at Mauna Kea. Hawai (Sellgrenetal.1992).," We obtained images of at $K'$ on the University of Hawaii (UH) 2.2 m telescope at Mauna Kea, Hawaii \citep{SWD92}."1045. Dome flats were taken and used for the flat field correction., Dome flats were taken and used for the flat field correction.1046 A sky image was constructed from a median-filtered combination of 8 images of offsets at aand ffrom200775., A sky image was constructed from a median-filtered combination of 8 images of offsets at and from.1047". The nebula image was constructed from a mosaic of 17 images at offsets of0""..307.71"".. 100"".. and ffrom200775."," The nebula image was constructed from a mosaic of 17 images at offsets of, and from."1048. The sky image was subtracted from each of the 17 nebula images before constructing the final mosaic image., The sky image was subtracted from each of the 17 nebula images before constructing the final mosaic image.1049 Bad pixel masks were used to correct cosmic rays and failed pixels during the combination., Bad pixel masks were used to correct cosmic rays and failed pixels during the combination.1050 We also obtained archival (HST) images with the Wide Field Planetary Camera 2 of citep*HST.GWROO in F606W (a wide V-band).," We also obtained archival ) images with the Wide Field Planetary Camera 2 \citep[WFPC2;][]{WFPC2}1051 of \\citep*{HST,GWR00} in F606W (a wide $V$ -band)."1052 Between two available WFPC2? fields. the one centered near W N of ccovers the field containing most nebulosity around the northwest filaments on two of its wide-field camera chips (WF3 and WFA).," Between two available WFPC2 fields, the one centered near W N of covers the field containing most nebulosity around the northwest filaments on two of its wide-field camera chips (WF3 and WF4)."1053 The adjacent high-resolution (PC) chip has significant instrumental artifacts from200775.. but these did not cross over to the WF chips.," The adjacent high-resolution (PC) chip has significant instrumental artifacts from, but these did not cross over to the WF chips."1054 For comparison among different images. each was geometrically transformed using IRAF routines to have the same orientation and scale as those of the final 3.29 limage.," For comparison among different images, each was geometrically transformed using IRAF routines to have the same orientation and scale as those of the final 3.29 image."1055 After these geometric registrations. the Images were measured to be aligned with each other betterthan (711 in," After these geometric registrations, the images were measured to be aligned with each other betterthan 1 in"1056"observational viewpoint, given the very low number of identified and well studied host galaxies.","observational viewpoint, given the very low number of identified and well studied host galaxies."1057 Only a few observational studies have attempted to address this question in the past few years., Only a few observational studies have attempted to address this question in the past few years.1058" ? used photometric redshift information in a field of 6’x containing the host galaxy of GRB 000210, and found no obvious galaxy concentration around the host."," \citet{Gorosabel_etal_2003} used photometric redshift information in a field of $6\arcmin\times6\arcmin$ containing the host galaxy of GRB 000210, and found no obvious galaxy concentration around the host."1059 ? analysed the cross-correlation function between host galaxies and surrounding field galaxies using, \citet{bor04} analysed the cross-correlation function between host galaxies and surrounding field galaxies using1060"where we use ος =0.2kin/s(T/10Is)!7.τη To solve the ODE in Equation (20)). we need a relation between r, aud r;.","where If we use $c_s = 0.2~\mathrm{km/s}\left(T/10\mathrm{K}\right)^{1/2}$, To solve the ODE in Equation \ref{drndx}) ), we need a relation between $r_n$ and $r_i$."1061 In the deuse interstellar meclitun. the main source of neutral ionization is Cosmic rays. while ious may recombine in the gas phase. or ou dust grains.," In the dense interstellar medium, the main source of neutral ionization is cosmic rays, while ions may recombine in the gas phase, or on dust grains."1062" The evolution of ion number deusity can be written as Comparing the orders of magnitude of the three coellicients. do,~101—10.19&+ for cosmic ray ionization (Shu1992:Draineetal. 1083)..ag;~LO*—107?ems.+ (TielensTable L11).. and ανω10Dem?s when T—101 (Weingartuer&Draine2001).."," The evolution of ion number density can be written as Comparing the orders of magnitude of the three coefficients, $\zeta_\mathrm{CR}\sim 10^{-17}-10^{-16}~\mathrm{s^{-1}}$ for cosmic ray ionization \citep{1992phas.book.....S, 1983ApJ...264..485D}, ,$\alpha_\mathrm{gas}\sim 10^{-7}-10^{-5}~\mathrm{cm^3 s^{-1}}$ \citep[][Table 4.11]{2005pcim.book.....T}, and $\alpha_\mathrm{grain}\sim 10^{-15}~\mathrm{cm^3 s^{-1}}$ when $T\sim 10~\mathrm{K}$ \citep{2001ApJ...563..842W}."1063" In clouds n;/1,710—* and n,~107—10*em* so we can drop the grain surface recombination term."," In moderate-density clouds $n_i/n_n\sim 10^{-5}-10^{-7}$ and $n_n\sim 10^2-10^3~\mathrm{cm^{-3}}$, so we can drop the grain surface recombination term."1064 The ion balance equation becomes Iu solving Equation (21)). oue possible approximation is to assume lonizatiou-recombiuation equilibrium⋅⋅⋅ everywhere.," The ion balance equation becomes In solving Equation \ref{equilibrium}) ), one possible approximation is to assume ionization-recombination equilibrium everywhere."1065 In this⋅ case. ον830gautt;. so that for where the coefficient. vij1—20 (Mclxeeetal.2010).. . 4⋅ 1/2 ⋅ ⊸ ∐∖∖↽↩⋯⇂∩↥↽≻↕↕↢↵≺↽↓⋯↕⋃∩∐⋜↜⊋⋅≻⊔⋅↕∐↩∐∣↝∣⋅∶∣↝⊔⋅⋜↕∐≺⊓∐≺↵∑≟∩∖⊽≺↵↕⋅∐⋯∑∸⋃∐↕↢↵∣⋈↵∢∙∩⋯≺↵," In this case, $\zeta_\mathrm{CR} n_n \approx \alpha_\mathrm{gas} n_i^2$, so that for where the coefficient $\chi_{i0}\sim 1-20$ \citep{2010ApJ...720.1612M}.. If we adopt Equation \ref{rec-ion}) ), then $r_i = r_n^{1/2}$ , and thegoverning ODE becomes"1066 In this⋅ case. ον830gautt;. so that for where the coefficient. vij1—20 (Mclxeeetal.2010).. . 4⋅ 1/2 ⋅ ⊸ ∐∖∖↽↩⋯⇂∩↥↽≻↕↕↢↵≺↽↓⋯↕⋃∩∐⋜↜⊋⋅≻⊔⋅↕∐↩∐∣↝∣⋅∶∣↝⊔⋅⋜↕∐≺⊓∐≺↵∑≟∩∖⊽≺↵↕⋅∐⋯∑∸⋃∐↕↢↵∣⋈↵∢∙∩⋯≺↵⊳," In this case, $\zeta_\mathrm{CR} n_n \approx \alpha_\mathrm{gas} n_i^2$, so that for where the coefficient $\chi_{i0}\sim 1-20$ \citep{2010ApJ...720.1612M}.. If we adopt Equation \ref{rec-ion}) ), then $r_i = r_n^{1/2}$ , and thegoverning ODE becomes"1067 In this⋅ case. ον830gautt;. so that for where the coefficient. vij1—20 (Mclxeeetal.2010).. . 4⋅ 1/2 ⋅ ⊸ ∐∖∖↽↩⋯⇂∩↥↽≻↕↕↢↵≺↽↓⋯↕⋃∩∐⋜↜⊋⋅≻⊔⋅↕∐↩∐∣↝∣⋅∶∣↝⊔⋅⋜↕∐≺⊓∐≺↵∑≟∩∖⊽≺↵↕⋅∐⋯∑∸⋃∐↕↢↵∣⋈↵∢∙∩⋯≺↵⊳∖," In this case, $\zeta_\mathrm{CR} n_n \approx \alpha_\mathrm{gas} n_i^2$, so that for where the coefficient $\chi_{i0}\sim 1-20$ \citep{2010ApJ...720.1612M}.. If we adopt Equation \ref{rec-ion}) ), then $r_i = r_n^{1/2}$ , and thegoverning ODE becomes"1068our moclels.,our models.1069 Since we are looking mostly at features around 5000.A.. we use the V-band luminosity (L4) aid mass-to-lightD ratio CAZ/ L4.) to obtain our luminosity weightings.," Since we are looking mostly at features around 5000, we use the V-band luminosity $L_V$ ) and mass-to-light ratio $M/L_V$ ) to obtain our luminosity weightings."10705 Fig., Fig.1071" 1 shows the results of the two reference models ""primordial ancl merger (or starburst).", 1 shows the results of the two reference models 'primordial' and 'merger' (or starburst).1072 The latter is a model with a starburst. sarting 5 Gyr ago.," The latter is a model with a starburst, starting 5 Gyr ago."1073 This figure illustrates the time evolution of the models., This figure illustrates the time evolution of the models.1074 The primordial model has a burst of star formation starting 17 Civr ago at a rate consistent with the imescale given by POOL for an elliptical galaxy (which is also similar to the timescale from. Silk 1986)., The primordial model has a burst of star formation starting 17 Gyr ago at a rate consistent with the timescale given by FG94 for an elliptical galaxy (which is also similar to the timescale from Silk 1986).1075 LE star ormation is stopped. after <2 Gyr. as depicted in Silk (1986). then very low metallicities and Ποstrengths result.," If star formation is stopped after $<2$ Gyr, as depicted in Silk (1986), then very low metallicities and line-strengths result."1076 Allowing for some inflow in the first. [ow 100 million vears (as might be expected. during the initial ealaxy formation) can increase the metallicity slightly., Allowing for some inflow in the first few 100 million years (as might be expected during the initial galaxy formation) can increase the metallicity slightly.1077 In the primordial model depicted in Fig., In the primordial model depicted in Fig.1078 1 star formation was allowed to continue up to recent times (still in proportion to the gas density) to allow the metallicity to increase., 1 star formation was allowed to continue up to recent times (still in proportion to the gas density) to allow the metallicity to increase.1079 Even allowing for later. more metal rich stars this initial burst (primordial) model produce the strong absorption lines observed in earlv-tvpe galaxies.," Even allowing for later, more metal rich stars this initial burst (primordial) model produce the strong absorption lines observed in early-type galaxies."1080 This is shown in Fig., This is shown in Fig.1081 2 where line-strengths from elliptical galaxies are compared wit1 these two reference models., 2 where line-strengths from elliptical galaxies are compared with these two reference models.1082 ‘Phe data shown are from. Fisver. Franx Ulineworth (1995 hereafter FEI95) and DS03.," The data shown are from Fisher, Franx Illingworth (1995 – hereafter FFI95) and DSP93."1083 The delaved starburst model (mimicking a later merger of wo gas-rich galaxies) is able to produce stronger lines since the main light is from stars mace from more metal rich material., The delayed starburst model (mimicking a later merger of two gas-rich galaxies) is able to produce stronger lines since the main light is from stars made from more metal rich material.1084 The delaved burst model shown in Fig., The delayed burst model shown in Fig.1085 1 and Fig., 1 and Fig.1086 2 starts 17 Civr ago with two identical She spirals (in terms of their pre-burst. SER)., 2 starts 17 Gyr ago with two identical Sbc spirals (in terms of their pre-burst SFR).1087 These spirals merge after 12 Gr. with aSER typical of üubiir starburst galaxies. and a factor two increase in total mass from in-llowing gas. enriched. to the same extent as the existing ISM.," These spirals merge after 12 Gyr, with a SFR typical of ir starburst galaxies, and a factor two increase in total mass from in-flowing gas, enriched to the same extent as the existing ISM."1088 Fig., Fig.1089 2 shows the time evolution of spectral. features and other stellar parameters in these two reference models ancl illustrates the stronger lines generated in the delaved )urst. (or merger) model (D) compared to the primordial model CX)., 2 shows the time evolution of spectral features and other stellar parameters in these two reference models and illustrates the stronger lines generated in the delayed burst (or merger) model (B) compared to the primordial model (A).1090 Figs 1: and 2 show examples of the stellar and easeous output we can obtain from our models., Figs 1 and 2 show examples of the stellar and gaseous output we can obtain from our models.1091 Phe spectral eatures output ignore any stars formed. more recently than 1.5 Cr ago. since the voungest SSP in W94 is 1.5 Gyr old.," The spectral features output ignore any stars formed more recently than 1.5 Gyr ago, since the youngest SSP in W94 is 1.5 Gyr old."1092" '""herefore. we note that variations shown in Fig.", Therefore we note that variations shown in Fig.1093 2 at the ime of onset of a starburst are unreliable., 2 at the time of onset of a starburst are unreliable.1094 In addition there are no νους («S Gyr). metal poor SSPs in W94. therefore he carly evolution of feature strengths is not shown here.," In addition there are no young $<8$ Gyr), metal poor SSPs in W94, therefore the early evolution of feature strengths is not shown here."1095 Later indices are unallected by this limitation since we are hen seeing old (2S Civr). metal-poor stars and voung (but -1.5 Gyr). metal-rich stars for which SSPs are available.," Later indices are unaffected by this limitation since we are then seeing old $>8$ Gyr), metal-poor stars and young (but $>1.5$ Gyr), metal-rich stars for which SSPs are available."1096 Another version of our main prograni allows the xwanmeter space of two variables to be stepped through: e.g. starburst time and SER., Another version of our main program allows the parameter space of two variables to be stepped through: e.g. starburst time and SFR.1097 By varving these parameters we can search For the best fit of our models to the data., By varying these parameters we can search for the best fit of our models to the data.1098 This is done in Section 4 for five parameters. fitting observed streneths in galaxies.," This is done in Section 4 for five parameters, fitting observed line-strengths in galaxies."1099 Others have compared the SSPs from various authors (V96: W94). so we only compare chemical evolution histories here.," Others have compared the SSPs from various authors (V96; W94), so we only compare chemical evolution histories here."1100 We compared our enrichment model with trat of V96., We compared our enrichment model with that of V96.1101 They have made a complete chemical evolution model. generating their own SSPs and following the chemical enrichment of their. closed. box model with time from. primordial abundances.," They have made a complete chemical evolution model, generating their own SSPs and following the chemical enrichment of their closed box model with time from primordial abundances."1102 Time histories of gas metallicity anc gas miss are compared in Fig., Time histories of gas metallicity and gas mass are compared in Fig.1103 3., 3.1104 To make this comparison we assumed no inflow. no SNla. a Salpeter LAL and an upper mass limit of 60 M. in order to correspond to the values used by V96.," To make this comparison we assumed no inflow, no SNIa, a Salpeter IMF and an upper mass limit of 60 $_{\odot}$ in order to correspond to the values used by V96."1105. We assumed. the same parameters as V96 for the IMS (case A. 5n=1/3 and a=0 from Renzini Voli 1981).," We assumed the same parameters as V96 for the IMS (case A, $\eta=1/3$ and $\alpha=0$ from Renzini Voli 1981)."1106 Three dillerent. SE. clliciencies were compared with those plotted by V96. their figure 11.," Three different SF efficiencies were compared with those plotted by V96, their figure 11."1107 These three SE elliciencies are characterised. by C'—1.92. 0.192 and 0.0192 per Gyr in the SER equation. corresponding tov= 19.2.1.92 and 0.192 per 10 Gar in the terminology of V96).," These three SF efficiencies are characterised by $C$ =1.92, 0.192 and 0.0192 per Gyr in the SFR equation, corresponding to $\nu=19.2$ ,1.92 and 0.192 per 10 Gyr in the terminology of V96)."1108 C'—0.192 is representative of the solar neighbourhood. (Arimoto Yoshii 1986)., $C$ =0.192 is representative of the solar neighbourhood (Arimoto Yoshii 1986).1109 “These input criteria were employed. for this comparison in order to correspond. as closely as possixe to those used by. V96., These input criteria were employed for this comparison in order to correspond as closely as possible to those used by V96.1110 Results from our models are shown at time steps of 0.1 Civr in Fig., Results from our models are shown at time steps of 0.1 Gyr in Fig.1111 3., 3.1112 Some results from the chemical evolution model of VOG are indicated. by. large open circles at a handful of times through the evolution history (estimated from their figure 11)., Some results from the chemical evolution model of V96 are indicated by large open circles at a handful of times through the evolution history (estimated from their figure 11).1113 We show two cases from our models in order also to compare predictions using the M92 ancl WAV massive star mocels: i) FLOSSLIAI=0.0 This means we are using only M92 models for SNILs., We show two cases from our models in order also to compare predictions using the M92 and WW massive star models: i) FLOSSLIM=0.0 This means we are using only M92 models for SNIIs.1114 ii) FLOSSLIAI=1.0 Means we are using WW cata for, ii) FLOSSLIM=1.0 Means we are using WW data for1115and HD 14282 relative to normal stars. matching the Ba stars distributions in both plots. while Nitrogen seems normal for the two stars.,"and HD 14282 relative to normal stars, matching the Ba stars distributions in both plots, while Nitrogen seems normal for the two stars."1116 Compared with the [C/H] vs. [Fe/H] relation of unpolluted stars from Fig., Compared with the [C/H] vs. [Fe/H] relation of unpolluted stars from Fig.1117 11 of Masseron et al. (, 11 of Masseron et al. (11182006) we also found that both are slightly overabundant. although they can be seemingly considered às normal stars 1n this plot.,"2006) we also found that both are slightly overabundant, although they can be seemingly considered as normal stars in this plot."1119 Among the light s-elements (hereafter /5) or first s-peak elements. Y is the purest. with a s-contribution of according to Arlandini et al. (," Among the light $s$ -elements (hereafter $ls$ ) or first $s$ -peak elements, Y is the purest, with a $s$ -contribution of according to Arlandini et al. ("11201999). followed by Sr and Zr. with and83%.. respectively. and for Mo. In Fig.,"1999), followed by Sr and Zr, with and, respectively, and for Mo. In Fig."1121 2 we can see that. except for Zr. the /s elements in HD 11397 and HD 14282 are mildly high when compared to normal stars. with Y showing the highest value in both stars (~ +0.6 dex).," 2 we can see that, except for Zr, the $ls$ elements in HD 11397 and HD 14282 are mildly high when compared to normal stars, with Y showing the highest value in both stars $\sim$ +0.6 dex)."1122 HD 14282 shows only slight enhancements for the other /s-elements., HD 14282 shows only slight enhancements for the other $ls$ -elements.1123 In Figs., In Figs.1124 3 and 4. the distributions for the heavy s-process elements (herafter 9) are depicted.," 3 and 4, the distributions for the heavy $s$ -process elements (herafter $hs$ ) are depicted."1125 For these elements we have found a clear distinction between the two stars. with HD 11397 showing high Ba. Ce. La. Hf. and Nd abundances compared to normal stars with similar metallicities.," For these elements we have found a clear distinction between the two stars, with HD 11397 showing high Ba, Ce, La, Hf, and Nd abundances compared to normal stars with similar metallicities."1126 HD 14282 depicts a normal behavior with only a slight enhancement in its [Ba/Fe] ratio., HD 14282 depicts a normal behavior with only a slight enhancement in its [Ba/Fe] ratio.1127 Bais the /75-element or second s-peak element (see Busso et al., Ba is the $hs$ -element or second $s$ -peak element (see Busso et al.1128" 1999, 2001) with the highest s-contribution. (81%)). followed by Ce (77%)) and La (62%)) according to Arlandinietal. (1999)."," 1999, 2001) with the highest $s$ -contribution ), followed by Ce ) and La ) according to \citet{arl99}."1129 Such elements are clearly above the normal stars distribution level in HD 11397., Such elements are clearly above the normal stars distribution level in HD 11397.1130 Lead is at the termination of the s-process path. or the third peak of the s-," Lead is at the termination of the $s$ -process path, or the third peak of the $s$ -process."1131 The s-contribution for Pb is still uncertain., The $s$ -contribution for Pb is still uncertain.1132 Theoretical estimations predict of yields from the s-process (Arlandini et al., Theoretical estimations predict of yields from the $s$ -process (Arlandini et al.1133" 1999), but observations for very low-metallicity stars indicate a strong s-contribution. although some contribution from the ;-process may still be present (e.g. Van Eck et al."," 1999), but observations for very low-metallicity stars indicate a strong $s$ -contribution, although some contribution from the $r$ -process may still be present (e.g. Van Eck et al."1134 2003. Aoki et al.," 2003, Aoki et al."1135 2000)., 2000).1136 Lead seems normal in both HD 11397 and HD 14282., Lead seems normal in both HD 11397 and HD 14282.1137 Praseodymium. Sm and Ru have a stronger r-contribution. with for Pr. for Sm. while Ru has59%.. but they also have an s- of ~ for Ru and Sm and for Pr.," Praseodymium, Sm and Ru have a stronger $r$ -contribution, with for Pr, for Sm, while Ru has, but they also have an $s$ -contribution of $\sim$ for Ru and Sm and for Pr."1138 Ru ts nearer the first s-peak while Sm and Pr are nearer the second s-peak., Ru is nearer the first $s$ -peak while Sm and Pr are nearer the second $s$ -peak.1139 These elements also show slight enhancements in HD 11397. while HD 14282 seems to have values similar to those of a normal star.," These elements also show slight enhancements in HD 11397, while HD 14282 seems to have values similar to those of a normal star."1140 Dysprosium. Gd and Eu are dominated by the +-process production.," Dysprosium, Gd and Eu are dominated by the $r$ -process production."1141 In. Fig., In Fig.1142 4 their. distributions. are. depicted. and we see that HD 11397 and HD 14282 show a marginally underabundant behavior for Gd and Dy. while Eu abundances overlap the lower envelope of the Ba-stars and the normal stars distributions. indicating a small r-contribution for these stars.," 4 their distributions are depicted and we see that HD 11397 and HD 14282 show a marginally underabundant behavior for Gd and Dy, while Eu abundances overlap the lower envelope of the Ba-stars and the normal stars distributions, indicating a small $r$ -contribution for these stars."1143 In order to improve our analysis on the abundance profile of the s-process elements in HD 11397 and HD 14282. we have compared our data with theoretical surface abundances of AGB stars from Goriely Mowlavi (2000) in Fig. 5..," In order to improve our analysis on the abundance profile of the $s$ -process elements in HD 11397 and HD 14282, we have compared our data with theoretical surface abundances of AGB stars from Goriely Mowlavi (2000) in Fig. \ref{mod_comp}."1144 If those stars inherited their abundance content from an AGB star. their surface s-elements abundances should mimic those of an AGB star.," If those stars inherited their abundance content from an AGB star, their surface $s$ -elements abundances should mimic those of an AGB star."1145 The models are for stars with metallicities similar to our sample stars., The models are for stars with metallicities similar to our sample stars.1146 Solid lines represent surface abundances after 10 dredge-ups. and dotted lines. after 30 dredge-ups.," Solid lines represent surface abundances after 10 dredge-ups, and dotted lines, after 30 dredge-ups."1147 The upper panel shows that most neutron-capture elemets in HD 14282 are below the abundances predicted for à 10 dredge-up model., The upper panel shows that most neutron-capture elements in HD 14282 are below the abundances predicted for a 10 dredge-up model.1148 Therefore that star seems to show a slight abundance anomaly for some neutron-capture elements (Y. Sr. Mo. Pb) while other elements are normal.," Therefore that star seems to show a slight abundance anomaly for some neutron-capture elements (Y, Sr, Mo, Pb) while other elements are normal."1149 Boyarchuk et al. (, Boyarchuk et al. (11502002) have also found some slight s-elements anomaly in normal field red giants.,2002) have also found some slight $s$ -elements anomaly in normal field red giants.1151 If these chemical anomalies in HD 14282 were inherited. from the proto-cloud of the star. such excess should be ascribed to a pristine contamination.," If these chemical anomalies in HD 14282 were inherited from the proto-cloud of the star, such excess should be ascribed to a pristine contamination."1152 If they are due to the beginning of the s-process operation and the dredge-up of material enriched in carbon and s-elements. HD 14282 s-anomaly may be ascribed to a mass-transfer mechanism.," If they are due to the beginning of the $s$ -process operation and the dredge-up of material enriched in carbon and $s$ -elements, HD 14282 $s$ -anomaly may be ascribed to a mass-transfer mechanism."1153 The abundances of HD 11397 for most of the elements in Fig., The abundances of HD 11397 for most of the elements in Fig.1154 5. are between the 10 and the 30 dredge-ups curves. in agreement with an AGB abundance profile.," \ref{mod_comp} are between the 10 and the 30 dredge-ups curves, in agreement with an AGB abundance profile."1155 Compared to normal disk stars. HD 11397 shows an overabundant behavior for most of the neutron capture elements and seems to share its chemical profile with the mild Ba stars.," Compared to normal disk stars, HD 11397 shows an overabundant behavior for most of the neutron capture elements and seems to share its chemical profile with the mild Ba stars."1156 We suggest that this is also a mild Ba-star., We suggest that this is also a mild Ba-star.1157 An apparent higher {1 content than the Is-elements is seen in that star., An apparent higher $hs$ content than the $ls$ -elements is seen in that star.1158 For both HD 11397 and HD 14282. the heavy element Pb. is underabundant relative to these models predictions.," For both HD 11397 and HD 14282, the heavy element Pb, is underabundant relative to these models predictions."1159 Radial velocities of the bulgelike stars have been derived using CORAVEL. and possible binaries have been discarded (Grenon. 1998).," Radial velocities of the bulgelike stars have been derived using CORAVEL, and possible binaries have been discarded (Grenon, 1998)."1160 Nevertheless. a detailed study of the radial velocities of HD 11397 and HD 14282 should be performed in order to have a robust statement about their single nature.," Nevertheless, a detailed study of the radial velocities of HD 11397 and HD 14282 should be performed in order to have a robust statement about their single nature."1161 If their non-binarity is confirmed. they could be used as templates for the study of s-enriched stars with à non-binary origin. anc therefore in à scenario different from that of the mass-transfer paradigm.," If their non-binarity is confirmed, they could be used as templates for the study of $s$ -enriched stars with a non-binary origin, and therefore in a scenario different from that of the mass-transfer paradigm."1162 We have performed a chemical abundance analysis of two dwarf stars with s-process anomalies. HD 11397 and HD 14282.," We have performed a chemical abundance analysis of two dwarf stars with $s$ -process anomalies, HD 11397 and HD 14282."1163 We aim to define if they can be considered as Ba-stars. and. if confirmed. to infer their Ba degree. 1.e.. to define if they are mild or strong Ba stars.," We aim to define if they can be considered as Ba-stars, and, if confirmed, to infer their Ba degree, i.e., to define if they are mild or strong Ba stars."1164 Abundances of 18 neutron-capture elements. with different 5 and/or r processes contributions. have been derived.," Abundances of 18 neutron-capture elements, with different $s$ and/or $r$ processes contributions, have been derived."1165 The resulting abundance ratios of the two stars have been compared to those of normal stars. to abundance ratios of mild and strong Ba stars and to theoretical predictions for AGB stars.," The resulting abundance ratios of the two stars have been compared to those of normal stars, to abundance ratios of mild and strong Ba stars and to theoretical predictions for AGB stars."1166 We have found that HD 11397 shows a mild enhancement for most of the s-process elements as well as for some r-process dominated elements., We have found that HD 11397 shows a mild enhancement for most of the $s$ -process elements as well as for some $r$ -process dominated elements.1167 This star seems to share its abundance profile with the mild Ba-stars., This star seems to share its abundance profile with the mild Ba-stars.1168 Although showing some slight chemical anomalies for Υ. Sr," Although showing some slight chemical anomalies for Y, Sr"1169stellar populations in the NIR requires a hard work (Riffel 2010).,stellar populations in the NIR requires a hard work \citep{riffel10}.1170". Nevertheless, it is clear from Fig."," Nevertheless, it is clear from Fig."1171 2 that even a small fraction of a MMyr population ( 5%)) detected in the NIR may be responsible for almost all the light observed in the NUV (~ 7050)., \ref{fractionsNIR} that even a small fraction of a Myr population $\sim$ ) detected in the NIR may be responsible for almost all the light observed in the NUV $\sim$ ).1172" Clearly, synthesis results should not be directly propagated from the NIR to the NUV/Optical, or vice versa."," Clearly, synthesis results should not be directly propagated from the NIR to the NUV/Optical, or vice versa."1173" Instead, Eqs."," Instead, Eqs."1174 to 4 should be used for this purpose., \ref{eqfrac} to \ref{eqfrac3} should be used for this purpose.1175" To help with such a comparison we have created an on-line form, the Panchromatic Averaged Stellar Population: and make available for download the tables with the results of the above equations (see Appendix A))."," To help with such a comparison we have created an on-line form, the nchromatic veraged tellar opulation: and make available for download the tables with the results of the above equations (see Appendix \ref{appen}) )."1176 Another important ingredient in stellar population fitting is the metallicities used., Another important ingredient in stellar population fitting is the metallicities used.1177" As shown by Chenetal.(2010), the results of the fitting have a weaker dependence on metallicity than age."," As shown by \citet{chen10}, the results of the fitting have a weaker dependence on metallicity than age."1178" The question which arises here is, does metallicity affect the propagation of the averaged stellar populations?"," The question which arises here is, does metallicity affect the propagation of the averaged stellar populations?"1179 We investigate this effect with MOS SSPs with 3 different metallicities (d5Z©; ZO; and Z©) and the same age grid as in Fig. 1.., We investigate this effect with M05 SSPs with 3 different metallicities $\frac{1}{50}Z\odot$; $Z\odot$; and $Z\odot$ ) and the same age grid as in Fig. \ref{fractionsOPT}.1180 The results are shown in Figs., The results are shown in Figs.1181 6 and 7.., \ref{metalopt} and \ref{metalnir}.1182 Clearly the propagation of the contributions has a negligible dependence on metallicity., Clearly the propagation of the contributions has a negligible dependence on metallicity.1183" Thus, one can use the condensed population vectors proposed by CidFernandes(2004,2005) to propagate the fitting results over all As. All the tests described above were made using the M05 models, but, as stated in Sec. ??,,"," Thus, one can use the condensed population vectors proposed by \citet{cid04,cid05} to propagate the fitting results over all $\lambda$ s. All the tests described above were made using the M05 models, but, as stated in Sec. \ref{epsmodels},"1184 there are more EPS models available in the literature covering simultaneously the spectral region between ~3500A aand2., there are more EPS models available in the literature covering simultaneously the spectral region between $\sim$ and.1185"5j:m.. Thus, it is necessary to test if the selection of EPS models will produce different results in the propagation of the synthesis results over different As. In Fig."," Thus, it is necessary to test if the selection of EPS models will produce different results in the propagation of the synthesis results over different $\lambda$ s. In Fig."1186 8 we compare the different models among each other., \ref{comparamodels} we compare the different models among each other.1187" It is clear that in the case of the optical normalisation point (5870A)), the four models produce very similar results in the interval between aandAA,, but a discrepancy between GALEV/MO5 and GRASIL/BCO3 models is observed in the NIR."," It is clear that in the case of the optical normalisation point ), the four models produce very similar results in the interval between and, but a discrepancy between /M05 and /BC03 models is observed in the NIR."1188" Such a discrepancy is due to the well known fact that and M05 models do include stars in the TP-AGB phase (see MOS, for example), which is more sensitive to the NIR than the optical, i.e TP-AGB stars account for 25 to of the bolometric light of an SSP, and for 40 to of the light emitted in the K-band (seeSchulzetal.2002;Maraston2005,andreferences therein).."," Such a discrepancy is due to the well known fact that and M05 models do include stars in the TP-AGB phase (see M05, for example), which is more sensitive to the NIR than the optical, i.e TP-AGB stars account for 25 to of the bolometric light of an SSP, and for 40 to of the light emitted in the K-band \citep[see][and references therein]{schulz02,maraston05}."1189" However, there is a difference between and M05 models, enhanced in the 100 Myr population."," However, there is a difference between and M05 models, enhanced in the 100 Myr population."1190 There are two possible explanations for this discrepancy: one is associated with the different onset age of the TP-AGB on the models., There are two possible explanations for this discrepancy: one is associated with the different onset age of the TP-AGB on the models.1191" A high TP-AGB contribution at MMyr, as applied by Schulzetal. (2002),, which is excessively high when compared to young Large Magellanic Cloud globular clusters (Maraston1998; 2008)."," A high TP-AGB contribution at Myr, as applied by \citet{schulz02}, which is excessively high when compared to young Large Magellanic Cloud globular clusters \citep{maraston98,marigo08}."1192. The other is associated with the way in which the TP-AGB treatment is made (Marastonetal.2006;Bruzual2007).," The other is associated with the way in which the TP-AGB treatment is made \citep{maraston06,bruzual07}."1193". includes TP-AGB by means of isochrones (Padova94+improvedetal. 2008),, while MOS is based on different approach, the fuel consumption theorem."," includes TP-AGB by means of isochrones \citep[Padova94 + improved TP-AGB models][]{bertelli94,girardi00,marigo08}, while M05 is based on a different approach, the fuel consumption theorem."1194" According to aMaraston(2005),, the stellar luminosity during the evolutionary phases that follow or suffer from mass loss cannot be predicted by stellar tracks, because there is no theory linking mass-loss rates to the basic stellar parameters, such as luminosity."," According to \citet{maraston05}, the stellar luminosity during the evolutionary phases that follow or suffer from mass loss cannot be predicted by stellar tracks, because there is no theory linking mass-loss rates to the basic stellar parameters, such as luminosity."1195"Let a;=Rj/Ry, be the dimension-less principal axes of the ellipsoid, whereR; with 1<i3 are its dimensional semi-major axes, and Ay, the size of a spherical top-hat corresponding to a mass M=(4n/3)pyR$. with the cosmological background density py.","Let $a_i=R_i/R_\mathrm{pk}$ be the dimension-less principal axes of the ellipsoid, where$R_i$ with $1\leq i\leq3$ are its dimensional semi-major axes, and $R_\mathrm{pk}$ the size of a spherical top-hat corresponding to a mass $M=(4\pi/3) \rho_\mathrm{b} R_\mathrm{pk}^3$ with the cosmological background density $\rho_\mathrm{b}$ ."1196" The evolution of the three principal axes a; with time f in a cosmology with a cosmological constant A=(81,G/c?)p, is given by (?),, where G is the gravitational constant and c the speed of light."," The evolution of the three principal axes $a_i$ with time $t$ in a cosmology with a cosmological constant $\Lambda=(8\pi G/c^2)\rho_\Lambda$ is given by \citep{Bond1996}, where $G$ is the gravitational constant and $c$ the speed of light."1197 The density contrast of the ellipsoid with respect to the background density is 6=(p—py)/pya[(a1a543)— 1., The density contrast of the ellipsoid with respect to the background density is $\delta=(\rho-\rho_\mathrm{b})/\rho_\mathrm{b}=a^3/(a_1a_2a_3)-1$ .1198 The parameters b; and οχι] denote the internal and external contributions to the gravitational tidal shear which occur because of the deviation from sphericity., The parameters $b_i$ and $\lambda_{\ext}$ denote the internal and external contributions to the gravitational tidal shear which occur because of the deviation from sphericity.1199" Generally, the total tidal field is described by the T with the elements Τη=Op/(Ox;0x;)Timij+Textij, where Gp denotes the peculiar gravitational potential, and Tint; and T,;; are the internal and external contributions to the shear, respectively."," Generally, the total tidal field is described by the $\tens{T}$ with the elements $T_{ij}=\partial^2\Phi_\mathrm{P}/(\partial x_i\partial x_j)=T_{\mathrm{int,}ij}+T_{\mathrm{ext,}ij}$, where $\Phi_\mathrm{P}$ denotes the peculiar gravitational potential, and $T_{\mathrm{int,}ij}$ and $T_{\mathrm{ext,}ij}$ are the internal and external contributions to the shear, respectively."1200" After a transformation into the ellipsoid's eigensystem, which is the same as the eigensystem of T in this model, the internal shear can be evaluated as while the external shear can be approximated by where D, is the linear growth factor, and the 4; are the eigenvalues of the Zel'dovich deformation tensor."," After a transformation into the ellipsoid's eigensystem, which is the same as the eigensystem of $\tens{T}$ in this model, the internal shear can be evaluated as while the external shear can be approximated by where $D_+$ is the linear growth factor, and the $\lambda_i$ are the eigenvalues of the Zel'dovich deformation tensor."

Showing the first 1,200 of 10287 lines. Download the file for the rest.