ReadingTimeMachine/rtm-sgt-ocr-v1

1source,target2 That the cnergctics are modest here imu be a result of the source being very close and viewed from a laree off-axis angle. aud that even the scattering material may have been somewhat off-axis.," That the energetics are modest here may be a result of the source being very close and viewed from a large off-axis angle, and that even the scattering material may have been somewhat off-axis."3 This is consistent with the otherwise coincidental situation that GRB 060218 is both one of the longest. closest. and softest gamuna rav bursts.," This is consistent with the otherwise coincidental situation that GRB 060218 is both one of the longest, closest, and softest gamma ray bursts."4 Thus. tle various distinguishing features of this GRB cau be attributed to a large off-axis viewing angele.," Thus, the various distinguishing features of this GRB can be attributed to a large off-axis viewing angle."5 We thank C. Wheeler. E. Ramirez-Ruiz. A. Soderberg. A. MeFaddeu. and D. Frail for helpful discussions.," We thank C. Wheeler, E. Ramirez-Ruiz, A. Soderberg, A. McFadden, and D. Frail for helpful discussions."6 This research was supported by the IsraclUS DBinational Science Foundation. the Israeli Academy of Science. and The Joan aud Robert Arnow Chair of Theoretical Astroplivsics. aud. via the IKavli institute of Theoretical Physics. the US NSF evant NSF PITYS05-5116I.," This research was supported by the Israel-US Binational Science Foundation, the Israeli Academy of Science, and The Joan and Robert Arnow Chair of Theoretical Astrophysics, and, via the Kavli institute of Theoretical Physics, the US NSF grant NSF PHYS05-51164."7"The orbits have been integrated for 10"" vears and for each of them a value of η has been computed (see Fig. 26)).",The orbits have been integrated for $10^{6}$ years and for each of them a value of $\eta$ has been computed (see Fig. \ref{phoebe-gap}) ).8 The y values indicate that the perturbations of Titan and lapetus are negligible and that test particles positioned in regions of high collision probability. (ie. for 710 and i 174°) ave not dynamically. The collisional origin of the Phoehe’s gap is. also confirmed. by observational data showing that the irregular satellites moving closer to. Phoebe. are those located in regions of the phase space where the impact probability with Phoebe is lower., The $\eta$ values indicate that the perturbations of Titan and Iapetus are negligible and that test particles positioned in regions of high collision probability (i.e. for $i \leq 10^{\circ}$ and $i \geq 174^{\circ}$ ) are not dynamically The collisional origin of the Phoebe's gap is also confirmed by observational data showing that the irregular satellites moving closer to Phoebe are those located in regions of the phase space where the impact probability with Phoebe is lower.9 In addition. the images of Phoebe taken by LSS on-board the Cassini spacecraft revealed a strongly cratered surface. with a continuous crater size distribution ranging [rom about 50 m. a ower value imposed. by. the resolution of ISS images. to an upper limit of about 100 km comparable to the dimension o ‘the satellite.," In addition, the images of Phoebe taken by ISS on-board the Cassini spacecraft revealed a strongly cratered surface, with a continuous crater size distribution ranging from about $50$ m, a lower value imposed by the resolution of ISS images, to an upper limit of about $100$ km comparable to the dimension of the satellite."10 A highly craterecl surface was predicted. by Nesvornyal. (2003).. who also suggested that the vast majority of Phoebes craters should be due to either impacts with other irregular satellites (mainly: prograde ones) or be the result of a past intense [lux of bodies crossing Saturn's. orbit.," A highly cratered surface was predicted by \cite{nes03}, who also suggested that the vast majority of Phoebe's craters should be due to either impacts with other irregular satellites (mainly prograde ones) or be the result of a past intense flux of bodies crossing Saturn's orbit."11" Comets give a negligible contribution having a frequency of collision with Phoebe of about 1 impact every 10"" vears (Zahneotal.2003).", Comets give a negligible contribution having a frequency of collision with Phoebe of about $1$ impact every $10^{9}$ years \citep{zah03}.12. Our results confirm those of Nesvornyetal.(2003) showing that indeed Phoebe had a major role in shapine the structure of Saturns irregular satellites., Our results confirm those of \cite{nes03} showing that indeed Phoebe had a major role in shaping the structure of Saturn's irregular satellites.13" The existence of a primordial now extinct. population of small irregular satellites or collisional shards between 11.2>.19° km and 14.9610"" km from Saturn could explain in a natural wav the abundance of craters on Phoebes IEroche’s sweeping ellect appears to have another major COLISCCuence. related to the existence of Phoebe's gap: it argues against the hypothesis of a Phoebe. family."," The existence of a primordial now extinct population of small irregular satellites or collisional shards between $11.22 \times 10^{6}$ km and $14.96 \times 10^{6}$ km from Saturn could explain in a natural way the abundance of craters on Phoebe's Phoebe's sweeping effect appears to have another major consequence, related to the existence of Phoebe's gap: it argues against the hypothesis of a Phoebe family."14 The existence of Phoebe's family has been a controversial subject since its proposition bv Gladmanetal.(2001)., The existence of Phoebe's family has been a controversial subject since its proposition by \cite{gla01}.15 Its existence was guessed on the close values of inclination of Phoebe and other retrograde satellites., Its existence was guessed on the close values of inclination of Phoebe and other retrograde satellites.16 Phe dynamical inconsistency of this criterion has been pointed. out by Nesvornyet.al.(2003 ). who showed that the velocity. dispersion. required to relate the retrograde satellites to Phoebe would be too high to be accounted as realistic in the context of the actual knowledge of fragmentation and disruption processes.," The dynamical inconsistency of this criterion has been pointed out by \cite{nes03}, who showed that the velocity dispersion required to relate the retrograde satellites to Phoebe would be too high to be accounted as realistic in the context of the actual knowledge of fragmentation and disruption processes."17 Our results suggest also that ifa breakup event involved Phoebe. the fragments would have been ejected within the Phoebe's eap for realistic ejection velocities.," Our results suggest also that if a breakup event involved Phoebe, the fragments would have been ejected within the Phoebe's gap for realistic ejection velocities."18 As a consequence. they would have been removed by impacting on Phoebe.," As a consequence, they would have been removed by impacting on Phoebe."19speed of light.,speed of light.20 After tag. the incoming black hole merges with the SMDII at the ceuter aud a new SMDIT is formed after having accreted σας for tace.," After $_{\rm df}$, the incoming black hole merges with the SMBH at the center and a new SMBH is formed after having accreted gas for $_{\rm acc}$."21" The accretion time and eficiency both implicitly encode the large-scale dvuaiics of the merecr aud the bulk gas accretion iuto the nuclearreeion. while fia, describes the accretion disk plivsics."," The accretion time and efficiency both implicitly encode the large-scale dynamics of the merger and the bulk gas accretion into the nuclearregion, while $t_{\rm sal}$ describes the accretion disk physics."22 As before. we set taa to describe sustained Eddineton- accretion with a efficiency of €20.1 (?)..," As before, we set $_{\rm sal}$ to describe sustained Eddington-limited accretion with a efficiency of $\epsilon = 0.1$ \citep{shakura:73}."23 Iu ?.. we found that a σας acerction triggered by major ucreers (4:1) and a Bovlan-INolchin dyvuaiuical friction xeseription produced a Ll«106 SMBIT from a 200 AL. seed in place by z=5.," In \citet{KHB:2008bhgrowth}, we found that a gas accretion triggered by major mergers (4:1) and a Boylan-Kolchin dynamical friction prescription produced a $4\times 10^6 \, {\rm M}_\odot$ SMBH from a 200 ${\rm M}_\odot$ seed in place by $=5$."24 We considerML.. this our preferred nodel as it produces a SAIBILT cousistent with the Ser A., We consider this our preferred model as it produces a SMBH consistent with the Sgr $^*$.25 Note. however. that the minor merger (10:1) Bovlan-Wolchin prescription also generates a 1.3«LOTAL. black tole from the same seed. aud can be considered a viable nodel for au M31-like SMDIT.," Note, however, that the minor merger (10:1) Boylan-Kolchin prescription also generates a $1.3 \times 10^7 \, {\rm M}_\odot$ black hole from the same seed, and can be considered a viable model for an M31-like SMBH."26 Iu order to compare our results with other published results. we follow the approach of ? to calculate the eravitational wave signal from the black hole iiergers iu our simulation volume.," In order to compare our results with other published results, we follow the approach of \citet{Sesana:04} to calculate the gravitational wave signal from the black hole mergers in our simulation volume."27 Iu this section. we provide a brief backeround on the theory aud outlue the method: for a 1nore in-depth description of eravitational waves from SMIDIIS see. for example. ?..," In this section, we provide a brief background on the theory and outline the method; for a more in-depth description of gravitational waves from SMBHs see, for example, \cite{Berti:2006}."28 The orbital motion ofthe binary svstei constitutes will excite a time-dependent mass quadrupole. which eoncrates eravitational radiation.," The orbital motion ofthe binary system constitutes will excite a time-dependent mass quadrupole, which generates gravitational radiation."29 The exavitational wave strain. P. imieasures the streneth of the propagating wave.," The gravitational wave strain, $h$, measures the strength of the propagating wave."30" The amplitude depends ou the comoving distance to the binary. d. the rest frame eravitational wave frequency. f,. aud the chirp mass of the binary. ντ."," The amplitude depends on the comoving distance to the binary, $d$, the rest frame gravitational wave frequency, $f_{r}$, and the chirp mass of the binary, $\mathcal{M}$."31" The rest-frame eravitational wave frequency is related tothe orbital period. P. of the binary. f,=2/P. aud the chirp mass depends ou the mass of cach binary component as follows: M—GnqioyOFUdams)."," The rest-frame gravitational wave frequency is related tothe orbital period, $P$, of the binary, $f_{r}= 2/P$, and the chirp mass depends on the mass of each binary component as follows: $\mathcal{M} \equiv (m_1 m_2)^{3/5}/(m_1 + m_2)^{1/5}$."32 When averaged over the sky position. polarization. aud period. the strain can be written as: This eravitational radiation will cause the binary conrponeuts to inspiral on ever tighter aud faster orbits.," When averaged over the sky position, polarization, and period, the strain can be written as: This gravitational radiation will cause the binary components to inspiral on ever tighter and faster orbits."33 When massive black hole binaries are widely separated and orbiting at low frequencies. the eravitational radiation enütted is relatively weak. aud therefore the frequency shift per orbit is minuscule.," When massive black hole binaries are widely separated and orbiting at low frequencies, the gravitational radiation emitted is relatively weak, and therefore the frequency shift per orbit is minuscule."34 Most of the evolution is speut in this phase. where a frequency shitt of order unity takes may orbits to achicve.," Most of the evolution is spent in this phase, where a frequency shift of order unity takes many orbits to achieve."35 As the binary separation slowly shrinks. ie eravitational radiation enuütted increases dramatically until the binary coalesces.," As the binary separation slowly shrinks, the gravitational radiation emitted increases dramatically until the binary coalesces."36 Close to coalescence. the binary rapidly sweeps through παν frequencies iu one orbit. and at the iunerinost stable circular orbit ISCO). the change in frequency per orbit is of order unity.," Close to coalescence, the binary rapidly sweeps through many frequencies in one orbit, and at the innermost stable circular orbit (ISCO), the change in frequency per orbit is of order unity."37 We set the nininmni observed frequeucy to [μου for a test particle orbiting around a single Sclavarzschild black hole., We set the minimum observed frequency to $f_{{\rm ISCO}}$ for a test particle orbiting around a single Schwarzschild black hole.38" In the black hole merecrs we consider. the mass ratios are always well outside this test particle limit: we adopt the conventional definition for [μου nonetheless: where e and G are the speed of light aud gravitational constant. respectively. 504 and o are the black hole πάθος, and + is the redshift."," In the black hole mergers we consider, the mass ratios are always well outside this test particle limit; we adopt the conventional definition for $f_{\rm ISCO}$ nonetheless: where $c$ and $G$ are the speed of light and gravitational constant, respectively, $m_{1}$ and $m_{2}$ are the black hole masses, and $z$ is the redshift."39 The frequency shift rate in the rest-frame is. to first order: The orbital periods of million. solar mass black hole binaries iu the eravitational radiation regine are on the order ofhours. which 1ieaus that 1uulti-xear observations could aceunuulate amy cvcles of an iuspiral atf a particular frequency.," The frequency shift rate in the rest-frame is, to first order: The orbital periods of million solar mass black hole binaries in the gravitational radiation regime are on the order of hours, which means that multi-year observations could accumulate many cycles of an inspiral at a particular frequency."40 LISA. for example. is expected to observe the skv for at least a vear. aud projections place the LISA lifespan at roughly a decade.," LISA, for example, is expected to observe the sky for at least a year, and projections place the LISA lifespan at roughly a decade."41 Million: solar mass black holes orbiting iu the LISA frequency baud of 10!1 Iz are years or less from coalescence. so LISA observations will be able to track the inspiral aud coalescence phases of supermassive black hole biuaries.," Million solar mass black holes orbiting in the LISA frequency band of $\sim 10^{-4} - 1$ Hz are years or less from coalescence, so LISA observations will be able to track the inspiral and coalescence phases of supermassive black hole binaries."42 The umber of eveles aceiunulated during the iuspiral in an observation of duration 7 depends on the observed eravitational wave frequency. 90=fr. where f=ffl| zh ," The number of cycles accumulated during the inspiral in an observation of duration $\tau$ depends on the observed gravitational wave frequency, $n=f \tau$, where $f=f_r/(1+{\rm z})$ ."43"Note that there are two reeimes: foonr. where the binary sweeps through many frequencies in au observation. aud f£p/r. where the signal builds frou several orbits at one frequency,"," Note that there are two regimes: $f > n/ \tau$, where the binary sweeps through many frequencies in an observation, and $f < n/ \tau$, where the signal builds from several orbits at one frequency."44" The observed characteristic strain. 7. is the strain acciunulated a single observation: Since we have the component mass and redshift of each black hole iierger i our simulation volume. we cau calculate fh aud P, over an observation span r before morecr."," The observed characteristic strain, $h_c$, is the strain accumulated a single observation: Since we have the component mass and redshift of each black hole merger in our simulation volume, we can calculate $h$ and $h_c$ over an observation span $\tau$ before merger."45 Most black holes in our volume will merge within a vear of reaching the LISA baud. so a 3-vear observation window should catch these merecrs in the act if this volume is a representative slice of the Universe.," Most black holes in our volume will merge within a year of reaching the LISA band, so a 3-year observation window should catch these mergers in the act – if this volume is a representative slice of the Universe."46 We distinguish between the mergers directly found in our voluue and the mergers extrapolated throughout the Universe in Table 2.., We distinguish between the mergers directly found in our volume and the mergers extrapolated throughout the Universe in Table \ref{tab:first}.47 Overall. most of the merecrs in our volume are between secd and intermediate mass black holes (O(L01) AL.) at redshifts ercater than 5 (Figure 1))," Overall, most of the mergers in our volume are between seed and intermediate mass black holes $O(10^4)\, {\rm M}_\odot$ ) at redshifts greater than 5 (Figure \ref{fig:histmap}) )."48 Though these nerecrs were critical in assenmibliug the eventual SMDII in our Milkv Wav analogue. the ecnerated LISA sigual-o-hnoise ratio of most of these low mass mergers at high redshift were unich less than 1.0.," Though these mergers were critical in assembling the eventual SMBH in our Milky Way analogue, the generated LISA signal-to-noise ratio of most of these low mass mergers at high redshift were much less than 1.0."49 Figure 7 demonstrates hat the overwhehuing majority of black hole mergers will fall below LISA’s detection limit.," Figure \ref{fig:histsnr}50 demonstrates that the overwhelming majority of black hole mergers will fall below LISA's detection limit."51 Although there are a few inergers that are resolvable by LISA at z>6. the iuuber of resolvable sources increases once the SMDII erows to a mass Within the LISA hand at zx 5. as can vo seen in Figure 8..," Although there are a few mergers that are resolvable by LISA at $>6$, the number of resolvable sources increases once the SMBH grows to a mass within the LISA band at $<5$ , as can be seen in Figure \ref{fig:histz}. ."52 This broadly agrees with ?.. which showed that ~10 of the precursors to 10?M. SMDIIS are expected to be observable with LISA out to redshift 10.," This broadly agrees with \citet{Sesana:05}, , which showed that $\sim10\%$ of the precursors to $10^9 \, {\rm M}_\odot$ SMBHs are expected to be observable with LISA out to redshift 10."53 However. Figure d. also reveals a substantial class of black hole mergers with very high mass ratios: these," However, Figure \ref{fig:histmap} also reveals a substantial class of black hole mergers with very high mass ratios; these"54" (Franx1993.Vogtοαἰ.1996.. (Faber&Jackson1976.TullyFisher1977)). S/N S/N Schade1999)). 1987)) (Jorgensen.Franx.&Kyjergaard~1993.Kelsonetaf.1997)). r,. c. (7)... s. 420% 7. £14% £12%(Kelsonefal1999)"," \cite{franx93,vogt}, \cite{fj76,tf77}) $S/N$ $S/N$ \cite{schade,vdfp98,simard99}) \cite{faber87,dd87}) \cite{jfk93,kelson}) $r_e$ $\sigma$ $\langle I\rangle_e$ $g$ $\pm 20\%$ $r_e$ $\pm 14\%$ \cite{lucey,jfk93}) $\pm 12\%$\cite{kelsonh0})"55" (Franx1993.Vogtοαἰ.1996.. (Faber&Jackson1976.TullyFisher1977)). S/N S/N Schade1999)). 1987)) (Jorgensen.Franx.&Kyjergaard~1993.Kelsonetaf.1997)). r,. c. (7)... s. 420% 7. £14% £12%(Kelsonefal1999))"," \cite{franx93,vogt}, \cite{fj76,tf77}) $S/N$ $S/N$ \cite{schade,vdfp98,simard99}) \cite{faber87,dd87}) \cite{jfk93,kelson}) $r_e$ $\sigma$ $\langle I\rangle_e$ $g$ $\pm 20\%$ $r_e$ $\pm 14\%$ \cite{lucey,jfk93}) $\pm 12\%$\cite{kelsonh0})"56" (Franx1993.Vogtοαἰ.1996.. (Faber&Jackson1976.TullyFisher1977)). S/N S/N Schade1999)). 1987)) (Jorgensen.Franx.&Kyjergaard~1993.Kelsonetaf.1997)). r,. c. (7)... s. 420% 7. £14% £12%(Kelsonefal1999))."," \cite{franx93,vogt}, \cite{fj76,tf77}) $S/N$ $S/N$ \cite{schade,vdfp98,simard99}) \cite{faber87,dd87}) \cite{jfk93,kelson}) $r_e$ $\sigma$ $\langle I\rangle_e$ $g$ $\pm 20\%$ $r_e$ $\pm 14\%$ \cite{lucey,jfk93}) $\pm 12\%$\cite{kelsonh0})"57signals detected oul to this frequency. (,signals detected out to this frequency. (58Lhe power spectra in Figure 2 are only plotted out to 10.? Lz to emphasize the structure at lower frequencies.),The power spectra in Figure 2 are only plotted out to $10^{-3}$ Hz to emphasize the structure at lower frequencies.)59 We therefore conclude that the light curve o£ TW Pic from this205 LL observation displays no intrinsic periodic signals. with a period. greater than about 3002 s. down to a limiting modulation of less thanVA.," We therefore conclude that the light curve of TW Pic from this HRI observation displays no intrinsic periodic signals, with a period greater than about 300 s, down to a limiting modulation of less than."60 The LIBRE count rate from PW Pie (0.20 ος +) is comparable to that of confirmed intermediate polars with similar optical Huxes., The HRI count rate from TW Pic (0.20 c $^{-1}$ ) is comparable to that of confirmed intermediate polars with similar optical fluxes.61 For instance AO Pse and V1223 Ser are both 13th. magnitude intermediate polars ancl exhibit ILRI count rates of 0:22 e | and O41 ος | respectively Clavlor et al 1997). whilst the 15th magnitude TX Col has n URL count rate of 0.07 ος (Norton et al 1997).," For instance AO Psc and V1223 Sgr are both 13th magnitude intermediate polars and exhibit HRI count rates of 0.22 c $^{-1}$ and 0.41 c $^{-1}$ respectively (Taylor et al 1997), whilst the 15th magnitude TX Col has an HRI count rate of 0.07 c $^{-1}$ (Norton et al 1997)."62 On 16 basis of its X-ray to optical [ux ratio therefore. PW Pic would seem to be a good. candidate for intermediate polar gaabus. as Tuohy et al (1986) originally suggested.," On the basis of its X-ray to optical flux ratio therefore, TW Pic would seem to be a good candidate for intermediate polar status, as Tuohy et al (1986) originally suggested."63 However. 16 one unambiguous signature of an intermediate polar is a 'oherent N-rav pulsation at a period significantly less than 16 binary orbital period.," However, the one unambiguous signature of an intermediate polar is a coherent X-ray pulsation at a period significantly less than the binary orbital period."64 Based on this most sensitive X-ray observation vet of TW. Pic. no such pulsation exists and so we have no evidence to support the proposed. classification of the system.," Based on this most sensitive X-ray observation yet of TW Pic, no such pulsation exists and so we have no evidence to support the proposed classification of the system."65 “PW Pieoney still be an intermediate polar. but one seen at a relatively low inclination angle such that a roughly constant N-ray. [lux is seen from the upper magnetic pole. and the lower one is permanently hidden.," TW Pic still be an intermediate polar, but one seen at a relatively low inclination angle such that a roughly constant X-ray flux is seen from the upper magnetic pole, and the lower one is permanently hidden."66 In this case. we would not expect the spin period of the white dwarf to be apparent in optical photometry or spectroscopy either.," In this case, we would not expect the spin period of the white dwarf to be apparent in optical photometry or spectroscopy either."67 So. whether or not PW Pic is an intermediate polar. we must therefore suggest. some other explanation for the periods previously reported in this system.," So, whether or not TW Pic is an intermediate polar, we must therefore suggest some other explanation for the periods previously reported in this system."68" The ~2 hr period cannot represent the spin period of an accreting magnetic white ονα, otherwise we would have seen evidence for it in ourROSA URI cata."," The $\sim 2$ hr period cannot represent the spin period of an accreting magnetic white dwarf, otherwise we would have seen evidence for it in our HRI data."69 Vhe original election of this period was in the optical spectroscopic ata of Buckley Tuohy (1990)., The original detection of this period was in the optical spectroscopic data of Buckley Tuohy (1990).70 Llowever. their subsequent etection of a similar period in theBNOSAL X-ray data is barely significant. ancl indeed went un-noticed in their varlicr analysis of the same data (Tuohy et al 1986).," However, their subsequent detection of a similar period in the X-ray data is barely significant, and indeed went un-noticed in their earlier analysis of the same data (Tuohy et al 1986)."71 Lt is oubtful whether the X-ray detection of the period would have been claimed without the prior discovery of the ~2 hr optical period., It is doubtful whether the X-ray detection of the period would have been claimed without the prior discovery of the $\sim 2$ hr optical period.72" Nonetheless. à 2 hr period has been clearly etected in. both optical spectroscopy (Buckley ""Tuohy 1900) and in optical photometry (Patterson Aloulden 1993). and is undoubtedly real."," Nonetheless, a $\sim 2$ hr period has been clearly detected in both optical spectroscopy (Buckley Tuohy 1990) and in optical photometry (Patterson Moulden 1993), and is undoubtedly real."73 We suggest. therefore. that 1.996 hr represents theorbilat period of the system and as such we would not necessarily expect to detect it in an X- observation. unless the svstem were at a relatively high inclination.," We suggest, therefore, that 1.996 hr represents the period of the system and as such we would not necessarily expect to detect it in an X-ray observation, unless the system were at a relatively high inclination."74 The reasons for disbelieving that the ~6 hr period represents the orbital motion have already been. discussed by Patterson Moulden (1993)., The reasons for disbelieving that the $\sim 6$ hr period represents the orbital motion have already been discussed by Patterson Moulden (1993).75 In their photometric data. this period was observed to drift over the course of 3 weeks bv up to 0.3 evcles.," In their photometric data, this period was observed to drift over the course of 3 weeks by up to 0.3 cycles."76 Moreover. they note that. the profile of the ~6 hr modulation simply does not look stable. and varies significantly from. evele to evele.," Moreover, they note that the profile of the $\sim 6$ hr modulation simply does not look stable, and varies significantly from cycle to cycle."77 “Phe spectroscopic and photometric observations reported by Buckley Tuohy (1990) both detected a ~6 hr period. but again the constancy of the period is not convincing. and in their casecould merely be a second. harmonic of the ~2 hr period. as neither of the periods are determined: very. accurately.," The spectroscopic and photometric observations reported by Buckley Tuohy (1990) both detected a $\sim 6$ hr period, but again the constancy of the period is not convincing, and in their case merely be a second harmonic of the $\sim 2$ hr period, as neither of the periods are determined very accurately."78 We sugeest instead that this quasi-periodie 6 hr modulation maw arise from a phenomenon associated with the accretion disc., We suggest instead that this quasi-periodic 6 hr modulation may arise from a phenomenon associated with the accretion disc.79 A comparable system may be. the. intermediate xar TV Col (llutchines et al 1981). which exhibits a spectroscopic (orbital) period of 5.49 hr but a photometric »riod Of 5.19 hr (in addition to the white dwarf spin period of 1911 s).," A comparable system may be the intermediate polar TV Col (Hutchings et al 1981), which exhibits a spectroscopic (orbital) period of 5.49 hr but a photometric period of 5.19 hr (in addition to the white dwarf spin period of 1911 s)."80 Ehe additional presence of à ~4 d beat! period (such that 1/5.49hr|1/4d=1/5.19 hr) led Barrett. O'Donoghue and Warner (1988) to interpret these multiple »eriodicities in terms of a retrogracely precessing accretion disc.," The additional presence of a $\sim 4$ d `beat' period (such that $1/5.49~{\rm hr} + 1/4~{\rm d} = 1/5.19~{\rm hr}$ ) led Barrett, O'Donoghue and Warner (1988) to interpret these multiple periodicities in terms of a retrogradely precessing accretion disc."81 “Phe disc is assumed to precess with a 4 d period. and he 5.19 hr period may then arise due to tidal interactions with the secondary (LMlellier 1993).," The disc is assumed to precess with a 4 d period, and the 5.19 hr period may then arise due to tidal interactions with the secondary (Hellier 1993)."82 Augusteijn et al (1994) noted that both the 5.19 hr and 4 d periods in PV Col are unstable and may vary monatonicallv., Augusteijn et al (1994) noted that both the 5.19 hr and 4 d periods in TV Col are unstable and may vary monatonically.83 Given the similarly unstable ~6 hr period in TW Pic. disc precession may be the explanation for the long period seen in this svstem also.," Given the similarly unstable $\sim 6$ hr period in TW Pic, disc precession may be the explanation for the long period seen in this system also."84 Finally. we note that i£ PW Pie really were à magnetic system. and if the spin period of the white dwarf really were around two hours. it would. represent. thesfowes? rotator amongst all the intermediate polars.," Finally, we note that if TW Pic really were a magnetic system, and if the spin period of the white dwarf really were around two hours, it would represent the rotator amongst all the intermediate polars."85 Phe only confirmed intermediate polar coming close to this is EX Ilya. with a spin period of 67 min.," The only confirmed intermediate polar coming close to this is EX Hya, with a spin period of 67 min."86 However. EX να is unusual in that its spin and orbital periods are in a ~2:8 ratio.," However, EX Hya is unusual in that its spin and orbital periods are in a $\sim$ 2:3 ratio."87 C. Wynn (private communication) has shown that high magnetic field intermediate polars will indeed evolve such that their periods end up in a 2:3 ratio. not the apparent 21:3 ratio which the previously. proposed. periods of PW Pie might indicate.," G. Wynn (private communication) has shown that high magnetic field intermediate polars will indeed evolve such that their periods end up in a $\sim$ 2:3 ratio, not the apparent $\sim$ 1:3 ratio which the previously proposed periods of TW Pic might indicate."88 1n conclusion. we find that there is no convincing evidence for regarding TW Pic as an intermediate. polar. and we suggest that the true orbital period of the binary is the shorter of the two previously identified. periods.," In conclusion, we find that there is no convincing evidence for regarding TW Pic as an intermediate polar, and we suggest that the true orbital period of the binary is the shorter of the two previously identified periods."89 The data analysis reported here was carried out using facilities provided. by PPARC. Starlink and the Open University Research Committec.," The data analysis reported here was carried out using facilities provided by PPARC, Starlink and the Open University Research Committee."90correlations are actually dilflieult to detect.,correlations are actually difficult to detect.91 One example of this is the use of a-enhancement to clock the interaction as we have already. discussed. in Section 3.1., One example of this is the use of $\alpha$ -enhancement to clock the interaction as we have already discussed in Section 3.1.92 Another expected correlation is that between the strength of the central starbursts and the metallicity dilution., Another expected correlation is that between the strength of the central starbursts and the metallicity dilution.93 In Fig., In Fig.94 13 we plot the slopes of the metallicity &radients and the central oxvgen abundances as a function of star formation rate [for wet low-redshift interactions., \ref{sfroh} we plot the slopes of the metallicity gradients and the central oxygen abundances as a function of star formation rate for wet low-redshift interactions.95 We can see that elobally there is a correlation so that the highest the star ormation rate. the strongest the metallicity dilution and he shallower the abundance profiles.," We can see that globally there is a correlation so that the highest the star formation rate, the strongest the metallicity dilution and the shallower the abundance profiles."96 However. the relations depend strongly on the orbital parameters.," However, the relations depend strongly on the orbital parameters."97 Siml and SimV iive weaker starbursts than Simll but nevertheless. they experienced. metallicity cülution in the central regions clue o the tidallvy-induced gas inflows.," SimI and SimV have weaker starbursts than SimII but nevertheless, they experienced metallicity dilution in the central regions due to the tidally-induced gas inflows."98 Hence. according to our indings. it might. be cillicult to establish this correlation observationally where galaxy pairs can be observed at dillerent. stages of evolution and with a varicty of orbital )nranmeters.," Hence, according to our findings, it might be difficult to establish this correlation observationally where galaxy pairs can be observed at different stages of evolution and with a variety of orbital parameters."99 Reearcding the gas-richness. from. Fig.l we can see hat. although there is a global trend. in diluting central metallicity and in flattening the mean gradients. there are also fluctuations which correlate with the triggering of inflows and the injection of fresh. oxvgen by the new SN II.," Regarding the gas-richness, from \ref{perfiles} we can see that, although there is a global trend in diluting central metallicity and in flattening the mean gradients, there are also fluctuations which correlate with the triggering of inflows and the injection of fresh oxygen by the new SN II."100 Nevertheless. the mean metallicity dilution is stronger than the less gaserich counterpart (Pig. 6)).," Nevertheless, the mean metallicity dilution is stronger than the less gas-rich counterpart (Fig. \ref{fit1}) )."101 Hence. the richness is another kev factor.," Hence, the gas-richness is another key factor."102 Depending on the astrophysical properties ancl orbital parameters. galaxies in pairs can be located: differently on the mean mass-metallicity relation depending on the stage of the interaction at which they are observed.," Depending on the astrophysical properties and orbital parameters, galaxies in pairs can be located differently on the mean mass-metallicity relation depending on the stage of the interaction at which they are observed."103 This ellect might be particularly relevant for the study of high redshift galaxies and could be at the origin of the apparent increase with redshift of the scatter in the mass-metallicity relation., This effect might be particularly relevant for the study of high redshift galaxies and could be at the origin of the apparent increase with redshift of the scatter in the mass-metallicity relation.104 It could also explain the presence of outliers in the local mass-metallicity relation., It could also explain the presence of outliers in the local mass-metallicity relation.105 For example. more eas-rich systems observed: just as they are first approaching might be dominated by the massively and elumpy star formation activity taking place at all radii. before the interaction can actually trigger larger scale inflows capable to transport Iow-metallicity eas into the central region.," For example, more gas-rich systems observed just as they are first approaching might be dominated by the massively and clumpy star formation activity taking place at all radii, before the interaction can actually trigger larger scale inflows capable to transport low-metallicity gas into the central region."106 Hence. these galaxies might show an excess of oxvgen (e.g.Michel-Dansacctal. 2008).," Hence, these galaxies might show an excess of oxygen \citep[e.g.][]{LMD08}."107.. ὃν using hvdrodynamical simulations of interacting pre-prepared galaxies of comparable masses. we study the chemical evolution of the interstellar medium cluring the interactions.," By using hydrodynamical simulations of interacting pre-prepared galaxies of comparable masses, we study the chemical evolution of the interstellar medium during the interactions."108 Our simulations include. self-consistently. SN feedback ancl a detailed. chemical model. which allows us to follow the evolution of the eas-phase metallicity from the initial to the final stages of an interaction.," Our simulations include self-consistently, SN feedback and a detailed chemical model, which allows us to follow the evolution of the gas-phase metallicity from the initial to the final stages of an interaction."109 These simulations constitute a unique set to study the effects of interactions on chemical properties., These simulations constitute a unique set to study the effects of interactions on chemical properties.110 We analyse the different chemical response to galaxy interactions. varying orbital configurations. streneth of οποιον SN feedback ane initial gas fractions in the galaxy discs.," We analyse the different chemical response to galaxy interactions, varying orbital configurations, strength of energy SN feedback and initial gas fractions in the galaxy discs."111 Particularly. we focused our study on wet interactions in the local Universe.," Particularly, we focused our study on wet interactions in the local Universe."112 For comparison. we also analysed a more gas-rich interaction in order to reproduce wet mergers typical of high redshift.," For comparison, we also analysed a more gas-rich interaction in order to reproduce wet mergers typical of high redshift."113 Our main results can be summarized as follows: 1) We find that a low-metaltlicity gas inflow developed from the first’ close passage dilutes the central oxygen, Our main results can be summarized as follows: 1) We find that a low-metallicity gas inflow developed from the first close passage dilutes the central oxygen114and imaging were performed. using standard: procedures within the National ltadio Astronomy Observatory’s(Alps).,and imaging were performed using standard procedures within the National Radio Astronomy Observatory's.115 Phe flux calibrators were ὃς48 and 3€286 (whichever was visible at the time of the observation)., The flux calibrators were 3C48 and 3C286 (whichever was visible at the time of the observation).116 Two phase calibrators. JIT44 3116 and 2524. were used to interpolate the eain corrections to the target.," Two phase calibrators, $-$ 3116 and $-$ 2524, were used to interpolate the gain corrections to the target."117 The latter is very weak ab 14.94CLlIz and could not be usec at CiCillz. so was only used at the lower frequencies.," The latter is very weak at GHz and could not be used at GHz, so was only used at the lower frequencies."118 Some of the observations lacked a primary calibrator. in which case the secondary calibrator was also used to set the {lux density scale. averaging the Dux density measured for the calibrator in previous observing runs.," Some of the observations lacked a primary calibrator, in which case the secondary calibrator was also used to set the flux density scale, averaging the flux density measured for the calibrator in previous observing runs."119 For observations including a primary calibrator. the secondary calibrator [Lux density was Found to vary by of order Imaging was carried out using a robust weighting scheme. biased. slightly. towards uniform. weighting (robust parameter set to -] or 0) in orcler to enhance the resolution.," For observations including a primary calibrator, the secondary calibrator flux density was found to vary by of order Imaging was carried out using a robust weighting scheme, biased slightly towards uniform weighting (robust parameter set to -1 or 0) in order to enhance the resolution."120 After a few iterations of imaging ancl phase-onlv self-calibration. the source appeared in many cases to be resolved into two components aligned. approximately East-West.," After a few iterations of imaging and phase-only self-calibration, the source appeared in many cases to be resolved into two components aligned approximately East-West."121 [t was then fit with a two point source mocol plus a background level ancl slope., It was then fit with a two point source model plus a background level and slope.122 X-ray monitoring data from the RNPESASAIT amd CORO/BATSE instruments were also obtained from the public archives ancl analysed in conjunction with the radio clata., X-ray monitoring data from the /ASM amd /BATSE instruments were also obtained from the public archives and analysed in conjunction with the radio data.123 AC variable point) source was detected. in cach ATCA observation and the resultant. lighteurve is. plotted. in the bottom panel of Fig. Ll., A variable point source was detected in each ATCA observation and the resultant lightcurve is plotted in the bottom panel of Fig. \ref{fig:lightcurves}.124 There is a relatively steep rise to a maximum [lux density of GOO my at 1384. MlIZz. Followed bv a more gradual decay.," There is a relatively steep rise to a maximum flux density of $\sim 600$ mJy at 1384 MHz, followed by a more gradual decay."125 The source rises again to 100 my at 4800 MIEIz approximately a month after the decay., The source rises again to $\sim 100$ mJy at 4800 MHz approximately a month after the decay.126 Compared. with outbursts of other sources (e.g. GRO 40. Harmon et al., Compared with outbursts of other sources (e.g. GRO $-$ 40 – Harmon et al.127 1995: NE J1859|226 Drocksopp et al., 1995; XTE J1859+226 -- Brocksopp et al.128 2002) the morphology. of the lighteurve is fairly straightforward., 2002) the morphology of the lightcurve is fairly straightforward.129 However. the peak flux. density reached is unusually high. compared with the more tvpical values of a few mdw eg. XTE J1720 318 (—5 mdlw: Brocksopp et al.," However, the peak flux density reached is unusually high, compared with the more typical values of a few mJy e.g. XTE $-$ 318 $\sim 5$ mJy; Brocksopp et al."130 2005). NPE J1859|226 (—120 mJy: AMocksopp et al.," 2005), XTE J1859+226 $\sim 120$ mJy; Brocksopp et al."131 2002)., 2002).132 The top panel of Fig., The top panel of Fig.133 1. shows the available BATSE data for this outburst., \ref{fig:lightcurves} shows the available BATSE data for this outburst.134 It is clearly in. decav by the time the observations were obtained. vet there is also a small temporary brightening before the source reaches its minimum.," It is clearly in decay by the time the observations were obtained, yet there is also a small temporary brightening before the source reaches its minimum."135 lt is not clear whether the hard. N-ray. emission remained high for the duration of the soft. X-ray. outburst or whether this was a rebrightening as part of the same event which led to the radio rebrightening on ΑΙ) 51017., It is not clear whether the hard X-ray emission remained high for the duration of the soft X-ray outburst or whether this was a rebrightening as part of the same event which led to the radio rebrightening on MJD 51017.136 RNTE/PCA observations by Bevnivtseyv. Frudolvuboy. Borozclin (2000) show a sharp dip in the 1530 keV emission (ALJD 50975 51000) followed by an increase. peaking on ALJD 51012.," /PCA observations by Revnivtsev, Trudolyubov Borozdin (2000) show a sharp dip in the 15–30 keV emission (MJD 50975 – 51000) followed by an increase peaking on MJD 51012."137 The second. plot shows the ASM lighteurve., The second plot shows the ASM lightcurve.138 Phere is, There is139Note that the above οσο can only mimic a systematically non-zero Aaja if the spectrograph temperature is systematically higher or lower for the ThaAr frames compared with the QSO frames.,Note that the above effect can only mimic a systematically non-zero $\da$ if the spectrograph temperature is systematically higher or lower for the ThAr frames compared with the QSO frames.140 This is possible if the ThAr exposures were always taken before or after the QSO frames ancl the temperature: evolves monotonically throughout the night., This is possible if the ThAr exposures were always taken before or after the QSO frames and the temperature evolves monotonically throughout the night.141 However. the effect is greatly. reduced. if VhAr exposures are taken near the time of the QSO observations.," However, the effect is greatly reduced if ThAr exposures are taken near the time of the QSO observations."142 This was the case in our observations., This was the case in our observations.143 We have used image header information to calculate the QSOThr. temperature.— dillerence. AT=(loseTypus). dnside ο for both the low and samples.," We have used image header information to calculate the QSO–ThAr temperature difference, $\Delta T \equiv \left<T_{\rm144QSO}\right> - \left<T_{\rm ThAr}\right>$, inside HIRES for both the low and samples."145 Here the average of Zoso and. Zh is taken over all exposures for cach object., Here the average of $T_{\rm QSO}$ and $T_{\rm ThAr}$ is taken over all exposures for each object.146 We find mean values of T=0.0430.02Ix and AY=0.240.1Wy for the low and samples respectively.," We find mean values of $\Delta T = 0.04 \pm 0.02{\rm ~K}$ and $\Delta T = 0.2147\pm 0.1{\rm ~K}$ for the low and samples respectively."148 Taking into account that. on average. rest-Lrame separations between transitions are £D300A then we can see that leniperature variations could not have mimicked any significant shift in Aa/a in cither sample.," Taking into account that, on average, rest-frame separations between transitions are $\la 300{\rm ~\AA}$ then we can see that temperature variations could not have mimicked any significant shift in $\da$ in either sample."149 The errors in ;Xafao presented in MOla ancl WOL ake into account errors from signal-to-noise ancl spectral resolution considerations and the velocity. structure of the xofile fits., The errors in $\da$ presented in M01a and W01 take into account errors from signal-to-noise and spectral resolution considerations and the velocity structure of the profile fits.150 The errors are also reduced when more lines are Που. simultaneously., The errors are also reduced when more lines are fitted simultaneously.151 However. we have assumed. that we rave deconvolved. cach absorption system into the correct number of velocity components.," However, we have assumed that we have deconvolved each absorption system into the correct number of velocity components."152 Phere may have been weak. interloping. unresolved lines which. if the interloping species were in the same absorption cloud. could have produced a shift in the fitted line wavelengths of all velocity components of one or more transitions.," There may have been weak, interloping, unresolved lines which, if the interloping species were in the same absorption cloud, could have produced a shift in the fitted line wavelengths of all velocity components of one or more transitions."153 We distinguish between blends andeimalic blends., We distinguish between blends and blends.154 blends may occur if many absorption clouds at dilferent. redshifts intersect the line of sight to a single QSO., blends may occur if many absorption clouds at different redshifts intersect the line of sight to a single QSO.155 Asyslemalie blend will occur when two species are in the same cloud and have absorption lines with similar rest-wavelengths., A blend will occur when two species are in the same cloud and have absorption lines with similar rest-wavelengths.156 Such an elfect could mimic a svstematic shift in a., Such an effect could mimic a systematic shift in $\alpha$.157 The importance of such an effect. diminishes as the number of transitions used in each fit is increased., The importance of such an effect diminishes as the number of transitions used in each fit is increased.158 Pherefore. the effect. if present. would be expected to be smaller in our sample.," Therefore, the effect, if present, would be expected to be smaller in our sample."159 In Section ?? we describe the results ofa detailed search for atomic (and not molecular) interlopers for all transitions of interest. ancl place constraints on their strengths aud positions.," In Section \ref{sec:sysblend} we describe the results of a detailed search for atomic (and not molecular) interlopers for all transitions of interest, and place constraints on their strengths and positions."160 As noted in MOlIa. all of the sample and many of the high redshift sample of QSOs were observed. before 1996 August at which time an image rotator was fitted to HIRES.," As noted in M01a, all of the sample and many of the high redshift sample of QSOs were observed before 1996 August at which time an image rotator was fitted to HIRES."161 Objects observed. prior to August 1996. were therefore not observed with the slit length (spatial direction) perpendicular to the horizon., Objects observed prior to August 1996 were therefore not observed with the slit length (spatial direction) perpendicular to the horizon.162 Atmospheric dispersion leads to a stretching of the target. spectrum relative to. the calibration spectrum., Atmospheric dispersion leads to a stretching of the target spectrum relative to the calibration spectrum.163 An additional consequence is the potential wavelength-dependent truncation of the QSO seeing. profile on the slit jaw edges., An additional consequence is the potential wavelength-dependent truncation of the QSO seeing profile on the slit jaw edges.164 This can introduce a wavelength dependent asvmmetry in the point-spread. function (PSE)., This can introduce a wavelength dependent asymmetry in the point-spread function (PSF).165 Tosether. these elfects can conspire to mimic a non- Aafa.," Together, these effects can conspire to mimic a non-zero $\da$."166 We provide detailed caleulations for these effects in Section 5 and show that they will tend to produce an apparent Aafa. and so cannot mimic our results.," We provide detailed calculations for these effects in Section 5 and show that they will tend to produce an apparent $\da$, and so cannot mimic our results."167 If the instrumental profile (1?) o£. LLRES shows significant asvnunetrics then we expect absorption line centroids to be incorrectly estimated., If the instrumental profile (IP) of HIRES shows significant asymmetries then we expect absorption line centroids to be incorrectly estimated.168 Lf the asymmetry varies with wavelength. this could mimic a non-zero Aa/a.," If the asymmetry varies with wavelength, this could mimic a non-zero $\da$."169 Valenti. Butler Marcy (1995) have determined the ΤΗ105 LP for several positions along a single order and they find that the LP is indeed asymmetric and that the asvmametry varies slightly along the echelle order.," Valenti, Butler Marcy (1995) have determined the HIRES IP for several positions along a single order and they find that the IP is indeed asymmetric and that the asymmetry varies slightly along the echelle order."170 Valenti et al. (, Valenti et al. (1711995). do not quantify the asvnimetry variation across orders Ithough it is likely to be comparable to the variation along 10 orders.,"1995), do not quantify the asymmetry variation across orders although it is likely to be comparable to the variation along the orders."172 As a by-product of the detailed: investigation into possible wavelength calibration errors. we find that. this ellect is negligible (see Section ?? for details).," As a by-product of the detailed investigation into possible wavelength calibration errors, we find that this effect is negligible (see Section \ref{sec:wavcal} for details)."173 We have therefore. eliminated. with some reliability. potential svstematic errors due to: laboratory wavelength errors. heliocentric velocity. variation. hyperfine structure ellects.. magnetic fields. kinematic cllects. airvacuum wavelength: conversion errors. and. temperature changes during observations.," We have therefore eliminated, with some reliability, potential systematic errors due to: laboratory wavelength errors, heliocentric velocity variation, hyperfine structure effects, magnetic fields, kinematic effects, air–vacuum wavelength conversion errors and temperature changes during observations."174 In the following sections we investigate in considerable detail the remaining effects: wavelength mis-calibration (and instrumental profile variations). line blending. atmospheric dispersion cllects. differential isotopic saturation ancl isotopic abundance variation.," In the following sections we investigate in considerable detail the remaining effects: wavelength mis-calibration (and instrumental profile variations), line blending, atmospheric dispersion effects, differential isotopic saturation and isotopic abundance variation."175 We also consider a simple test [or any simple but unidentified: systematic errors for the sample in Section ??.., We also consider a simple test for any simple but unidentified systematic errors for the sample in Section \ref{sec:apns}.176 In this section. we investigate the possibility. that wavelength calibration errors could have lead to an apparent non-zero Aaf/a.," In this section, we investigate the possibility that wavelength calibration errors could have lead to an apparent non-zero $\da$."177 To quantify this directly. we analyse sets of Thr emission lines in the calibration lamp spectra in the same wav as cach set of QSO absorption lines has been analvsed.," To quantify this directly, we analyse sets of ThAr emission lines in the calibration lamp spectra in the same way as each set of QSO absorption lines has been analysed."178 Ε no calibration error is present then we expect that (Aefo)puas=0 For all clouds (1.6. for all z)., If no calibration error is present then we expect that $(\da)_{\rm ThAr} = 0$ for all clouds (i.e. for all $z$ ).179The polarization. properties of the radiation can be described using different vectors.,The polarization properties of the radiation can be described using different vectors.180 The Jones vector describes the information related to amplitude and phase of the observed signal components. thus it is the preferred one to describe mathematically the receiver internal devices.," The Jones vector describes the information related to amplitude and phase of the observed signal components, thus it is the preferred one to describe mathematically the receiver internal devices."181 The Stokes vector reflects the relation between the signal components and detines the overall polarization of the signal. thus it is the preferred one to describe the source physical characteristics.," The Stokes vector reflects the relation between the signal components and defines the overall polarization of the signal, thus it is the preferred one to describe the source physical characteristics."182 The Jones and Stokes vectors carry the same information and are linked by a set of transformation equations., The Jones and Stokes vectors carry the same information and are linked by a set of transformation equations.183 As a rule of thumb. the basis for the transformation should be chosen according to the polarization of the outputs delivered by the telescope under examination.," As a rule of thumb, the basis for the transformation should be chosen according to the polarization of the outputs delivered by the telescope under examination."184 The following equations are related to the circular LR frame. which we adopted in this paper following ? and ?..," The following equations are related to the circular LR frame, which we adopted in this paper following \citet{Kraus} and \citet{Heiles}."185 The Jones vector. J. and the Stokes vector. S. are defined as follows where the asterisk standsfor complex conjugate and A=Op—01.," The Jones vector, $\overrightarrow{J}$, and the Stokes vector, $\overrightarrow{S}$, are defined as follows where the asterisk standsfor complex conjugate and $\Delta=\delta_R-\delta_L$."186 The Miilller matrix relates the polarimetric properties of the source. 1.9. the Stokes parameters. of the source with the measurements from the polarimeter at the telescope.," The Mülller matrix relates the polarimetric properties of the source, i.e. the Stokes parameters, of the source with the measurements from the polarimeter at the telescope."187 Hence it describes the instrument polarization characteristics., Hence it describes the instrument polarization characteristics.188" An ideal instrument that measures Stokes parameters perfectly. would be characterized by a unitary Mülller matrix $,,2MS, where Jy. On. U, and V, are the measured Stokes parameters and /.. Q.. U, and V, are the true Stokes parameters describing the source polarization."," An ideal instrument that measures Stokes parameters perfectly, would be characterized by a unitary Mülller matrix $\overrightarrow{S}_{m}=\mathbf{M}\overrightarrow{S}_{s}$ where $I_{m}$, $Q_{m}$ , $U_{m}$ and $V_{m}$ are the measured Stokes parameters and $I_{s}$, $Q_{s}$, $U_{s}$ and $V_{s}$ are the true Stokes parameters describing the source polarization."189 When handling linear polarization. the polarization angle is measured with respect to some instrumental zero point. usually related to a reference direction on the celestial sphere.," When handling linear polarization, the polarization angle is measured with respect to some instrumental zero point, usually related to a reference direction on the celestial sphere."190 During the source tracking. the reference system of an alt-azimuth mounted telescope rotates with respect to the source. that is. the parallactic angle of the sources (and consequently its polarization angle) rotates during the observation.," During the source tracking, the reference system of an $-$ azimuth mounted telescope rotates with respect to the source, that is, the parallactic angle of the sources (and consequently its polarization angle) rotates during the observation."191 Following ? the time-dependent rotation component B of the Mülller matrix can be extracted and applied separately from. the instrumental component of the Mülller matrix. T is assumed to be constant with time.," Following \citet{Turlo} the $-$ dependent rotation component $\mathbf{B}$ of the Mülller matrix can be extracted and applied separately from the instrumental component of the Mülller matrix, $\mathbf{T}$ is assumed to be constant with time."192 The overall Mülller matrix can be then expressed as M=TB. where B is the rotation matrix of a linear rotating system (rotation of 9) ? and ? published major works on how to manage polarimetry with a radio interferometer and ? adapted their study to single-dish antennas equipped with linear dipoles.," The overall Mülller matrix can be then expressed as $\mathbf M = \mathbf{TB}$, where $\mathbf{B}$ is the rotation matrix of a linear rotating system (rotation of $\vartheta$ ) \citet{McKinnon} and \citet{Hamaker1} published major works on how to manage polarimetry with a radio interferometer and \citet{Johnston} adapted their study to single-dish antennas equipped with linear dipoles."193 Nowadays telescopes are commonly equipped with scalar feeds and hybrid or wave-guide elements that supply circularly polarized outputs., Nowadays telescopes are commonly equipped with scalar feeds and hybrid or wave-guide elements that supply circularly polarized outputs.194 In the following we describe. step by step. how this instrumentation can affect. in terms of instrumental polarization. the measurement of the incoming radiation.," In the following we describe, step by step, how this instrumentation can affect, in terms of instrumental polarization, the measurement of the incoming radiation."195 A typical receiving chain is sketched in Fig. 1.., A typical receiving chain is sketched in Fig. \ref{p:receiver}. .196" In Table 1. we summarize ournotation,", In Table \ref{t:notation} we summarize ournotation.197 The targets of single-dish observations are. typically. point- with respect to the antenna beam and the observables are, The targets of single-dish observations are typically point-like with respect to the antenna beam and the observables are198employed ppreviously on 512? simulations (e.g.?)..,employed previously on $512^3$ simulations \citep[e.g.][]{Knollmann2008}.199 'To summarize the choice of recommended parameters:, To summarize the choice of recommended parameters:200siiall correlations at low-£).,small correlations at $\ell$ ).201 The resultiug baud-powers are compared with the unu-differenced. baud-powers to check for svstematic deviations., The resulting band-powers are compared with the un-differenced band-powers to check for systematic deviations.202 This test is particularly scusitive to auv residual chopper offsets., This test is particularly sensitive to any residual chopper offsets.203 We find no siguificaut deviation iun the LAIT sui banud-powers from the un-differenced baud-powers., We find no significant deviation in the LMT sum band-powers from the un-differenced band-powers.204 When compared with the mode ACDM power spectrum. we do notice a slight rise in the LMT suma baud-powers for (22300.," When compared with the model $\Lambda$ CDM power spectrum, we do notice a slight rise in the LMT sum band-powers for $\ell>2300$."205 It is difficult to assess the significance or the origin of this low level treud., It is difficult to assess the significance or the origin of this low level trend.206 ILlowever. even if it is caused by a residual svsteimatic effect. the contribution to the joiut baud-powers would be sialler than the statistical uncertainty in the reported baud powers in this paper after accounting for the factor of 3 amplification.," However, even if it is caused by a residual systematic effect, the contribution to the joint band-powers would be smaller than the statistical uncertainty in the reported band powers in this paper after accounting for the factor of 3 amplification."207 At frequencies below the peak of the CMD intensity (~200 CITz). the coutribution of extra-galactic radio poiut- sources to the observed CAIB temperature anisotropy decreases rapidly with increasing observing frequency.," At frequencies below the peak of the CMB intensity $\sim 200\,$ GHz), the contribution of extra-galactic radio point sources to the observed CMB temperature anisotropy decreases rapidly with increasing observing frequency."208" Iu addition to the negative spectral indices of the majority of the radio sources. the flux-to-temperature conversion factor. (dB,dieaipn) reaches a nuit as the observing frequency approaches the peak of the CMD."," In addition to the negative spectral indices of the majority of the radio sources, the flux-to-temperature conversion factor, $(dB_\nu/dT_{CMB})^{-1}$, reaches a minimum as the observing frequency approaches the peak of the CMB."209 Iu particular. this factor is nearly 15 times smaller at 150€IIz than at 30 GIIz.," In particular, this factor is nearly 15 times smaller at $150\,$ GHz than at $30\,$ GHz."210 The measurements of ACBAR are therefore mmch less susceptible to contamination bv radio poiut sources than experiments operating at 30 GITz such as CDI DIMÁA. and ΧΑ.," The measurements of ACBAR are therefore much less susceptible to contamination by radio point sources than experiments operating at $30\,$ GHz such as CBI, BIMA, and VSA."211 We coustruct templates using the positious of the known radio sources frou the L85 GIIz Parkes-MIT-NBRAO (PAIN) strvey (?).. aud project out their coutributions to the power spectruni estimations.," We construct templates using the positions of the known radio sources from the $4.85\,$ GHz Parkes-MIT-NRAO (PMN) survey \citep{wright94}, and project out their contributions to the power spectrum estimations."212 Using the method described by IL we remove thei from the data without making assunuptions about their fluxes.," Using the method described by K04, we remove them from the data without making assumptions about their fluxes."213 Of 200 PAIN sources in the observed CXMB fields. we detected the euicding quasars and six additional sources with significance ercater than >2.86.," Of 200 PMN sources in the observed CMB fields, we detected the guiding quasars and six additional sources with significance greater than $>2.8\sigma$."214 These sources tend to have shallow. and in some cases inverted spectral iudices.," These sources tend to have shallow, and in some cases inverted spectral indices."215 Table 2. lists the parameters of the PAIN sources that are detected in the ACBAR fields: the detection threshold of >2.56. corresponds to a false detection rate of 1.," Table \ref{tab:pmnsources} lists the parameters of the PMN sources that are detected in the ACBAR fields; the detection threshold of $>2.8\sigma$, corresponds to a false detection rate of 1."216 The uncertainties are calculated using Monte-Carlo sinulatious. and are cominatec by contributions from CAB primary anisotropics.," The uncertainties are calculated using Monte-Carlo simulations, and are dominated by contributions from CMB primary anisotropies."217 With the exception of the enicing quasar in cach of the CMD ποιά». he effect of removing the point sources on the baud-»owers is not sienificaut.," With the exception of the guiding quasar in each of the CMB fields, the effect of removing the point sources on the band-powers is not significant."218 Thermal enmüssiou from interstellar dust also las4. =< xteutial to contaminate the measured power spectrin., Thermal emission from interstellar dust also has the potential to contaminate the measured power spectrum.219 The ACBAR CMD fields are located in the regions of low Galactic dust emission., The ACBAR CMB fields are located in the regions of low Galactic dust emission.220 2? (FSD) combined observations youn IRAS. COBE/DIRBE. aud COBE/FIRAS to eenerate a multi-component dust model that predicts the hermal cuuission at CAMB frequencieswith an angular resolution of 6 arciuiuutes.," \citet{finkbeiner99} (FSD) combined observations from IRAS, COBE/DIRBE, and COBE/FIRAS to generate a multi-component dust model that predicts the thermal emission at CMB frequencieswith an angular resolution of 6 arcminutes."221 We apply the ACBAR filtering o the predicted dust maps for 150 GIIz. and fined the expected RMS to be at the pA level.," We apply the ACBAR filtering to the predicted dust maps for $150\,$ GHz, and find the expected RMS to be at the $\mu K$ level."222 Assuniug the ACBAR maps contain the FSD dust templates with auplitudes paramctrized by the quantity £. the observed maps Z can be written as the stun {ελ|ofrsp.," Assuming the ACBAR maps contain the FSD dust templates with amplitudes parametrized by the quantity $\xi$, the observed maps $T$ can be written as the sum $T_{CMB}+\xi T_{FSD}$."223 After cross-corrclating the dust template maps TEgsp with the observed maps. we fud that the Calactic dust is undetectable in the ACBAR 150 GIIz data.," After cross-correlating the dust template maps $T_{FSD}$ with the observed maps, we find that the Galactic dust is undetectable in the ACBAR $150\,$ GHz data."224 The I-0 upper lait on the amplitude paramcter is $«2.6. consistent with the FSD predictions(46. £=1).," The $\sigma$ upper limit on the amplitude parameter is $\xi< 2.6$, consistent with the FSD predictions, $\xi=1$ )."225 As in the case for the radio source flux measurements. the uncertainty ia & is dominated bv the CAIB primary anisotropies.," As in the case for the radio source flux measurements, the uncertainty in $\xi$ is dominated by the CMB primary anisotropies."226 Therefore. dust with the morphology of the FSD aaps docs not sienificantly contribute to the observed anisotropy power.," Therefore, dust with the morphology of the FSD maps does not significantly contribute to the observed anisotropy power."227 Nonetheless. the reported baud-powers are calculated with the dust template mode projected out in each of the fields.," Nonetheless, the reported band-powers are calculated with the dust template mode projected out in each of the fields."228 Dust onüsson from high redshift star forming ealaxies can be a significant foreground contaminant im wullimeter wavelength CAIB maps., Dust emission from high redshift star forming galaxies can be a significant foreground contaminant in millimeter wavelength CMB maps.229 Recent observations with SCUBA/JOAMT ην. Bolocaan/CSO (?).. and MAMBO/IRAM. (2)... provide constraints on both the source counts and the spectral dependence of these proto-ealaxies.," Recent observations with SCUBA/JCMT \citep{smail02,chapman02,borys03}, Bolocam/CSO \citep{laurent05}, and MAMBO/IRAM \citep{greve04}, provide constraints on both the source counts and the spectral dependence of these proto-galaxies."230 However. despite the tremendous progress made in studyiug these sources. their contributions at 150 Gz are still liehly uncertain.," However, despite the tremendous progress made in studying these sources, their contributions at $150\,$ GHz are still highly uncertain."231 The uucertainties come frou the low number statistics dn source counts and spectral dependence. difficulties iu modeling the survey. bias aud conrpleteness. and poorly studied augular correlations.," The uncertainties come from the low number statistics in source counts and spectral dependence, difficulties in modeling the survey bias and completeness, and poorly studied angular correlations."232 Iu the absence of more decisive measurements. such as iüght be produced by the ongoing survey and the experiment. we ignore the clustering uoise conrponeut and estimate the Poisson contribution frou these proto-galaxies at the ACBAR observing frequencies.," In the absence of more decisive measurements, such as might be produced by the ongoing survey and the experiment, we ignore the clustering noise component and estimate the Poisson contribution from these proto-galaxies at the ACBAR observing frequencies."233 The SCUBA results coustrain the SbOjuu source counts with an uncertainty of ~—1405., The SCUBA results constrain the $850\mu$ m source counts with an uncertainty of $\sim 40\%$.234/ There are considerable uucertaimties in extrapolating this result to ower frequencies because the process depends not ouly ou the dust properties. but also ou the cosmic star ormation historv and the source evolution.," There are considerable uncertainties in extrapolating this result to lower frequencies because the process depends not only on the dust properties, but also on the cosmic star formation history and the source evolution."235" Observatious carried out at two different waveleneths. 1.22unu with MAMBO. aud {ντι with Dolociun. can potentiallv xovide this extrapolation plenomenologically,"," Observations carried out at two different wavelengths, $1.2\,$ mm with MAMBO, and $1.1\,$ mm with Bolocam, can potentially provide this extrapolation phenomenologically."236 ? πια hat the MAMDO and SCUBA source counts agree if he MAMDBO counts are scaled up in flux by a factor of 2.5. corresponding to a spectral depeudeuce of ὧνx ye ," \citet{greve04} find that the MAMBO and SCUBA source counts agree if the MAMBO counts are scaled up in flux by a factor of 2.5, corresponding to a spectral dependence of $S_\nu\propto \nu^{2.65}$ ."237We scale the fit for the SCUBA source count results (7) 150C€IIz (2112) according this spectral dependence.," We scale the fit for the SCUBA source count results \citep{borys03} to $150\,$ GHz $2\,$ mm) according this spectral dependence."238" Usine the formulae giveu in ?.. we fud the contribution of these sources to the CAIB power spectruui at 150 GIIz to be D,~τί2500)?I7."," Using the formulae given in \citet{scott99}, we find the contribution of these sources to the CMB power spectrum at $150\,$ GHz to be ${\cal D}_{\ell}\sim 37(\ell/2500)^2 \mu {\rm K}^2$."239 Since most of this fluctuation power comes frou sources with fluxes between 0.1 aud 1014Jw. the result is nisensitive to a flux cut- ereater than 1014Jy.," Since most of this fluctuation power comes from sources with fluxes between $0.1$ and $10\,$ mJy, the result is insensitive to a flux cut-off greater than $10\,$ mJy."240 The Dolocan team (?) found fewersources at τα than MAMBO did at 1.2 uuu. iurplviug a steeper spectrum for the sources.," The Bolocam team \citep{laurent05} found fewersources at $1.1\,$ mm than MAMBO did at $1.2\,$ mm, implying a steeper spectrum for the sources."241" A fluctuation analysis of the same data also suggested lower source counts, especially for fluxes « ιν (7)."," A fluctuation analysis of the same data also suggested lower source counts, especially for fluxes $< 1$ mJy \citep{maloney05}."242" From Figure 15 of 7.. we estimate the spectra between 850 yan aud Ldn to seale as S,x vl."," From Figure 15 of \citet{laurent05}, we estimate the spectra between 850 $\mu$ m and $1.1\,$ mm to scale as $S_\nu\propto \nu^{4}$ ."243 Compared with the 3zpIK? at (=2500 from the ATAMIBO/SCUBA extrapolation. which is just below the instrmucutal noise of ACDAR. the Bolocain/SCUBA model results in a negligible value of tls?.," Compared with the $37\mu{\rm K}^2$ at $\ell=2500$ from the MAMBO/SCUBA extrapolation, which is just below the instrumental noise of ACBAR, the Bolocam/SCUBA model results in a negligible value of $4 \mu{\rm K}^2$ ."244 Future sub-iillimeter aud millimeter observations are needed to fully characterize the properties of these sources., Future sub-millimeter and millimeter observations are needed to fully characterize the properties of these sources.245 In the interpretation of the ACDAR data. we," In the interpretation of the ACBAR data, we"246field. across the brighter main core is more complex. in the northeast of the core. the magnetic field has a northeast-southwest direction. but towards the southwest of the core. it is clirected northwest-southeast. and in the very southwest of the region there is a polarisation null.,"field across the brighter main core is more complex, in the northeast of the core, the magnetic field has a northeast-southwest direction, but towards the southwest of the core, it is directed northwest-southeast, and in the very southwest of the region there is a polarisation null."247 Phe abrupt change in the orientation of the magnetic field may be due to two cores being present. although unresolved in the JCME beam.," The abrupt change in the orientation of the magnetic field may be due to two cores being present, although unresolved in the JCMT beam."248 Previous observations of OLI maser emission have. vielded estimates of the line of sight magnetic field. strength of ~ 0.35 mG toward the source. with maximum strengths of 0.5 m (Alavoetal.2004).," Previous observations of OH maser emission have yielded estimates of the line of sight magnetic field strength of $\sim$ 0.35 mG toward the source, with maximum strengths of 0.5 mG \citep{mayo}."249. Phe polarimetry presented here lead to estimates of the plane of the sky. component of the field strength to be ~ 0.9 mC. consistent with the previous line of sight measurements.," The polarimetry presented here lead to estimates of the plane of the sky component of the field strength to be $\sim$ 0.9 mG, consistent with the previous line of sight measurements."250 The molecular complex Coronae Australis is  129 pe away from the Sun (Marraco&Itvdgren1981).. and is dominated bv the centrally condensed core centred. near the emission line star It Cr X. Nutter.Ware-Phompson&André(2006) have recently studied this source in the submillimetre. and reveal three submillimetre peaks within the main source. named SMM-LAD and €. SAIAL-LA is located in the south-eastern part on the core. with SALM-1B and C being located in the north-east and northwest of the core.," The molecular complex Coronae Australis is $\sim$ 129 pc away from the Sun \citep{marraco}, and is dominated by the centrally condensed core centred near the emission line star R Cr A. \citet{nutter} have recently studied this source in the submillimetre, and reveal three submillimetre peaks within the main source, named SMM-1A,B and C. SMM-1A is located in the south-eastern part on the core, with SMM-1B and C being located in the north-east and northwest of the core."251 lt Cr A is one of the more evolved. sources in the polarimetry sample. and is classified as a Llerbig Ae star (Alarraco&Rwegren 1981).," R Cr A is one of the more evolved sources in the polarimetry sample, and is classified as a Herbig Ae star \citep{marraco}. ."252. There is a CO bipolar outflow with a position angle of ~90° in the region of It CrX (Walker.Lada&Llartigan1985) but more recent. molecular line mapping by Andersonetal.(1997) has cast doubt on whether 1t CrA is the driving source., There is a CO bipolar outflow with a position angle of $\sim$ in the region of R CrA \citep{walker85} but more recent molecular line mapping by \citet{anderson} has cast doubt on whether R CrA is the driving source.253 The polarimetry of the 1t CrX region (fig., The polarimetry of the R CrA region (fig.254 LL) indicate that the magnetic field is roughly parallel to the direction of the outflow. (seealsoClarketal2000)., \ref{fig1}l l) indicate that the magnetic field is roughly parallel to the direction of the outflow \citep[see also][]{clark}.255. Phere are also polarisation nulls across this source. in two regions one in the north of the source. and one in the south.," There are also polarisation nulls across this source, in two regions — one in the north of the source, and one in the south."256 A plane of sky field component of 1: mG has been calculated., A plane of sky field component of 1 mG has been calculated.257 5140 is located in the Cepheus ring. at a distance of 900 pc (Preibisceh&Smith2002).," S140 is located in the Cepheus ring, at a distance of 900 pc \citep{preibisch}."258. 8140 has two outflows. one with a position angle of ~160°.. which is bipolar in nature. and another smaller scale outllow which has a position angle of EO (Preibisch&Smith2002).," S140 has two outflows, one with a position angle of $\sim$, which is bipolar in nature, and another smaller scale outflow which has a position angle of $\sim$ \citep{preibisch}."259. Phe higher resolution Ix'-xuxd data of Weigeltetal(2002). show arc-like structures ootruding from the northeast of the source. which are ooposed to trace outflow: cavities carved: out by material lowing away from 8140 Πο]., The higher resolution K'-band data of \citet{weigelt} show arc-like structures protruding from the northeast of the source which are proposed to trace outflow cavities carved out by material flowing away from S140 IRS1.260 The resolution of their daI is 240 milli-aresec and covers an area of approximately one beam width of the SCUBA data oesented here., The resolution of their data is 240 milli-arcsec and covers an area of $\times$ – approximately one beam width of the SCUBA data presented here.261 Phe SCUBA cata (lig., The SCUBA data (fig.262 1mm) show that he magnetic field vectors are ordered. in a north-south direction to the western side of the source. whereas in the east. the magnetic field is east-west orientated.," \ref{fig1}m m) show that the magnetic field vectors are ordered, in a north-south direction to the western side of the source, whereas in the east, the magnetic field is east-west orientated."263 OLI Zeeman observations (Baudryetal.LOOT) have revealed line of sight field estimates of |2.8 mG. CE estimates of the plane of sky component of the field lead to strengths of ~ 0.4 mG. which either indicates the magnetic field is mainlv in the line of sight. or. that the field is stronger on smaller scales.," OH Zeeman observations \citep{baudry97} have revealed line of sight field estimates of +2.8 mG. CF estimates of the plane of sky component of the field lead to strengths of $\sim$ 0.4 mG, which either indicates the magnetic field is mainly in the line of sight, or, that the field is stronger on smaller scales."264 There is à faint viclec of gas and dust extending from the east of the source. Curling northwards. which the magnetic field vectors seem to follow. running parallel to it (in the plane of the skv).," There is a faint ridge of gas and dust extending from the east of the source, curling northwards, which the magnetic field vectors seem to follow, running parallel to it (in the plane of the sky)."265 lt may be possible that the are-like structure in the SCUBA image is related in some wav to the smaller scale outllow cavities seen by Weigeltetal(2002).., It may be possible that the arc-like structure in the SCUBA image is related in some way to the smaller scale outflow cavities seen by \citet{weigelt}.266 58146 is located at a distance of 5.2 kpe (Wuetal.2005)., S146 is located at a distance of 5.2 kpc \citep{wu05}.267. A bipolar molecular outllow. in a north-south direction (north lobe blueshifted). driven by a star of spectral type O6.5 or earlier. (required to account for the ionisation of the region).," A bipolar molecular outflow, in a north-south direction (north lobe blueshifted), driven by a star of spectral type O6.5 or earlier (required to account for the ionisation of the region)."268 In the submillimetre data (fis., In the submillimetre data (fig.269 Inn) there are (wo cores in à north-south configuration., \ref{fig1}n n) there are two cores in a north-south configuration.270 There is à ridge of gas and dust seemingly. connecting the two cores., There is a ridge of gas and dust seemingly connecting the two cores.271 The polarimetry appears to be almost randomly distributed. with several polarisation nulls on the northern core. and to he north ancl south of the southern core. although there does not seem to be a relation between the intensity and »olarisation percentage.," The polarimetry appears to be almost randomly distributed, with several polarisation nulls on the northern core, and to the north and south of the southern core, although there does not seem to be a relation between the intensity and polarisation percentage."272 Phe large scatter of the polarimetry vector position angles mav suggest that the magnetic field is weak across this region., The large scatter of the polarimetry vector position angles may suggest that the magnetic field is weak across this region.273 It may also be explained. if he magnetic field was preclominantly in the line of sight (the outllow is mainly in the line of sigh see table 3)). which can cause random. polarisation patterns and/or low »olarisation percentages.," It may also be explained if the magnetic field was predominantly in the line of sight (the outflow is mainly in the line of sight – see table \ref{tab:pol}) ), which can cause random polarisation patterns and/or low polarisation percentages."274 Phe plane of sky component of the ield strength is caleulated to be 0.1 mC. which is one of the weakest measured for this sample.," The plane of sky component of the field strength is calculated to be 0.1 mG, which is one of the weakest measured for this sample."275 8157 is a dilluse nebula located: towards the Perseus arm at a distance of ~ 2.5 kpe. and is surrounded bv regions anc young open clusters. (Shirleyοἱal. 2003).," S157 is a diffuse nebula located towards the Cassiopeia-Perseus arm at a distance of $\sim$ 2.5 kpc, and is surrounded by regions and young open clusters \citep{shirley03}."276. Phe submillimetre data (fig., The submillimetre data (fig.277 loo) show the region is only slightly. more extended than a point source. with bright submiullimetre emission. extending southwareds [roni the main core.," \ref{fig1}o o) show the region is only slightly more extended than a point source, with bright submillimetre emission extending southwards from the main core."278 The polarimetry of the core show that the magnetic Geld vectors have an east-west direction to the south of the core. but in the north. the vectors are aligned roughly northwest-southeast.," The polarimetry of the core show that the magnetic field vectors have an east-west direction to the south of the core, but in the north, the vectors are aligned roughly northwest-southeast."279 Phere are two regions of null polarisation — one in the northwest of the source. the other in the southeast.," There are two regions of null polarisation — one in the northwest of the source, the other in the southeast."280 The plane of sky field strength component is estimated to be ~ 0.2 mC. W49 is in the galactie plane at a distance of 11.4 kpe (Ciwinn.&Reid 1992).. and is one of the most luminous regions in the Galaxy Smithctal.(~J10'..," The plane of sky field strength component is estimated to be $\sim$ 0.2 mG. W49 is in the galactic plane at a distance of 11.4 kpc \citep{gwinn}, and is one of the most luminous regions in the Galaxy \citet[$\sim$ 10$^{7}$,."281 There are two bright cores in this region (fig., There are two bright cores in this region (fig.282 Ipp) one in the northwest. W49N. and one in the southeast. WAOSL. with a third fainter source along the ridge inbetween the brighter two cores. \W49E. This ridge cillers from. those previously mentioned. as it is much more extended and less concentrated. (fainter).," \ref{fig1}p p) – one in the northwest, W49N, and one in the southeast, W49SE, with a third fainter source along the ridge inbetween the brighter two cores, W49E. This ridge differs from those previously mentioned, as it is much more extended and less concentrated (fainter)."283 There is a CO bipolar outflow from WON. with the redshifted lobe to the north and blueshifted," There is a CO bipolar outflow from W49N, with the redshifted lobe to the north and blueshifted"284eutities.,entities.285 Whether the upper branch of the stream in Fig., Whether the upper branch of the stream in Fig.286 or 3 ids the extension of the lower branch of stream is unclear in these data., \ref{fig_M31} or \ref{fig_cutout} is the extension of the lower branch of stream is unclear in these data.287 The upper branch certainly fits within the same ellipse aud has a shared metallicity distribution at the precision we can measure it., The upper branch certainly fits within the same ellipse and has a shared metallicity distribution at the precision we can measure it.288 However. the upper branch is less a stream than three cdominaut seenieuts separated by gaps of comparable size.," However, the upper branch is less a stream than three dominant segments separated by gaps of comparable size."289 It is of ercat interest to note that the seemeuts are tilted with respect to the gcucral path of the stream., It is of great interest to note that the segments are tilted with respect to the general path of the stream.290 Simulations show that tilted segments are some of the dominant structures frequently seen in chopped streams. as first noted iu Yoon.Johnston&Ποσο(2010).. aud as we show below.," Simulations show that tilted segments are some of the dominant structures frequently seen in chopped streams, as first noted in \citet{YJH:10}, and as we show below."291 The total huuinosity of the lower brauch of the stream is found by summing the over-deusities i one magnitude bius within a range of |-70.-10| ecliptic angle. to find a stream segineut luminosity of L3«I0 to AM;c0 mae.," The total luminosity of the lower branch of the stream is found by summing the over-densities in one magnitude bins within a range of [-70,-10] ecliptic angle, to find a stream segment luminosity of $4.3\times 10^5 \lsun$ to $M_i\simeq 0$ mag."292 The hunuinositv is the sumi over the ROBL.. stars alone., The luminosity is the sum over the RGB stars alone.293 To estima he total luminosity we use the cumulative stellar luminosity function of a stellar population of siuilar age and imoetalliitv. for which we chose the elobular cluster MI2.," To estimate the total luminosity we use the cumulative stellar luminosity function of a stellar population of similar age and metallicity, for which we chose the globular cluster M12."294 MI2 is an iuterimiediate metallicity cluster with a well studied stellar luminosity function (IIargisetal.2001)., M12 is an intermediate metallicity cluster with a well studied stellar luminosity function \citep{Hargis:04}.295. Using their distance modulus of LLOQ mag aud their I baud huuinositv fuuctiou we fiud about the light above M;=0 mag., Using their distance modulus of 14.0 mag and their I band luminosity function we find about of the light above $M_i=0$ mag.296 Therefore the corrected total Iuninosity is 7.1«107LL..., Therefore the corrected total luminosity is $7.4\times 10^5 \lsun$.297 The mean surface brightuess is 1060L..kpc3," The mean surface brightness is $1060\lsun\, \kpc^{-2}$."298 Tf we assume the stream is a wuiform deusitv cvlinder of 5 kpe radius aud a stellar mass-to-lieht ratio of. sav. 3 in the /biud. then the vohune stellar mass density is SOOMEkpe7.," If we assume the stream is a uniform density cylinder of 5 kpc radius and a stellar mass-to-light ratio of, say, 3 in the $i$ -band, then the volume stellar mass density is $500 \msun\, \kpc^{-3}$."299 This can be compared to the mass density iu the dark matter halo at sav 90 kpc. which for our adopted halo paramicters is 1.2«10Mkpe that is. about a factor of about 210 higher than the stellar stream mass density.," This can be compared to the mass density in the dark matter halo at say 90 kpc, which for our adopted halo parameters is $1.2\times 10^5\msun\, \kpc^{-3}$, that is, about a factor of about 240 higher than the stellar stream mass density."300 At 30 kpe the ratio is 3600 times more background halo dark matter than stream stellar ass., At 30 kpc the ratio is 3600 times more background halo dark matter than stream stellar mass.301 If dark matter from the progenitor is maxed iuto the stream then the mass deusitv of the streams would be proportionally increased. lowering the local lnass ratios but the stream is unlikely to be siguificantly selt-eravitating anywhere.," If dark matter from the progenitor is mixed into the stream then the mass density of the stream would be proportionally increased, lowering the local mass ratios but the stream is unlikely to be significantly self-gravitating anywhere."302 Quillen&Comparetta(2010) have raised the possibility that Jeans iustabilities cau be the source of of density variations iu the stream., \citet{QC:10} have raised the possibility that Jeans instabilities can be the source of of density variations in the stream.303" The Jeans leneth iu the stream, Ay=\/ro?/Cp evaluated with the stellar mass density and assumune an internal velocity dispersion of 10 iis about 100 kpc."," The Jeans length in the stream, $\lambda_J=\sqrt{\pi \sigma^2/G\rho}$ evaluated with the stellar mass density and assuming an internal velocity dispersion of 10 is about 400 kpc."304 Ou the other hand if the velocity dispersion were ax low ax 1 tthen the formal Jeaus leugth would be 10 kpc. comparable to the longest scale of density variations in the stream.," On the other hand if the velocity dispersion were as low as 1 then the formal Jeans length would be 40 kpc, comparable to the longest scale of density variations in the stream."305 Very low velocity dispersious are expected for slowly tidally stripped elobular clusters., Very low velocity dispersions are expected for slowly tidally stripped globular clusters.306 The preseuce of several globular clusters along the course of the stream (Mackeyetal.2010) aud the possibility that AudNNVIT (Richardsonetal.2011) as the source of the stream sugeests that the internal velocity dispersion ds at the evel that a chwart galaxv would create. ~καιum nore.," The presence of several globular clusters along the course of the stream \citep{Mackey:10} and the possibility that AndXXVII \citep{Richardson:11} is the source of the stream suggests that the internal velocity dispersion is at the level that a dwarf galaxy would create, $\sim 3 \kms$ or more."307 The resulting Jeans leugth is then roughly 120 spe. the size of the stream.," The resulting Jeans length is then roughly 120 kpc, the size of the stream."308 It is iuportaut to note that he stability analvsis is far from complete., It is important to note that the stability analysis is far from complete.309 First. the disrupted object may have its own dark matter which jas identical kanematics to the visible stars aud would Increase the mass density.," First, the disrupted object may have its own dark matter which has identical kinematics to the visible stars and would increase the mass density."310 Second. the stability analysis reeds to account for the very strong tidal feld of the ADS] dark halo.," Second, the stability analysis needs to account for the very strong tidal field of the M31 dark halo."311 Strong tides ecnerally act to suppress uuch instability. particularly in Llureelv radially oricuted streams. which is likely the case for the NW streai.," Strong tides generally act to suppress much instability, particularly in largely radially oriented streams, which is likely the case for the NW stream."312 Another mechanism that generates clumps iu tidal strealus is simply the pileup of stars at low velocity points of their epievclic orbits (INüpporotal.2008)., Another mechanism that generates clumps in tidal streams is simply the pileup of stars at low velocity points of their epicyclic orbits \citep{KMH:08}.313. The luups appear at 12a times the tidal radius., The lumps appear at $12\pi$ times the tidal radius.314 Since no progenitor object for the stream has been coufideutly identified. the tidal radius is verv uncertain.," Since no progenitor object for the stream has been confidently identified, the tidal radius is very uncertain."315 However if we take a low uuinosity dwarf galaxy of stellar plus a relatively low clark mass for a total progenitor dwart mass of AZ;=105NL... then at a galactic radius of ry=100 kpe. the tidal radius. CM° κ approximately 3.2 kpc.," However if we take a low luminosity dwarf galaxy of stellar plus a relatively low dark mass for a total progenitor dwarf mass of $M_d=10^8 \msun$, then at a galactic radius of $r_g=100$ kpc, the tidal radius, $r_g(M_d/(3M_g))^{1/3}$, is approximately 3.2 kpc."316 The epievelerj pileup(ΛΙ) separation will then be a a separation of 120 kpc., The epicycle pileup separation will then be at a separation of 120 kpc.317 That scale compares to the sizc of the stream., That scale compares to the size of the stream.318" The stellar mass in the stream alone. 2«106NE, for a fairly niuimal azimued mnass-to-lieh ratio of 3. gives about LO kpe epicvelic chuuping scale. which is at the upper eud of rauge of what we sec. bu still leaves the sinaller scale variations to be explained."," The stellar mass in the stream alone, $2\times 10^6 \msun$ for a fairly minimal assumed mass-to-light ratio of 3, gives about 40 kpc epicyclic clumping scale, which is at the upper end of range of what we see, but still leaves the smaller scale variations to be explained."319 We conclude that epicvclic pile-p is not a significa chunping factor for the NW streau., We conclude that epicyclic pile-up is not a significant clumping factor for the NW stream.320. If the progenitor svstem had large internal deusitv variations in a highlv order velocity Ποια they could be fed out iuto the stream to create lumps., If the progenitor system had large internal density variations in a highly order velocity field they could be fed out into the stream to create lumps.321 However. this idea ruus mto trouble with our astroplivsical knowledge of low mass galaxies.," However, this idea runs into trouble with our astrophysical knowledge of low mass galaxies."322 Dwarf spheroidal sealaxies are conrised largely of stars on randomly oricuted orbits aud have esseutially uo substructure. which when tidallv disrupted would lead to a smooth stream.," Dwarf spheroidal galaxies are comprised largely of stars on randomly oriented orbits and have essentially no substructure, which when tidally disrupted would lead to a smooth stream."323 Darf imegular ealaxies do have substantial substructure in their eas and voune stellar populations., Dwarf irregular galaxies do have substantial substructure in their gas and young stellar populations.324 The visible huuinositv (corrected for light below our observational lint) iu the lower branch of the stream is oulv 7&10?E..., The visible luminosity (corrected for light below our observational limit) in the lower branch of the stream is only $7\times 10^5 \lsun$.325 Dwarfs with such low huuinositics usually have comparable rotational and random velocities which would quickly blur out stellar structures deposited iuto a stream., Dwarfs with such low luminosities usually have comparable rotational and random velocities which would quickly blur out stellar structures deposited into a stream.326 Caven that Mackeyetal.(2010) associate 3 or E elobular clusters with the NW stream. which would lead to à unumsually lnieh ratio of elobular cluster huunmositv to progenitor Iuiuinositv. it it possible that the NW stream is only part of a larger streain that has vet to be clearly ideutified.," Given that \citet{Mackey:10} associate 3 or 4 globular clusters with the NW stream, which would lead to a unusually high ratio of globular cluster luminosity to progenitor luminosity, it it possible that the NW stream is only part of a larger stream that has yet to be clearly identified."327 The structures within star forming dwarf galaxies are often low mass stellar associations aud clusters that fairly quickly dissolve as a result of stellar mass loss and their own internal dynamics., The structures within star forming dwarf galaxies are often low mass stellar associations and clusters that fairly quickly dissolve as a result of stellar mass loss and their own internal dynamics.328 Spiral patterns are at best very chaotic in low mass chwart ivreenlars., Spiral patterns are at best very chaotic in low mass dwarf irregulars.329 Our assessment is that although carefully sub-structured voung progenitor disk system could produce some of the stream Iunipiuess. iypropriate astroplivsical svstenis do not exist.," Our assessment is that although carefully sub-structured young progenitor disk system could produce some of the stream lumpiness, appropriate astrophysical systems do not exist."330 Star streams uormally originate from the tidal dissolution of either dwarf galaxies or elobular clusters., Star streams normally originate from the tidal dissolution of either dwarf galaxies or globular clusters.331 Iu both cases tidal fields pull off outer stars or sometimes dissolve the cutire svstem and distribute the stars in, In both cases tidal fields pull off outer stars or sometimes dissolve the entire system and distribute the stars in332no positive average cross-correlations in a sky region of area more than half of the full angular coverage.,no positive average cross-correlations in a sky region of area more than half of the full angular coverage.333" From all these analyses, we cannot exclude the value of Iorc| being compatible with zero for any 0 within field-to-field fluctuations."," From all these analyses, we cannot exclude the value of $|\omega334_{TG}|$ being compatible with zero for any $\theta $ within field-to-field fluctuations."335 Thus we conclude that there is no significant cross-correlation detection., Thus we conclude that there is no significant cross-correlation detection.336 This situation is similar to that found for the SDSS 3D self-correlation by Sylos-Labini et al. (, This situation is similar to that found for the SDSS 3D self-correlation by Sylos-Labini et al. (337"2009), who also demonstrated that the field-to-field fluctuations are of the order of the signal in the previously announced discovery of baryon acoustic oscillations and scale anticorrelations.","2009), who also demonstrated that the field-to-field fluctuations are of the order of the signal in the previously announced discovery of baryon acoustic oscillations and large-scale anticorrelations."338 Other authors who calculated the cross-correlation between WMAP and SDSS galaxy counts measureda significant signal., Other authors who calculated the cross-correlation between WMAP and SDSS galaxy counts measured a significant signal.339" For instance, Cabré et al. ("," For instance, Cabré et al. ("3402006) measured a value of 0.5 uK and a significant positive ωτο for all angles lower than 20 degrees for the subsample 20«r21 over 5500 square degrees of SDSS-DR4.,2006) measured a value of $\omega _{TG}(3^\circ )\sim 0.5$ $\mu $ K and a significant positive $\omega _{TG}$ for all angles lower than 20 degrees for the subsample $20<r<21$ over 5500 square degrees of SDSS-DR4.341 Giannantonio et al. (, Giannantonio et al. (342"2008) obtained a value of wrg(3°)~0.3 wK for the subsample 18«r<21 of SDSS-DR6, excluding the southern Galactic hemisphere and high Galactic extinction regions.","2008) obtained a value of $\omega _{TG}(3^\circ )\sim3430.3$ $\mu $ K for the subsample $18<r<21$ of SDSS-DR6, excluding the southern Galactic hemisphere and high Galactic extinction regions."344" In addition, they found a significant positive signal out to 0—8 degrees."," In addition, they found a significant positive signal out to $\theta =8$ degrees."345" Their values are more or less compatible with our estimate of the cross-correlation function, within the re-sampling error bars and taking into account that their subsamples are slightly different."," Their values are more or less compatible with our estimate of the cross-correlation function, within the re-sampling error bars and taking into account that their subsamples are slightly different."346" However we do not measure significant cross-correlations, whereas these author do a result we cannot explain."," However we do not measure significant cross-correlations, whereas these author do a result we cannot explain."347 Cabré et al. (, Cabré et al. (3482006) and Giannantonio et al. (,2006) and Giannantonio et al. (349"2008) performed Monte Carlo simulations using mock maps, and obtained similar values or ones only slightly larger than a jack-knife.","2008) performed Monte Carlo simulations using mock maps, and obtained similar values or ones only slightly larger than a jack-knife."350 We do not know whether these disagreement are caused by mistakes in their calculations or whether their claim is that re-sampling errors represent the full errors., We do not know whether these disagreement are caused by mistakes in their calculations or whether their claim is that re-sampling errors represent the full errors.351" Other authors used only jack-knife technique errors (e.g., Sawangwit et al."," Other authors used only jack-knife technique errors (e.g., Sawangwit et al."352 2009)., 2009).353 A similar problem may affect the results of Raccanelli et al. (, A similar problem may affect the results of Raccanelli et al. (354"2008), who measured the cross-correlation between NVSS radio sources and WMAP anisotropies.","2008), who measured the cross-correlation between NVSS radio sources and WMAP anisotropies."355 Raccanelli et al. (, Raccanelli et al. (356"2008) calculate the error in simulating 1,000 mock NVSS maps by randomly distributing the unmasked pixels of the true NVSS maps.","2008) calculate the error in simulating 1,000 mock NVSS maps by randomly distributing the unmasked pixels of the true NVSS maps."357" We are concerned that this process might destroy part of the self-correlation of each map, and that the errors might not represent the full field-to-field fluctuations."," We are concerned that this process might destroy part of the self-correlation of each map, and that the errors might not represent the full field-to-field fluctuations."358 There has been considerable discussion of these errors (Cabréetal.2007;Giannantonioetal. 2008)..," There has been considerable discussion of these errors \cite{Cabre07,Giannantonio08}."359" However, against their claims one can infer from the analyses of this paper that: i) jack-knife or bootstrap methods do not calculate the whole error; ii) the level"," However, against their claims one can infer from the analyses of this paper that: i) jack-knife or bootstrap methods do not calculate the whole error; ii) the level"360its motion cireularized: The value of À is uncertain. but recent simulations by Alig et ((2011) suggest that Ac03. 0.,"its motion circularized: The value of $\lambda$ is uncertain, but recent simulations by Alig et (2011) suggest that $\lambda\sim 0.3$ – $0.4$."361 The disk radius is determined by the initial angular momentum of fluid elements with impact parameter by. for which eq (4)) vields Strikingly. (he disk size corresponds to the upper envelope of the maser disks with the canonical model parameters ez225kms|! and Az0.3.," The disk radius is determined by the initial angular momentum of fluid elements with impact parameter $b_0$ , for which eq \ref{eq:r-vs-b}) ) yields and so the mean surface density of the disk is Strikingly, the disk size corresponds to the upper envelope of the maser disks with the canonical model parameters $v \approx 225 \kms$ and $\lambda\approx 0.3$."362 In other words. the kinematics of the model naturally produces the observed extent of the maser disks. the empirical scaling with black hole mass.," In other words, the kinematics of the model naturally produces the observed extent of the maser disks, the empirical scaling with black hole mass."363 To estimate the surface density profile of the disk we note that eq (4)) implies that a thin evlindrical ring of cloud material with impact parameters in the range |b.b+db] ends up in the annulus [r.r+dr] in the disk.," To estimate the surface density profile of the disk we note that eq \ref{eq:r-vs-b}) ) implies that a thin cylindrical ring of cloud material with impact parameters in the range $[b,b+db]$ ends up in the annulus $[r,r+dr]$ in the disk."364 Thenmass conservation relates the cloud and disk surface densities via ὀπύμΕΝ)db=2xrX(r)dr implving that the disk has a L/r surface densitv prolile: where in the final expression rp. is the cvlindrical radius in parsecs.," Thenmass conservation relates the cloud and disk surface densities via $2\pi b\,\kappa(1.4 \, m_\mathrm{H} N_\mathrm{H})\,db\, = 2\pi r\,\Sigma(r)\,dr$ implying that the disk has a $1/r$ surface density profile: where in the final expression $\rpc$ is the cylindrical radius in parsecs."365 For this profile the disk mass enclosed within radius r of the DII is linearly. proportional to r. ancl the density al (he disk edge is 0.5X.," For this profile the disk mass enclosed within radius $r$ of the BH is linearly proportional to $r$, and the density at the disk edge is $ 0.5\,\Sigma_d$."366 This simple model disk must meet (hiree requirements in addition to its size (o guarantee (hat masers may be present over (he radial extent of the disk., This simple model disk must meet three requirements in addition to its size to guarantee that masers may be present over the radial extent of the disk.367 First. the mass given by eq (5)) must not exceed the mass of the incoming cloud. otherwise the disk will be truncated ab a smaller radius where the enclosed disk mass equals the cloud mass Ma.," First, the mass given by eq \ref{eq:Rd}) ) must not exceed the mass of the incoming cloud, otherwise the disk will be truncated at a smaller radius where the enclosed disk mass equals the cloud mass $M_\mathrm{cl}$."368 second. collisional inversion of the GGIIz sO transition requires densiües of LO! 10!em. 7. temperatures in the range KIX. and a sufficient column ofwater to," Second, collisional inversion of the GHz $_2$ O transition requires densities of $10^7$ $10^{11}\percc$ , temperatures in the range K, and a sufficient column ofwater to"369e100 km JF. reE kpe) the typical values are (in the same order as in eq. 2)).,"$v\sim100$ km $^{-1}$, $r\sim1$ kpc) the typical values are (in the same order as in eq. \ref{pdotprime}) )."370 The observed period change is dominated bv the intrinsic value. and all other terms are (tiny corrections of similar magnitudes.," The observed period change is dominated by the intrinsic value, and all other terms are tiny corrections of similar magnitudes."371 Thus ordinary radio pulsars are not useful for constraining the solar svstem acceleration on the order !., Thus ordinary radio pulsars are not useful for constraining the solar system acceleration on the order $a_{\odot}/c\la 10^{-18}$ $^{-1}$.372 An additional complication is that normal pulsars concentrate strongly toward the Galactic plane and therefore cannot be used to constrain the component of the acceleration perpendicular to the plane., An additional complication is that normal pulsars concentrate strongly toward the Galactic plane and therefore cannot be used to constrain the component of the acceleration perpendicular to the plane.373 The period distribution of radio pulsars is bimodal (Phinney&Ixulkarni1994)... with a minimun) at about 25 ms: there are only a few tens of objects with shorter periods.," The period distribution of radio pulsars is bimodal \citep{phin94}, with a minimum at about 25 ms; there are only a few tens of objects with shorter periods."374" These nillisecond pulsars (MSDPs) tvpicallv have P10? s. DPod4x107,5100 km |. and ro] kpe."," These millisecond pulsars (MSPs) typically have $P\sim10^{-2}$ s, $\dot{P}\sim4\times 10^{-20}$, $v\sim100$ km $^{-1}$, and $r\sim1$ kpc."375 In Chis case. The proper motion anc the possible acceleration-related corrections are comparable to (he intrinsic value of (he period derivative. whereas (he radial velocity term can be neglected.," In this case, The proper motion and the possible acceleration-related corrections are comparable to the intrinsic value of the period derivative, whereas the radial velocity term can be neglected."376 In Section 2.3. we implement a statistical approach to determining the solar svstem barvcenter acceleration using MSPs., In Section \ref{sec_msps} we implement a statistical approach to determining the solar system barycenter acceleration using MSPs.377 This approach is based on the plausible assumption that the intrinsic P distribution is independent of the position on the sky., This approach is based on the plausible assumption that the intrinsic $\dot{P}$ distribution is independent of the position on the sky.378 Therefore. if the solar svstem accelerates relative to the AISP population. there would be a svstematic dependence of the observed P on the position of the pulsars in the Galaxy (this idea was first explored by Harrison1977. [or a small sample of ordinary. pulsars).," Therefore, if the solar system accelerates relative to the MSP population, there would be a systematic dependence of the observed $\dot{P}'$ on the position of the pulsars in the Galaxy (this idea was first explored by \citealt{harr77} for a small sample of ordinary pulsars)."379 Such a change can be detected. provided that it is not too small compared to the dispersion of values 7 in the population.," Such a change can be detected, provided that it is not too small compared to the dispersion of values $\dot{P}'$ in the population."380 Therefore. in contrast to ordinary pulsars. MSPs offer the hope of measuring a solar acceleration.," Therefore, in contrast to ordinary pulsars, MSPs offer the hope of measuring a solar acceleration."381 From the ATNF pulsar (Manchesterοἱal.2005) as of January 2005. we selected all objects with P<25 ms.," From the ATNF pulsar \citep{manc05} as of January 2005, we selected all objects with $P'<25$ ms."382 We rejected all objects associated with globular clusters (as per Harris 1996)) since their accelerations relative to the solar svstem are likely to be, We rejected all objects associated with globular clusters (as per \citealt{harr96}) ) since their accelerations relative to the solar system are likely to be383carly and [ate morphological types separately in Figure 5..,early and late morphological types separately in Figure \ref{fig:mag_comp_sub}.384" Both figures show a rapid. drop in completeness. bevond b,;-—15. with approximately 50 per cent. completeness. at he surveys magnitude limit."," Both figures show a rapid drop in completeness beyond $\bJ$ =15, with approximately 50 per cent completeness at the survey's magnitude limit."385 Also shown in the figures are the best-fit) polynomial functions approximating the completeness variation over the magnitude range (dashed ines)., Also shown in the figures are the best-fit polynomial functions approximating the completeness variation over the magnitude range (dashed lines).386 These functions are used in applying the magnitucde-dependent completeness corrections in later papers., These functions are used in applying the magnitude-dependent completeness corrections in later papers.387 In early May anc mic April 1994. 34 redshifts were obtained with the Double Beam Spectrograph (DBS) on the ANU 2.3m telescope at SSO.," In early May and mid April 1994, 34 redshifts were obtained with the Double Beam Spectrograph (DBS) on the ANU 2.3m telescope at SSO."388 The target galaxies were a selection of objects observed: with FLAIR. for which redshifts were obtained., The target galaxies were a selection of objects observed with FLAIR for which redshifts were obtained.389 The purpose ofthese further observations was to explore the possibility that the ΕΤΕΛΙ survey might be biased against obtaining recshifts for some subset of objects in the target catalogue on the basis of their spectral tvpe or recshitt., The purpose of these further observations was to explore the possibility that the FLASH survey might be biased against obtaining redshifts for some subset of objects in the target catalogue on the basis of their spectral type or redshift.390 The top panel of FigureD 7 shows the numbermagnitude5 distribution of galaxies for which redshifts were obtained with FLAIR or the ANU 2.3m in the same fields., The top panel of Figure \ref{fig:ks} shows the number–magnitude distribution of galaxies for which redshifts were obtained with FLAIR or the ANU 2.3m in the same fields.391" Clearly the objects that failed to have a redshift measured. with FLAIR: (the ANU 2.3m sample) are at the faint end. with typical apparent magnitudes fainter than mi ,=15.5. consistent with Figure 4.."," Clearly the objects that failed to have a redshift measured with FLAIR (the ANU 2.3m sample) are at the faint end, with typical apparent magnitudes fainter than $\mbJ$ =15.5, consistent with Figure \ref{fig:mag_comp_all}."392 The bottom. panel of Figure 7 shows the numberredshift) distribution. of the ΕΤΕΛΙ and. 2.3m samples. along with the cumulative clistributions from cach sample.," The bottom panel of Figure \ref{fig:ks} shows the number--redshift distribution of the FLAIR and 2.3m samples, along with the cumulative distributions from each sample."393 The numberredshift’ cistributions for the FLALR sample ancl 2.3m. sample appear quite Consistent., The number–redshift distributions for the FLAIR sample and 2.3m sample appear quite consistent.394 We use the sstatistic to quantitatively test. the hypothesis that. the ealaxies with and without FLALII redshifts come from the same underlving redshift population., We use the statistic to quantitatively test the hypothesis that the galaxies with and without FLAIR redshifts come from the same underlying redshift population.395 We find that the two redshift distributions are consistent. at. the 74. per cent confidence level., We find that the two redshift distributions are consistent at the 74 per cent confidence level.396 We can conclude. therefore. that the galaxies observed with PLALR for which no redshifts were obtained have the same redshift. clistribution as those for which we obtained FLAIR. redshifts.," We can conclude, therefore, that the galaxies observed with FLAIR for which no redshifts were obtained have the same redshift distribution as those for which we obtained FLAIR redshifts."397 Phe only redshift. bias in the. FLASLI survey is that due to maenituce-cependent incompleteness. and can therefore be corrected in a straightforward manner.," The only redshift bias in the FLASH survey is that due to magnitude-dependent incompleteness, and can therefore be corrected in a straightforward manner."398 Vhe ΕΙΛΙ survey includes a total of redshifts over 605 cee? of skv., The FLASH survey includes a total of redshifts over 605 $^2$ of sky.399 An example subset of the FLASLI survey catalogue is. presented in. ‘Table 2.., An example subset of the FLASH survey catalogue is presented in Table \ref{tab:catalogue}.400 The full survey catalogue is available from NASA's Astrophysical Data Centre (ADC: http:/f/acde.gslenasa.gov/adc.html) and he Centre de. Donnéces astronomiques de. Strasbourg (CDS: http://edsweb.u-strasbg.fr/)., The full survey catalogue is available from NASA's Astrophysical Data Centre (ADC; http://adc.gsfc.nasa.gov/adc.html) and the Centre de Donnéees astronomiques de Strasbourg (CDS; http://cdsweb.u-strasbg.fr/).401 Each entry in the table contains: the Schmidt field. the galaxy was observed. in. he ealaxw’s aand ((DB1950). the magnitude (b;). major axis diameter (arcmin). morphological tvpe. veliocentric radial velocity ancl error. 1)). and source reference for the radial velocity (FLAIR. 2.3m. JD. ZCAT).," Each entry in the table contains: the Schmidt field the galaxy was observed in, the galaxy's and (B1950), the magnitude $\bJ$ ), major axis diameter (arcmin), morphological type, heliocentric radial velocity and error ), and source reference for the radial velocity (FLAIR, 2.3m, NED, ZCAT)."402 The magnitudes were extinction-corrected according to Durstein HILleiles (1982).. uxing lp—θαV).," The magnitudes were extinction-corrected according to Burstein Heiles \shortcite{burstein:82}, using $A_B =4034.0*E(B-V)$."404 Figure S. displavs those galaxies in the ΕΕΛΙ survey catalogue with measured. redshifts in an equal-area Aitolf projection., Figure \ref{fig:skyzplot} displays those galaxies in the FLASH survey catalogue with measured redshifts in an equal-area Aitoff projection.405 Clusters and filamentary structures are clearly evident across the survey region: the ddominates the survey region near (/.56)-(312.. 31)): the Ηνάνα cluster of galaxies is clearly visible at the lower right of the plot: a number of other prominent Abell clusters are indicated on the figure.," Clusters and filamentary structures are clearly evident across the survey region: the dominates the survey region near $(l,b)$, ); the Hydra cluster of galaxies is clearly visible at the lower right of the plot; a number of other prominent Abell clusters are indicated on the figure."406 The fattened structure spanning the survey regionH north-south at about 137NT! (see also FigureEM 1)) is thePlane.. the projection onto the sky of the Local Supercluster.," The flattened structure spanning the survey region north-south at about $13^h$ (see also Figure \ref{fig:hcc}) ) is the, the projection onto the sky of the Local Supercluster."407 The redshift distribution of the ΕΤΕΛΑΡΙ survey is shown in Figure 9.., The redshift distribution of the FLASH survey is shown in Figure \ref{fig:nz}. .408 Ht extends out to about 30000 aand has a median depth of about 10000., It extends out to about 30000 and has a median depth of about 10000.409ο Vhe APAI-Stromlo (Lovedayetal.1992) and Durham/UIlST (Ratelilfectal.1998) redshift) surveys cover factors of 6 and 2 times more sky respectively to a similar depth., The APM-Stromlo \cite{loveday:1992} and Durham/UKST \cite{ratcliffe:1998} redshift surveys cover factors of 6 and 2 times more sky respectively to a similar depth.410 Llowever both these surveys use. sparse- strategies (1-in-20 ancl l-in-3 respectively). which allow greater sky. coverageat the expense of reducing the," However both these surveys use sparse-sampling strategies (1-in-20 and 1-in-3 respectively), which allow greater sky coverageat the expense of reducing the"411"From (he model we find Q,=0.005640.0007. which can be compared wilh the Fukugita. Hogan Peebles (1998.FIIP) value of Q,=0.0037JHMM μμ.","From the model we find $\Omega_{\star}=0.0056\pm0.0007$, which can be compared with the Fukugita, Hogan Peebles (1998,FHP) value of $\Omega_{\star}=0.0037^412{+0.0019}_{-0.0013}$ $^{-1}_{0.66}$."413 Although the error bars overlap. our value is 50% higher.," Although the error bars overlap, our value is $\sim50\%$ higher."414 Further. the model shows that Mj; Missoig=0.90.2 compared to ~0.3 [rom FUP. ~1 from Schechter Dressler (1987). and ~1.840.2 from Benson et (( 2002).," Further, the model shows that $_{disk}$ $_{spheroid}=4150.9\pm0.2$ compared to $\sim0.3$ from FHP, $\sim1$ from Schechter Dressler (1987), and $\sim1.3\pm0.2$ from Benson et ( 2002)."416 Finally. the gas from the Population ΕΤΗ phase that did not go into making the galaxy is considered to be a diffuse component and contributes 0.002. (," Finally, the gas from the Population III phase that did not go into making the galaxy is considered to be a diffuse component and contributes $\Omega_{diffuse}=0.044\pm0.002$ . ("417This assumes that the mass in dark condensed remnants is negligible).,This assumes that the mass in dark condensed remnants is negligible).418 Here il is considered separate [from the WILIIN aidtentatively identified with (he gas responsible for producing the local Lyman à. forest whieh apparently could significantly exceed of all local barvons (Stocke et 22003)., Here it is considered separate from the WHIM andtentatively identified with the gas responsible for producing the local Lyman $\alpha$ forest which apparently could significantly exceed of all local baryons (Stocke et 2003).419 Two additional consistency cheeks on our model are now considered., Two additional consistency checks on our model are now considered.420 The first is concerned with the origin of the damped Lyman a systems (DLAs) lor which any complete model of chemical evolution applied to the entire universe should account., The first is concerned with the origin of the damped Lyman $\alpha$ systems (DLA's) for which any complete model of chemical evolution applied to the entire universe should account.421 Figure 6 shows the asstuned age-metallicitv relations (converted to metallicitv-redshilt relations) on which are superimposed a selection of observations of DLA's [rom Ixulkorni Fall (2002) and Prochaska et ((2003)., Figure 6 shows the assumed age-metallicity relations (converted to metallicity-redshift relations) on which are superimposed a selection of observations of DLA's from Kulkarni Fall (2002) and Prochaska et (2003).422 Recall that the two relations apply to each of the metal poor clusters. and metal rich clusters.," Recall that the two relations apply to each of the metal poor clusters, and metal rich clusters."423 Ia order to make a robust statistical estimate of the probability. (hat anv particular component is intersected by à random line of sight. one needs to know the distribution in area as well as the number ancl size of the protogalactic clumps. parameters which are not specified by the model.," In order to make a robust statistical estimate of the probability that any particular component is intersected by a random line of sight, one needs to know the distribution in area as well as the number and size of the protogalactic clumps, parameters which are not specified by the model."424" At most. we can calewlate the amount of gas remaining in each component as a function of [Fe/Il],"," At most, we can calculate the amount of gas remaining in each component as a function of [Fe/H]."425 The range over which this amount is >1% of the original amounts of 3.5. and 8.5x LOMAL . is indicated by the thickened parts of the lines in Figure 6.," The range over which this amount is $>$ of the original amounts of 3.5, and $\times10^{10}$ M $_{\odot}$ is indicated by the thickened parts of the lines in Figure 6."426 As it stands the model does not predict either an age-metallicity relation nor a metallicity distribution function for the disk component. but only the mass of gas that is available. 3.2x LOMAL..," As it stands the model does not predict either an age-metallicity relation nor a metallicity distribution function for the disk component, but only the mass of gas that is available, $3.2\times10^{10}$ $_{\odot}$."427 This comparison with the observations of DLAs is necessarily crude. but there appear to be no major inconsistencies.," This comparison with the observations of DLA's is necessarily crude, but there appear to be no major inconsistencies."428 secondly. we note that both the large reservoir of low angular momentum gas released alter the first collapse and a conduit to the center of the galaxy. (along (he rotation axis) are ingredients required for Duelling and possibly creating a black hole.," Secondly, we note that both the large reservoir of low angular momentum gas released after the first collapse and a conduit to the center of the galaxy (along the rotation axis) are ingredients required for fuelling and possibly creating a black hole."429 Thus. the model naturally provides an environment conducive to the establishment of a relationship between the mass of a central black hole and of the surrounding bulge stars.," Thus, the model naturally provides an environment conducive to the establishment of a relationship between the mass of a central black hole and of the surrounding bulge stars."430 If Ας ΑΓ=0.002 (Magorrian οἱ 11998). then the mass of the black hole in the model is Me=2οd0.5x10* M... comparable to the black hole mass in the nucleus of M31.," If ${\bullet}$ $_431{bulge}=0.002$ (Magorrian et 1998), then the mass of the black hole in the model is ${\bullet}=2.6\pm0.5\times10^{7}$ $_{\odot}$, comparable to the black hole mass in the nucleus of M31."432" When divided by (he assumed comoving volume one finds p,=2.630.5x LOAM. /Mpc for the local black hole density which is in excellent agreement. with the recent determination of this quantity by Yu Tremaine (2002).", When divided by the assumed comoving volume one finds $\rho_{\bullet}=2.6\pm0.5\times10^{5}$ $_{\odot}$ $^{3}$ for the local black hole density which is in excellent agreement with the recent determination of this quantity by Yu Tremaine (2002).433 Further. the maximum SFR of the bulge component occurs at reclshilt zc 2.3. which is close to the redshift at which the space density of luminous quasars is a," Further, the maximum SFR of the bulge component occurs at redshift $\sim2.3$ , which is close to the redshift at which the space density of luminous quasars is a"434van Paradijs.J. IXouveliotou.C Wijers.RAALJ. 2000. Anni.,"van Paradijs,J, Kouveliotou,C Wijers,R.A.M.J, 2000, Annu."435Rev.Astron.,Rev.Astron.436Astrophys. Wooslev.s. 2001. Proc.,"Astrophys, 38, Woosley,S. 2001, Proc."437 Rome Svimp..," Rome Symp.,"438 Eds., Eds.439 E Costa. E. Frontera. and J. Hjorth (Springer:Berlin) p. Zhang.W. Frver.C.L. 2001. ApJ. 550. 357.," E. Costa, F. Frontera, and J. Hjorth (Springer:Berlin) p. Zhang,W. Fryer,C.L. 2001, ApJ, 550, 357."440 Zhuge.X.. Centrella.J.M. MeMillan.L.W. 1994. Phys.," Zhuge,X., Centrella,J.M. McMillan,L.W. 1994, Phys."441 Rev. D. 1994.," Rev. D, 1994,"442in Fig. H1..,"in Fig. \ref{fig:res},"443 where the percentages of galaxies classitied correctly is shown for each method., where the percentages of galaxies classified correctly is shown for each method.444" In the case of the +) spectral classification and the Fisher diseriminant it is possible to change the relative success rates between types simply by adopting different ""cuts! in these continuous parameters.", In the case of the $\eta$ spectral classification and the Fisher discriminant it is possible to change the relative success rates between types simply by adopting different `cuts' in these continuous parameters.445 This is demonstrated in Fig., This is demonstrated in Fig.446 |] where the success rates are shown for both an +<1.4 cut (2dFGRS default) and an jj<<2 cut., \ref{fig:res} where the success rates are shown for both an $\eta<-1.4$ cut (2dFGRS default) and an $\eta<-2$ cut.447 In general. the most successful method of correlating galaxy morphology with spectra appears to be the Artificial Neural Network. which can achieve consistently high success rates for both types of galaxies.," In general, the most successful method of correlating galaxy morphology with spectra appears to be the Artificial Neural Network, which can achieve consistently high success rates for both types of galaxies."448 However. in deciding whether to implement such an algorithm the relative pay-off must be weighed against the additional complexity of this algorithm.," However, in deciding whether to implement such an algorithm the relative pay-off must be weighed against the additional complexity of this algorithm."449 This is particularly true for this data-set since all three methods of classitication make roughly similar distinctions between morphological types., This is particularly true for this data-set since all three methods of classification make roughly similar distinctions between morphological types.450 However. with the advent of higher resolution (and signal-to-noise) spectra. it is possible that such advanced methods will become much more successful.," However, with the advent of higher resolution (and signal-to-noise) spectra, it is possible that such advanced methods will become much more successful."451 Note that in practical situations the interpretation of Fig., Note that in practical situations the interpretation of Fig.452 is not as straight-forward as it might at first seem., \ref{fig:res} is not as straight-forward as it might at first seem.453 One must also bear in mind the relative fractions of different galaxy types in the sample under consideration., One must also bear in mind the relative fractions of different galaxy types in the sample under consideration.454 In the case of the 2dFGRS. there are over 2 times as many Late type galaxies as Early types (2610 and 1289 galaxies respectively). so a decrease in the success rate of classifying Late types by implies there will be an additional 250 galaxies classitied as Early type. which is at least a contamination of the Early type sample.," In the case of the 2dFGRS, there are over 2 times as many Late type galaxies as Early types (2610 and 1289 galaxies respectively), so a decrease in the success rate of classifying Late types by implies there will be an additional 250 galaxies classified as Early type, which is at least a contamination of the Early type sample."455 This is particularly important if one wants to create a relatively ‘pure’ sample of a particular type of galaxy (e.g. for efficient use of telescope time during observational follow-ups)., This is particularly important if one wants to create a relatively `pure' sample of a particular type of galaxy (e.g. for efficient use of telescope time during observational follow-ups).456Studies of stellar populations are of high importance for understanding the ormation and evolution of the Allkv Way.,Studies of stellar populations are of high importance for understanding the formation and evolution of the Milky Way.457 Iu this context. it has been discussed whether the Calactic halo consists of more than one population.," In this context, it has been discussed whether the Galactic halo consists of more than one population."458 The uonolithic collapse model of Egecu ct al. (1962)), The monolithic collapse model of Eggen et al. \cite{eggen62}) )459 corresponds to a single population. but from a study. of globular clusters. Searle and Ziuu (1978)) sugeested that the halo coutaims two populations: /) au inner. old. fattened population with a sheh prograde rotation formed during a dissipative collapse. aud 1) au outer. vounecr. spherical population accreted from dwarf ealaxies.," corresponds to a single population, but from a study of globular clusters, Searle and Zinn \cite{searle78}) ) suggested that the halo contains two populations: $i)$ an inner, old, flattened population with a slight prograde rotation formed during a dissipative collapse, and $ii)$ an outer, younger, spherical population accreted from dwarf galaxies."460 The presence of this dichotomy was supported by a study of ~20000 stars in the SDSS survey performed by Carollo et al. (2007)).," The presence of this dichotomy was supported by a study of $\sim \! 20\,000$ stars in the SDSS survey performed by Carollo et al. \cite{carollo07}) )."461 Elemental abuudances of stars du the solar neighborhood maw provide additional information about the halo populations., Elemental abundances of stars in the solar neighborhood may provide additional information about the halo populations.462 Iu this context. the ratio/Fe]. where a refers o the average abundance of Mg. Si. Ca. and Ti. is of particular interest.," In this context, the ratio, where $\alpha$ refers to the average abundance of Mg, Si, Ca, and Ti, is of particular interest."463 The o-clemoeuts are produced iaiulv caving Type II superuovae (SNe) explosious ou a short nuescale. (—10* vers) whereas iron ds also produced by Type Ia SNe on a much longer inescale (~LOP years).," The $\alpha$ -elements are produced mainly during Type II supernovae (SNe) explosions on a short timescale $\sim \! 10^7$ years), whereas iron is also produced by Type Ia SNe on a much longer timescale $\sim \! 10^9$ years)."464" Hence. ccan be used as a ""clock to probe the star formation ustory of a Galactic component."," Hence, can be used as a `clock' to probe the star formation history of a Galactic component."465 Several previous studies have focused on thepossible differences in for stars in the solar neighborhood., Several previous studies have focused on thepossible differences in for stars in the solar neighborhood.466 Fulbright (2002)). Stephens Bocsegaard (2002)). and Caatton et al. (2003))," Fulbright \cite{fulbright02}) ), Stephens Boesgaard \cite{stephens02}) ), and Gratton et al. \cite{gratton03}) )"467 all finel evidence that stars associated with the outer halo jave lower than stars connected to the inner halo., all find evidence that stars associated with the outer halo have lower than stars connected to the inner halo.468 The differences iu found iu these studies are. however. uot larecr than .lddex. aud it is unclear whether the distribution of lis continuous or bimodal.," The differences in found in these studies are, however, not larger than dex, and it is unclear whether the distribution of is continuous or bimodal."469 Nissen Schuster (1997)) achieved a higher precision measurement of and found evidence of a bimodal distribution of ffor 13 alo stars with 1.3«|Fe/Tl]< O.L but a larger sample of these *umetalrich’ halo stars needs to be observed to confirin these fiudines aud study possible correlations with kinematies.," Nissen Schuster \cite{nissen97}) ) achieved a higher precision measurement of and found evidence of a bimodal distribution of for 13 halo stars with $-1.3 < \feh < -0.4$, but a larger sample of these `metal-rich' halo stars needs to be observed to confirm these findings and study possible correlations with kinematics."470 In this Letter. we present the first results of such a study.," In this Letter, we present the first results of such a study."471 Stars were selected from the Schuster et al. (2006)), Stars were selected from the Schuster et al. \cite{schuster06}) )472 weby-3 catalogue of high-velocity aud metal-poor stars., $uvby$ $\beta$ catalogue of high-velocity and metal-poor stars.473 To ensure that a star has a high probability of belonging to the halo population. the total space velocity.Έναν with respect to the local standard of rest (LSR) was coustrained to be larger than ((Veun et al. 2001)).," To ensure that a star has a high probability of belonging to the halo population, the total space velocity, with respect to the local standard of rest (LSR) was constrained to be larger than (Venn et al. \cite{venn04}) )."474 Furthermore. the Stromuueren indicesgy). ing and eq were used to select dsvarfs aud subegiauts with 5200Ta G300I.I& and |Fe/TI]Z1.6.," Furthermore, the Strömmgren indices, $m_{\rm 1}$ and $c_{\rm 1}$ were used to select dwarfs and subgiants with $5200 < \teff < 6300$ K and $\feh \simgt -1.6$."475 This produced a list of about 200 stars. of which 37 jiwe VLT/UVES spectra available iu the ESO/ST-ECF Science Archive (Table 1)).," This produced a list of about 200 stars, of which 37 have VLT/UVES spectra available in the ESO/ST-ECF Science Archive (Table \ref{table:UVES.RV}) )."476 Furherimore. 16 stars with hick-disk kinematics and UVES spectra were included.," Furthermore, 16 stars with thick-disk kinematics and UVES spectra were included."477 Du addition. 53 randomly selected stars were observed with he FIbre fed Echelle Spectrograph (FIES) at the Nordic Optical Telescope (NOT) (Table 2)).," In addition, 53 randomly selected stars were observed with the FIbre fed Echelle Spectrograph (FIES) at the Nordic Optical Telescope (NOT) (Table \ref{table:FIES.RV}) )."478 Six were found to e double-lined spectroscopic binaries aud excluded., Six were found to be double-lined spectroscopic binaries and excluded.479 All FIES stars aud most of the UVES stars are brighter than ΈτILI. three having V. — 11.2. 12.2. aud 12.8.," All FIES stars and most of the UVES stars are brighter than $V = 11.1$, three having $V$ = 11.2, 12.2, and 12.8."480 The average distance is ppc with Dj44;=335 ppc., The average distance is pc with $D_{\rm max} = 335$ pc.481 The UVES spectra cover the spectral region 15060 - aud have resolutions A!2:55000 aud signal-to-noise ratios (S/N) from 250 to 500.," The UVES spectra cover the spectral region 4800 - and have resolutions $R\simeq \! 55\,000$ and signal-to-noise ratios $S/N$ ) from 250 to 500."482 The FIES spectra range frou 1000 to . but only the 1700 -," The FIES spectra range from 4000 to , but only the 4700 -"483to generate suitable ancillary response Iunction files and these were used in conjunetion with (he most up-to-date response matrix (swxwt0to2.20010101v00T.1i).,to generate suitable ancillary response function files and these were used in conjunction with the most up-to-date response matrix $\_$ 20010101v007.rmf).484 All spectra were grouped to a minimum of 20 counts per bin in order to facilitate (he use of the 4? statistic in XSPEC. the results of which are discussed below.," All spectra were grouped to a minimum of 20 counts per bin in order to facilitate the use of the $\chi^2$ statistic in XSPEC, the results of which are discussed below."485 AIL fits included an eneregv scale offset that was allowed to vary. between -Q.1 and O keV (ο account Lor possible CCD bias measurement uncertanties., All fits included an energy scale offset that was allowed to vary between -0.1 and 0 keV to account for possible CCD bias measurement uncertanties.486 Errors are given al the 90 per cent level (e.g.. NA? = 2.7 for one degree of freedom).," Errors are given at the 90 per cent level (e.g., $\Delta\chi^{2}$ = 2.7 for one degree of freedom)."487 Figure 6. shows a sample of the resulting spectra at selected epochs and Table 2. gives further information. including the results of Che model fits described below.," Figure \ref{spectra} shows a sample of the resulting spectra at selected epochs and Table \ref{tab1} gives further information, including the results of the model fits described below."488 The initial count rale ab /=3.17 days doubled in less than two davs to peak at around 30 counts !. then fell again until the emergence of a new soft component beginuing al Observation 8 (/=26 days) which dominated (he XRT spectrum by Observation 9 (Osborneοἱal.2006a./=29days) - see Figure 6..," The initial count rate at $t = 3.17$ days doubled in less than two days to peak at around 30 counts $^{-1}$, then fell again until the emergence of a new soft component beginning at Observation 8 $t = 26$ days) which dominated the XRT spectrum by Observation 9 \citep[$t = 29$ days]{osb06a} - see Figure \ref{spectra}."489 The gross evolution of (he spectrum was a progressive softening curing the period of these X-ray observations., The gross evolution of the spectrum was a progressive softening during the period of these X-ray observations.490 In the earliest phase of the outburst. the ejecta from the WD surface are expected (o traverse the binary svstem on a timescale /e 2d for v;=3000 km ! (Dobrzecka& 2000).," In the earliest phase of the outburst, the ejecta from the WD surface are expected to traverse the binary system on a timescale $t \sim 2$ d for $v_{ej} = 3000$ km $^{-1}$ \citep{dob94,fek00}."491. Subsequent (o (his. using models previously applied to the evolution of SNR. Bode&Ixahn(1985) found that Phase I (where (the ejecta were still important in supplving energy {ο the shocked stellar wind) ended in the first few davs after the 1935 outburst.," Subsequent to this, using models previously applied to the evolution of SNR, \citet{bod85} found that Phase I (where the ejecta were still important in supplying energy to the shocked stellar wind) ended in the first few days after the 1985 outburst."492 Indeed. using the parameters derived by OBrienetal.(1992) in Equations 5- and 9 of Bode&Ixahn(1935).. we derive /ον 6.2d for the duration of Phase I (see Figures 7. and 83)).," Indeed, using the parameters derived by \citet{obr92} in Equations 5 and 9 of \citet{bod85}, we derive $t \simeq 6.2$ d for the duration of Phase I (see Figures \ref{NH} and \ref{vs}) )."493 Fherealter. the remnant evolution can be well modelled by the instantaneous release of energv al a point.," Thereafter, the remnant evolution can be well modelled by the instantaneous release of energy at a point."494 The Bode&Kahn(1985). analvtical model of the 1985 outburst led to the conclusion that at the time the ENOSAT observations began (/= 55d). the remmant was in the transition between Phase II (where a blast wave is being driven into (he wind and is so hot as to be effectively adiabatic) and Phase HI (where the shocked material is well cooled by radiation).," The \citet{bod85} analytical model of the 1985 outburst led to the conclusion that at the time the EXOSAT observations began $t = 55$ d), the remnant was in the transition between Phase II (where a blast wave is being driven into the wind and is so hot as to be effectively adiabatic) and Phase III (where the shocked material is well cooled by radiation)."495 Initially. a double shock svstem is established. with the forward shock being driven into the stellar wind and a reverse shock being driven into the otherwise unshocked ejecta.," Initially, a double shock system is established, with the forward shock being driven into the stellar wind and a reverse shock being driven into the otherwise unshocked ejecta."496 In Phase IL. the Primakolf similarity solution for this type of explosion into an r7 density distribution (where r is the racial distance Hom the site of the explosion) gives," In Phase II, the Primakoff similarity solution for this type of explosion into an $r^{-2}$ density distribution (where $r$ is the radial distance from the site of the explosion) gives"497contains characteristic information about the complete set e of model parameters. and in order to reveal accurate estimates for all of them. each of them requires appropriate coverage.,"phases contains characteristic information about the complete set of model parameters, and in order to reveal accurate estimates for all of them, each of them requires appropriate coverage."498 While the mid-phase gets squeezed for impact angles (8p of the order of the angular Einstein radius 8p or larger. the part of the light curve with ηZ»1 becomes indistinguishable from the baseline for strongly-blended events. so that on the approach to baseline. the offset magnification is proportional tov+ rather than #4. given that still & πο ΕΕ“.1.," While the mid-phase gets squeezed for impact angles $u\,\theta_\rmn{E}$ of the order of the angular Einstein radius $\theta_\rmn{E}$ or larger, the part of the light curve with $u \gg 1$ becomes indistinguishable from the baseline for strongly-blended events, so that on the approach to baseline, the offset magnification is proportional to $u^{-1}$ rather than $u^{-4}$ , given that still $u \ll 1$ for $|t-t_0| \gg 1$."499 It is well-known that an appropriate estimate of event parameters at early event stages is not feasible. and in particular the peak magnification is regularly overpredicted by à maximum-likelihood estimate corresponding to minimizing the sum of normalized squared deviations. X7.," It is well-known that an appropriate estimate of event parameters at early event stages is not feasible, and in particular the peak magnification is regularly overpredicted by a maximum-likelihood estimate corresponding to minimizing the sum of normalized squared deviations $\chi^2$."500" Just for this reason. ? had suggested to use a maximum-a-posteriori estimate instead. with a suitable prior,"," Just for this reason, \citet{Albrow:MAP} had suggested to use a maximum-a-posteriori estimate instead, with a suitable prior."501 While this brings the estimate closer to its expectation value. it does not get around the uncertainty.," While this brings the estimate closer to its expectation value, it does not get around the uncertainty."502 A closer eximation shows that the light curve is compatible with an infinite peak flux until roughly a third of the true offset magnification is reached., A closer eximation shows that the light curve is compatible with an infinite peak flux until roughly a third of the true offset magnification is reached.503" Moreover. it is a wing region 1.5νο5|ffu]S3byes that is best suited to determine the blend ratio g""=£pfe (and with it the time-scale /p. rather than the immediate vicinity of the peak."," Moreover, it is a wing region $1.5~t_{1/2} \la |t-t_0| \la 3~t_{1/2}$ that is best suited to determine the blend ratio $g^{(k)} = F_\rmn{B}^{(k)}/F_\rmn{S}^{(k)}$ (and with it the time-scale $t_\rmn{E}$, rather than the immediate vicinity of the peak."504 The rather long duration of this phase allows to obtain a suitable measurement without the need for very dense sampling., The rather long duration of this phase allows to obtain a suitable measurement without the need for very dense sampling.505 For an accurate prediction of the observed flux. a proper determination of the full set of model parameters is not required. so that local approximations can provide a reasonable substitute.," For an accurate prediction of the observed flux, a proper determination of the full set of model parameters is not required, so that local approximations can provide a reasonable substitute."506" In sharp contrast. proper knowledge of /p. which implies knowledge of the blend ratio gi""! and the magnification ο(ή). is a requirement for determining the event detection efficiency to planets (2) as well as for prioritising ongoing events in order to maximize it (22)."," In sharp contrast, proper knowledge of $t_\rmn{E}$, which implies knowledge of the blend ratio $g^{(k)}$ and the magnification $A(t)$, is a requirement for determining the event detection efficiency to planets \citep{fivelong} as well as for prioritising ongoing events in order to maximize it \citep{Han:priorities,webPLOP}."507 Without such information. one neither knows the amplitude. nor the duration. nor the location of potentially arising planetary signals.," Without such information, one neither knows the amplitude, nor the duration, nor the location of potentially arising planetary signals."508 Therefore. an efficient campaign for inferring the planet population from observed microlensing events needs to invest time into observations that allow to properly determine the event parameters. rather than just trying to detect planets in poorly determined events. where unsuitable assumptions about model parameters may yield to bad choices. or efforts could even turn out to be wasted if the planet detection efficieney cannot be assessed.," Therefore, an efficient campaign for inferring the planet population from observed microlensing events needs to invest time into observations that allow to properly determine the event parameters, rather than just trying to detect planets in poorly determined events, where unsuitable assumptions about model parameters may yield to bad choices, or efforts could even turn out to be wasted if the planet detection efficiency cannot be assessed."509 Building upon the findings presented in this paper. a more detailed study of event (Cunipredictability taking into account the specific capabilities of observing campaigns could hence provide importantclues towards optimizing strategies for detecting planets and determining their population statistics.," Building upon the findings presented in this paper, a more detailed study of event (un)predictability taking into account the specific capabilities of observing campaigns could hence provide importantclues towards optimizing strategies for detecting planets and determining their population statistics."510Stepkinetal.(2007) argue that errors in fitting or spectral baseline. removal alone cannot account for the line-width discrepancy: however including the known low-frequency spectral turnover of (e.g..Helmboldt&Kassim.2009) lowers the amount of predicted broadening and brings the model and observations into agreement.,"\cite{2007MNRAS.374..852S} argue that errors in fitting or spectral baseline removal alone cannot account for the line-width discrepancy; however including the known low-frequency spectral turnover of \citep[e.g.,][]{hk09} lowers the amount of predicted broadening and brings the model and observations into agreement."511 Most observations have focussed on sightlines towards known bright background sources or passing through gas-rich regions along the inner Galactic plane. but the detection rates have been low.," Most observations have focussed on sightlines towards known bright background sources or passing through gas-rich regions along the inner Galactic plane, but the detection rates have been low."512 Only two broader surveys for the lines have been published., Only two broader surveys for the lines have been published.513 Ericksonetal.(1995) found carbon RRL absorption complexes on roughly 30 sightlines between Galactic longitudes of aand 20°.. with Galactic latitudes |b]<2°.," \cite{1995ApJ...454..125E} found carbon RRL absorption complexes on roughly 30 sightlines between Galactic longitudes of and , with Galactic latitudes $|b| < 2\degr$."514 They concluded that there is no evidence for pressure broadening in their data. and that the linewidths reflect Doppler broadening and the presence of many clouds in their bbeam.," They concluded that there is no evidence for pressure broadening in their data, and that the linewidths reflect Doppler broadening and the presence of many clouds in their beam."515 Kantharia&Anantharamaiah(2001) found RRLs on 9 sightlines. with six of those in the Galactic longitude range- 352°«κ17°.," \cite{2001JApA...22...51K} found RRLs on 9 sightlines, with six of those in the Galactic longitude range $352\degr < \ell < 17\degr$."516 Combining their data with that of Ericksonetal.(1995) and information from observations on the same sightlines at higher frequencies. they conclude that the clouds likely have sizes of aand occur in photodissociation regions.," Combining their data with that of \cite{1995ApJ...454..125E} and information from observations on the same sightlines at higher frequencies, they conclude that the clouds likely have sizes of and occur in photodissociation regions."517 There is growing interest in the use of the 21-cm hyperfine transition of as a cosmological and astrophysical probe of the distant Universe., There is growing interest in the use of the 21-cm hyperfine transition of as a cosmological and astrophysical probe of the distant Universe.518 For comprehensive reviews. see Furlanettoetal.(2006) and Pritchard&Loeb(2008).," For comprehensive reviews, see \cite{fob06} and \cite{pl08}."519. For a portion of the Universe's history (1100xz 7). the baryonic content of the intergalactic medium (IGM) ts dominated byHt. and interactions between the adiabatic expansion of the IGM and various heating sources (e.g.. first stars. first black holes. dark matter decay) can produce either an absorption or emission signal relative to the cosmic microwave background (CMB).," For a portion of the Universe's history $1100 \lesssim520z \lesssim 7$ ), the baryonic content of the intergalactic medium (IGM) is dominated by, and interactions between the adiabatic expansion of the IGM and various heating sources (e.g., first stars, first black holes, dark matter decay) can produce either an absorption or emission signal relative to the cosmic microwave background (CMB)."521 At least three distinct epochs have been identified. which have corresponding spectral windows for the redshifted signal.," At least three distinct epochs have been identified, which have corresponding spectral windows for the redshifted signal."522 Estimates of the strength of these signals is both epoch and model dependent but are in the range 10-100mK., Estimates of the strength of these signals is both epoch and model dependent but are in the range 10–100.523. One of the key challenges to detecting any of these signals are the various foregrounds., One of the key challenges to detecting any of these signals are the various foregrounds.524" These foregrounds include radio frequency interference (RFI) from terrestrial radio transmitters. ionospheric phase corruptions. the Galactic synchrotron emission. and emission from extragalactic sources,"," These foregrounds include radio frequency interference (RFI) from terrestrial radio transmitters, ionospheric phase corruptions, the Galactic synchrotron emission, and emission from extragalactic sources."525 The various spectral windows for these signals also include frequencies where RRLs have been observed on Galactic plane sightlines., The various spectral windows for these signals also include frequencies where RRLs have been observed on Galactic plane sightlines.526 It is currently unknown at what strength RRLs might exist on sightlines at the high Galactic latitudes where most of the cosmological studies will be made., It is currently unknown at what strength RRLs might exist on sightlines at the high Galactic latitudes where most of the cosmological studies will be made.527 While 21-cm cosmological observations will clearly be conducted at high Galactic latitudes. in order to minimize the contribution of the Galactic synchrotronemission. even residual effects from RRLs might still vitiate the measurements.," While 21-cm cosmological observations will clearly be conducted at high Galactic latitudes, in order to minimize the contribution of the Galactic synchrotronemission, even residual effects from RRLs might still vitiate the measurements."528"IMF have fneb/fstars£25 and =3, respectively.","IMF have $f_\mathrm{neb}/f_\mathrm{stars}\approx 5$ and $\approx 3$, respectively."529" To bring nebular emission down to a level where it no longer affects the broadband fluxes of €10 Myr old pop III objects in any significant way (fneb/fstars~ 0.1), the ionizing escape fraction would have to be fes.20.95."," To bring nebular emission down to a level where it no longer affects the broadband fluxes of $\leq 10$ Myr old pop III objects in any significant way $f_\mathrm{neb}/f_\mathrm{stars}\sim 0.1$ ), the ionizing escape fraction would have to be $f_\mathrm{esc}\gtrsim 0.95$."530 It is doubtful whether pop III galaxies can be identified based on colour criteria for much longer than ~107 yr., It is doubtful whether pop III galaxies can be identified based on colour criteria for much longer than $\sim 10^7$ yr.531" 'The lifetimes of massive pop III stars are very short (=2-3x106yryearsforzz50-500stars;Schaerer 2002), and as soon as the first supernovaeMo explode, metals will be dispersed into the surrounding medium."," The lifetimes of massive pop III stars are very short \citep[$\approx 2$--$3\times 10^6$ yr years for $\approx 50$--$500\ M_\odot$ stars; ][]{Schaerer a}, and as soon as the first supernovae explode, metals will be dispersed into the surrounding medium."532" While the ejecta from the first pop III supernovae may take ~10° yr to cool sufficiently to be used in the subsequent formation of pop II and pop I stars (Greifetal. 2007), the time it takes before metal emission lines start to emerge from the surrounding nebula may be much shorter, perhaps no more than a few Myr."," While the ejecta from the first pop III supernovae may take $\sim 10^8$ yr to cool sufficiently to be used in the subsequent formation of pop II and pop I stars \citep{Greif et al. a}, the time it takes before metal emission lines start to emerge from the surrounding nebula may be much shorter, perhaps no more than a few Myr."533" At that point, the broadband signatures of metal-free nebulae discussed in Sect."," At that point, the broadband signatures of metal-free nebulae discussed in Sect."534 5 would be jeopardized., \ref{typeA} would be jeopardized.535" If feedback from the first supernovae clears the galaxy of photoionized gas, the SED would be dominated by direct star light from pop III stars until pop II/I star formation sets in (after ~105 yr), but even then, the unique spectral characteristics of pop III stars are only retained for ~107 yr (see Sect. 6))."," If feedback from the first supernovae clears the galaxy of photoionized gas, the SED would be dominated by direct star light from pop III stars until pop II/I star formation sets in (after $\sim 10^8$ yr), but even then, the unique spectral characteristics of pop III stars are only retained for $\sim 10^7$ yr (see Sect. \ref{typeC}) )."536 While Fig., While Fig.537 1 and Fig., \ref{spectra} and Fig.538" 2 would suggest that the SEDs of pop III galaxies are almost completely nebular in nature, there are a number of mechanisms that could diminish the impact of nebular emission."," \ref{fneb} would suggest that the SEDs of pop III galaxies are almost completely nebular in nature, there are a number of mechanisms that could diminish the impact of nebular emission."539" As shown in the simulations by Johnsonetal.(2009),, stellar feedback can drive the photoionized gas present in these systems outward, eventually pushing it beyond the virial radius of the dark matter halo."," As shown in the simulations by \citet{Johnson et al. b}, stellar feedback can drive the photoionized gas present in these systems outward, eventually pushing it beyond the virial radius of the dark matter halo."540" Once the HII region breaks out of its host halo, the gas density will be extremely low, resulting in a huge intergalactic nebula with very long recombination timescales."," Once the HII region breaks out of its host halo, the gas density will be extremely low, resulting in a huge intergalactic nebula with very long recombination timescales."541" This nebula, while very important for cosmic reionization, is probably of little importance for JWST observations of individual galaxies."," This nebula, while very important for cosmic reionization, is probably of little importance for JWST observations of individual galaxies."542" When attempting aperture photometry of galaxies surrounded by such large, low-surface brightness nebulae, the nebular contribution (appearing in both the object and sky annulus) may be almost completely subtracted away, effectively leading to the detection of a SED dominated by the more concentrated stellar component."," When attempting aperture photometry of galaxies surrounded by such large, low-surface brightness nebulae, the nebular contribution (appearing in both the object and sky annulus) may be almost completely subtracted away, effectively leading to the detection of a SED dominated by the more concentrated stellar component."543" Whether or how quickly the gas gets expelled depends on the star formation efficiency within a given halo, and also the form of the IMF."," Whether or how quickly the gas gets expelled depends on the star formation efficiency within a given halo, and also the form of the IMF."544" Having a very low ratio of stellar to dark matter mass, and/or an IMF without too many high-mass stars, would likely result in a situation in which the HII region stays completely confined within the CDM halo (forthelimitinge.g.Kitayamaetal. 2004)."," Having a very low ratio of stellar to dark matter mass, and/or an IMF without too many high-mass stars, would likely result in a situation in which the HII region stays completely confined within the CDM halo \citep[for the limiting case of just one pop III star in halos with different masses, see e.g.][]{Kitayama et al.}."545". In Fig. 3,,"," In Fig. \ref{schematic},"546 we have schematically defined three classes of objects in different stages of gas expulsion., we have schematically defined three classes of objects in different stages of gas expulsion.547" For type A, the ionization-bounded HII region (grey region) remains compact and confined well inside the virial radius (dashed line)."," For type A, the ionization-bounded HII region (grey region) remains compact and confined well inside the virial radius (dashed line)."548 This implies a Lyman-continuum escape fraction fec&0 and a maximal contribution from nebular emission to the overall SED of the galaxy., This implies a Lyman-continuum escape fraction $f_\mathrm{esc}\approx 0$ and a maximal contribution from nebular emission to the overall SED of the galaxy.549" In the case of type B, stellar feedback has pushed the HlILregion partly outside the virial radius (~1 kpc for a 1085Mg halo at zzz giving rise to partial leakage of ionizing radiation into 7-13),the intergalactic medium:"," In the case of type B, stellar feedback has pushed the HII-region partly outside the virial radius $\sim 1$ kpc for a $10^8\ M_\odot$ halo at $z\approx 7$ –13), giving rise to partial leakage of ionizing radiation into the intergalactic medium:"550"In the model of Usov(1998a.2001) the thermalized electron-positron pair [lux is given by where # is the radius of the star. Arp the thickness of the emitting region and ""m is the characteristic time of thermalization of the electrons.","In the model of \citet{Us98a,Us01} the thermalized electron-positron pair flux is given by where $R$ is the radius of the star, $\Delta r_{E}$ the thickness of the emitting region and $t_{th}^{-1}$ is the characteristic time of thermalization of the electrons."551" For /,e/,! the expression (8). with €=2ynfz(zp/T). was assumed (Usov19982.2001)."," For }$ the expression $t_{th}^{-1}\approx \left( 3/2\pi552\right) \left( \alpha /\sqrt{\pi }\right) \left( T^{2}/\varepsilon553_{F}\right) J\left( \xi \right) $ , with $\xi =2\sqrt{\alpha /\pi554}\left( \varepsilon _{F}/T\right) $, was assumed \citep{Us98a,Us01}."555. The function J(£) is defined bx Eq. (3))., The function $J\left( \xi \right) $ is defined by Eq. \ref{j}) ).556 In order to obtain a more realistic description of the tühermalized electron-positron flux of the electrosphere which takes into account the inhomogeneities of the electron and electric field distributions. we assume that. due to its dependence on the Fermi energy T). the thermalization time is also a function of the distance z to the quark star surface.," In order to obtain a more realistic description of the thermalized electron-positron flux of the electrosphere which takes into account the inhomogeneities of the electron and electric field distributions, we assume that, due to its dependence on the Fermi energy $\varepsilon _{F}\approx \mu _e=V(z,T)$ , the thermalization time is also a function of the distance $z$ to the quark star surface."557 Therefore we define (he (hermalized electron-positron flux from the quark star's surface in the generalized emission model as The variations of the thermalized electron-positron fIuxes in the Usov model aud in the eeneralized electron-positron emission mechanism are presented in Fig., Therefore we define the thermalized electron-positron flux from the quark star's surface in the generalized emission model as The variations of the thermalized electron-positron fluxes in the Usov model and in the generalized electron-positron emission mechanism are presented in Fig.558 12., 12.559" In the limit of low temperatures the condition T/V,—0 holds with a verv good By approximation."," In the limit of low temperatures the condition $T/V_q\rightarrow5600$ holds with a very good approximation."561assuming that the number densitv of thee.-e pairs can be approximated bv Eq. (33)).," By assuming that the number density of the $e^{+}-e^{-}$ pairs can be approximated by Eq. \ref{dens}) ),"562 in which all the parameters are estimated near the quark star surface zzz(. the electron-positron thermalized flux from the electrosphere of the quark stars can be represented in an approximate [orm as For T=0 the electron-positron flux from the quark star surface is zero.," in which all the parameters are estimated near the quark star surface $z\approx 0$, the electron-positron thermalized flux from the electrosphere of the quark stars can be represented in an approximate form as For $T=0$ the electron-positron flux from the quark star surface is zero."563 Generally. FL is a funetionof the electrostatic potential at the quark star surface V; and of the temperature T.," Generally, $F_{\pm }$ is a functionof the electrostatic potential at the quark star surface $V_q$ and of the temperature $T$."564 The presence of à surface magnetic Ποια //can also enhance the pair production rate (Nikishoy 1970).., The presence of a surface magnetic field $H$can also enhance the pair production rate \citep{Ni70}. .565 The magnetic fiekl increases the pair production rate by a factor, The magnetic field increases the pair production rate by a factor566the profile transitious from r> in the inuer parts to r3 dn the outer parts. but we parauictrize it iustead through the coucentration parameter. which is defined as the ratio of a halo’s virial radius to itsscale radius (c=Reiss).,"the profile transitions from $r^{-1}$ in the inner parts to $r^{-3}$ in the outer parts, but we parametrize it instead through the concentration parameter, which is defined as the ratio of a halo's virial radius to itsscale radius $c\equiv \Rvir/r_s$ )."567 Finally. we adopt the concentration - mass relation given by Zheuectal.(2007) for the modification of Bullocketal.(2001): 6=HiN(Mmi 5. where ey=11. M. is the non-linear mass at the median redshitt (AL=1.2.10!LAL.+ 0286) of the siuuple for our choice of cosmology. aud .=.13.," Finally, we adopt the concentration - mass relation given by \citet{zheng07} for the modification of \citet{bullock01}: $c=\frac{c_{0}}{(1 + z)}\times(\frac{M_\mathrm{halo}}{M_{\ast}})^{-\beta}$ , where $c_{0}=11$, $M_{\ast}$ is the non-linear mass at the median redshift $\Mstar = 1.2\times 10^{12} \hMsun, z = 0.286$ ) of the sample for our choice of cosmology, and $\beta = .13$."568 We can caleulate the galaxy umimber deusity for a eiven HOD bv inteerating over halo mass and weighting the abundance of halos by the mean occupation of galaxies CNG ag: where dofdAI is the ciffereutial halo mass function., We can calculate the galaxy number density for a given HOD by integrating over halo mass and weighting the abundance of halos by the mean occupation of galaxies $\langle N \rangle _{M}$ : where $dn/dM$ is the differential halo mass function.569" We adopt the Warrenctal.(2006) halo mass function with the following cosmological model: ,,=0.25.Ὃν0.275.0,=0.01.fyOT.σς0.5... 1.0."," We adopt the \citet{warren06} halo mass function with the following cosmological model: $\Omega_{m}=0.25, \Omega_{\Lambda}=0.75, \Omega_{b}=0.04, h_{0}=0.7, \sigma_{8}=0.8, n_{s}=1.0$ ."570" Iu the halogg model. the galaxy two-point correlation function is given as the sium of the ""1-halo aud ""2-halo* terms (Zheng2001).. The 2-halo term is the coutribution of galaxy pairs found in separate dark matter halos."," In the halo model, the galaxy two-point correlation function is given as the sum of the “1-halo” and “2-halo” terms \citep{zheng04a}, The 2-halo term is the contribution of galaxy pairs found in separate dark matter halos."571 Therefore. at scales sanaller than the virial diameter of the smallest halos considered. the l-halo teri completely dominates the correlation function.," Therefore, at scales smaller than the virial diameter of the smallest halos considered, the 1-halo term completely dominates the correlation function."572" Zhengctal.(2008) found that the minium halo mass for LRGs is Afgan,1017AJ... which corresponds to virial radii of OS09)!AIpe."," \citet{zheng08} found that the minimum halo mass for LRGs is $\Mmin\cong10^{13.7}M_{\odot}$, which corresponds to virial radii of $0.8-0.9\hmpc$."573 Therefore. it is sufficient to oulv consider the I-halo teri. when modeling €(1) for our simall-scale data.," Therefore, it is sufficient to only consider the 1-halo term when modeling $\xi(r)$ for our small-scale data."574" The I-halo term of the two-point correlation function can be written as an integral over halo mass. where at cach mass we add the coutribution of ceutral-satellite aud satellite-satellite pairs: where Fite) aud Εν} are the pair separation distributions of centrablsatellite and satellite-satellite pairs. respectively,"," The 1-halo term of the two-point correlation function can be written as an integral over halo mass, where at each mass we add the contribution of central-satellite and satellite-satellite pairs: where $F_{\mathrm{cs}}(r)$ and $F_{\mathrm{ss}}(r)$ are the pair separation distributions of central-satellite and satellite-satellite pairs, respectively."575 The pair distribution for ceutral-satellite pairs is esseutially the same as the density profile of satellite galaxies. whereas the satellite-satellite pair distribution is equivalent to a couvolutiou of the density profile with itself.," The pair distribution for central-satellite pairs is essentially the same as the density profile of satellite galaxies, whereas the satellite-satellite pair distribution is equivalent to a convolution of the density profile with itself."576 We use the Shethetal.(2001) calculation for the couvolutioun of a truncated NEW profile with itself., We use the \citet{sheth01} calculation for the convolution of a truncated NFW profile with itself.577" As we discussed in αν, centralsatellite pairs will dominate the LRG correlation fiction duc to the large halo masses involved."," As we discussed in \ref{HOD}, central-satellite pairs will dominate the LRG correlation function due to the large halo masses involved."578 Therefore. the shape of the ziall-scale LRG correlation function should be very close to the shape of the satellite galaxy deusity profile.," Therefore, the shape of the small-scale LRG correlation function should be very close to the shape of the satellite galaxy density profile."579 AIOG measured the projected correlation function. so we lust trausforni our theoretical €(r) iuto ayy.) by integrating along the line of sight (Davis&PechlesZehavietal.2001111 Since we need to integrate €(r) to large radii in order to get a(ry). we cannot completely ignore the 2-halo term.," M06 measured the projected correlation function, so we must transform our theoretical $\xi(r)$ into $\wpp\rp$ by integrating along the line of sight \citep{davis83,zehavi04a}: Since we need to integrate $\xi(r)$ to large radii in order to get $\wpp\rp$, we cannot completely ignore the 2-halo term."580 However. (7) is a rapidly declining function of + and so a7) at the small scales we are cousideriug is not sensitive to variations in the 2-halo term.," However, $\xi(r)$ is a rapidly declining function of $r$ and so $\wpp\rp$ at the small scales we are considering is not sensitive to variations in the 2-halo term."581 For this reason. we use the best-fit 2-halo term from Zhengctal.(2008) instead of calculating it explicitly cach time we vary our HOD parameters.," For this reason, we use the best-fit 2-halo term from \citet{zheng08} instead of calculating it explicitly each time we vary our HOD parameters."582 We tested this by shifting the amplitude of the 2-halo term: by in cach direction aud we found that πρ) is not appreciably affected., We tested this by shifting the amplitude of the 2-halo term by in each direction and we found that $\wpp\rp$ is not appreciably affected.583" Now that we have mieasurenieuts of i, and a(7). as well as a iechanisin to predict these quantities from a eiven set of TOD parameters (e.e.. My. Ady. Ady. a). we can probe the parameter space aud find the region that gives a good fit to the data."," Now that we have measurements of $\ngal$ and $\wpp\rp$, as well as a mechanism to predict these quantities from a given set of HOD parameters (e.g., $\Mmin$, $\Mzero$, $\Mone$, $\alpha$ ), we can probe the parameter space and find the region that gives a good fit to the data."584 We use a Markov Chain Moute Carlo (AICAIC) iiethod for this purpose., We use a Markov Chain Monte Carlo (MCMC) method for this purpose.585 Specifically. we adopt the Metropolis-Tastines aleoritlin. which works as follows.," Specifically, we adopt the Metropolis-Hastings algorithm, which works as follows."586 The free paraictcrs are given initial values and 7 is computed for this startine »omt iu parameter space., The free parameters are given initial values and $\chi^2$ is computed for this starting point in parameter space.587 Steps are then chosen for the owanmeters and 4? is computed for this new location., Steps are then chosen for the parameters and $\chi^2$ is computed for this new location.588" The new location is added to the chain if ντAZ, or àου(QaN24)0/2]. where n is a random umber )etween 0 and 1."," The new location is added to the chain if $\chi^{2}_{new}<\chi^{2}_{old}$ or $n < \expon[-(\chi^{2}_{new}-\chi^{2}_{old})/2]$, where n is a random number between 0 and 1."589 If these conditions are not met. then he old location is repeated in the chain.," If these conditions are not met, then the old location is repeated in the chain."590 This process heu repeats until the chain has couverged., This process then repeats until the chain has converged.591 We tes for convereence bv starting three chains with οποσο initial paraincter values aud checking whether their final ραο. distribution functions are dn agreement., We test for convergence by starting three chains with different initial parameter values and checking whether their final parameter distribution functions are in agreement.592 We assune uniform priors for all parameters aud do no iurpose restrictions ou their rauges., We assume uniform priors for all parameters and do not impose restrictions on their ranges.593 Once a given chain has converged. we cau plot the jistoerani of paramicter values in the chain.," Once a given chain has converged, we can plot the histogram of parameter values in the chain."594 The mos ikelv value for cach parameter is given by the mean of its distribution. aud the associated errors are given x the extrema of the middle 68:34. of the distribution.," The most likely value for each parameter is given by the mean of its distribution, and the associated errors are given by the extrema of the middle $68.3 \%$ of the distribution."595 The best-fit parameters correspond to the combination of parameter values for which \7 is a nimium., The best-fit parameters correspond to the combination of parameter values for which $\chi^{2}$ is a minimum.596 For more details on MCMC techniques. see Duuklevetal.(2005).," For more details on MCMC techniques, see \citet{dunkley05}."597. As discussed in όν the Zhenectal.(2008). fit to the intermecdiate-scale LRG correlation fiction does not match the MOG data points when extrapolated imwards.," As discussed in \ref{intro}, , the \citet{zheng08} fit to the intermediate-scale LRG correlation function does not match the M06 data points when extrapolated inwards."598 We first investigate whether we can achieve a good fit to these new MOG small-scale data points while oulv varvingthe PON[M) part of the TOD., We first investigate whether we can achieve a good fit to these new M06 small-scale data points while only varyingthe $P(N|M)$ part of the HOD.599 We vary the following paralucters (also defined iu ??)):, We vary the following parameters (also defined in \ref{HOD}) ):600For S LO917. the short time scale aud the non-periocic behaviour of the times of vanishing amplitude secu to rule out the infiuence of a third component.,"For S 10947, the short time scale and the non-periodic behaviour of the times of vanishing amplitude seem to rule out the influence of a third component."601 We therefore suppose that these eclipses are caused by. circiunstellar matter or changes iu the accretion disk. variable in size and/or brightness," We therefore suppose that these eclipses are caused by circumstellar matter or changes in the accretion disk, variable in size and/or brightness."602 This supports earlier statements by Tall Ramsey (1992) that exteuded matter in RS CVn systelus mav play a ereater role than hitherto assuued., This supports earlier statements by Hall Ramsey (1992) that extended matter in RS CVn systems may play a greater role than hitherto assumed.603" Because of (1) the σπα. difference to the best-fit N-ray position of only —5"" ((far below the ταν position error). (2) the lack of ay other optical objects within the N-rav error circle brighter than 19 mae."," Because of (1) the small difference to the best-fit X-ray position of only $\sim$ (far below the X-ray position error), (2) the lack of any other optical objects within the X-ray error circle brighter than 19 mag."604 and (3) a N-vay-to-optical huninositv ratio of Lx/Lo©0.01 which is in the range exhibited by RS CV stars. we are quite certain about the ideutification of RN J2009.8|1557 with S 10917 Aql.," and (3) a X-ray-to-optical luminosity ratio of $L_{\rm X}/L_{\rm opt} \approx 0.01$ which is in the range exhibited by RS CVn stars, we are quite certain about the identification of RX J2009.8+1557 with S 10947 Aql."605 S 10917. Aql shows interesting properties dn its eclipsing light curve which nav contribute to a hetter understanding of the RS CVin systems., S 10947 Aql shows interesting properties in its eclipsing light curve which may contribute to a better understanding of the RS CVn systems.606 This work demonstrates the dmuportauce of plate archives where mauy still unknown secrets and information are hidden., This work demonstrates the importance of plate archives where many still unknown secrets and information are hidden.607 Detailed spectroscopic observations are urgently needed to further the uuderstauding of this cnigmatic source S 10091 Αι., Detailed spectroscopic observations are urgently needed to further the understanding of this enigmatic source S 10947 Aql.608 Also. continued long-term monitoring should determine the further evolution of the Oο curve. and thus cau prove/cdisprove he influence of a third body.," Also, continued long-term monitoring should determine the further evolution of the $O-C$ curve, and thus can prove/disprove the influence of a third body."609 Arne Tenden. following our request iu response to a sugecstionOO of the referce. has performed a CCD-based photometric calibration as given below.," Arne Henden, following our request in response to a suggestion of the referee, has performed a CCD-based photometric calibration as given below."61000.97412 0.97553 0.97555 ὃς e e 0.99360 Sd e - 0.97139,0.97412 0.97553 0.97555 5s ... ... 0.99360 5d ... ... 0.97139611"we write a=vA? l. and use a normalization function C'u(n.o) in place of Csr): Delining n,—Szig/3. we find: We note thatgoing from (50)) to (73)) takes three simple steps: (73)) becomes (50)) in the limit o70. as it should.","we write $\alpha = \sqrt{k^2 - 1}$, and use a normalization function $C_4612(n,\sigma )$ in place of $C_2 (n)$: Defining $n_p = -8 \pi {\mathrm i} \sigma /3$, we find: We note thatgoing from \ref{eq:A2basic}) ) to \ref{eq:A2basicperm}) ) takes three simple steps: \ref{eq:A2basicperm}) ) becomes \ref{eq:A2basic}) ) in the limit $\sigma \rightarrow 0$, as it should."613 For the Fourier synthesis we need. in general. to use a more complicated contour than the one shown in figure 3.. because wehave to take account of thebranch points of both & and o.," For the Fourier synthesis we need, in general, to use a more complicated contour than the one shown in figure \ref{fig:contour2}, because wehave to take account of thebranch points of both $k$ and $\alpha$."614 With our choice ofa=1/2z. however. the branch points ofa coalesce into a single point at n=i. and the contour of figure 3. is adequate.," With our choice of$\sigma = 1/2\pi$, however, the branch points of $\alpha$ coalesce into a single point at $n=-{\mathrm i} $, and the contour of figure \ref{fig:contour2} is adequate."615The result is shown in figure &..,The result is shown in figure \ref{fig:AT2EPS}.616 We can show easilv that in the limit jj5ox. clo satisfies a diffusion equation.," We can show easily that in the limit $\eta \rightarrow \infty$, $A_2$ satisfies a diffusion equation."617 The main terms come from the integral down the cut. and in this limit the integrand is concentrated close to p—0.," The main terms come from the integral down the cut, and in this limit the integrand is concentrated close to $n=0$."618" Note that this integral involves only A... the ""Held term: the 4, (7scalar potential”) term. being even in n. does not contribute."," Note that this integral involves only ${\cal K_{\sigma}}$, the “field term”; the ${\cal A_{\sigma}}$ (“scalar potential”) term, being even in $n$ , does not contribute."619 From this integral we obtain a simple analytic expression for the asvmptotic form: Whenx is also large. of order η. theterm involving the error function is negligible compared to the others.," From this integral we obtain a simple analytic expression for the asymptotic form: When$\chi$ is also large, of order $\sqrt{\eta}$, theterm involving the error function is negligible compared to the others."620 In figure 9 we plot i.i» against €= with ele computed using the asymptotic formula.," In figure \ref{fig:AT3EPS} we plot $\sqrt{\eta} A_2$ against $\xi = \chi/\sqrt{\eta}$ , with $A_2$ computed using the asymptotic formula."621 We wn.see. ids tending to a constant form: with our choice of abscissa. this is à Gaussian for large x and ap.," We see $\sqrt{\eta} A_2$ tending to a constant form; with our choice of abscissa, this is a Gaussian for large $\chi$ and $\eta$."622 For az4. this graph is incistinguishable from the one using integration around the contotr.," For $\eta \ge 4$, this graph is indistinguishable from the one using integration around the contour."623 We will now investigate the effect of à non-zero conductivity on the temperature of the plasma., We will now investigate the effect of a non-zero conductivity on the temperature of the plasma.624 We start [from ?.. equation 4.7.9: This expression is zero for any value of 7. butthe most important one for us is v=0.," We start from \citet{wein2}, equation 4.7.9: This expression is zero for any value of $\nu$ , butthe most important one for us is $\nu = 0$."625 We treat the plasma as a perfect [uid at rest in EV coordinates 9. X. 6. 6. so that the velocity {TE) has= 0.7= 12.3.," We treat the plasma as a perfect fluid at rest in FRW coordinates $\eta$ , $\chi$ , $\theta$ , $\phi$ , so that the 4-velocity $\left( U^0 , U^i \right)$ has$U^i = 0$ , $i=1,2,3$ ."626 Since land guo=I (ap).UU =P., Since $g_{\mu \nu } U^{\mu} U^{\nu} = -1$ and $g_{00} = R^2 (\eta)$ $U^0 = 1/R$ .627 Alultiplv (18)) by ο. sunm over g. and set the resulting expression equal to zero: weget," Multiply \ref{eq:wein4_7_9}) ) by $\sqrt{g}\,{\mathrm d} ^4 x\,U_{\nu}$ , sum over $\nu$ , and set the resulting expression equal to zero; weget"628aud references therein).,and references therein).629" Cabreraetal.(2007) confirmed the existence of the Eyes - Eis, correlation in the rest-frame for Ly Swift GRBs.", \cite{cab07} confirmed the existence of the $\Epkin$ - $E_{iso}$ correlation in the rest-frame for 47 Swift GRBs.630" These studies considered bright loug CRBs with known redshifts enabliug £;.., to be determined.", These studies considered bright long GRBs with known redshifts enabling $E_{iso}$ to be determined.631 This causes strong selection effect in the studied samples., This causes strong selection effect in the studied samples.632 It is possible that this selection effect cause c.g. tjo entire DATSE siuupe to follow the Amati-relation cither ouly in a modified version or even not at all. even though the relation holds for the truncated sample of bright GRBs (Nakar&Piran.2005:Butleretal..2007).," It is possible that this selection effect cause e.g. the entire BATSE sample to follow the Amati-relation either only in a modified version or even not at all, even though the relation holds for the truncated sample of bright GRBs \citep{napi05,but07}."633". DBOG obtainec that it is better to use EyNEj. with suitable « and bj for the BATSE sample (7,4.τα| :)).", BB06 obtained that it is better to use $\Epkin \propto E_{iso}^{a_1}\tau_{intr}^{b_1}$ with suitable $a_1$ and $b_1$ for the BATSE sample $\tau_{intr} =\tau/(1+z)$ ).634 Hence. if bi#0. then the Amati-relation is altered.," Hence, if $b_1 \neq 0$, then the Amati-relation is altered."635 BBOG Xoposes. as the optimal choice. by=0.3.," BB06 proposes, as the optimal choice, $b_1 = -0.3$."636 Some papers even rejec the Aimati-relation both in the BATSE sample (Nakar&Piran.2005) and in the Swift sample (Butleretaον2007)., Some papers even reject the Amati-relation both in the BATSE sample \citep{napi05} and in the Swift sample \citep{but07}.637. The most radical soution even challenges the meaning of Eyes itself in the spectra of CRBs (Ryde.2005b)., The most radical solution even challenges the meaning of $\Epkin$ itself in the spectra of GRBs \citep{ry05b}.638". For our purposes. it is essential statistically that the correlation between og aud logE does not imply tha there is a linear conucctionF between logE;;,, aud log [μον "," 	For our purposes, it is essential statistically that the correlation between $\log \fl$ and $\log \Epk$ does not imply that there is a linear connection between $\log E_{iso}$ and $\log \Epkin$ ."639BBOG aso arrived at the conclusion that a relation of the form should exist with some suitable non-zero constants ay.by. and ej.," BB06 also arrived at the conclusion that a relation of the form should exist with some suitable non-zero constants $a_1, b_1,$ and $c_1$."640" We note that Του and 7 strouglv correlates with each other. ic. in this equation either 75,54 0r 735,44: Cal ο use."," We note that $\tem$ and $\tau$ strongly correlates with each other, i.e. in this equation either $\tau_{intr}$ or ${\cal T}_{50;intr}$ can be used."641 The factor loadings nuplv that logF is explained vasically by the first aud third factors., 	The factor loadings imply that $\log \fl$ is explained basically by the first and third factors.642 Siuce in and log7 :uid log£j; ave very strong. respectively. it suggests that should hold with some suitable 05.05.605. aud d. non-zero constants (£7... is the isotropic peak Iuninositv).," Since in and $\log \tau$ and $\log \Epk$ are very strong, respectively, it suggests that should hold with some suitable $a_2, b_2, c_2,$ and $d$ non-zero constants $L_{iso}$ is the isotropic peak luminosity)."643 We rote that a similar relation was also proposed by etal. (2006)., We note that a similar relation was also proposed by \cite{Fir06}.644 The correlation between logF aud logEy is madly determined byFuector3., 	The correlation between $\log \fl$ and $\log \Epk$ is mainly determined by.645 It follows from the loadings of the first and third factors that the relationship bxTween logF and log£g is as important as with the variables dominatingF, It follows from the loadings of the first and third factors that the relationship between $\log \fl$ and $\log \Epk$ is as important as with the variables dominating.646"uetoert, This fact disfavors a simple linear relatioushiponly between logEis aud logE;,."," This fact disfavors a simple linear relationship between $\log \Epkin$ and $\log647E_{iso}$."648 The (letaed study of Eq., The detailed study of Eq.649 Lo (cf., \ref{epkinmod} (cf.650 determination of ay.by.ονd. and alternative equations) is bevoud the aim of this paper.," determination of $a_2,651b_2, c_2, d$, and alternative equations) is beyond the aim of this paper."652 Even from this conclusion. it however follows that the Alarelation in its original fornis disfavored aud soe uocdified version proposed by BBOG is also supporteLhere.," Even from this conclusion, it however follows that the Amati-relation in its original form is disfavored and some modified version proposed by BB06 is also supported here."653 The fourth factor is defined by low frequency spectral index o aud Tye., 	The fourth factor is defined by low frequency spectral index $\alpha$ and $\lag$.654 This nuplies that the direct corre‘lation eCTWOCLL Τις and V ds ueeleible. and hence there is no direct support for the luminosity estimators based on hese two variables (Ramurez-Ruiz&Fenimore.ichartetal.2001:Norris. 2002).," This implies that the direct correlation between $\lag$ and $V$ is negligible, and hence there is no direct support for the luminosity estimators based on these two variables \citep{rafe00,re01,nor02}."655. The fifth factor is doinated bv 7 aud F., 	The fifth factor is dominated by $\tau$ and $\fl$.656 With the first factor this demoistrates. fvat Toy aud 7534 ave not completely equivalent. although ου characterizes a burst nore closely.," With the first factor this demonstrates that $\td$ and $\tem$ are not completely equivalent, although $\tem$ characterizes a burst more closely."657 Iu our opinion. the most remarkable result is that so few qtantities are needed. ic. that all nine quautitics can be characterized by five variables.," 	In our opinion, the most remarkable result is that so few quantities are needed, i.e. that all nine quantities can be characterized by five variables."658 Because all of these conclusions are derivec fre1u the measured data alone. all inodels o “CRBs must resect theseexpectations.," Because all of these conclusions are derived from the measured data alone, all models of GRBs must respect theseexpectations."659 The uuuber of esseutiU variables is in accordancee with BBOG., 	The number of essential variables is in accordance with BB06.660 Thev clained that 3-5 PCs should be used. aud we coustrained the uunber o: jurportant quantities to be 5.," They claimed that 3-5 PCs should be used, and we constrained the number of important quantities to be 5."661" The results of the paper may be sununuarized as folOWns,", The results of the paper may be summarized as follows.662the mass of 2M 1101-7732D“Bascal... The probability that two of these brown dwarts have a projected separation ofaxM 5«10,the mass of 2M 1101-7732B. The probability that two of these brown dwarfs have a projected separation of $a\leq1\farcs44$ is $5\times10^{-5}$.663 on this low probability. I couclude that 2\0 1101-7732 A aud B comprise a bound binary svsteii rather than two unrelated members of Chamacleon I. To test the binarity of this pair through common proper motions. such measurcuents would need to have extremely high precision since unrelated members of the cluster share the same motion to within ~2 kins J|. or ~Q/0025 vr|.," Based on this low probability, I conclude that 2M 1101-7732 A and B comprise a bound binary system rather than two unrelated members of Chamaeleon I. To test the binarity of this pair through common proper motions, such measurements would need to have extremely high precision since unrelated members of the cluster share the same motion to within $\sim2$ km $^{-1}$, or $\sim0\farcs0025$ $^{-1}$."664 Extinctions for 2M 1101-7732 A and D have been estimated Boni the spectra in the manner described by Lulman (20012)..., Extinctions for 2M 1101-7732 A and B have been estimated from the spectra in the manner described by \citet{luh04a}. .665 The resulting values of 44;=0.15 and 0 (E0.2) are consistent with the extinctions inplied bv the color excesses of EUR—D)=0.2 and 0.1 produced when the observed colors are combined with estimates of intrinsic values of Rf for vouug late-type objects (bulinianetal.2003b)., The resulting values of $A_J=0.45$ and 0 $\lesssim0.2$ ) are consistent with the extinctions implied by the color excesses of $E(R-I)=0.2$ and 0.1 produced when the observed colors are combined with estimates of intrinsic values of $R-I$ for young late-type objects \citep{luh03b}.666. When the observed near-IR colors of A are dereddened by ο=0.15. there remain small excesses of ECSIT)=0.06 xb E(IT.IN.)—0.11 relative to chwart values.," When the observed near-IR colors of A are dereddened by $A_J=0.45$, there remain small excesses of $E(J-H)=0.06$ and $E(H-K_s)=0.11$ relative to dwarf values."667 The IR colors of D exhibit larger excesses of ECF[T)=0.12 id EUWY)=0.18., The IR colors of B exhibit larger excesses of $E(J-H)=0.12$ and $E(H-K_s)=0.18$.668 These excesses could result from circtuustelar material or from deviations of the intrinsic colors frou the average divarf values., These excesses could result from circumstellar material or from deviations of the intrinsic colors from the average dwarf values.669 The fact that these excesses remain when computed relative to other late-type members of Cliunacleou I instead of field dwarfs. as illustrated in Figure L. tends to support the former possibility.," The fact that these excesses remain when computed relative to other late-type members of Chamaeleon I instead of field dwarfs, as illustrated in Figure \ref{fig:jhhk}, tends to support the former possibility."670 The spectral types of 241 1101-7732 A aud D are converted to effective temperatures with the temperature scale that was designed by Lulunanctal.(2003b) to be compatible with the models of Daraffectal.(1998). aud Chabrieretal. (2000).., The spectral types of 2M 1101-7732 A and B are converted to effective temperatures with the temperature scale that was designed by \citet{luh03b} to be compatible with the models of \citet{bar98} and \citet{cha00}. .671" Bolometric Wuuinositics are estimated by combining dereddened J-hand measurements, a distance of 168 pe (Whittetetal.1997:Wichmann1998:Bertout 1999).. aud bolometric corrections described in buluuau(1999) and from Reidctal.(2001a)."," Bolometric luminosities are estimated by combining dereddened $J$ -band measurements, a distance of 168 pc \citep{whi97,wic98,ber99}, and bolometric corrections described in \citet{luh99} and from \citet{rei01a}."672. The combined uncertainties in ly. J. and DC; (6~0.2. 0.03. 0.1) correspond to errors of £0.09 in the relative values of log μμ.," The combined uncertainties in $A_J$, $J$, and $_J$ $\sigma\sim0.2$, 0.03, 0.1) correspond to errors of $\pm0.09$ in the relative values of log $L_{\rm bol}$ ."673 When au uncertainty in the distance is included (0~10 pe). the total uncertaiuties are +0.1.," When an uncertainty in the distance is included $\sigma\sim10$ pc), the total uncertainties are $\pm0.1$."674 The extinctions. effective temperatures. and bolometric luminosities for 2M. 1101-7732 A aud D iue listed in Table 1," The extinctions, effective temperatures, and bolometric luminosities for 2M 1101-7732 A and B are listed in Table \ref{tab:data}."675 The temperatures and huuinositics derived in the previous section cau be interpreted in terms of ages ancl masses via theoretical evolutionary models., The temperatures and luminosities derived in the previous section can be interpreted in terms of ages and masses via theoretical evolutionary models.676 For this analysis. IE select the models of Daraffeetal.(1998) and Chabrieretal.(2000) because they provide the best agreciment with observational coustraiuts 2003b).," For this analysis, I select the models of \citet{bar98} and \citet{cha00} because they provide the best agreement with observational constraints \citep{luh03b}."677. The adopted temperature scale was designed to produce coevality for the componcuts of the voung quadruple CG Tau aud the same ages for the stellar and substellar ποπος of Taurus aud IC 3Ls with these evolutionary models (Luluetal.90020)., The adopted temperature scale was designed to produce coevality for the components of the young quadruple GG Tau and the same ages for the stellar and substellar members of Taurus and IC 348 with these evolutionary models \citep{luh03b}.678 As shown in the Π-Π diagram in Figure 5.. the colmponents of 2M 1101-7732ÀD. exhibit nearly perfect coevality as well. providing strong support for the validity of this combination of temperature scale and models.," As shown in the H-R diagram in Figure \ref{fig:hr}, the components of 2M 1101-7732AB exhibit nearly perfect coevality as well, providing strong support for the validity of this combination of temperature scale and models."679 The cocvality of 241 1101-7732 ABis additioual evidence that these sources comprise a true binary system., The coevality of 2M 1101-7732AB is additional evidence that these sources comprise a true binary system.680 The IER ciagraim in Figure 5. nuaplies masses of 0.0530.01 and 0.025270.005. AZ... for A and D. both of which are below the hydrogenburning mass lianit.," The H-R diagram in Figure \ref{fig:hr} implies masses of $0.05\pm0.01$ and $0.025\pm0.005$ $M_\odot$ for A and B, both of which are below the hydrogenburning mass limit."681 The projected separation of ULL of these brown dwarfs correspouds to 210 AU at the distance of Chamacleou 1. Thus.2M 1101-7732ÀD is the first known binary browndwarf with a separation ereater than ~20 AU.," The projected separation of $1\farcs44$ of these brown dwarfs corresponds to 240 AU at the distance of Chamaeleon I. Thus,2M 1101-7732AB is the first known binary browndwarf with a separation greater than $\sim20$ AU."682and the selected: prevalent models in figure 1.. figure 2. and figure 3.. respectively.,"and the selected prevalent models in figure \ref{fig1}, figure \ref{fig2} and figure \ref{fig3}, respectively."683" In figure 4. and figure 5.. we show the portraits of xz,—8 phase of parameterizations of family 1 and family Le and the selected prevalent mioclels in the light of the results obtained. by using 157 Gold SNla data and newly compiled 192 SNla data. respectively."," In figure \ref{fig6} and figure \ref{fig7}, we show the portraits of $\chi^2_{\rm min}-s$ phase of parameterizations of family I and family II and the selected prevalent models in the light of the results obtained by using 157 Gold SNIa data and newly compiled 192 SNIa data, respectively."684" As ds shown in leure 4.. it ds clear that [or porameterizations in family L the minima of vz,4, and s are coincide with each other at à=1. which corresponds to the widely used CPL parametrization."," As is shown in figure \ref{fig6}, it is clear that for parameterizations in family I, the minima of $\chi^2_{\rm min}$ and $s$ are coincide with each other at $n=1$, which corresponds to the widely used CPL parametrization."685" However. in family LH. the münimum of x2, does not coincide with the one of s."," However, in family II, the minimum of $\chi^2_{\rm min}$ does not coincide with the one of $s$ ."686" Phe minimum of x5, is located at ο=4. while the minimum of sis located at i—3."," The minimum of $\chi^2_{\rm min}$ is located at $n=4$, while the minimum of $s$ is located at $n=3$."687" Note that pn=1 parametrization (i.e. CPL) have neither the minimum xz, nor the minimum s.", Note that $n=1$ parametrization (i.e. CPL) have neither the minimum $\chi^2_{\rm min}$ nor the minimum $s$.688 Therefore. if we have to choose a parametrization in family II. η—1 parametrization will not be preferred. and the two most competitive paramctrizations are n=3 and n=4.," Therefore, if we have to choose a parametrization in family II, $n=1$ parametrization will not be preferred, and the two most competitive parametrizations are $n=3$ and $n=4$."689" But as the dillerence between two xz, is much less than the difference between (wo s. we would prefer the η=3 parameterization."," But as the difference between two $\chi^2_{\rm min}$ is much less than the difference between two $s$, we would prefer the $n=3$ parameterization."690 For the prevalent models we investigated here. the value of xz of GCG model is much greater than those of CPL. UIS. Linear and P2 models. so it is not preferred. in the sense of data fitting. but it may still be interesting because of its physical meaning.," For the prevalent models we investigated here, the value of $\chi^2_{\rm min}$ of GCG model is much greater than those of CPL, UIS, Linear and P2 models, so it is not preferred in the sense of data fitting, but it may still be interesting because of its physical meaning."691 As an improvement and extension of earlier data. the newly compiled data set (Davisοἱa£2007). provide us more sample data.," As an improvement and extension of earlier data, the newly compiled data set \citep{Davis07} provide us more sample data."692 Compare figure 5. with figure 4.. we find that although there exit minor changes between the results based on the 192 newly compiled SNla data and those based on the 157 gold data. the main results remain unchanged.," Compare figure \ref{fig7} with figure \ref{fig6}, we find that although there exit minor changes between the results based on the 192 newly compiled SNIa data and those based on the 157 gold data, the main results remain unchanged."693 Firstly. or parametrizations in family L. the best one is still η=1 due to its smallest area of ws) band. albeit the minimum value of 47 for nm=4 parametrization is slightly smaller ian that of Á&=1.," Firstly, for parametrizations in family I, the best one is still $n=1$ due to its smallest area of $w(z)$ band, albeit the minimum value of $\chi^2$ for $n=4$ parametrization is slightly smaller than that of $n=1$."694 Secondly. as is also shown in figure 4.. »wametrizations in family LL. as a whole. have both smaller areas of the (2) band and lower minimunr value of us than rose in family 1. ancl among the parametrizations in family ll. n=3 and »=4 are still the two most. competitive ones.," Secondly, as is also shown in figure \ref{fig6}, parametrizations in family II, as a whole, have both smaller areas of the $w(z)$ band and lower minimum value of $\chi^2$ than those in family I, and among the parametrizations in family II, $n=3$ and $n=4$ are still the two most competitive ones."695 Finally. among the selected prevalent models. although 1e area of wí(z) band of P2 model becomes relative larger. 1e relative locations of these models does not. changed significantly. and compared with CPL. UlS. Linear and. P2 models. GCG model is still not preferred. due to its relative weer Van," Finally, among the selected prevalent models, although the area of w(z) band of P2 model becomes relative larger, the relative locations of these models does not changed significantly, and compared with CPL, UIS, Linear and P2 models, GCG model is still not preferred due to its relative larger $\chi^2_{\rm min}$."696 Lhe minor dillerence between figure 4+ and igure Ὁ mav arise from the cata calibration of cilferent cata sets among which we shall leave in a future work about a COMprehensive ecompariscM., The minor difference between figure \ref{fig6} and figure \ref{fig7} may arise from the data calibration of different data sets among which we shall leave in a future work about a comprehensive comparison.697 ‘Traditionally. forming the so-called. Baves factor (likelihood ratio for [requentists) By=LOU)/LGU;). where L(M;) is called. likelihood for the model A; to obtain the data if the model is true. ds used in comparison of the cosmologicalfand/or dark energv) models (John&Narlikar2002:Nesserisοἱaf 2005).," Traditionally, forming the so-called Bayes factor (likelihood ratio for frequentists) $B_{ij}\equiv L(M_i)/L(M_j)$, where $L(M_{i})$ is called likelihood for the model $M_{i}$ to obtain the data if the model is true, is used in comparison of the cosmological(and/or dark energy) models \citep{baye1,baye2}."698. Generally. LCM;) is dependent on the prior probability and the Likelihood.which is determined by X7. for the model parameters.," Generally, $L(M_i)$ is dependent on the prior probability and the likelihood,which is determined by $\chi^2$, for the model parameters."699 And when one has no prior to the model parameter. everything is determined.H by X7.2 whichH isH à measure ofB the fit. to the data.," And when one has no prior to the model parameter, everything is determined by $\chi^2$, which is a measure of the fit to the data."700 However. according to the above analysis based on the atest SNla data. there exist lots of cosmological models aud xwametrizations of dark energv which lead to very. similar Vue ," However, according to the above analysis based on the latest SNIa data, there exist lots of cosmological models and parametrizations of dark energy which lead to very similar $\chi_{\rm min}^2$."701dhis is especially trae for the results we obtained owed on the newly compiled. 192 data., This is especially true for the results we obtained based on the newly compiled 192 data.702 The best fitting of any parametrization or model we consider here is within he le bound of those of others. see table 2 or figure 5..," The best fitting of any parametrization or model we consider here is within the $1\sigma$ bound of those of others, see table \ref{table2} or figure \ref{fig7}."703 Under this circumstance. how do we compare them?," Under this circumstance, how do we compare them?"704 Or what parametrization approach should be used to. probe he nature of dark energye in the future experiments?, Or what parametrization approach should be used to probe the nature of dark energy in the future experiments?705 The Daves approach only works in the condition that fittings of models are distinctly different., The Bayes approach only works in the condition that fittings of models are distinctly different.706 When the dillerence of X? is very small. one should. pursue other figures of merit.," When the difference of $\chi^2$ is very small, one should pursue other figures of merit."707 The above introduced area of ws)τς band. we think. is such a ligure of merit and our point is that both 47 and. the area of ws)2 band should be svnthetically considered or choosing a better parametrization of dark energy in the uture experiments.," The above introduced area of $w(z)-z$ band, we think, is such a figure of merit and our point is that both $\chi^2$ and the area of $w(z)-z$ band should be synthetically considered for choosing a better parametrization of dark energy in the future experiments."708 Bearing this point in mind and according to the results presented. in the above section. we find that he widely used CPL parametrization. which has a very simple interpretation in terms of the scale factor. is not statistically “special” among the two-parameter xwametrization [umilies. instead η=3 in family H. which ooks like a variation on the CPL parametrization. is more referred.," Bearing this point in mind and according to the results presented in the above section, we find that the widely used CPL parametrization, which has a very simple interpretation in terms of the scale factor, is not statistically ""special"" among the two-parameter parametrization families, instead $n=3$ in family II, which looks like a variation on the CPL parametrization, is more preferred."709 Note that CPL parametrization corresponds to he n=lI ease in both family E anc family HE and if we also take n as a free parameter. family | and family LL are just two three-parameter parametrizations.," Note that CPL parametrization corresponds to the $n=1$ case in both family I and family II and if we also take $n$ as a free parameter, family I and family II are just two three-parameter parametrizations."710 Llowever. in this work we only consider two-parameter parametrizations and mis not treated as a [rec parameter.," However, in this work we only consider two-parameter parametrizations and $n$ is not treated as a free parameter."711 son =1 and n=3 actually denote two cistinet parametrizations. in which the evolution of w(2) is qualitatively different.," so $n=1$ and $n=3$ actually denote two distinct parametrizations, in which the evolution of $w(z)$ is qualitatively different."712 There is an interesting question that whether the dillerences among the area of w(z) band are significant enough to single out once parametrization., There is an interesting question that whether the differences among the area of $w(z)$ band are significant enough to single out one parametrization.713 In our opinion. the answer is somewhat depended on the observational data.," In our opinion, the answer is somewhat depended on the observational data."714 As far as the data we used here. the dillerences among the areas of (2) band are so significant that we can pick out n=3 of the family HE as the best parametrization among the models we consider in this work.," As far as the data we used here, the differences among the areas of $w(z)$ band are so significant that we can pick out $n=3$ of the family II as the best parametrization among the models we consider in this work."715 However. this does mean that the other parametrizations are completely. ruled out.," However, this does mean that the other parametrizations are completely ruled out."716 For example. the simple CPL parametrization and 2 model still do well to a certain extent.," For example, the simple CPL parametrization and P2 model still do well to a certain extent."717 It should. be also pointed out that. the dillerences among the areas of ws) band ave much more significant than those among xou for both 157 gold data and latest 192 data.," It should be also pointed out that, the differences among the areas of $w(z)$ band are much more significant than those among $\chi_{\rm min}^2$ for both 157 gold data and latest 192 data."718" This fact indicates that the area of w(z) band is likely to be a good figure of merit. especially in the situation that the value of xz, for different parametrizations are very close."," This fact indicates that the area of $w(z)$ band is likely to be a good figure of merit, especially in the situation that the value of $\chi^2_{\rm min}$ for different parametrizations are very close."719 Generally speaking. the motivation from a physical point of view should be at the top priority when we choose cosmological models.," Generally speaking, the motivation from a physical point of view should be at the top priority when we choose cosmological models."720 However. it is perfectly clear that in the absence of any compelling dark energy. model. the sugeested parametrizations are phenomenological.," However, it is perfectly clear that in the absence of any compelling dark energy model, the suggested parametrizations are phenomenological."721" Phen the reason why people might prefer à given parametrization is because of its simplicity ancl also because they feelthat it allows us to extract useful information for a very large class of models. and hopefully the ""true"" model is one of them."," Then the reason why people might prefer a given parametrization is because of its simplicity and also because they feelthat it allows us to extract useful information for a very large class of models, and hopefully the ""true"" model is one of them."722 Anyway. to estimate the elfects of dark energy. one needs to quantify them ancl parametrization of a has turned out to be an ellictent tool in this respect.," Anyway, to estimate the effects of dark energy one needs to quantify them and parametrization of $w$ has turned out to be an efficient tool in this respect."723 Therefore. there is a subtle balance between motivation [rom a physical. point," Therefore, there is a subtle balance between motivation from a physical point"724Iu order to characterize the persistent cussion. we calculated average soft (3.75.1 keV/2.3.3.7 keV) and hare (8.7l8 keV/5.18.7 keV) X-ray colors for 256 s intervals using background-subtracted light curves.,"In order to characterize the persistent emission, we calculated average soft (3.7–5.1 keV/2.3–3.7 keV) and hard (8.7–18 keV/5.1–8.7 keV) X-ray colors for 256 s intervals using background-subtracted light curves."725 Location on au N-rav color-color diagram is a good tracer of the accretion rate AL (seco van der Islis 1995 aud references therein)., Location on an X-ray color-color diagram is a good tracer of the accretion rate $\dot M$ (see van der Klis 1995 and references therein).726 and (Franco2001:vauStraatenetal.2001) exhibit bursts over the widest range of AL. while cursory checks of the color-color diagrams for the other sources suggest that bursts were oulv observed at lugh accretion rates because few or no observations were made dunrug low AT intervals.," \\citep{mun00} and \citep{fra00,vst00} exhibit bursts over the widest range of $\dot M$, while cursory checks of the color-color diagrams for the other sources suggest that bursts were only observed at high accretion rates because few or no observations were made during low $\dot M$ intervals."727 To search for oscillations. we created twicc-oversauiled power spectra for 2 s intervals of data every 0.25 s for the uration of cach burst.," To search for oscillations, we created twice-oversampled power spectra for 2 s intervals of data every 0.25 s for the duration of each burst."728 If oscillations are detected within +3 Tz of the expected frequency and within 10 seconds after the start of a burst with a probability «2«10! that the sigual is due to Poisson noise in a single trial. we consider the oscillation to be a detection.," If oscillations are detected within $\pm 3$ Hz of the expected frequency and within 10 seconds after the start of a burst with a probability $< 2\times10^{-4}$ that the signal is due to Poisson noise in a single trial, we consider the oscillation to be a detection."729 We also used this technique to search all ofthe bursts for oscillations at the fist three harmonics of the slow oscillations. aud at 0.5. 1.5. and 2.0 times the frequency of the fast oscillations.," We also used this technique to search all of the bursts for oscillations at the first three harmonics of the slow oscillations, and at 0.5, 1.5, and 2.0 times the frequency of the fast oscillations."730 We found no evidence for oscillations at these frequencies in any of the bursts., We found no evidence for oscillations at these frequencies in any of the bursts.731 Using power spectra of 1 s intervals. we can place upper limits on the fractional RAIS amplitude of oscillations at these frequencies of during a burst.," Using power spectra of 1 s intervals, we can place upper limits on the fractional RMS amplitude of oscillations at these frequencies of during a burst."732 These are not very stringent upper limits. so application of iiore seusitive search techuiques such as those described in Miller(1999).. Minoetal. (2000)... aud Strobiuawver (2001) will be useful for more detailed studies.," These are not very stringent upper limits, so application of more sensitive search techniques such as those described in \citet{mil99}, \cite{mun00}, and Strohmayer (2001) will be useful for more detailed studies."733 We produced euergv spectra for each 0.25 s interval from cach burst using available combinations of data modes which provide at least 32 cnerey chaunels, We produced energy spectra for each 0.25 s interval from each burst using available combinations of data modes which provide at least 32 energy channels.734 We subtracted spectra from 15 s of cuission from before the burst to account for backeround. and fit each spectrum between 2.520 keV with a model cousistiug of a blackbody 1uultiplied by a constant interstellar absorption (determined from the mean value frou fits using variable absorption).," We subtracted spectra from 15 s of emission from before the burst to account for background, and fit each spectrum between 2.5–20 keV with a model consisting of a blackbody multiplied by a constant interstellar absorption (determined from the mean value from fits using variable absorption)."735 The model provides au apparent temperature (aps) aud a normalization equal to the square of the apparent radius (Rapp) of the burst emission surface. aud allows us to estimate the bolometric fux as a function of time.," The model provides an apparent temperature $T_{\rm app}$ ) and a normalization equal to the square of the apparent radius $R_{\rm736app}$ ) of the burst emission surface, and allows us to estimate the bolometric flux as a function of time."737 The peak flux of bursts from a eiveu source can vary by a factor of 510., The peak flux of bursts from a given source can vary by a factor of 5–10.738" Iu niu bursts. photospheric radius expansion is evident at the start of the burst. during which 24,4, increases and Taj, decreases such that the bolometric fiux remains constaut. prestunably at the Eddington limit (see Lewin et al."," In many bursts, photospheric radius expansion is evident at the start of the burst, during which $R_{\rm app}$ increases and $T_{\rm app}$ decreases such that the bolometric flux remains constant, presumably at the Eddington limit (see Lewin et al."739 1995)., 1995).740 Our definition of radius expansion includes bursts which exhibit a second lucrease m radius immediately after the nuum which follows the expansion phase., Our definition of radius expansion includes bursts which exhibit a second increase in radius immediately after the minimum which follows the expansion phase.741 We find several such bursts from (seealsovanStraatenetal.2O01).. ancl oue such burst from both aud42.," We find several such bursts from \citep[see also][]{vst00}, and one such burst from both and."742 A stumary of our results for the nine burst oscillation sources is given iu Table 1., A summary of our results for the nine burst oscillation sources is given in Table 1.743 There is a tight connection between the presence of oscillatious and of radius expansion idu fast (2.600 IIZ) sources from every perspective., There is a tight connection between the presence of oscillations and of radius expansion in fast $\simeq 600$ Hz) sources from every perspective.744 Fast oscillations occur predominantly daring musts with radius expansion. and almost all bursts with radius expansion exhibit oscillations.," Fast oscillations occur predominantly during bursts with radius expansion, and almost all bursts with radius expansion exhibit oscillations."745 At the same time uost of the bursts without fast oscillations also lack radius expansion., At the same time most of the bursts without fast oscillations also lack radius expansion.746 In slow (25300 Iz) sources. there is 10 preference for whether bursts with or without radius expansion exhibit oscillations.," In slow $\simeq 300$ Hz) sources, there is no preference for whether bursts with or without radius expansion exhibit oscillations."747 This suggests that the yequencics of burst oscillatious and the properties of ursts are connected., This suggests that the frequencies of burst oscillations and the properties of bursts are connected.748 Although the sample of bursts from an individual source Is du sole cases quite snall. the correlatious for fast aud slow sources as groups are quite significaut.," Although the sample of bursts from an individual source is in some cases quite small, the correlations for fast and slow sources as groups are quite significant."749 Iu order to quantity the sienificance of our results. we hivpothesize that radius expansion occurs in bursts with a fixed probability. f.," In order to quantify the significance of our results, we hypothesize that radius expansion occurs in bursts with a fixed probability, $f$."750 Given a sample of iH bursts. the probability of observing a number η of bursts with radius expausion 1s We can then compute the probability deusity for f given 0 radius expansion bursts dn ai sample of js bursts. μη).," Given a sample of $m$ bursts, the probability of observing a number $n$ of bursts with radius expansion is We can then compute the probability density for $f$ given $n$ radius expansion bursts in a sample of $m$ bursts, $p(f| n, m)$."751 We have plotted these probability densities in Figure 3. for fast and slow sources. considering as our sample population either all observed bursts (dashed lue) aud only those bursts which exhibit oscillations (solid line).," We have plotted these probability densities in Figure \ref{dist} for fast and slow sources, considering as our sample population either all observed bursts (dashed line) and only those bursts which exhibit oscillations (solid line)."752 Although fast aud slow sources exhibit radius expansion in about equal fractions of bursts im general. fast sources exhibit radius expansion during bursts with oscillations far more often of the time) than slow sources (31!ο “)).," Although fast and slow sources exhibit radius expansion in about equal fractions of bursts in general, fast sources exhibit radius expansion during bursts with oscillations far more often $^{+4}_{-8}$ of the time) than slow sources $^{+10}_{-6}$ )."753 The trends for individual sources are interesting in that they followthe behavior expected from considering the sources as groups (Figure 1D)., The trends for individual sources are interesting in that they followthe behavior expected from considering the sources as groups (Figure \ref{whendo}) ).754 Ou the other haud. slow sources are more likely to show radius . κ ∙∙ ∙≝≱ than fast sources (20!iU η).," On the other hand, slow sources are more likely to show radius expansion during bursts oscillations $^{+9}_{-8}$ ) than fast sources $^{+10}_{-5}$ )."755 We can rule out some observational selection effects as causes for these correlations., We can rule out some observational selection effects as causes for these correlations.756 For example. we are not systematically iuissing oscillations from weak bursts. as nuelt occur if all oscillatious had the same fractional auplitude.," For example, we are not systematically missing oscillations from weak bursts, as might occur if all oscillations had the same fractional amplitude."757 We observe oscillatious iu some of the weakest bursts from260..53.053. and 31. while no oscillations are detected duiug some of the strouger bursts from these sources.," We observe oscillations in some of the weakest bursts from, and , while no oscillations are detected during some of the stronger bursts from these sources."758 Therefore. our correlations are based upon gemune variations in the streneths of the oscillations.," Therefore, our correlations are based upon genuine variations in the strengths of the oscillations."759 ILlowewer. since bursts were observed frou onlv two sources at low M aand 231). there is a remote chauce that these correlations are an artifact of the higher AT at which the remade sources were observed.," However, since bursts were observed from only two sources at low $\dot M$ and ), there is a remote chance that these correlations are an artifact of the higher $\dot M$ at which the remaining sources were observed."760 For instance. the bursts with the longest time scales take place at low fluxes in (NMuralguuietal.1980). and at low iuferred Af in (vauderKlisctal.1990).," For instance, the bursts with the longest time scales take place at low fluxes in \citep{mur80} and at low inferred $\dot M$ in \citep{vdk90}."761 It is reasonable to believe that these loug bursts do uot exhibit radius expausion (as in260: soo for example Nino et al., It is reasonable to believe that these long bursts do not exhibit radius expansion (as in; see for example Muno et al.762 2000). so there would be no strict relationship between radius expansion and fast oscillations if these bursts exhibit oscillations.," 2000), so there would be no strict relationship between radius expansion and fast oscillations if these bursts exhibit oscillations."763" Ποπονο, we would consider this a surprise given the absence of oscillatious at low accretion rates in and 31."," However, we would consider this a surprise given the absence of oscillations at low accretion rates in and ."764 Ou the other haud. the time scales of bursts from the slow oscillators (Syauk.," On the other hand, the time scales of bursts from the slow oscillators (Swank,"765equivalent width and velocity οπου. Cropper AlMason 1990) and the contamination will be severe.,"equivalent width and velocity Hellier, Cropper Mason 1990) and the contamination will be severe."766 More generally. there is no model-independent way of attributing variations to the beat evele rather than the spin evele. once hese are allowed to be functions of orbital phase.," More generally, there is no model-independent way of attributing variations to the beat cycle rather than the spin cycle, once these are allowed to be functions of orbital phase."767 A further caveat is ai standard. one applying to omography. in that it assumes that optical depth effects are not changing the intensity of line enission over the different eveles.," A further caveat is a standard one applying to tomography, in that it assumes that optical depth effects are not changing the intensity of line emission over the different cycles."768 From the large changes in equivalent width. xuwticularlv over the spin evcle (Fig.," From the large changes in equivalent width, particularly over the spin cycle (Fig."769 2). we know that this assumption is violated inCen.," 2), we know that this assumption is violated in."770. In Fig., In Fig.771 7 we present. beat-phasc-resolved. orbital-evcle omogranis of the dedata.," 7 we present beat-phase-resolved, orbital-cycle tomograms of the data."772 The changes in features over beat phase will be caused by (1) components whose velocities genuinely vary with beat phase. (2) components whose velocities vary with spin phase. and (3) material whose illumination by EUY and X-ray photons is a function of beat phase.," The changes in features over beat phase will be caused by (1) components whose velocities genuinely vary with beat phase, (2) components whose velocities vary with spin phase, and (3) material whose illumination by EUV and X-ray photons is a function of beat phase."773 For comparison. Fig.," For comparison, Fig."774 S shows the equivalent tomograms computed from the diamagnetie-blob model., 8 shows the equivalent tomograms computed from the diamagnetic-blob model.775 These will contain the elfects 1 and 2 just mentioned. but not elfect 3. since irradiation is not incluclec," These will contain the effects 1 and 2 just mentioned, but not effect 3, since irradiation is not included."776 oth the cata and the model tomograms show the hook-like features varving with beat phase., Both the data and the model tomograms show the hook-like features varying with beat phase.777 In the moclel. the ‘tail’ of the hook becomes less pronounced. during the ejection events. re-forming during accretion episodes.," In the model, the `tail' of the hook becomes less pronounced during the ejection events, re-forming during accretion episodes."778" ""This disappearance and re-Forming Is also seen clearly in the data.", This disappearance and re-forming is also seen clearly in the data.779 However. the variation is greater in the data. and occurs once per beat evele. whereas in the model it occurs twice per beat evele.," However, the variation is greater in the data, and occurs once per beat cycle, whereas in the model it occurs twice per beat cycle."780 The dillerence can plausibly be explained by the Lack of X-rav irradiation in the model., The difference can plausibly be explained by the lack of X-ray irradiation in the model.781" Given the sinusoidal X-ray spin pulse. regions fixed in the orbital frame will receive a sinusoidal evele of irradiation over the beat evcle. enhancing he line emission once per beat ονο]ο,"," Given the sinusoidal X-ray spin pulse, regions fixed in the orbital frame will receive a sinusoidal cycle of irradiation over the beat cycle, enhancing the line emission once per beat cycle."782 Given that irradiation is not included in the model. and the other caveats expressed above. it is not surprising that there are major dilferences »etween the data and the current model.," Given that irradiation is not included in the model, and the other caveats expressed above, it is not surprising that there are major differences between the data and the current model."783 Jo complete the discussion of the. beat-resolved omogranms (Lig., To complete the discussion of the beat-resolved tomograms (Fig.784 7) we should consider how they could. be interpreted in the partial-disc model., 7) we should consider how they could be interpreted in the partial-disc model.785 Phe emission. near velocity (0. 300) 1oseen at all beat phases. would likely x [rom the secondary star.," The emission near velocity (0, 300), seen at all beat phases, would likely be from the secondary star."786 Phe hook! feature would be due o the stream flowing from the Lagrangian point. colliding with the dise (near velocity 500. 400 19). and then overllowing the disc (moving to the lower-left quadrant of he tomogram).," The `hook' feature would be due to the stream flowing from the Lagrangian point, colliding with the disc (near velocity –500, 400 ), and then overflowing the disc (moving to the lower-left quadrant of the tomogram)."787 The fact that the velocities in the blower-Icft, The fact that the velocities in the lower-left788"to an alpha model witl a72—25x10"".","to an alpha model with $\alpha\sim 2-5789\times10^{-3}$."790 Thermal euergy is removect from the disk gas by radiative cooling due to passive blackbody emission Grom the disks photospliere surfaces., Thermal energy is removed from the disk gas by radiative cooling due to passive blackbody emission from the disk's photosphere surfaces.791 The blackbody temperature is calculaed at each time step aud for each SPH particle., The blackbody temperature is calculated at each time step and for each SPH particle.792 This treatinent remedies a major shortcoming of yrevious models (Nelsonefαἱ.1993:Pickettal.1998:Boss1997) which usecl a ‘locally isotherual or ‘locally adiabatic’ approximation to show that relatively low tmass disks can undergo fraguentation aud/or collapse. despite earlier claims (Podolakefal1993) that a very Imnassive disk is requirec.," This treatment remedies a major shortcoming of previous models \citep{DynamI,Pick98,Boss97} which used a `locally isothermal' or `locally adiabatic' approximation to show that relatively low mass disks can undergo fragmentation and/or collapse, despite earlier claims \citep{PodPP3} that a very massive disk is required."793 The system is evolved for eight binary orbits. or 2700 vr 23320 vr).," The system is evolved for eight binary orbits, or 2700 yr 350 yr)."794 Fi[n]we] shows the system shortly before ad alter the fourth periapse passage of the two compouenhl (1neasured from the beginuiug of the sijiulation)., Figure \ref{fig:disk-pair} shows the system shortly before and after the fourth periapse passage of the two components (measured from the beginning of the simulation).795 Before each periapse. tle two disks are 1000 aud exhibit uo ‘sible spiral structure. although they are no longer perfectly ‘round’.," Before each periapse, the two disks are smooth and exhibit no visible spiral structure, although they are no longer perfectly `round'."796 During ad alter periapse. ach disk develops stro&. two armect spiral structures due to the mutual tidal ieractions of the inary.," During and after periapse, each disk develops strong, two armed spiral structures due to the mutual tidal interactions of the binary."797 The structures |ecay to a smooth couditiou like that in the top panel over je nent 0.57pig., The structures decay to a smooth condition like that in the top panel over the next $0.5$.798 he cycle repeats with ittle variation as the system again approaches periapse. ad we expect tliat rther evolution will be similar.," The cycle repeats with little variation as the system again approaches periapse, and we expect that further evolution will be similar."799 Te spiral structres clecay jecause internal heating in the cisk increases te stability of the cisks aealust spiral aiu growtlü as measured by the Tooure stability (fig.," The spiral structures decay because internal heating in the disk increases the stability of the disks against spiral arm growth, as measured by the Toomre stability (fig."800 2 lop) of each disk., \ref{fig:temp-prof} top) of each disk.801 The miniur value of Q iiCreases [rom its iulial =11.5 to ~IE before periaose alid ~~2 afterwards., The minimum value of $Q$ increases from its initial 1.5 to $\sim4$ before periapse and $\sim5$ afterwards.8021 These valtes are the same before atd alter each successive perlapse passage aid both are well above the (QX) values for which spira Srucures a'e expected to grow., These values are the same before and after each successive periapse passage and both are well above the $Q\lesssim3$ values for which spiral structures are expected to grow.803 Therelore. if we suppose that Jovian pl:jets form via gravitatiola colapse or [ragmeutati¢1 of spiral structure iu disks. their formation will be unlikely i1 this systelu.," Therefore, if we suppose that Jovian planets form via gravitational collapse or fragmentation of spiral structure in disks, their formation will be unlikely in this system."804 Tie. high stabili is due to au increase iL (ie disk temperature., The high stability is due to an increase in the disk temperature.805 I act. the temperatures Hὁ high enough to cause some graln s»ecies to be vaJOLIZEC|.," In fact, the temperatures are high enough to cause some grain species to be vaporized."806 This is impolant because the core accretion model for platet formation requires tla solid grains can coagu€ale aud are not insteacl repeatedly returied to vapor state., This is important because the core accretion model for planet formation requires that solid grains can coagulate and are not instead repeatedly returned to vapor state.807 Water ice may “be particularly importart because it composes by mass ofthe solic inaterial in the disk aud is alone te most volatile erain species 1).. vaporizing at ~150 Ex. The bottom panel o ‘figure 2 shows the clisk miplane temperatures obalnec iu this simulation.," Water ice may be particularly important because it composes by mass of the solid material in the disk and is among the most volatile grain species \citep{Pol94}, vaporizing at $\sim150$ K. The bottom panel of figure \ref{fig:temp-prof} shows the disk midplane temperatures obtained in this simulation."808 Only in a region with rZ10 AU does the temperattre reach low enough νvalues that water ice cau form. even temporarily.," Only in a region with $r\gtrsim 10$ AU does the temperature reach low enough values that water ice can form, even temporarily."809 However iu this region. the ualter is most subjec to shocks generated by tlie spiral structure produced by the binary iuteractiyu. Which raise its tem»erature by as much as a [actor three over tlie azimutli average.," However in this region, the matter is most subject to shocks generated by the spiral structure produced by the binary interaction, which raise its temperature by as much as a factor three over the azimuth average."810 The spiral pateris co-rotate with tle orbit of the binary aud the orbital period at 15 AU is ~82 vr. so material everywhere in the cisk will have time to travel through the spiral arius several times belore they «ecay.," The spiral patterns co-rotate with the orbit of the binary and the orbital period at 15 AU is $\sim$ 82 yr, so material everywhere in the disk will have time to travel through the spiral arms several times before they decay."811 A simulation rtu with zero eccentricity, A simulation run with zero eccentricity812ratio fooredfeone Is not determüued. e. &.. no choice can be made among either 3/2. 1/2 or 1/2.,"ratio $f_{core}/ f_{cone}$ is not determined, e. g., no choice can be made among either $3/2$ , $1/\sqrt{2}$ or $1/2$."813" For zero offset cases when a=30°507 range. due to the niultivalueduess of ran infinite muuber of pairs of kr, and 7,4, eive the same A."," For zero offset cases when $\alpha \cong 30^{\circ} - 50^{\circ}$ range, due to the multivaluedness of $r$ an infinite number of pairs of $r_{core}$ and $r_{cone}$ give the same $\Delta_{cc}$."814 But if either Foore OF Keane Is found to be on the rising brauch (sav by usiuz DCW) aud such that its Ay is small οποιο] to be possible ouly at oue altitude (uinltivalueduess is avoided) then core aud conal altitudes must Le close., But if either $r_{core}$ or $r_{cone}$ is found to be on the rising branch (say by using BCW) and such that its $\Delta \varphi$ is small enough to be possible only at one altitude (multivaluedness is avoided) then core and conal altitudes must lie close.815" Consider the pulsars PSR 1917|00 and PSR 2028|22 for which A, respectively eives Ar as =GOR. aud =100Π.."," Consider the pulsars PSR 1917+00 and PSR 2028+22 for which $\Delta_{cc}$ respectively gives $\Delta r$ as $\cong 60 \, R_*$ and $\cong 100 \, R_*$."816" It is obvious that altitudes given by RL R6 or IIN are of no help here because they can not eive the correct sign of A,"," It is obvious that altitudes given by R4, R6 or HX are of no help here because they can not give the correct sign of $\Delta_{cc}$."817" We thus try to find fF,/fiu m these cases using the 1/3 rule.", We thus try to find $f_{core}/ f_{cone}$ in these cases using the 1/3 rule.818 Assuimine muni posible altitude for the conal component. ie. 1/8... and using Eqs.," Assuming minimum posiible altitude for the conal component, i.e., $1 \, R_*$, and using Eqs."819" 21 aud 22 for these two pulsus we find. f£,ἡfeone=1/16 for PSR 1917|00 and 1/20 for PSR 2028|22. (", \ref{wcor} and \ref{wcon} for these two pulsars we find $f_{core}/ f_{cone} = 1/16$ for PSR 1917+00 and $1/20$ for PSR 2028+22. (820The solid angle fillime factors will be squares of these.),The solid angle filling factors will be squares of these.)821" Other combinations consistent with A, values will give higher altitudes and further reduce this ratio.", Other combinations consistent with $\Delta_{cc}$ values will give higher altitudes and further reduce this ratio.822 Iu the above discussion of lags. leads aud zero offset cases we have tried to reconcile the various altitude estimates provided the core/cone distinction is not ignored.," In the above discussion of lags, leads and zero offset cases we have tried to reconcile the various altitude estimates provided the core/cone distinction is not ignored."823" Tutrinsically the DCW inethod is au excellent method for altitude determination and eujovs the advantages that use of A,,. also eutails.", Intrinsically the BCW method is an excellent method for altitude determination and enjoys the advantages that use of $\Delta_{cc}$ also entails.824 In both BCW and IN a puzzling feature is that the derived altitudes have large uncertainties despite waving quality data aud improved plivsics., In both BCW and HX a puzzling feature is that the derived altitudes have large uncertainties despite having quality data and improved physics.825 Also iu some pulsars the BCW offsets conie out with a wroue sien (BCW. IDEN).," Also in some pulsars the BCW offsets come out with a wrong sign (BCW, HX)."826 We feelthat this may be a consequence of an insuffiicicut discutaneling of core aud conal features., We feel that this may be a consequence of an insufficient disentangling of core and conal features.827 For example. in BOW he position augle curve is dominated bv the conal wings while the iteusity xofile is dominated by the central core in some pulsars.," For example, in BCW the position angle curve is dominated by the conal wings while the intensity profile is dominated by the central core in some pulsars."828 The DCW iethod as also used by IIX. utilises the linear polarization position angle sweep. a typical conal property. despite the refereuce to pulsars as core-domüuated ete.," The BCW method as also used by HX utilises the linear polarization position angle sweep, a typical conal property, despite the reference to pulsars as 'core-dominated' etc."829 The o)sition angle sweeps for cores are well known to be usually not so clean., The position angle sweeps for cores are well known to be usually not so clean.830 It secnus that BOW? offsets in effect lead to altitudes ouly of conal componcuts mit in cases where the core is or is not seen., It seems that BCW offsets in effect lead to altitudes only of conal components but in cases where the core is or is not seen.831 Therefore. for core conrponeuts using the circular polarization signatures may provide the wav.," Therefore, for core components using the circular polarization signatures may provide the way."832" Though there is some overlap in their aud our pulsar lists. the reconciliation of altitude values quoted for specific pulsars by either DCW or IIX. with our Ar here will not be meanineful: certainly so ifwe use f,""/fone=1/2 which relates ouly to mean values of altitudes in BCW or ΗΝ."," Though there is some overlap in their and our pulsar lists, the reconciliation of altitude values quoted for specific pulsars by either BCW or HX with our $\Delta r$ here will not be meaningful; certainly so if we use $f_{core}/ f_{cone} = 1/\sqrt{2}$ which relates only to mean values of altitudes in BCW or HX."833 A case by case analysis can be done. but it maynot prove of umch value. because our Ar cau be determined precisely ouly when the core aud coual components have been completely separated.," A case by case analysis can be done, but it maynot prove of much value, because our $\Delta r$ can be determined precisely only when the core and conal components have been completely separated."834 Therefore.," Therefore,"835reader to a more indepth study of the influence of these parameters on the results in Section 3..,reader to a more indepth study of the influence of these parameters on the results in Section \ref{sec:testing}.836 We will now describe the approach taken to parallelize the halo-finding., We will now describe the approach taken to parallelize the halo-finding.837 In refsubsubsec:domdecomp we describe the way the volume is decomposed into smaller chunks and we then elaborate in refsubsubsec:lb how the load-balancing is achieved before ending in refsubsubsec:caveats with some cautionary notes on the applicability of this scheme., In \\ref{subsubsec:domdecomp} we describe the way the volume is decomposed into smaller chunks and we then elaborate in \\ref{subsubsec:lb} how the load-balancing is achieved before ending in \\ref{subsubsec:caveats} with some cautionary notes on the applicability of this scheme.838 There are many ways to distribute the workload to multiple processes., There are many ways to distribute the workload to multiple processes.839 In some halo finders this is done by first generating particle groups with a ffinder which can then independently processed (e.g., In some halo finders this is done by first generating particle groups with a finder which can then independently processed \citep[e.g.][]{Kim2006}.840" However, we chose to ‘blindly’ split the whole computational7T). volume to multiple CPUs."," However, we chose to `blindly' split the whole computational volume to multiple CPUs."841 This is done because we plan to implement aas part of the simulation code aand not only as a stand-aloneversion®;; the same as iis also integrated into aand can perform the halo finding ‘on the fly’ during the runtime of a simulation., This is done because we plan to implement as part of the simulation code and not only as a stand-alone; the same as is also integrated into and can perform the halo finding `on the fly' during the runtime of a simulation.842" For the simulation code, a proper volume decomposition scheme is required and hence we are using the approach described here."," For the simulation code, a proper volume decomposition scheme is required and hence we are using the approach described here."843" However, it should be noted that, in principle, iis capable of working on any user defined sub-set of a simulation box and as such is not tied to the employed decomposition scheme."," However, it should be noted that, in principle, is capable of working on any user defined sub-set of a simulation box and as such is not tied to the employed decomposition scheme."844" The volume decomposition is done by using a space-filling curve which maps the three spatial coordinates (x,y,2) (SFC),into a one-dimensional SFC-index i; an illustration is given in the left panel of figure 2.."," The volume decomposition is done by using a space-filling curve (SFC), which maps the three spatial coordinates $(x,y,z)$ into a one-dimensional SFC-index $i$; an illustration is given in the left panel of figure \ref{fig:decomposition}."845 As many other workers have done before (e.g.??) we use the Hilbert-curve.," As many other workers have done before \citep[e.g.][]{Springel2005, Prunet2008} we use the Hilbert-curve."846" A new parameter,LB,, of rregulates on which level this ordering is done; we then use 2LB cells per dimension; it is important to remark that the grid used for load-balancing is distinct from the domain grid and therefore an additional parameter."," A new parameter, of regulates on which level this ordering is done; we then use $2^{\LB}$ cells per dimension; it is important to remark that the grid used for load-balancing is distinct from the domain grid and therefore an additional parameter."847" Each process will then receive a segment, described by a consecutive range of indices istart.fstop, of the SFC curve and hence a sub-volume of the whole computational box."," Each process will then receive a segment, described by a consecutive range of indices $i_\mathrm{start}\ldots i_\mathrm{stop}$, of the SFC curve and hence a sub-volume of the whole computational box."848 The Hilbert curve has the useful property to preserve locality and also to minimize the surface area., The Hilbert curve has the useful property to preserve locality and also to minimize the surface area.849" It is in this respect superior to a simple slab decomposition, however at the cost of volume chunks with a complicated shape."," It is in this respect superior to a simple slab decomposition, however at the cost of volume chunks with a complicated shape."850" In addition to its segment of the SFC curve, each process will also receive copies of the particles in a buffer zone around its volume with the thickness of one cell of the LB--grid tthe right panel of figure 2))."," In addition to its segment of the SFC curve, each process will also receive copies of the particles in a buffer zone around its volume with the thickness of one cell of the -grid the right panel of figure \ref{fig:decomposition}) )."851" With these particles, each CPU can then follow the standard recipe described above refsubsec:ahf)) in finding halos."," With these particles, each CPU can then follow the standard recipe described above \\ref{subsec:ahf}) ) in finding halos."852 The thickness of the boundary is hence an important quantity that is supposed to counteract the situation in which one halo is situated close to the boundary of two processes., The thickness of the boundary is hence an important quantity that is supposed to counteract the situation in which one halo is situated close to the boundary of two processes.853" As there is no further communication between the tasks, for reasons we will allude to below, the buffer zones need to be large enough to encompass all relevant particles."," As there is no further communication between the tasks, for reasons we will allude to below, the buffer zones need to be large enough to encompass all relevant particles."854" 'To fulfill this requirement, the thickness of the boundary zone, given by the pparameter, should obey the relation where B is the size of the simulation box and RIE* is the radius of the largest objects of interest."," To fulfill this requirement, the thickness of the boundary zone, given by the parameter, should obey the relation where $B$ is the size of the simulation box and $R_\mathrm{vir}^\mathrm{max}$ is the radius of the largest objects of interest."855 Note that each process only keeps those halos whose centers are located in the proper volume (as given by the SFC segment) of the process., Note that each process only keeps those halos whose centers are located in the proper volume (as given by the SFC segment) of the process.856" To subdivide the SFC curve, different schemes can be employed, influencing the quality of the achieved load-balancing; this is ultimately measured by the difference of the run times of the different processes."," To subdivide the SFC curve, different schemes can be employed, influencing the quality of the achieved load-balancing; this is ultimately measured by the difference of the run times of the different processes."857" We chose to use a scheme that distributes the particles evenly between the processes, e.g. each CPU will require the same amount of storage for the particles."," We chose to use a scheme that distributes the particles evenly between the processes, e.g. each CPU will require the same amount of storage for the particles."858 This can of course lead to vastly different volumes assigned to each task f, This can of course lead to vastly different volumes assigned to each task (cf.859igure but we will show in (cf. refsubsec:res;bthatthissimpleschemecanprovidereasonableresults.," figure \ref{fig:decomposition}) ), but we will show in \\ref{subsec:res_lb} that this simple scheme can provide reasonable results."860"2)), We therefore segment the SFC curve into chunks containing the same (within the precision allowed by the coarseness of the load balance grid given by LB)) number of particles.", We therefore segment the SFC curve into chunks containing the same (within the precision allowed by the coarseness of the load balance grid given by ) number of particles.861" As with this scheme any scheme based on segmenting the volume along a (orSFC curve) it can happen that objects are cut between different processes, the grid tree construction would be very communication intensive."," As with this scheme (or any scheme based on segmenting the volume along a SFC curve) it can happen that objects are cut between different processes, the grid tree construction would be very communication intensive."862" To circumvent this, we use the inclusion of a buffer zone alluded to above."," To circumvent this, we use the inclusion of a buffer zone alluded to above."863" Note that after the duplication of boundary particles the balance of particles might shift, which is especially prominent for very inhomogeneous particle distributions, as found, for example, in simulations with a small box size."," Note that after the duplication of boundary particles the balance of particles might shift, which is especially prominent for very inhomogeneous particle distributions, as found, for example, in simulations with a small box size."864" Since the decision of how to segment the SFC curve is modular, it is very easy to implement a different scheme, we plan on investigating into further criteria in future work."," Since the decision of how to segment the SFC curve is modular, it is very easy to implement a different scheme, we plan on investigating into further criteria in future work."865CGranato et al.,Granato et al.866 1999)., 1999).867 I£ all the gas had been ejected. such a long aceretion phase would be dillicult to sustain unless the ejection were somehow anisotropic: ejection might for example occur along the hole's rotation axis. leaving a torus of centrifugallv-supported. accretable material in the plane of the accretion disk.," If all the gas had been ejected, such a long accretion phase would be difficult to sustain unless the ejection were somehow anisotropic: ejection might for example occur along the hole's rotation axis, leaving a torus of centrifugally-supported, accretable material in the plane of the accretion disk."868 Such a scenario may sipultaneouslvy aclelress the problem of the excess soft SRB produced by the niocel. as alluded to in section 3.," Such a scenario may simultaneously address the problem of the excess soft XRB produced by the model, as alluded to in section 3."869 This torus will plausibly lead to mocderate absorption in lines of sight passing through it. as in orientation-based. ACN unification. schemes. (see Antonucci 1993 for a review).," This torus will plausibly lead to moderate absorption in lines of sight passing through it, as in orientation-based AGN unification schemes (see Antonucci 1993 for a review)."870 The number of sources with absorbed X-ray. spectra ancl narrow optical emission lines. compared to those appearing as classical quasars will thus depend upon the solid angle subtended by the outflow Cor. ecquivalently. on the ratio of the times spent in the ejection phase and quasar phase proper. in those cases where the gas expulsion is isotropic and. ultimately complete).," The number of sources with absorbed X-ray spectra and narrow optical emission lines, compared to those appearing as classical quasars will thus depend upon the solid angle subtended by the outflow (or, equivalently, on the ratio of the times spent in the ejection phase and quasar phase proper, in those cases where the gas expulsion is isotropic and ultimately complete)."871 Note that even anisotropic ejection is likely to clear the spheroid. of all hyerostaticallv-supported gas. leaving only that which is centrifugally supported.," Note that even anisotropic ejection is likely to clear the spheroid of all hydrostatically-supported gas, leaving only that which is centrifugally supported."872 This means that the overall picture in which ejection ultimately terminates both black hole and spheroid growth is retained., This means that the overall picture in which ejection ultimately terminates both black hole and spheroid growth is retained.873 We can also compute the mechanical energy. deposited bv the AGN winds in the surrounding gas during the obscured growth phase., We can also compute the mechanical energy deposited by the AGN winds in the surrounding gas during the obscured growth phase.874 Averaging over all objects in the simulation and taking a wind power of the form Ly=eL (where @ is constant). we find that the winds deposit a mean specific» energy of⋅ 1.4107a ⊥⋠into the surrounding. eas of haloes (assuming that all the wind power is absorbed).," Averaging over all objects in the simulation and taking a wind power of the form $L_{\rm{W}}=aL_{\rm{Edd}}$ (where $a$ is constant), we find that the winds deposit a mean specific energy of $1.4 \times 10^{51}a$ $^{-1}$ into the surrounding gas of haloes (assuming that all the wind power is absorbed)."875 If instead the energy ds distributed. evenly among all the barvons in the Universe. the specific energy input is slightly lower at 2.3.107a l," If instead the energy is distributed evenly among all the baryons in the Universe, the specific energy input is slightly lower at $2.3 \times 10^{50}a$ $^{-1}$."876opor a—0.1 (as suggested. bv. Fabian 1999) these values are equivalent Oo 40 and ~7 per particle (for fully ionized. gas with cosmic abunances).). respectively. ancl are thus more han capable of providing the excess specific energy of yor particle which Wu. Fabian Nulsen (2000) ind is recuired to xealk the sell similarity of X-ray cluster properties and. match the observed. LxT relation.," For $a \sim 0.1$ (as suggested by Fabian 1999) these values are equivalent to $\sim 40$ and $\sim 7$ per particle (for fully ionized gas with cosmic abundances), respectively, and are thus more than capable of providing the excess specific energy of per particle which Wu, Fabian Nulsen (2000) find is required to break the self similarity of X-ray cluster properties and match the observed $L_{\rm{X}}-T$ relation."877 That AGN winds may be important in this regard has been recognise bv Ensslin et al. (, That AGN winds may be important in this regard has been recognised by Ensslin et al. (878L998). Wu et al. (,"1998), Wu et al. ("8792000) and Bower et al. (,2000) and Bower et al. (8802000).,2000).881 A discussion of the impact of this energy release on the general (ie. non-cluster) intergalactie mecdiunr is deferred to a later paper., A discussion of the impact of this energy release on the general (i.e. non-cluster) intergalactic medium is deferred to a later paper.882that X-ray activity rather depends on the amount of generated magnetic flux but not on the mechanism responsible for its creation.,that X-ray activity rather depends on the amount of generated magnetic flux but not on the mechanism responsible for its creation.883 Future X-ray observations of ultracool dwarfs are highly desirable to extend and deepen our understanding of magnetic activity and coronae in the regime of the coolest stars at the end of the main sequence., Future X-ray observations of ultracool dwarfs are highly desirable to extend and deepen our understanding of magnetic activity and coronae in the regime of the coolest stars at the end of the main sequence.884Although Waléuu's original suggestionOO (Waléu1919) that torsional azimuthal Alfvéun oscillatious ay exist in stellar interiors. the energy source sustainig these oscillations is not vot known (Rosner&Weiss1992).,"Although Walénn's original suggestion \citep{wal} that torsional azimuthal Alfvénn oscillations may exist in stellar interiors, the energy source sustaining these oscillations is not yet known \citep{ros}."885. Witrout au CLOYSV source. he oscillations ave believed to be damped * phase mixing (Clharbonuceatt&MacCiregor1993) or transport of magneic flux towards he surface.," Without an energy source, the oscillations are believed to be damped by phase mixing \citep{cha} or transport of magnetic flux towards the surface."886 Several different τιrechanismis have con sugeested from time to time tfo support orsional oscillations (Lavzercta.1979:Mcelutvre1991:Lanzaetal.1998:Cougho 2002).. however ione of them may account for the energv source.," Several different mechanisms have been suggested from time to time to support torsional oscillations \citep{la,ma,lan,ga}, however none of them may account for the energy source."887" The main problem ids that tesional /Alfvéuu waves are pure electromaguotic* oscillatious iu naenetolydrocdwnamics and onc| needs tfo. have au azuuuthal force which may"" support them against camping.", The main problem is that torsional Alfvénn waves are pure electromagnetic oscillations in magnetohydrodynamics and one needs to have an azimuthal force which may support them against damping.888 The abscuce of such a force suppressed the iuterest iu these oscillations as a possible mechanisgi for stellar activity cvcles., The absence of such a force suppressed the interest in these oscillations as a possible mechanism for stellar activity cycles.889 ILowever no one paid attfion to he fact that it is nof necessary to have| a direct diiviusg force or the amplification of the oscillations., However no one paid attention to the fact that it is not necessary to have a direct driving force for the amplification of the oscillations.890 The oscillation can also he zuuplifie through the veriodical variation of the svstel parameters and the process is calleL parametric xSOLLALLCC., The oscillation can also be amplified through the periodical variation of the system parameters and the process is called parametric resonance.891 The well-known example of such SVSeni ds the nathematical penduhiun witl periodically varius eusth., The well-known example of such system is the mathematical pendulum with periodically varying length.892 When the freqchev of hne parameter variation is twice the frequency of the svstem oscillation. then the oscillations wil exponentially erow in time.," When the frequency of the parameter variation is twice the frequency of the system oscillation, then the oscillations will exponentially grow in time."893 The pliase s»eed of torsiona Alfvén waves depends ou the maenctic field aud medi density., The phase speed of torsional Alfvénn waves depends on the magnetic field and medium density.894 So if some exterval force (or oscillation) cases a periodical variaion of either cnautitics (or both together) then tιο Alfvéun wavess will be zupliied., So if some external force (or oscillation) causes a periodical variation of either quantities (or both together) then the Alfvénn waves will be amplified.895 The pioucering work of Zacjarashvili sugeestedOO that the periodica clensity, The pioneering work of \citet{zaq} suggested that the periodical density896see in the DCOs Guaxiniun outflow speeds of at least 1500 kins +) are extremely rare in composite obscured-AGN/starburst svsteiis and not seen at all in “pure” obscured AGN.,"see in the DCOs (maximium outflow speeds of at least 1500 km $^{-1}$ ) are extremely rare in composite obscured-AGN/starburst systems and not seen at all in ""pure"" obscured AGN."897 They couclude that the ACN can drive very high-speed outflows. but these flows are confined to a sinall region verv near the black hole (aud are hence undetectable in Type 2 AGN).," They conclude that the AGN can drive very high-speed outflows, but these flows are confined to a small region very near the black hole (and are hence undetectable in Type 2 AGN)."898 They also conclude that the AGN plavs little or no role in driving the large-scale outflows seen in the composite svstenis., They also conclude that the AGN plays little or no role in driving the large-scale outflows seen in the composite systems.899 It therefore seclus unlikely that the high velocity outflows iu the DCOs are driven by an obscured ACN., It therefore seems unlikely that the high velocity outflows in the DCOs are driven by an obscured AGN.900 Tremonti et al. (, Tremonti et al. (9012007) have detected outflows at velocities ranging from 500 to 2000 kaa + in massive post-starburst galaxies at z ~ 0.6.,2007) have detected outflows at velocities ranging from 500 to 2000 km $^{-1}$ in massive post-starburst galaxies at z $\sim$ 0.6.902 They speculate that such high velocities may require an AGN-driven outflow., They speculate that such high velocities may require an AGN-driven outflow.903 It is also possible that the these objects are more massive versions of the DCOs in this paper., It is also possible that the these objects are more massive versions of the DCOs in this paper.904 This could be confirmed by TST imaging of their galaxies., This could be confirmed by HST imaging of their galaxies.905 The properties of the Lyra profile are. highly. scusitive to the kinematics aud distribution of the outfiowiug gas and dust in star-forming galaxies (e.g. Steidel et al., The properties of the $\alpha$ profile are highly sensitive to the kinematics and distribution of the outflowing gas and dust in star-forming galaxies (e.g. Steidel et al.906 2010: Tansen Oh 2006: Verhanune et al., 2010; Hansen Oh 2006; Verhamme et al.907 2006. 2008: I&ornuci et al.," 2006, 2008; Kornei et al."908 2010)., 2010).909 In Figure 5 we plot the Ίσα profiles for the eight LBAs with COS data. aud highlight one very sugeestive trench.," In Figure 8 we plot the $\alpha$ profiles for the eight LBAs with COS data, and highlight one very suggestive trend."910 Iu the five cases iun which the inferred relative escape fraction is simall. there is very little Gf amv) Lye ciission blueward of the systemic velocity.," In the five cases in which the inferred relative escape fraction is small, there is very little (if any) $\alpha$ emission blueward of the systemic velocity."911 Instead. the profiles show Lvo emission recdhward of the systemic velocity ancl either pure absorption or a mix of absorption aud weak enission to the blue.," Instead, the profiles show $\alpha$ emission redward of the systemic velocity and either pure absorption or a mix of absorption and weak emission to the blue."912 Such profiles are typical of high-z star forming galaxies (e.g. Shaplev et al., Such profiles are typical of high-z star forming galaxies (e.g. Shapley et al.913 2003: Steidel et al., 2003; Steidel et al.914 2010)., 2010).915" In contrast. for the three objects in which we infer a hieh relative escape fraction (LBA0213|12. 0505120, and 0921115) a significant fraction of the Ίνα emission is blucshifted with respect to the svstemic velocity."," In contrast, for the three objects in which we infer a high relative escape fraction (LBA0213+12, 0808+39, and 0921+45) a significant fraction of the $\alpha$ emission is blueshifted with respect to the systemic velocity."916 To quautify this. we have measured the net equivalent widths (cuiission - absorption) blueward aud redward of the svstemic velocity iu these cight LBAs.," To quantify this, we have measured the net equivalent widths (emission - absorption) blueward and redward of the systemic velocity in these eight LBAs."917 We define net enmüssion (absorption) to have a positive (negative) equivalent width., We define net emission (absorption) to have a positive (negative) equivalent width.918 We then take the ratio of the red aud blue equivalent widths (Regia)., We then take the ratio of the red and blue equivalent widths $R_{eqw}$ ).919 We list the results in Table 2., We list the results in Table 2.920" For the five galaxies with low inferred escape fractions Rega is either negative (οὐμασα cussion aud blicshitted absorption) or sinall and positive (strong redshifted cussion and weals blueshifted net emission): LBAQOSS- 00 CR,= 0.21). 0150|13 (-0.78). 0926|15 (0.1). 093815E qu(20.17). and 2103-07 (0.01)."," For the five galaxies with low inferred escape fractions $R_{eqw}$ is either negative (redshifted emission and blueshifted absorption) or small and positive (strong redshifted emission and weak blueshifted net emission): LBA0055- 00 $R_{eqw} =$ -0.21), 0150+13 (-0.78), 0926+45 (0.14), 0938+54 (-0.17), and 2103-07 (0.04)."921" For the three galaxies with high inferred escape fractions Roya is oforder oue aud positive (similar amounts of red- and bluc-hüfted cuuission): 0213|12 (1,ge= 0.36). 080829 (0.27). aud 0921|15 (0.75)."," For the three galaxies with high inferred escape fractions $R_{eqw}$ is of-order one and positive (similar amounts of red- and blue-shifted emission): 0213+12 $R_{eqw} =$ 0.36), 0808+39 (0.27), and 0921+45 (0.75)."922" Profiles with sieui&cant| blueshifted enüssion are ""uuiconmion but not uuxnuown in star- foruiug galaxies at D 2-3.", Profiles with significant blueshifted emission are uncommon but not unknown in star- forming galaxies at $z \sim$ 2 - 3.923 Verhanuuae et al. (, Verhamme et al. (924"2008) discuss two such cases and interpret thei as arising due to radiative transfer effects m a static οςτα,",2008) discuss two such cases and interpret them as arising due to radiative transfer effects in a static medium.925 This model cau not be correct for the three DCOs in Figure 8 because the metal lines demonstrate that the gas is rapidly outfhowing (Figures land 6)., This model can not be correct for the three DCOs in Figure 8 because the metal lines demonstrate that the gas is rapidly outflowing (Figures 4 and 6).926 Erb et al. (, Erb et al. (9272010) discuss the case of a high-z ealaxy with both an outflow detected in the metal lines and a significant amount of blicshifted Ίσα emission.,2010) discuss the case of a high-z galaxy with both an outflow detected in the metal lines and a significant amount of blueshifted $\alpha$ emission.928 They argue that in this case the optical depth to Lya photons is uuusuallv low in the foreground outflowing material aud that this is naturally related to the low, They argue that in this case the optical depth to $\alpha$ photons is unusually low in the foreground outflowing material and that this is naturally related to the low929under the assumption that our noise is) Ciaussian distributed.,under the assumption that our noise is Gaussian distributed.930 We have weakly tested this assumption in 4?? and77., We have weakly tested this assumption in $\S~\ref{chi}$ $\mathrm{and} ~\ref{invert}$.931 PCAL) acts as an overall normalisation in equation (1)). and does not depend upon y as long as the noise is not correlated: with sources on the sky.," $P(M)$ acts as an overall normalisation in equation \ref{eqn:pd}) ), and does not depend upon $S_{\mathrm{p}}$ as long as the noise is not correlated with sources on the sky."932" Strictly speaking. DPOMB]|8,) should be altered from the form we have used to account for the fact that we are examining the probability at a location where we have found a peak."," Strictly speaking, $P(M|S_{\mathrm{p}})$ should be altered from the form we have used to account for the fact that we are examining the probability at a location where we have found a peak."933" La practice. at SuoS,ο8a the full expression converges to the simpler form we have used (Bond&Efstathiou 1987))."," In practice, at $S_{\mathrm{m}}-S_{\mathrm{p}} \geq 3\,\sigma$ the full expression converges to the simpler form we have used \citealt{Bond}) )."934" The posterior (us probability. distribution. P(59,) is shown as a solid histogram in the panels of Fig."," The posterior flux probability distribution, $P(S_{\mathrm{p}})$ is shown as a solid histogram in the panels of Fig."935 4 for each of the 5 sources for which we also have follow-up xhotometrv. information., \ref{fig:boost} for each of the 5 sources for which we also have follow-up photometry information.936" The dot-dashed Gaussian in each inel is PCM[9,). which is often incorrectly adopted as the lux estimate of a map source."," The dot-dashed Gaussian in each panel is $P(M|S_{\mathrm{p}})$, which is often incorrectly adopted as the flux estimate of a map source."937 In Fig., In Fig.938" we have placed the individual (5,5) plots in order of increasing4. map SNR."," \ref{fig:boost}939 we have placed the individual $P(S_{\mathrm{p}})$ plots in order of increasing map SNR."940 Ht is clear that one expects Oo measure à non-zero Lux value a significant fraction of he time only for sources with relatively high SNRs., It is clear that one expects to measure a non-zero flux value a significant fraction of the time only for sources with relatively high SNRs.941 Sources with modest SNRs are much. more likely to have non-zero hotometry. results than lower SNR sources., Sources with modest SNRs are much more likely to have non-zero photometry results than lower SNR sources.942 The peak in he à posteriori distribution at zero flux dominates for SNlt 3.5., The peak in the a posteriori distribution at zero flux dominates for SNR $\lesssim3.5$.943 This confirms the usual prejudice towards 1igh SNR. sources if a source is bright. it needs to be detected with a SNR2do in order to be deemed a secure detection.," This confirms the usual prejudice towards high SNR sources – if a source is bright, it needs to be detected with a $\mathrm{SNR}\gtrsim4\,\sigma$ in order to be deemed a secure detection."944 Moreover. we also find that at the same SNI level. apparently. brighter sources (vith consequently. higher fux uncertainty) are more likely to be spurious (i.e. lux boosted [rom +0 mJ) than fainter sources.," Moreover, we also find that at the same SNR level, apparently brighter sources (with consequently higher flux uncertainty) are more likely to be spurious (i.e. flux boosted from $\simeq0\,\mathrm{mJy}$ ) than fainter sources."945 Thus at a given SNR. low flux sources are more likely to be real than hieh [lux SOULCCS.," Thus at a given SNR, low flux sources are more likely to be real than high flux sources."946 We performed independent photometry observations on 5 sources with SNRs ranging from 4.10 and the results are shown in Table 1..," We performed independent photometry observations on 5 sources with SNRs ranging from $4.1\,\sigma$ and the results are shown in Table \ref{tab:sources}."947 We wish to know if the photometry measurements are consistent with the mape-detected Luxes., We wish to know if the photometry measurements are consistent with the map-detected fluxes.948" In other words. we want to answer the question: “What is the probability hat we will measure 5, in photometry mode given the map-«etected Dux (Sy) and uncertainty (σι) and the uncderlving source count model?'."," In other words, we want to answer the question: `What is the probability that we will measure $S_{\mathrm{p}}$ in photometry mode given the map-detected flux $S_{\mathrm{m}}$ ) and uncertainty $\sigma_{\mathrm{m}}$ ) and the underlying source count model?'."949 Phe main point is that the à posteriori wobability of finding a bright source will be cdown-weighted ον the a priori probability coming from the source counts., The main point is that the a posteriori probability of finding a bright source will be down-weighted by the a priori probability coming from the source counts.950 A compeariscon of the dashed curves and solid histograms in Fig., A comparison of the dashed curves and solid histograms in Fig.951" 4 Erows that our photometry results are consistent with PCS,). even though the photometry is often Inconsistent wil athe raw (i.e. Dux boosted) map readings η."," \ref{fig:boost} shows that our photometry results are consistent with $P(S_{\mathrm{p}})$, even though the photometry is often inconsistent with the raw (i.e. flux boosted) map readings $P(M|S_{\mathrm{p}})$."952 We can now assess the probability. of obtaining cach of the. photometry measurements using the distributions in Vig. 4., We can now assess the probability of obtaining each of the photometry measurements using the distributions in Fig. \ref{fig:boost}.953 For Gsss50.6. à photometry result. than what we measured is expected 43 per cent of the time.," For GSS850.6, a photometry result than what we measured is expected 43 per cent of the time."954 For the remaining sources GSSS50.1. GSS850.2. GSSS850.11 and," For the remaining sources GSS850.1, GSS850.2, GSS850.11 and"955"between 9 and jm toa new plateau) is masked by a similar behavior in the H5O0 and CI, opacilies.","between 9 and $\,\mu$ m to a new plateau) is masked by a similar behavior in the $_2$ O and $_4$ opacities."956 Nevertheless. (he cloud opacity is large enough to appear in two windows where the gas is less opaque in models with 1200ziTirS2000 Ix: from ~4 to pau (allecting IRAC bands 1 through 3) and from 9 to yan. The effect is strongest al Tay1600 IN. which corresponds to the largest cloud optical depth above the 10; photosphere.," Nevertheless, the cloud opacity is large enough to appear in two windows where the gas is less opaque in models with $1200 \wig< \Teff957\wig< 2000\,$ K: from $\sim 4$ to $\,\mu$ m (affecting IRAC bands 1 through 3) and from 9 to $\,\mu$ m. The effect is strongest at $\Teff \sim 1600\,$ K, which corresponds to the largest cloud optical depth above the $\,\mu$ m photosphere."958 Because the silicate absorption flattens the spectrum noticeably in the 9 to pm range. its identification should be fairly straightforward in IRS spectra of mid-L. cdwarls.," Because the silicate absorption flattens the spectrum noticeably in the 9 to $\,\mu$ m range, its identification should be fairly straightforward in IRS spectra of mid-L dwarfs."959 For models with a higher condensate sedimentation efficiency (freq=5). the thinner cloud laver remains optically thin above the photosphere and is invisible αἱ jam. This feature can thus serve as a diagnostic of the vertical structure of the cloud model as its strength depends strongly on the cloud's vertical extent and on (he particle size.," For models with a higher condensate sedimentation efficiency $f_{\rm sed}=5$ ), the thinner cloud layer remains optically thin above the photosphere and is invisible at $\,\mu$ m. This feature can thus serve as a diagnostic of the vertical structure of the cloud model as its strength depends strongly on the cloud's vertical extent and on the particle size."960 The IRAC will image brown dwarfs and provide photometry in 4 broad bandpasses (Fig., The IRAC will image brown dwarfs and provide photometry in 4 broad bandpasses (Fig.961 1). labeled Band 1 through 4.," 1), labeled Band 1 through 4."962 Bands 1 ancl 2 cover wavelengths not accessible to the IRS., Bands 1 and 2 cover wavelengths not accessible to the IRS.963 As can be seen in Fig., As can be seen in Fig.964 1. Dand 2 covers the bands of CO and PIT; but is nevertheless the least affected by strong molecular bands over the full range of Tog of interest.," 1, Band 2 covers the bands of CO and $_3$ but is nevertheless the least affected by strong molecular bands over the full range of $\Teff$ of interest."965" Dands 1 and 4 are increasinglv affected by Cll, absorption below Zig~1900 1. Strong Ε.Ο absorption appears in Dand 3 below Tiο1200 Ix. A study of color-color diagrams shows that HIRAC colors will be most useful as diagnostics ol brown dwarf physical properties for Zigzi1400 Ix. At higher Zi. the colors change little with Zur or the Zar and gravity dependences become hopelessly (angled."," Bands 1 and 4 are increasingly affected by $_4$ absorption below $\Teff \sim 1500\,$ K. Strong $_2$ O absorption appears in Band 3 below $\Teff \sim 1200\,$ K. A study of color-color diagrams shows that IRAC colors will be most useful as diagnostics of brown dwarf physical properties for $\Teff \wig< 1400\,$ K. At higher $\Teff$, the colors change little with $\Teff$ or the $\Teff$ and gravity dependences become hopelessly tangled."966 For brown clwarls with known distances. color-magnitude diagrams provide a good cdiscriminant of both Toy and gravitv.," For brown dwarfs with known distances, color-magnitude diagrams provide a good discriminant of both $\Teff$ and gravity."967 The IR spectra of giant planets reveal the presence of trace carbon species such as CO». Calls. and HCN inemission.," The IR spectra of giant planets reveal the presence of trace carbon species such as $_2$, $_2$ $_2$, and HCN in."968 These arise primarily [rom the photochemistry of CIL; (also seen in enission) in the stratosphere. ie. above a temperature inversion in the atmosphere (Moses 2000).," These arise primarily from the photochemistry of $_4$ (also seen in emission) in the stratosphere, i.e. above a temperature inversion in the atmosphere (Moses 2000)."969 No published brown cdwarf model shows a temperature inversion. mainlv because the present knowledge of the plvsics of brown clwarl atmospheres is too primitive (o model a credible stratosphere.," No published brown dwarf model shows a temperature inversion, mainly because the present knowledge of the physics of brown dwarf atmospheres is too primitive to model a credible stratosphere."970 We nevertheless anticipate that brown clwarls also may have stratospheres (Yelle 2000) and observations in the mil-IR are a powerlul way to discover, We nevertheless anticipate that brown dwarfs also may have stratospheres (Yelle 2000) and observations in the mid-IR are a powerful way to discover971discontinuity correction was applied to the original observations with the CCD camera on the University of Torouto’s Las Campanas 0.611 telescope.,discontinuity correction was applied to the original observations with the CCD camera on the University of Toronto's Las Campanas 0.6m telescope.972 Iu coutrast. the TST WFEDPC?2 observations were tied to calibrations by Holtziuzuetal.(1995) aud Dolphin(2000).. although there are mudicatious that the scusitivity of the camera degrades temporally (Sahaetal.2006).," In contrast, the HST WFPC2 observations were tied to calibrations by \citet{ho95} and \citet{do00}, although there are indications that the sensitivity of the camera degrades temporally \citep{sa06}."973. It is possible to adjust observations frou one svsteni to those of the other. so both svstems were treated separately as the standard reference frame and the results examined individually.," It is possible to adjust observations from one system to those of the other, so both systems were treated separately as the standard reference frame and the results examined individually."974 When the observations of Shorlinetal.(2001) aud Carraro&Costa(2000). are standardized in such fashion to the IST system. of Wallaceetal.(2005).. the resulting data for likely cluster members Wine within the cluster nucleus (see below) eenerate the color-color aud color-maguitude diagrams of Fie. 3..," When the observations of \citet{sh04} and \citet{cc09} are standardized in such fashion to the HST system of \citet{wa05}, the resulting data for likely cluster members lying within the cluster nucleus (see below) generate the color-color and color-magnitude diagrams of Fig. \ref{fig3}."975 Tf the Shorlinetal.(2001). svsten is used as the standard reference frame. theCDV. color-color and coloranaegnitude diagrams are as shown in Fie. L.," If the \citet{sh04} system is used as the standard reference frame, the color-color and color-magnitude diagrams are as shown in Fig. \ref{fig4}."976 The scatter in the observations is large in both cases. typically raugiug between £0.10 and +O.10 in the colors.," The scatter in the observations is large in both cases, typically ranging between $\pm0^{\rm m}.10$ and $\pm0^{\rm m}.40$ in the colors."977 Although secuinely consisteut with he cited uncertainties iu the three studies. such laree scatter may also be linked to differential reddening in the field. as suggested by the systematic teudency for faint stars to be bluer on average than brighter stars. the expected bias caused by imterstellar extinction.," Although seemingly consistent with the cited uncertainties in the three studies, such large scatter may also be linked to differential reddening in the field, as suggested by the systematic tendency for faint stars to be bluer on average than brighter stars, the expected bias caused by interstellar extinction."978 It does μανάτwhat one can infer from the observatious., It does limitwhat one can infer from the observations.979 The differeuces between Figs., The differences between Figs.980 3 and L are generallv simall. however. except for à magnitude offset. the Shorlinetal.(2001) system being brighter than that of Wallaceetal.(2005).," \ref{fig3}~ and \ref{fig4} are generally small, however, except for a magnitude offset, the \citet{sh04} system being brighter than that of \citet{wa05}."981. The(BY observations of Fies., The observations of Figs.982 3 aud bo can be analyzed by standard techuiques (e.g.παπα1976a: 2005).. but an essential first step is to confirm the reddening relation for the field," \ref{fig3}~ and \ref{fig4} can be analyzed by standard techniques \citep[e.g.,][]{tu76a,te80,te05}, , but an essential first step is to confirm the reddening relation for the field"983"""Phe new catalog on egamua-ray pulsars obtained. bv the ""formiLATLAT instrumentiu) (Abd∖CXbdoet'al.2010a) ---increased.á nthe number of⋅ gamma-ray pulsars from⋅ seven to about fifty.",The new catalog on gamma-ray pulsars obtained by the Fermi-LAT instrument \citep{2010ApJS..187..460A} increased the number of gamma-ray pulsars from seven to about fifty.984⋅⋅ Since» then. new pulsars are discovered. regularly.," Since then, new pulsars are discovered regularly."985 uasThis allows for⋅ the first⋅ time. a reasonable statisticalui analysis⋅⋠ on the high-energy. emissionD. properties. of. these objects. like» spectral shape. cut-olf⋅ energy. ancl comparison. between radio: and eamnmia-ray raciiation if available.," This allows for the first time a reasonable statistical analysis on the high-energy emission properties of these objects like spectral shape, cut-off energy, and comparison between radio and gamma-ray radiation if available."986 For some high Iuminosity pulsars. this analysis has been completed by à phase-resolved studyB likeboe fordr theη CrabM (AbdoetDal..DJ 2010b).. VelaTale and CGeminga CXbdoctal.2010€)...," For some high luminosity pulsars, this analysis has been completed by a phase-resolved study like for the Crab \citep{2010ApJ...708.1254A}, , Vela and Geminga \citep{2010ApJ...720..272A}."987 Raclio pulses and ganinia-ray photons are expected to be produced in. dilferentqe emissionD. sites.: probably. close to the neutron ⋅ ⋅ ⋅ ⊳∖⋜⊔⋅⊳∖⊔↓⋅⇂⋯∼⋖⋅⇂∪↓⋅⇂↓↕⋖⋅⇂∩↓⋅⊔↓∢⋅↓⋅⋡∠⇂∢⋅≱∖≼∙↓⋅↓∣⋊⋅∠⇂∣⋡∙∖⇁⋜↧↓≻∪↓⋜⊔⋅≼⇍⋜↧↓≻⊔↓⋯⇂⋖⊾↓. ↿∖↘⋯∐↥⋜↧↳↓⋅↓⊳∖↓⊔⋯⊔⊾∖↽≺∩∪↓∡∢⋅↓≤⋗≼≻≤∩↣⋜⋯∠⇂↓⊔↿↓↥⋖⋅∖⇁⊔⇍↓⊔∐∙∖⇁∪⇂↥↓↥⋖⋅>. . s ⋅ : ⋅ ↓↓⋏∙≟⇂∐⊣⇍∙∖⇁↓↓⊔∠⇂⋖⋅↓⋅⇂∪↓⋅∣↓∐⋅↥⋜⊔⊓⋅↓⋅⋡∢⋅⇀∖↓≻↓⋜⊔⊔∢⋅∠⇂∣⋡∙∖⇁∪⋯∢⋅↓⋅⋏∙≟⋜↧↓≻⊳∖↿∖≺↓↕∢⊾⊔⋏∙≟. etal.1986).⋅ or alternatively⊀ by the striped. wind. (PétriYu Y09)..," Radio pulses and gamma-ray photons are expected to be produced in different emission sites, probably close to the neutron star surface for the former, described by a polar cap model \citep{1969ApL.....3..225R}, and in the vicinity of the light-cylinder for the latter, explained by outer gaps \citep{1986ApJ...300..500C} or alternatively by the striped wind \citep{2009A&A...503...13P}."988 Polarization properties at multi-wavelength: would certainly. help to constrain. the geometry of⋅ the sites.. of emission (Dyvksetal.2004:Pétri&Wirk2005:2009)..," Polarization properties at multi-wavelength would certainly help to constrain the geometry of the sites of emission \citep{2004ApJ...606.1125D, 2005ApJ...627L..37P,989 2009A&A...503...13P}."990 Canuna-ray light-curves alone can already give good insight into the magnetosphere (Romani&Watters2010).., Gamma-ray light-curves alone can already give good insight into the magnetosphere \citep{2010ApJ...714..810R}.991 The new sample of Fermi-LAT ganima-ray pulsars increased interest into modeling of gamma-ray emission., The new sample of Fermi-LAT gamma-ray pulsars increased interest into modeling of gamma-ray emission.992be turbulent. stably stratified aud inagjetizecl.,"be turbulent, stably stratified and magnetized."993 The tachocline is believed to play a crucial role in the generation of the eleven year solar cycle (see Tobias&Weiss (2007))) and areular Inomentunm: tansport through the tachoeline may be respolsible for spiuniig down the solar interior., The tachocline is believed to play a crucial role in the generation of the eleven year solar cycle (see \cite{tw:2007}) ) and angular momentum transport through the tachocline may be responsible for spinning down the solar interior.994 Oje Crucia issue to be resolved is thereore the role of turoileuce in transporting augular 1nolnueium in the tachocline. aud this lias bee1 adressecl both iu the bydrocdylane and magueencodyllc :tines (see e.g. Spiegel&Zahn(1992):Crout&Melutyre{1998):MclutyP'e(2003):Tobiasetal. (2007))).," One crucial issue to be resolved is therefore the role of turbulence in transporting angular momentum in the tachocline, and this has been addressed both in the hydrodynamic and magnetohydrodynamic settings (see e.g. \cite{spza:1992,gm:1998,mci:2003,Tobias:2007p116}) )."995 What has been slowl is that whereas aulsotropic |veldrodsnaiie two-dimeINLOLalt wbueuce leads to the ellicleut formation of zonal lows via Reyuods stresses: e aclclitLOL OL a magietic: field leads to Maxwell stresses t1al cali oppose the formation o ‘jets., What has been shown is that whereas anisotropic hydrodynamic two-dimensional turbulence leads to the efficient formation of zonal flows via Reynolds stresses; the addition of a magnetic field leads to Maxwell stresses that can oppose the formation of jets.996" The SUppLesslot of jets is a finction of the streiotLol the large-scale inagnetic ield ancl the ocal inaguetic Rey10ds number £2,,,.", The suppression of jets is a function of the strength of the large-scale magnetic field and the local magnetic Reynolds number $R_m$.997 The paper is orgaiisecl in the ollowiugoO nanner: in the next section we jut'oduce the general inethod axl the compttational savines tlal can be achieved for the case of Sjoherieal sviunetry in two clitjensious., The paper is organised in the following manner: in the next section we introduce the general method and the computational savings that can be achieved for the case of spherical symmetry in two dimensions.998 In section 3 the particular moclel of MHD turbulence oi a rotatiug spherical surface is jutroduced aid a comparison of he large-scale οναός of DNS aid DSS ismace., In section 3 the particular model of MHD turbulence on a rotating spherical surface is introduced and a comparison of the large-scale dynamics of DNS and DSS is.999.. We conclude |yy discussing extensions to the 1jethod. aud speculating ou the range of problems where such a tecinique may* be of use., We conclude by discussing extensions to the method and speculating on the range of problems where such a technique may be of use.1000 Iu this section we describe the derivation of a general fully spectral algorituu for the direct statistical simulation of astrophysical flows, In this section we describe the derivation of a general fully spectral algorithm for the direct statistical simulation of astrophysical flows.1001 We develop the method for the typical case of equations with quad‘atic nonlinearities. before specialising to systems with splerical sviunetry ist two cimensious.," We develop the method for the typical case of equations with quadratic nonlinearities, before specialising to systems with spherical symmetry in two dimensions."1002 Consider a systeu that is represented by partial cillereial evolution equations (PDEs) for a liuuber r of scalar fields., Consider a system that is represented by partial differential evolution equations (PDEs) for a number $r$ of scalar fields.1003 Typically sucht a system may be solved directly by clise'etising the PDEs using a fiuite-dilIererce. finite volume or finite element mehod or by deriving equations for the uplitude of modes i Laspectral expausion.," Typically such a system may be solved directly by discretising the PDEs using a finite-difference, finite volume or finite element method or by deriving equations for the amplitude of modes in a spectral expansion."1004 Formally. this t‘ausloriuus the PDEs into a finite set of ‘inary dillerential equations (ODEs) that may be integrated forward in time.," Formally, this transforms the PDEs into a finite set of ordinary differential equations (ODEs) that may be integrated forward in time."1005 It the discretisation is performect at. s discrete poiuts (or [or s spectral modes) hen the evolution equations cau take, If the discretisation is performed at $s$ discrete points (or for $s$ spectral modes) then the evolution equations can take1006asviuuetrice drift corrections are applicable to collisionless stellar systems for which the magnitude of the random motions is auch smaller than that ofthe rotation velocity.,asymmetric drift corrections are applicable to collisionless stellar systems for which the magnitude of the random motions is much smaller than that of the rotation velocity.1007 However. it is often used even for gaseous disks. where the asstuuption beiug made is that the pressure support cau be approximated as the eas deusity times the square of the raudom velocity.," However, it is often used even for gaseous disks, where the assumption being made is that the pressure support can be approximated as the gas density times the square of the random velocity."1008 In the absence of any measurement for A. for DDO210. we have assmmed d(In(2i))/dr.=0 (ic. that the scale height docs not change with radius).," In the absence of any measurement for $h_z$ for DDO210, we have assumed $d(\ln(h_z))/dr=0$ (i.e. that the scale height does not change with radius)."1009 Also. using the fact that σ is constant across the galaxy. we get: Using the fitted Gaussian profile to the radial surface density distribution. (see equ 1)) we obtain The observed III velocity dispersion was corrected for the instrmucutal broadcuing as well as for the broadening due to the velocity eradient over the finite size of the beam.," Also, using the fact that $\sigma$ is constant across the galaxy, we get: Using the fitted Gaussian profile to the radial surface density distribution, (see eqn \ref{eqn:hisb}) ) we obtain The observed HI velocity dispersion was corrected for the instrumental broadening as well as for the broadening due to the velocity gradient over the finite size of the beam."1010" The applied correction is G2truc=a""olsAi?PVD where Gra. is the true velocity dispersion. Ac is the channel width. & characterizes the beam width (i.c. the bem ds assmmed to be of form e 7!) and «, is the observed rotation velocity."," The applied correction is $\sigma_{\rm true}^{2}=\sigma_{\rm obs}^2-\Delta v^2-1011 \frac{1}{2}{b}^2{({\nabla}v_{\rm{o}})}^2$, where $\sigma_{\rm true}$ is the true velocity dispersion, $\Delta v$ is the channel width, $b$ characterizes the beam width (i.e. the beam is assumed to be of form $e^{-x^2/b^2}$ ) and $v_{\rm{o}}$ is the observed rotation velocity."1012 After putting the appropriate values in the above equation. we ect σTrucD2~7736.0∙⋅ kii?D D7.," After putting the appropriate values in the above equation, we get $\sigma^2_{\rm true} \approx 36.0$ $^2$ $^{-2}$."1013 Finally.H substitutingHB thisB value back iuto- the Equ. (1))," Finally, substituting this value back into the Eqn. \ref{eqn:corr_curve}) )"1014 the vasvunmetric drift” corrected curve obtained is given as a dotted line in Fig. 9..," the “asymmetric drift"" corrected curve obtained is given as a dotted line in Fig. \ref{fig:v_asy}."1015 As discussed above. the “asvuumetric αν correction was derived under the assumption that the disk scale height of DDO210 does not chanee with radius.," As discussed above, the “asymmetric drift” correction was derived under the assumption that the disk scale height of DDO210 does not change with radius."1016 To quautify the effect of this assiuuption. the correction was recalculated assuuiug a linear increase of in the scale height. from qy=0.25 at the ceuter to qu20.5 at the edee of the galaxy.," To quantify the effect of this assumption, the correction was recalculated assuming a linear increase of in the scale height, from $_0$ =0.25 at the center to $_0$ =0.5 at the edge of the galaxy."1017 The chanee in the asvuuuectric dift correction was found to be <1«, The change in the asymmetric drift correction was found to be $<1$.1018 The asstuuption that the scale heigl is constant hence have a negligible effect ou the derived halo parameters., The assumption that the scale height is constant hence have a negligible effect on the derived halo parameters.1019" lu this section we use the ""asviuuetmÉe drift” corrected rotation curve derived iu the last section to derive lass uodels for DDO210."," In this section we use the “asymmetric drift"" corrected rotation curve derived in the last section to derive mass models for DDO210."1020 As discussed in Sect., As discussed in Sect.1021 3.1. the optical eiission is not axtsvuuuetric but is instead patchy aud elongated., \ref{ssec:HI_dis} the optical emission is not axi-symmetric but is instead patchy and elongated.1022" For ie purpose of mass modeling however. we approximate 16 optical ejuission to be αλποΊο, but we return o this issue iu Sect 3.5."," For the purpose of mass modeling however, we approximate the optical emission to be axi-symmetric, but we return to this issue in Sect \ref{ssec:discuss}."1023" Assume that the optical nissbon was axisviunuetrie. Leeetal.(1999) estimate 1ο D bud. scale length. (65) of DDO210 to be 35.0"" (~OF kpce)"," Assuming that the optical emission was axi-symmetric, \cite{lee99}1024 estimate the B band scale length $\alpha_B$ ) of DDO210 to be ${''}$ $\sim 0.17$ kpc)."1025 This estimate of ap agrees to within with the value of 39 estimated by vanZee(2000)., This estimate of $\alpha_B$ agrees to within with the value of $^{''}$ estimated by \cite{vanzee00}.1026. Leeetal.(1999) also ound the colours of the galaxy to be nearly constant with the ealactocentric radius., \cite{lee99} also found the colours of the galaxy to be nearly constant with the galactocentric radius.1027 Hence. we computed the contribution of the stellar mass to the observed rotation curve by assmuiug it to be an exponcutial disk of constant /Lj ratio (Y p) and with an intrinsic thickness ratio (qu) of 0.25.," Hence, we computed the contribution of the stellar mass to the observed rotation curve by assuming it to be an exponential disk of constant $_B$ ratio $\Upsilon_B$ ) and with an intrinsic thickness ratio $_0$ ) of 0.25."1028 For the vertical density distribution of the stellar disk. we assted a sech?(2/29) profile. with τρ independent of galactocentrie radius.," For the vertical density distribution of the stellar disk, we assumed a $^2(z/z_0)$ profile, with $z_0$ independent of galactocentric radius."1029 These are reasonable assuniptious for disk galaxies (c.c. van der Rrui Searle 1981. de Cais Peleticr," These are reasonable assumptions for disk galaxies (e.g. van der Kruit Searle 1981, de Grijs Peletier"1030—]. Set T»contains the jobs that wereassigned toAZ;ator,"After the job removal step each machine $M_j$ , $1\leq j\leq m$, has a load of at most $\max\{\beta(j)L^*,(\alpha_m-1)L\}$."1031 before time /* —1and arela, We first observe that this load is at most $\alpha_mL$.1032rgeatime/*—1. Finally 75isthe setof jobs," If $(\alpha_m-1)L\geq \beta(j)L^* $, there is nothing to show."1033 assigned, We evaluate $\beta(j)L^*$.1034 toAZ; atoralter time/*.Weshowanum," If $j> \lfloor m/\alpha_m\rfloor$ , then $\beta(j)=\alpha_m$ and $\beta(j)L^* = \alpha_m L^*1035\leq \alpha_m L$."1036ber of claims that we will use inthe furtherproof. Claim4.1.Eachjob in75 UT is large at thetim," If $j\leq \lfloor m/\alpha_m\rfloor$, then $\beta(j) = (\alpha_m-1)m/(m-j) 1037\leq (\alpha_m-1)m/(m-\lfloor m/\alpha_m\rfloor) = (\alpha_m-1)m/\lceil (1-1/\alpha_m)m\rceil)\leq \alpha_m$ and thus $\beta(j)L^* \leq \alpha_m L$."1038"e itisassigned toÀj. 4.2.There holds 57 , PES O(J)L$. whereJ;isthe job"," Hence $M_j$ 's load is upper bounded by $\alpha_m {\mathit OPT}$, where ${\mathit OPT}$ denotes the value of the optimum makespan for the job sequence $\sigma$."1039of Tjthat assigned last Ad;. Claim chu," The followinglemma ensures thatafter the reassignment step, each machine still has aload of at most $\alpha_m {\mathit OPT}$ ."10402001) in the IRAF/STSDASeither package.,2001) in the IRAF/STSDAS package.1041 The data were calibrated in the VEGAMACG photometric svstem. using the current best values for the zero points (cf.," The data were calibrated in the VEGAMAG photometric system, using the current best values for the zero points (cf."1042 HST/WEDPC? Instrument Handbook. V.5. 2000).," /WFPC2 Instrument Handbook, V.5, 2000)."1043 Figure 1 shows color images of each of the three WFPC?2 fields of view. with blue representing FOOGW V. red representing FES14W J. and green a mean image of the two.," Figure 1 shows color images of each of the three WFPC2 fields of view, with blue representing F606W $V$, red representing F814W $I$, and green a mean image of the two."1044 The three WEPCO? fields include a total of 26 X-ray sources listed in the 300 ksec Chandra catalog., The three WFPC2 fields include a total of 26 X-ray sources listed in the 300 ksec Chandra catalog.1045 Figure 2 shows the three WFPCY fields of view overlaid on a map of the 300 Κου N-rav sources. coded for intensity. as detected by Tozzi et al. (," Figure 2 shows the three WFPC2 fields of view overlaid on a map of the 300 ksec X-ray sources, coded for intensity, as detected by Tozzi et al. ("10462001) in their 0.5—7 keV detection hand.,2001) in their $0.5-7$ keV detection band.1047 Since we chose IIST pointings to optimize the number of X-ray sources observed. (he density of sources in our fields is higher than would be expected for random WEPC?2 fields.," Since we chose HST pointings to optimize the number of X-ray sources observed, the density of sources in our fields is higher than would be expected for random WFPC2 fields."1048 Note that Tozzi et al., Note that Tozzi et al.1049" find 197 sources in their 0.104 deg? field. with clustering on scales up to100""."," find 197 sources in their 0.104 ${}^2$ field, with clustering on scales up to."1050.. Our 26 sources in a total area. of zz17 arcmin? represents 3 limes the average source density., Our 26 sources in a total area of $\approx\!17$ ${}^2$ represents $\sim\!3$ times the average source density.1051" Figure 3 is a collage of 20""xIIST W+J/ image sections. centered on each Chandra source in our fields and overlaid with the 0.5—? keV X-ray contours [rom the 300 ksee data (Tozzi et al."," Figure 3 is a collage of $20\arcsec\times20\arcsec$ $V\!+\!I$ image sections, centered on each Chandra source in our fields and overlaid with the $0.5-7$ keV X-ray contours from the 300 ksec data (Tozzi et al."1052 2001)., 2001).1053 We show. for reference. the expectedCXO point source response size (FWIIM) at each source location.," We show, for reference, the expected point source response size (FWHM) at each source location."1054 We first created a detection image for each WEDPC? field by adding the Ἐ and J images., We first created a detection image for each WFPC2 field by adding the $V$ and $I$ images.1055 We then extracted sources with the SExtractor program (version 2.1.6: Bertin Arnouts 1996)., We then extracted sources with the SExtractor program (version 2.1.6; Bertin Arnouts 1996).1056 In light of the correlated noise ancl the z3-pixel EWIIM in the 07005/pixel drizzled images. we adopted conservative SExtractor detection criteria: 10 contiguous pixels al >1.5 limes the RAIS (222.3 counts in the WFC regions and zz5.3 counts in the PC regions).," In light of the correlated noise and the $\approx\!3$ -pixel FWHM in the 05/pixel drizzled images, we adopted conservative SExtractor detection criteria: 10 contiguous pixels at $\geq\!1.5$ times the RMS $\approx\!2.3$ counts in the WFC regions and $\approx\!5.8$ counts in the PC regions)."1057 For a source wilh equivalent counts in each filter (V—7= 1.2). this corresponds to a WEC detection limit of J<28.2 and ji;<24.2 mag 7.," For a source with equivalent counts in each filter $V\!-\!I=1.2$ ), this corresponds to a WFC detection limit of $I \leq 28.2$ and $\mu_I \leq105824.2$ mag ${}^{-2}$."1059" The corresponding limits for the PC are J<27.2 and pry<23.2 mag 7, "," The corresponding limits for the PC are $I1060\leq 27.2$ and $\mu_I \leq 23.2$ mag ${}^{-2}$ ."1061We ran SExtractor twice for each field using a single detection image as described above. but performing the photometry separately on the V. and / images.," We ran SExtractor twice for each field using a single detection image as described above, but performing the photometry separately on the $V$ and $I$ images."1062 This ensures that (he source pixels are identically matched between the two filters., This ensures that the source pixels are identically matched between the two filters.1063 Our resulting optical source catalog contains 3834 objects among the three fields., Our resulting optical source catalog contains 3834 objects among the three fields.1064 We plot their J magnitudes and V—1 colors (based on the SExtractor DBBEST parameter) as the small dots in Figure 2.., We plot their $I$ magnitudes and $V\!-\!I$ colors (based on the SExtractor BEST parameter) as the small dots in Figure \ref{fig:col_mag}.1065 These magnitudes do not include correction for the WEPC? eharge-transler inelliciency (Whitmore. Hever. Casertano 1999 and references therein). which we expect to be «0.03 mag given theappreciable background in our exposures.," These magnitudes do not include correction for the WFPC2 charge-transfer inefficiency (Whitmore, Heyer, Casertano 1999 and references therein), which we expect to be $<\!0.03$ mag given theappreciable background in our exposures."1066"The presence of CO gas in at least some of these systems is known independently through mm-wave spectroscopic observations (Yan et al.,","The presence of CO gas in at least some of these systems is known independently through mm-wave spectroscopic observations (Yan et al.,"1067 in prep.)., in prep.).1068" In particular, MIPS16080 and"," In particular, MIPS16080 and"1069(he effective area of the detector.,the effective area of the detector.1070" The triggers operate off the peak [αν (2?,,4)). so the limiting [Inence will depend on the effective duration (5544/17,4,). Which can vary widely from burst to burst."," The triggers operate off the peak flux $P_{max}$ ), so the limiting fluence will depend on the effective duration $S_{bolo}/P_{max}$ ), which can vary widely from burst to burst."1071" Thus. the limit due to trigger thresholds will be ""Tuzzv. with no sharp edge but rather a gradient as ο is reduced."," Thus, the limit due to trigger thresholds will be `fuzzy', with no sharp edge but rather a gradient as $S_{bolo}$ is reduced."1072" Approximately. the trigger threshold will produce a horizontal eutoff at the bottom of the S,4;5bol—necespeak.olops diagram."," Approximately, the trigger threshold will produce a horizontal cutoff at the bottom of the $S_{bolo} - E_{peak,obs}$ diagram."1073e In principle. the exact trigger thresholds can be ealeulated lor every detector and burst.," In principle, the exact trigger thresholds can be calculated for every detector and burst."1074 In practice. the conditions (Linearops. background flux. incidence angle. burst light curve) vary eveally [rom burst to burst. creating substantial scatter in the thresholds.," In practice, the conditions $E_{peak,obs}$, background flux, incidence angle, burst light curve) vary greatly from burst to burst, creating substantial scatter in the thresholds."1075" For (his paper. we do not need an accurate distribution for the 55,5; threshold. so instead we caleulate the ivpical 55,4; threshold as a function of Eyck, lor average conditions."," For this paper, we do not need an accurate distribution for the $S_{bolo}$ threshold, so instead we calculate the typical $S_{bolo}$ threshold as a function of $E_{peak,obs}$ for average conditions."1076 In. particular. we adopt an average spectral shape as the Dand funcüon (Band et al.," In particular, we adopt an average spectral shape as the Band function (Band et al."1077 1993) with a low-energy power law index of -1.0 and a hieh-enereyv power law index of -2.0., 1993) with a low-energy power law index of -1.0 and a high-energy power law index of -2.0.1078 We also take the effective duration of the peak (ος μμ) such that it fits the observed distribution of (he detectors., We also take the effective duration of the peak $S_{bolo}/P_{max}$ ) such that it fits the observed distribution of the detectors.1079 For each detector. we take its trigger energv range. [ace-on effective area. and the average backeround flux.," For each detector, we take its trigger energy range, face-on effective area, and the average background flux."1080 The formalism and many of the input parameters were taken from Band (2003)., The formalism and many of the input parameters were taken from Band (2003).1081 The result is a lower limit in the Spor—Epeat.obs diagram. as displaved in Figure 7 [or DATSE. Figure 8. for IEEE. and Figure 9. lorSif...," The result is a lower limit in the $S_{bolo} - E_{peak,obs}$ diagram, as displayed in Figure \ref{fig:NaPBATSE} for BATSE, Figure \ref{fig:NaPHETE} for HETE, and Figure \ref{fig:NaPSwift} for."1082" We do not have enough information to calculate trigger thresholds for some satellites. but the threshold is usually fairly obvious (e.g.. Figure 12 has a nearly flat ancl moderately sharp lower limit to 55,4)."," We do not have enough information to calculate trigger thresholds for some satellites, but the threshold is usually fairly obvious (e.g., Figure \ref{fig:NaPKonus} has a nearly flat and moderately sharp lower limit to $S_{bolo}$ )."1083 These thresholds are nol sharp. so bursts can easily appear somewhat below the threshold.," These thresholds are not sharp, so bursts can easily appear somewhat below the threshold."1084 Indeed. by varvine ihe input conditions somewhat. the trigger threshold lines can be (trauslated up and down substantially.," Indeed, by varying the input conditions somewhat, the trigger threshold lines can be translated up and down substantially."1085" A second detector selection effect is that the burst must have enough photons recorded for the analvst to be able to determine the Ej, value."," A second detector selection effect is that the burst must have enough photons recorded for the analyst to be able to determine the $E_{peak,obs}$ value."1086" This will depend on both Spey, and {ροκος as well as the detector properties."," This will depend on both $S_{bolo}$ and $E_{peaks,obs}$ as well as the detector properties."1087" For example. a burst just above the trigger threshold. will have just enough. photons to be detected but not enough photons to allow any constraints on the £E,,4,,/5; value. so this burst will not be included in a sample Lor plotting on the 5,44—Epeatoos diagram."," For example, a burst just above the trigger threshold will have just enough photons to be detected but not enough photons to allow any constraints on the $E_{peak,obs}$ value, so this burst will not be included in a sample for plotting on the $S_{bolo} - E_{peak,obs}$ diagram."1088" For another example. consider a burst with E44, al the upper edge of the measured spectral range for a detector. such (hat a very bright burst will have a well-measured turnover that accurately defines the fitted Byronops value. whereas a fainter burst will have poor photon statistics near the turnover in (he spectrum and the Z,,,,,; value will remain unmeasured and the burst will not be included in any of our samples."," For another example, consider a burst with $E_{peak,obs}$ at the upper edge of the measured spectral range for a detector, such that a very bright burst will have a well-measured turnover that accurately defines the fitted $E_{peak,obs}$ value, whereas a fainter burst will have poor photon statistics near the turnover in the spectrum and the $E_{peak,obs}$ value will remain unmeasured and the burst will not be included in any of our samples."1089is alleviated by employing equations of state for star-forming gas that are stiffer than isothermal.,is alleviated by employing equations of state for star-forming gas that are stiffer than isothermal.1090 Using a suite of isolated galaxy formation simulations. we have explicitly demonstrated the consequences our treatment for the ISM has on the formation of disk galaxies.," Using a suite of isolated galaxy formation simulations, we have explicitly demonstrated the consequences our treatment for the ISM has on the formation of disk galaxies."1091 Simulations with an effectively isothermal ISM yield) gravitationally unstable disks and fail to produce large. smooth stellar disks.," Simulations with an effectively isothermal ISM yield gravitationally unstable disks and fail to produce large, smooth stellar disks."1092 In contrast. the multiphase model supplies enough pressure support to the star-forming gas to prevent catastrophic Toomre instability. allowing stable galactic disks to form.," In contrast, the multiphase model supplies enough pressure support to the star-forming gas to prevent catastrophic Toomre instability, allowing stable galactic disks to form."1093 We have also verified that these conclusions do not depend on details of ournumerical integration., We have also verified that these conclusions do not depend on details of our numerical integration.1094 Our work differs in à number of respects from earlier numerical attempts to study disk galaxy formation., Our work differs in a number of respects from earlier numerical attempts to study disk galaxy formation.1095 We have used a novel formulation of smoothed particle hydrodynamics that. by construction. conserves both energy and entropy simultaneously. even when smoothing lengths vary in response to changes in density (Springel&Hernquist2002)..," We have used a novel formulation of smoothed particle hydrodynamics that, by construction, conserves both energy and entropy simultaneously, even when smoothing lengths vary in response to changes in density \citep{sh02a}."1096 While we consider it unlikely that previous numerical work on galaxy formation with SPH was strongly affected by inaccuracies in e.g. entropy conservation. it is clearly preterable to use a fully conservative scheme that mitigates against any numerical “overcooling.”," While we consider it unlikely that previous numerical work on galaxy formation with SPH was strongly affected by inaccuracies in e.g. entropy conservation, it is clearly preferable to use a fully conservative scheme that mitigates against any numerical “overcooling.”"1097 More important. we have formulated the consequences of feedback using a subresolution model for star formation in a multiphase ISM.," More important, we have formulated the consequences of feedback using a subresolution model for star formation in a multiphase ISM."1098 In this scheme. the local supernova feedback modifies the effective equation of state for star-forming gas. leading to an efficient self-regulation of star formation and a stabilizing effect on highly overdense gas.," In this scheme, the local supernova feedback modifies the effective equation of state for star-forming gas, leading to an efficient self-regulation of star formation and a stabilizing effect on highly overdense gas."1099 This aspect of our modeling feeds directly into the dynamics of forming disks by modifying the pressure gradient term in the hydrodynamie equations of motion., This aspect of our modeling feeds directly into the dynamics of forming disks by modifying the pressure gradient term in the hydrodynamic equations of motion.1100 Our formulation of the ISM physics has à number of advantages over earlier treatments of star formation and feedback., Our formulation of the ISM physics has a number of advantages over earlier treatments of star formation and feedback.1101 Other workers have implemented techniques to allow feedback to have a significant dynamical impact by. e.g. depositing thermal energy into gas surrounding star-forming regions and artificially delaying the radiative loss of this energy by the gas for an ad hoe time interval.," Other workers have implemented techniques to allow feedback to have a significant dynamical impact by, e.g. depositing thermal energy into gas surrounding star-forming regions and artificially delaying the radiative loss of this energy by the gas for an ad hoc time interval."1102 While similar in spirit to our method. these approaches make the comparison of results from different calculations challenging because the consequences of feedback are difficult to summarize in à concise. quantitative form.," While similar in spirit to our method, these approaches make the comparison of results from different calculations challenging because the consequences of feedback are difficult to summarize in a concise, quantitative form."1103 When expressed in terms of an effective equation of state. as done here. the dynamical implications of feedback become easier to interpret.," When expressed in terms of an effective equation of state, as done here, the dynamical implications of feedback become easier to interpret."1104 A discussion in terms of the effective equation of state is also relatively insensitive to the detailed physics responsible for pressurizing the gas. and enables one to relate the description of feedback to observations of the ISM.," A discussion in terms of the effective equation of state is also relatively insensitive to the detailed physics responsible for pressurizing the gas, and enables one to relate the description of feedback to observations of the ISM."1105" In order to demonstrate the sensitivity of galaxy formation to the physics of star-forming gas. we employed a simple multiphase model of the ISM (Springel&Hernquist2003a).. where the gas is imagined to be thermally unstable at sufficiently high densities and exists in two distinct phases. in pressure equilibrium,"," In order to demonstrate the sensitivity of galaxy formation to the physics of star-forming gas, we employed a simple multiphase model of the ISM \citep{sh03a}, where the gas is imagined to be thermally unstable at sufficiently high densities and exists in two distinct phases, in pressure equilibrium."1106 In detail. it is not plausible that this is a true picture of the ISM of disk galaxies. because this model ignores. e.g. turbulent motion. magnetic fields. and cosmic rays.," In detail, it is not plausible that this is a true picture of the ISM of disk galaxies, because this model ignores, e.g. turbulent motion, magnetic fields, and cosmic rays."1107 However. we believe that the basic points of our modeling do not depend on these complications.," However, we believe that the basic points of our modeling do not depend on these complications."1108 The evolution of the different models shown in 6 depends mainly on the effective equation of state for the star-forming gas on scales which characterize the Toomre instability., The evolution of the different models shown in \ref{sec:sim:halo:intro} depends mainly on the effective equation of state for the star-forming gas on scales which characterize the Toomre instability.1109 According to the Toomre criterion. a thin sheet of gas in differential rotation will be gravitationally unstable if where ος is the sound speed. # is the epicyclic frequency. and X 1s the surface density.," According to the Toomre criterion, a thin sheet of gas in differential rotation will be gravitationally unstable if where $c_s$ is the sound speed, $\kappa$ is the epicyclic frequency, and $\Sigma$ is the surface density."1110 In terms of an effective temperature of the gas. where 37As/(Skmsee!kpe) and X54=MCISM.pe-)τω are values characteristic of the Milky Way in the solar neighborhood. and 7;=T/(10?K).," In terms of an effective temperature of the gas, where $\kappa_{35} \equiv \kappa/ (35\,{\rm km \, sec^{-1}kpc^{-1}})$ and $\Sigma_{75} \equiv \Sigma / (75\,{\rm M}_\odot\,{\rm pc}^{-2})$ are values characteristic of the Milky Way in the solar neighborhood, and $T_5 \equiv T/(10^5\,{\rm K})$."1111" If the gas obeyed an isothermal equation of state with 7320.1. then Q~0.3. and the disk would be violently unstable. as for example in the case of our ""1so-ism"" simulations described in $6."," If the gas obeyed an isothermal equation of state with $T_5 = 0.1$, then $Q\sim 0.3$, and the disk would be violently unstable, as for example in the case of our “iso-ism” simulations described in 6."1112" With our equation of state. however. the effective temperature at densities characteristic of a forming galaxy would be 73~1 (seeFig.|ofSpringel&Hernquist2003a).. so Q>| and the disk would be stable. explaining why our ""multiphase-ism"" models in 6 produce realistic disks."," With our equation of state, however, the effective temperature at densities characteristic of a forming galaxy would be $T_5 \sim 1$ \citep[see Fig.~11113of][]{sh03a}, so $Q>1$ and the disk would be stable, explaining why our “multiphase-ism” models in 6 produce realistic disks."1114 Moreover. for thin disks of gas. the fastest growing unstable mode has a wavelength which ts a significant fraction of the disk size (e.g.Binney&Tremaine1987) and. therefore. the growth of the instability is sensitive only to the macroscopic equation of state for the ISM.," Moreover, for thin disks of gas, the fastest growing unstable mode has a wavelength which is a significant fraction of the disk size \citep[e.g.][]{bt87} and, therefore, the growth of the instability is sensitive only to the macroscopic equation of state for the ISM."1115 Whether or not the effective equation of state we have adopted (i.e. Figure 1)) for the ISM ts appropriate 1n detail is uncertain. but this question is largely independent of the exact properties of star-forming gas.," Whether or not the effective equation of state we have adopted (i.e. Figure \ref{fig:eos}) ) for the ISM is appropriate in detail is uncertain, but this question is largely independent of the exact properties of star-forming gas."1116 In particular. we expect our results to generalize to other models for the ISM with similar effective equations of state.," In particular, we expect our results to generalize to other models for the ISM with similar effective equations of state."1117 Having identified one exponential disk galaxy im our simulation. future work is needed to extend our analysis to other galaxies.," Having identified one exponential disk galaxy in our simulation, future work is needed to extend our analysis to other galaxies."1118 For example. a good statistical sample of simulated disks is needed before we can characterize the distribution of disk sizes and their morphological types. as well as the photometric and kinematic properties of the galaxies.," For example, a good statistical sample of simulated disks is needed before we can characterize the distribution of disk sizes and their morphological types, as well as the photometric and kinematic properties of the galaxies."1119 This work will require larger. simulations of substantial dynamic range in order to obtain adequate numbers of simulated disk galaxies. but we find the initial successes described here and recently by other authors highly encouraging.," This work will require larger simulations of substantial dynamic range in order to obtain adequate numbers of simulated disk galaxies, but we find the initial successes described here and recently by other authors highly encouraging."1120 A comparison with the wealth of observational data on the structural parameters of galaxies in the local universe will then show whether hydrodynamical simulations are finally producing realistic disk galaxies., A comparison with the wealth of observational data on the structural parameters of galaxies in the local universe will then show whether hydrodynamical simulations are finally producing realistic disk galaxies.1121" This work was supported in part by NSF grants ACI 96-19019. AST 98-02568. AST 99-00877. AST 00-71019. and AST-0206299, and NASA ATP grants NAGS-12140. NAGS-13292. and NAGS-13381."," This work was supported in part by NSF grants ACI 96-19019, AST 98-02568, AST 99-00877, AST 00-71019, and AST-0206299, and NASA ATP grants NAG5-12140, NAG5-13292, and NAG5-13381."1122 NY acknowledges support from JSPS Special Research Fellowship (02674)., NY acknowledges support from JSPS Special Research Fellowship (02674).1123 The simulations were performed at the Center for Parallel Astrophysical Computing at the Harvard-Smithsonian Center for Astrophysics., The simulations were performed at the Center for Parallel Astrophysical Computing at the Harvard-Smithsonian Center for Astrophysics.1124High-resolution spectroscopy of far-infrared and sub-millimetre water lines is an important tool for studying the physical and chemical properties of the interstellar medium (ISM).,High-resolution spectroscopy of far-infrared and sub-millimetre water lines is an important tool for studying the physical and chemical properties of the interstellar medium (ISM).1125 Absorption by terrestrial atmospheric water vapour has ground-based studies of water in extragalactic systems limited to radio maser transitions (such as the famous GGHz water line) or to a few systems with significant redshift (e.g. Combes Wiklind 1997.. Cernicharo citepepeQ6.. Menten citementen0S8)).," Absorption by terrestrial atmospheric water vapour has ground-based studies of water in extragalactic systems limited to radio maser transitions (such as the famous GHz water line) or to a few systems with significant redshift (e.g. Combes Wiklind \cite{combes97}, Cernicharo \\cite{pepe06}, Menten \\cite{menten08}) )."1126. Earlier. satellite missions. such as ODIN and SWAS. did not have enough collecting area to detect the relatively faint ground transitions of water in. external galaxies.," Earlier satellite missions, such as ODIN and SWAS, did not have enough collecting area to detect the relatively faint ground transitions of water in external galaxies."1127 ISO and. more recently.Spitzer have provided the first systematic studies of water in the far infrared (IR) regime (e.g. Fischer citefischer99... Gonzalez-Alfonso 2004))," ISO and, more recently, have provided the first systematic studies of water in the far infrared (IR) regime (e.g. Fischer \\cite{fischer99}, Gonzalez-Alfonso \cite{alfonso04}) )."1128 These missions. however. did not cover the frequencies of the water ground transitions and other low-level water lines.," These missions, however, did not cover the frequencies of the water ground transitions and other low-level water lines."1129 Only with the launch of the satellite. with its large collecting area. have these transitions become accessible in the nearby universe (e.g. van der Werf citepaullO)).," Only with the launch of the satellite, with its large collecting area, have these transitions become accessible in the nearby universe (e.g. van der Werf \\cite{paul10}) )."1130 As part of the HEXGAL guaranteed time key program Güssten). we are surveying the low-level water lines in different nuclear environments. performing velocity-resolved spectroscopy with the Heterodyne Instrument for the Far Infrared (HIFI. de Graauw citedeGraauw )).," As part of the HEXGAL guaranteed time key program Güssten), we are surveying the low-level water lines in different nuclear environments, performing velocity-resolved spectroscopy with the Heterodyne Instrument for the Far Infrared (HIFI, de Graauw \\cite{deGraauw}) )."1131 In this letter we report on our first observations towards the central region of the archetypal starburst galaxy M$2., In this letter we report on our first observations towards the central region of the archetypal starburst galaxy M82.1132 We adopt a distance of MMpe (Sakai Madore 1999))., We adopt a distance of Mpc (Sakai Madore \cite{sakai99}) ).1133" Using the HIFI instrument onboard Herschel. we have observed the ground transitions of ortho and para water. aand-Ooo).. as well as the line. towards the centre of M82 (RA=09'55""52*22 Dec=69°40'46"".9 J2000)."," Using the HIFI instrument onboard Herschel, we have observed the ground transitions of ortho and para water, and, as well as the line, towards the centre of M82 $09^h55^m52^s.22$ $69^\circ40'46''.9$ J2000)."1134 Observations were carried out in fast-chopping dual-beamswitch mode using a wobbler throw of 3’ for all observations., Observations were carried out in fast-chopping dual-beamswitch mode using a wobbler throw of $3'$ for all observations.1135 The wobbler frequency was 0.8. L4. and 2.0 Hz for the 557GGHz. GGHz. and I113GGHz observations. respectively.," The wobbler frequency was 0.8, 1.4, and 2.0 Hz for the GHz, GHz, and GHz observations, respectively."1136 Nodding was performed every ~40 seconds., Nodding was performed every $\sim40$ seconds.1137 Calibration was achieved through hot/cold absorber measurements every 20 minutes., Calibration was achieved through hot/cold absorber measurements every 20 minutes.1138 The data were recorded using the wide-band acousto-optical spectrometer. consisting of four units with a bandwidth of GGHz each. covering the GGHz IF for each polarization with spectral resolution of MMHz.," The data were recorded using the wide-band acousto-optical spectrometer, consisting of four units with a bandwidth of GHz each, covering the GHz IF for each polarization with spectral resolution of MHz."1139" The recent work of EKováccs Jurcsik (Ilxováces Juresik 1996 [KJOG|. 1997 ντ, .Juresik 1998 [JO5]) proposes an almost purely empirical iuiethod of extracting the absolute maenitudes. colors aud effective teniperatures )) aud incetallicities directly from the observed periods aud lieht curves of RR Lyrac stars."," The recent work of Kováccs Jurcsik (Kováccs Jurcsik 1996 [KJ96], 1997 [KJ97], Jurcsik 1998 [J98]) proposes an almost purely empirical method of extracting the absolute magnitudes, colors and effective temperatures ) and metallicities directly from the observed periods and light curves of RR Lyrae stars."1140 The ouly theoretical input appears iu the transformation from My- and V. Iv to L aud vvia [Nurucz* static model atmospheres., The only theoretical input appears in the transformation from $_V$ and V – K to L and via Kurucz's static model atmospheres.1141 The very enipirical nature aud potential usefulness of the approach has attracted a ereat deal of attention from observers., The very empirical nature and potential usefulness of the approach has attracted a great deal of attention from observers.1142 The most recent work (J98) compiles and analyzes a sizeable set of observational data of RB. Lyrac cluster variables and RR Lyrac field) stars., The most recent work (J98) compiles and analyzes a sizeable set of observational data of RR Lyrae cluster variables and RR Lyrae field stars.1143 This study thus includes RR Lyrac stars with mctallicitics raugimg frou Z-—Uü.00001 to almost solar Z=0.020., This study thus includes RR Lyrae stars with metallicities ranging from Z=0.00001 to almost solar Z=0.020.1144 The eud products of her analysis that we are most concerned about here are aadlL pplots. andL the byproductsP which are relations between DIuniuositv L. mass M and metallicity Z. The J98 aad ddata are LshownP in Figure 1 as small cdots. circles for RRab aud triangles for RRe stars.," The end products of her analysis that we are most concerned about here are and plots, and the byproducts which are relations between luminosity L, mass M and metallicity Z. The J98 and data are shown in Figure 1 as small dots, circles for RRab and triangles for RRc stars."1145 The fieures indicate well defined slopes for the fuudinueutal aud overtone blue edees aud red edges. all four of which have approximately the same values. although there nay be a slight broadening of the instability strip with Iuninositv.," The figures indicate well defined slopes for the fundamental and overtone blue edges and red edges, all four of which have approximately the same values, although there may be a slight broadening of the instability strip with luminosity."1146 IIereafter we call this slope = (=A L/ A Tig)., Hereafter we call this slope $\Xi$ $=\Delta$ / $\Delta$ ).1147 JOS poiuted out that no model calculations can explain the strong depeucdence of the temperature on the huuinosity., J98 pointed out that no model calculations can explain the strong dependence of the temperature on the luminosity.1148 Tn 822 we first confini that indeed the slope = of this enipirical ddiagraiài is in nrecoucilable disagreement with that of radiative models., In 2 we first confirm that indeed the slope $\Xi$ of this empirical diagram is in irreconcilable disagreement with that of radiative models.1149 Next. in 832 we show that the inclusion of turbulent couvection in the mocels ouly shifts the lue. but does uot change the slope.," Next, in 3 we show that the inclusion of turbulent convection in the models only shifts the line, but does not change the slope."1150 The discrepancy is therefore very basic because it is already iuherent in uodels that are as fundamental as purcly radiative ones., The discrepancy is therefore very basic because it is already inherent in models that are as fundamental as purely radiative ones.1151 Tn 811 we reexanune the uncertainties of both radiative and turbulent model calculations aud conclude that the heorctical Z is very robust aud cannot be changed very uuch without introducing new plysics., In 4 we reexamine the uncertainties of both radiative and turbulent model calculations and conclude that the theoretical $\Xi$ is very robust and cannot be changed very much without introducing new physics.1152 In 855 we ciscuss he theory that goes into the empirical relations of Jurcsik. in particular the iufiueuce of the use of the static I&urucz color temperature transformation.," In 5 we discuss the theory that goes into the empirical relations of Jurcsik, in particular the influence of the use of the static Kurucz color – temperature transformation."1153 We conclude in 866., We conclude in 6.1154 Iu the following we examine how well RR Lyrae models aeree with the J9S data., In the following we examine how well RR Lyrae models agree with the J98 data.1155 We first exinuue radiative models. aud thus linut ourselves to the vicinity of the blue edges.," We first examine radiative models, and thus limit ourselves to the vicinity of the blue edges."1156 Figure 1l shows a pplot on the left., Figure 1 shows a plot on the left.1157 On these we superpose the luear fundamental aud first overtone blue edees of radiative RR Lyrae models., On these we superpose the linear fundamental and first overtone blue edges of radiative RR Lyrae models.1158 We recall here that amone the linear οσος of the instability strip ouly the overtone liuear blue edee and the fundamental linear red edge coincide with the observable edees of the instability strip., We recall here that among the linear edges of the instability strip only the overtone linear blue edge and the fundamental linear red edge coincide with the observable edges of the instability strip.1159" Nonlinear dvuamical effects shift tle observable fundamental blue edee aud the overtone red edge to the red aud to the blue. respectively. compared to their corresponding linear οσος, bv up to several LOOT Buchler 2000)."," Nonlinear dynamical effects shift the observable fundamental blue edge and the overtone red edge to the red and to the blue, respectively, compared to their corresponding linear edges, by up to several K Buchler 2000)."1160 Strictly speaking. for radiative models. it is only the liuear overtone blue edge that is relevant. but the lear fundamental blue edge can eive a rough indication of the slope =.," Strictly speaking, for radiative models, it is only the linear overtone blue edge that is relevant, but the linear fundamental blue edge can give a rough indication of the slope $\Xi$."1161 Finthermore. the red edges are determined by convection. and radiative red edges are not relevant aud are therefore uot shown in Fie.," Furthermore, the red edges are determined by convection, and radiative red edges are not relevant and are therefore not shown in Fig."1162 1., 1.1163In this work. the DA white dwarf evolutionary. tracks used to derive the spectroscopic masses of the DAVs have been emplovecl to infer the asteroseismological masses.,"In this work, the DA white dwarf evolutionary tracks used to derive the spectroscopic masses of the DAVs have been employed to infer the asteroseismological masses."1164 Thus. a comparison between both sets of values is worth doing.," Thus, a comparison between both sets of values is worth doing."1165 We compare in Fig., We compare in Fig.1166 9 the spectroscopic and asteroseismological masses., \ref{massmass} the spectroscopic and asteroseismological masses.1167 The dotted line is the 1:1 correspondence., The dotted line is the 1:1 correspondence.1168 The plot, The plot1169for NGC 2976) to cuspy (a=1.20 for NGC 5963).,for NGC 2976) to cuspy $\alpha=1.20$ for NGC 5963).1170 Note that NGC 5963 has an inner bright disk. and it may therefore not be dark-matter-dominated all the way to the center. making the value of its mass-density slope uncertain (see also ?)).," Note that NGC 5963 has an inner bright disk, and it may therefore not be dark-matter-dominated all the way to the center, making the value of its mass-density slope uncertain (see also \citealt{Bosma:1988p1467}) )."1171 The average slope they derive is a=0.73+0.44., The average slope they derive is $\alpha = 0.73 \pm 0.44$.1172 Their analysis method differs in a few aspects from the other studies referenced in this paper., Their analysis method differs in a few aspects from the other studies referenced in this paper.1173 Firstly. their inner slope values are derived. from a single power-law fit to the entire rotation curve.," Firstly, their inner slope values are derived from a single power-law fit to the entire rotation curve."1174 Most of their models do not take into account the gas component due to a lack of HI observations., Most of their models do not take into account the gas component due to a lack of HI observations.1175 In. low-mass galaxies the gas can dynamically be more important than the stars (especially in the outer parts). and correcting for this component could potentially change the derived slopes.," In low-mass galaxies the gas can dynamically be more important than the stars (especially in the outer parts), and correcting for this component could potentially change the derived slopes."1176 More importantly. the mass models in ?. are derived under the explicit assumption of a constant inclination and position angle for each galaxy.," More importantly, the mass models in \citet{Simon:2005p68} are derived under the explicit assumption of a constant inclination and position angle for each galaxy."1177 Most velocity fields of nearby galaxies show radial inclination and position angle trends. especially in the outer parts (e.g. ?))," Most velocity fields of nearby galaxies show radial inclination and position angle trends, especially in the outer parts (e.g., \citealt{deBlok:2008p155}) )."1178 ?— attribute these to radial velocities that give the Impression of changes in inclination and position angle.," \citet{Simon:2005p68}1179 attribute these to radial velocities that give the impression of changes in inclination and position angle."1180 Nevertheless. the non-circular velocities derived in this way are typically less than ~20 km s7!. and in a few galaxies less than ~5 km s!," Nevertheless, the non-circular velocities derived in this way are typically less than $\sim 20$ km $^{-1}$ , and in a few galaxies less than $\sim 5$ km $^{-1}$."1181 Whether these velocities are real or whether inclination and position angle changes are preferred would be an interesting topic of further study., Whether these velocities are real or whether inclination and position angle changes are preferred would be an interesting topic of further study.1182 ? also derive harmonic decompositions of the velocity fields to study non-circular motions. but they do not list the values of the harmonie coefficients.," \citet{Simon:2005p68} also derive harmonic decompositions of the velocity fields to study non-circular motions, but they do not list the values of the harmonic coefficients."1183 It would be similarly interesting to compare these with results derived for other disk and LSB galaxies., It would be similarly interesting to compare these with results derived for other disk and LSB galaxies.1184 Direct measurements of non-circular motions are presented in 2.., Direct measurements of non-circular motions are presented in \citet{Gentile:2005p89}.1185 Using high-resolution HI observations. they make a harmonic decomposition of the velocity field. and show that non-circular motions are only present at the level of a few km !.," Using high-resolution HI observations, they make a harmonic decomposition of the velocity field, and show that non-circular motions are only present at the level of a few km $^{-1}$."1186 This is a factor of ~10 lower than is needed for the non- motions to wipe out the kinematical signature of a cusp and to give the impression of a core., This is a factor of $\sim 10$ lower than is needed for the non-circular motions to wipe out the kinematical signature of a cusp and to give the impression of a core.1187 An analysis by ? of another gas-rich dwarf galaxy. NGC 3741. based on the entire 3D HI data cube. showed circular motions around 5-10 km s. and a strong preference for a core model.," An analysis by \citet{Gentile:2007p90} of another gas-rich dwarf galaxy, NGC 3741, based on the entire 3D HI data cube, showed non-circular motions around $-$ 10 km $^{-1}$, and a strong preference for a core model."1188 NFW models could be accommodated. but with the usual caveat of the fit parameters not being consistent with the cosmological (c.Vigo) relation.," NFW models could be accommodated, but with the usual caveat of the fit parameters not being consistent with the cosmological $(c, V_{200})$ relation."1189 ? present harmonic decompositions of the velocity fields of galaxies from The HI Nearby Galaxy Survey (THINGS; ?))., \citet{Trachternach:2008p152} present harmonic decompositions of the velocity fields of galaxies from The HI Nearby Galaxy Survey (THINGS; \citealt{Walter:2008p154}) ).1190 The THINGS survey covers a large range in galaxy properties. from luminous early-type disk galaxies to late-type dwarfs.," The THINGS survey covers a large range in galaxy properties, from luminous early-type disk galaxies to late-type dwarfs."1191 ? find a relation between the median strength of the non- motions and the luminosity of the galaxies. indicating that the non-cireular motions are associated with the baryons.," \citet{Trachternach:2008p152} find a relation between the median strength of the non-circular motions and the luminosity of the galaxies, indicating that the non-circular motions are associated with the baryons."1192 Luminous disk galaxies have non-circular motions up to ~30 km s!. mostly associated with bars and spiral arms.," Luminous disk galaxies have non-circular motions up to $\sim 30$ km $^{-1}$, mostly associated with bars and spiral arms."1193 These then decrease rapidly to a level of only a few km s! for dwarf galaxies like DDO 154., These then decrease rapidly to a level of only a few km $^{-1}$ for dwarf galaxies like DDO 154.1194 They also found that offsets between photometric and kinematic centers were typically ~200 pe or less., They also found that offsets between photometric and kinematic centers were typically $\sim 200$ pc or less.1195 This low level of non-circular motions seems inconsistent with observations by 2.. who find significant. non-circular motions in à sample of 4 LSB galaxies.," This low level of non-circular motions seems inconsistent with observations by \citet{Pizzella:2008p35}, who find significant non-circular motions in a sample of 4 LSB galaxies."1196 However. their galaxies all contain bright bulges and are therefore probably more representative of the class of giant LSB galaxies. rather than the late-type ones discussed in this paper (in this regard see also 2)).," However, their galaxies all contain bright bulges and are therefore probably more representative of the class of giant LSB galaxies, rather than the late-type ones discussed in this paper (in this regard see also \citealt{Coccato:2008p62}) )."1197 Finally. and?) show that the core/cusp issue is not limited to the2? very inner parts of galaxies.," Finally, \citet{McGaugh:2007p64} and \citet{Gentile:2007p83} show that the core/cusp issue is not limited to the very inner parts of galaxies."1198 As mentioned before. the shape of the NFW curve is fundamentally different from that of observed curves.," As mentioned before, the shape of the NFW curve is fundamentally different from that of observed curves."1199 One can try and work out the implications in two ways: ?. match the observed outer rotation velocities with those of corresponding cosmological NFW halos (effectively identifying NFW halos that have a similar dark matter density at these outer radii as real galaxies). and find that these halos are too massive.," One can try and work out the implications in two ways: \citet{McGaugh:2007p64} match the observed outer rotation velocities with those of corresponding cosmological NFW halos (effectively identifying NFW halos that have a similar dark matter density at these outer radii as real galaxies), and find that these halos are too massive."1200 Due to the cusp mass excess their average density 1s then also a factor of 2-3 too high (see also ?))., Due to the cusp mass excess their average density is then also a factor of 2–3 too high (see also \citealt{deBlok:2008p155}) ).

Showing the first 1,200 of 10362 lines. Download the file for the rest.