Datasetpublic

nakas/nbm-conus-analysis

NOAA NBM CONUS Daily Analysis (Zarr) Daily best-estimate analysis derived from NOAA NBM (National Blend of Models) CONUS forecasts, on the native ~2.5 km Lambert conformal grid (2345 x 1597). Built by nbm-to-zarr, dynamical.org-style. Variables: tmean / tmax / tmin (degC), precip (mm), srad (MJ/m2/day) Construction: best estimate for day D = lead-day 1 of that day's 00z NBM init Coverage: rolling backfill from 2020-10-01 (AWS NBM archive floor) to present Layout: one standalone… See the full description on the dataset page: https://huggingface.co/datasets/nakas/nbm-conus-analysis.

sourceHugging Facecc-by-4.0updated 7h agoView on Hugging Face

0likes58kdownloads

Dataset Card

NOAA NBM CONUS Daily Analysis (Zarr)

Daily best-estimate analysis derived from NOAA NBM (National Blend of Models) CONUS forecasts, on the native ~2.5 km Lambert conformal grid (2345 x 1597). Built by nbm-to-zarr, dynamical.org-style.

—Variables: tmean / tmax / tmin (degC), precip (mm), srad (MJ/m2/day)
—Construction: best estimate for day D = lead-day 1 of that day's 00z NBM init
—Coverage: rolling backfill from 2020-10-01 (AWS NBM archive floor) to present
—Layout: one standalone Zarr per ISO week under YYYY-Www/

Usage

python

import xarray as xr
ds = xr.open_zarr("https://huggingface.co/datasets/nakas/nbm-conus-analysis/resolve/main/2026-W23")
ds.tmean.isel(time=0).plot()

Concatenate weeks along time to span longer periods.

Source: NOAA NBM via the AWS Open Data Program (public domain). Derived data CC-BY-4.0.