easysnowdata

A Python package to easily retrieve data relevant to snow science.

easysnowdata unifies access to a wide range of snow-relevant geospatial datasets — weather stations, satellite imagery, snow products, climate reanalysis, DEMs and a hillshade, basins, and boundaries from countries and counties to mountain ranges and glacier outlines — under a consistent API that returns xarray objects (and GeoDataFrames for vector products). The emphasis is on minimising downloads and local computation by leveraging cloud-optimised data formats wherever possible.

Gallery

One executed script per product — browse them, each with the code, the figure, and a downloadable notebook.

Data Source Status

Every route of every product is probed weekly. A failure opens an issue labelled data-source and a recovery closes it; latency and DMR++ readiness are on the status page.

What a probe cannot see — a reprocessed collection, a republished file, a new release (the next Census year, a new RGI version) — is caught by a weekly upstream watch that opens one digest issue labelled upstream-watch and closes the previous one, carrying its unticked to-dos forward. The same digest lists datasets nobody has told us about yet: new snow, ice and frozen-ground collections in NASA's CMR, and new collections on Planetary Computer, Earth Search and both Earth Engine catalogs, under "Worth adding" when they are about the cryosphere.

Last updated: 2026-10-26 08:22 UTC
⚠️ = skipped (credentials not available in this run). Latency and virtualization probes are on the status page.

Stations (esd.stations)

Snow (esd.snow)

SAR (esd.sar)

Optical imagery (esd.optical)

Terrain (esd.terrain)

Land cover (esd.land)

Hydrography (esd.hydro)

Boundaries (esd.boundaries)

Climate (esd.climate)

Installation

pip install easysnowdata

conda install -c conda-forge easysnowdata

mamba install -c conda-forge easysnowdata

Development install (with pixi)

git clone https://github.com/egagli/easysnowdata.git
cd easysnowdata
pixi install --all                    # every environment (plain `pixi install` sets up only the default one)
pixi run -e test-py313 test-unit      # offline tests (no network, no credentials)
pixi run -e test-py313 test-live      # live tests against the data providers (credentialed ones skip without secrets)
pixi run -e docs docs-fast            # build the site without running the gallery
pixi run -e docs docs-serve           # serve the built site at http://localhost:8000

Services that require account setup

Some data sources need free accounts and credentials passed as environment variables:

35 of the 41 products have at least one route that needs no account; each product's catalog page says which.

What is in it

42 products across 9 themes, each with one or more access routes:

Every product's routes, resolution, credentials, licence, and health are on its own page: https://egagli.github.io/easysnowdata/catalog/.

Quick Start

import easysnowdata as esd

aoi = (-121.94, 46.72, -121.54, 46.99)          # Mount Rainier; any AOI form works

# Snow stations: which are here, then one water year of observations
inv_gdf = esd.stations.inventory(aoi, daily_only=True)
obs_ds = esd.stations.load(inv_gdf, variables=["swe", "snwd"], time="2023-10/2024-09")

# Terrain, SAR, and snow water equivalent — lazy, Dask-backed, CRS attached
dem_da = esd.terrain.dem.load(aoi)                          # Copernicus GLO-30 (default)
dem_da = esd.terrain.dem.load(aoi, product="3dep")          # or NASADEM, SRTM, 3DEP, ALOS
dem_da = esd.terrain.dem.load(aoi, product="gedtm30")       # global bare-earth DTM (OpenTopography)
s1_ds = esd.sar.sentinel1.load(aoi, "2024-03", units="dB")  # Sentinel-1 RTC
snodas_ds = esd.snow.snodas.load(aoi, "2024-03")            # SNODAS, no account

# Boundaries as GeoDataFrames: states, counties, mountain ranges, glaciers
wa_gdf = esd.boundaries.admin.states(aoi)                   # Washington (US Census)
ranges_gdf = esd.boundaries.mountains.load(aoi)             # GMBA Mountain Inventory v2
glaciers_gdf = esd.boundaries.glaciers.load(aoi)            # RGI 7.0 (Earthdata username + password); RGI 6.0 from source="oggm-mirror" needs none

# Optical, masked, and a snow index written out rather than hidden in a helper
s2_ds = esd.optical.sentinel2.load(aoi, "2024-03", mask="scl-default")
ndsi_da = (s2_ds["green"] - s2_ds["swir16"]) / (s2_ds["green"] + s2_ds["swir16"])

# Maps with equal aspect, a scale bar, and a graticule; legends from CF flags
ax = esd.plotting.map(dem_da, cmap="terrain")
esd.plotting.add_outline(ax, glaciers_gdf)                  # vectors in the map's CRS
esd.plotting.categorical(esd.land.landcover.load(aoi))
esd.plotting.timeseries(obs_ds["swe"])                      # calendar dates, units in [ ]

# What is available, and what it needs
esd.catalog.search("swe")
esd.catalog.describe("snodas")
esd.auth.status()

Coming from 0.0.x? The old module API (remote_sensing.get_*, automatic_weather_stations.StationCollection, …) was deprecated in 0.2 and removed in 0.3; the migration guide maps every old name to its replacement.

Documentation

Executed example gallery, data catalog, and API reference: https://egagli.github.io/easysnowdata

Contributing

Contributions welcome! See CONTRIBUTING for guidelines.

Citing

If you use easysnowdata in your research, please cite the Zenodo archive: