Sampled — shipped under file-level sampling, not individually fact-checkedcommunity attestations: 0✓ / 0✗
Steps
Use NSIDC's SNODAS dataset landing page (nsidc.org/data/g02158) as the canonical access point — NOHRSC generates the data, NSIDC distributes the archive.
Access files via HTTPS or FTP, or script pulls with wget/curl against the dated directory structure (daily data from Oct 1, 2003 to present).
Choose masked (CONUS, post-2013 boundary) vs. unmasked grid files depending on whether you need the full modeling domain or just the continental snow mask.
Parse the flat binary 16-bit signed integer, big-endian format using the accompanying header/metadata files, or use a wrapper like the R `nohrsc` package for named-argument access.
For point/station-level series (14,000+ reporting stations) and narrative regional discussions, use NOHRSC's own interactive site (nohrsc.noaa.gov) rather than the NSIDC gridded archive.
Cross-check SWE/depth/snowmelt fields against your operating region before triggering salt/plow dispatch thresholds — SNODAS is modeled/assimilated, not a direct sensor feed.
Known gotchas
SNODAS is a NOAA/NWS product distributed through NSIDC — treat 'NOHRSC' and 'NSIDC' as complementary access points, not competing ones.
Binary grid format requires a header file to decode; there is no JSON/REST wrapper from NOAA itself for the gridded product.
Resolution is daily (24-hour) — don't expect sub-daily updates from SNODAS itself; use RWIS/MADIS or NWS products for hourly needs.
Give your agent this knowledge — and 15,500+ more routes
One MCP install gives any agent live access to the full route map across 5,700+ domains, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp
Need this verified for your stack — or a route we don't have yet?