Data Information
User Access
Provider Information
Related Links
  MADIS Snow Dataset

Snow Dataset



Reports of snow depth, new snowfall, and the snow water equivalent of these observations are available from automated networks, and from manual reports taken by volunteer observers. The MADIS snow dataset includes reports from many observing networks run by different "providers". The Hydrometeorological Automated Data System (HADS)/Automated Flood Warning System (AFWS) were integrated into operational MADIS and went live at the end of 2016. MesoWest was providing some HADS data before this integration. HADS/AFWS is broken up into three different areas of MADIS: Atmospheric, Hydrological and Snow. The atmospheric data is a part of the Integrated Mesonet and includes temperature, winds (direction, speed and gust), pressure, dewpoint, relative humidity, and accumulated precip. The hydrological data is a part of the Hydrological Surface dataset and includes river stage and flow. Lastly the snow data is a part of the Snow dataset and includes snow depth and snow water equivalent. The HADS sites can change per owner operational needs.

MADIS started ingesting SNOw TELemetry (SNOTEL) data directly from Natural Resources Conservation Service (NRCS) at the end of 2016. MesoWest was providing some SNOTEL data before this integration. MADIS is now the pathway for providing SNOTEL observations to NOAA. SNOTEL provides a reliable, cost-effective way to collect snowpack and other meteorological data needed to produce water supply forecasts.

MADIS has been collecting snow data since October 2005.

Geographic Coverage

The geographic coverage of the automated data is the United States. Some of the manual reports taken by CWOP and GLOBE volunteers are global.

For HADS: Alaska, Hawaii, Mexico, Canada, Puerto Rico, and the CONUS, please click here for a coverage map and site information by location identifiers provided by the HADS website.

For SNOTEL: Alaska and Western United States

Data Schedule

Data arrive on a continuous, asynchronous schedule, and newly arriving data are processed every 5 minutes. The data are segmented into hourly files, with the file for hour HH containing data for HH00 through HH59. Many of the manually-taken reports are reported hours, days, even weeks late, so you should set up your ingest schedule based on whether you only need data in near real-time, or on whether you can accept reports that have been made long after the fact.


Typical daily volume for all MADIS datasets can be seen here.


No restrictions. All observations are publicly accessible.

Variable Details

Related Links

Last updated 29 August 2018