Ground snow load case study: what it is, and when you need one

A ground snow load case study is a site-specific estimate of how much snow a location should be designed for. The ground snow load is the weight of snow on flat, open ground; roof loads are calculated from it. Instead of reading that number from a national map, a case study works it out for one site from the snow records of nearby weather stations and the site’s own elevation.

ASCE 7, the standard US building codes use for structural loads, has long spelled out what the study must rest on. In ASCE 7-16 the wording is: an extreme-value statistical analysis of data available in the vicinity of the site, using the value with a 2% annual probability of being exceeded (the 50-year ground snow load), approved by the authority having jurisdiction.

A national map works where snow changes gradually across the landscape. In mountains it doesn’t: two sites a few miles apart can differ by hundreds of feet of elevation, and their snow loads by far more than the map could show. ASCE 7 describes the CS areas as places where “extreme local variations” in snow loads make mapping at a national scale impossible.

Through ASCE 7-16, those areas were large: much of the Rocky Mountain West, the mountains of California and the Northwest, and parts of New England. For any site inside one, there is no mapped value to fall back on; a case study is the prescribed path.

ASCE 7-22 replaced the old map with reliability-targeted values served from an online geodatabase (the ASCE Hazard Tool). Built from about thirty more years of snow data, it covers far more locations, and the case-study regions shrank by more than 90%. The ones that remain are essentially sites higher than any nearby measurement location, where there is nothing to interpolate from.

Case studies did not disappear. Where the geodatabase indicates one is needed, ASCE 7-22 still requires the value to be based on an analysis of data near the site and approved by the authority having jurisdiction. And jurisdictions still enforcing the 2021 or 2018 IBC reference ASCE 7-16, where the old CS areas remain in force. For the full picture, see ASCE 7-22 ground snow loads: what changed.

One caveat that matters. A case study under ASCE 7-22 must deliver values on that edition’s strength-level basis. A 50-year value is the case-study answer under ASCE 7-16 and earlier; under 7-22 it is background reference, not a substitute for the mapped value.

• Your site is in a CS area on the map your jurisdiction uses, or the Hazard Tool says a case study is required for the location.
• Your building official asks for one. Local officials in mountain areas often have the authority to require or accept site-specific studies. California building officials, for example, published guidance in 2025 confirming that case studies remain available under the 2024 California Building Code.
• The mapped value looks wrong for your site, typically because the site sits well above or below the stations the map was built from. A documented study is how you make that case to the official, in either direction.

Check local rules first. Some counties and towns publish their own snow load tables or minimums, and some states require a state-specific study instead of the national method. Where a local rule exists, it decides what a case study can and can’t change.

The federal reference method is ERDC/CRREL SR-20-1, “Site-Specific Case Studies for Determining Ground Snow Loads in the United States” (2020), from the Army Corps of Engineers’ cold regions laboratory. It comes with a database of 15,104 stations, from about 200 National Weather Service stations that weigh the snowpack to thousands of cooperative stations that measure only its depth. The procedure, in plain steps:

1. Pin the site: latitude, longitude, and elevation.
2. Gather the stations around it, each with its elevation and its own 50-year ground snow load.
3. Keep the comparable ones: close enough to the site, with a long enough record, and within a sensible elevation range.
4. Relate snow load to elevation. In mountains, load usually rises with elevation; plotting the stations shows how fast, and a fitted line captures it.
5. Read the value at the site’s elevation, and check it against the nearest stations.
6. Document every choice: which stations were kept, which were dropped and why, and the fit the answer rests on.

SR-20-1 walks through ten worked examples and sixteen practice sites. The steps are simple; the judgment is in step three, and that is why the documentation in step six matters.

The official approves the value, so the study has to let someone else retrace it. In practice that means:

• The site: exact coordinates and elevation.
• The data: every station considered, with its distance, elevation, and record length.
• The selection: which stations were used, and the reason each excluded one was left out.
• The method, named, and the code edition it answers to (a 50-year value for ASCE 7-16 and earlier).
• The result and its basis: the value, the elevation relationship behind it, and the plots.
• Who stands behind it: the project professional who applies the number to the design, not the software that produced it.

Nivometra runs the SR-20-1 method as software. Pin a site anywhere in the lower 48 states or Alaska, see the nearby stations and which ones the filters keep, adjust the selection yourself, and preview the result for free. When the analysis is right, a documented report costs $29: the estimated 50-year ground snow load with the site inputs, every station, and the elevation fits it rests on.

Reports can’t be purchased in Colorado, Idaho, Montana, New Hampshire, Oregon, and Washington: their building codes set ground snow loads from their own state sources, so a report from us would give a number your building department is unlikely to accept. The report is a 50-year reference value, not an ASCE 7-22 case study, not an engineering service, and not a guarantee that an official will accept it.