ASCE 7-22 ground snow loads: what changed, and where site-specific studies still fit

From its early editions through ASCE 7-16, the standard published a national ground snow load map of uniform-hazard values: the load with a 2% annual probability of being exceeded (a 50-year mean recurrence interval), used as a service-level load with a 1.6 factor in strength design. The map was built largely from CRREL’s station analysis, with data ending in the early 1990s.

In mountainous terrain the map gave up: large regions — much of the Rocky Mountain West, plus parts of New England and the Northwest — were shaded as case-study (CS) zones, where the designer was directed to establish the value from a site-specific study. Several states responded by publishing their own studies, which ASCE 7-16 incorporated as state tables; values could jump discontinuously at state lines.

ASCE 7-22 replaced that map with reliability-targeted ground snow loads — the first loads in ASCE 7 to target uniform reliability rather than uniform hazard. The practical differences:

• Strength-level values. The mapped loads are used with a load factor of 1.0, not 1.6. They are calibrated directly to the safety targets in Chapter 1 of the standard.
• No more snow importance factor. Separate values are published per risk category instead.
• A geodatabase, not a figure. The values come from the ASCE Hazard Tool’s ground snow load geodatabase — the printed figures are explicitly non-design illustrations.
• A new underlying study. The values were developed by Bean, Maguire and colleagues from roughly thirty additional years of snow data and updated depth-to-load conversion, with spatial mapping that removes the old discontinuities at state boundaries.
• Case-study regions cut by more than 90%. The remaining CS zones are essentially limited to sites higher in elevation than any nearby measurement location.

The 2024 IBC adopts ASCE 7-22, so these loads arrive wherever the 2024 code is adopted. Jurisdictions on the 2021 or 2018 IBC continue to reference ASCE 7-16 and the old map.

A reasonable first read of 7-22 is that site-specific work is over. In practice there are four situations where it remains exactly what the code expects:

• Projects under older editions. Many jurisdictions still enforce codes referencing ASCE 7-16 or earlier, where CS zones are live and a case study is the prescribed path.
• The commentary’s standing rule. Every edition of ASCE 7 has said that a detailed study of a site may produce a value lower or higher than the map, and that the study’s result should be followed. That language survives in 7-22: a defensible site-specific value, accepted by the authority having jurisdiction, can supersede the mapped one.
• Jurisdictions that say so explicitly. California building officials, for example, published guidance confirming the continued availability of site-specific case studies under 2024 CBC §1608.2.
• Terrain the grid can’t resolve. Gridded values are interpolations. In steep terrain, where snow loads can change dramatically over short distances, a station-based study of the specific site remains the honest answer — the same reason the remaining CS zones are elevation-driven.

One honest caveat. A case study submitted under ASCE 7-22 must deliver values on the standard’s strength-level, reliability-targeted basis. A 50-year uniform-hazard estimate, the kind SR-20-1 produces, is the prescribed case-study answer under ASCE 7-16 and earlier; under 7-22 it is background reference, not a value you can substitute for the mapped one.

ERDC/CRREL SR-20-1 (2020) is the federal case-study methodology: CRREL’s database of more than ten thousand measurement stations and its procedure for estimating a site’s 50-year ground snow load from the surrounding stations — filtering by distance, record length, and elevation, then regressing load against elevation. It is the same family of analysis CRREL has performed for federal agencies for decades, and it meets ASCE 7’s requirement that a case study rest on an extreme-value analysis of nearby station data rather than a single point. Its output is a 50-year uniform-hazard value: the form the case-study provisions of ASCE 7-16 and earlier call for.

Nivometra runs that methodology as software: pin a site, see the stations the filters select, and generate a documented case-study report — the estimated 50-year ground snow load with the inputs, stations, and elevation regressions it rests on — for $29. The workspace is free to explore.