Snowfall Extremes at Alta, Part I

Michael Wasserstein and I have a new paper out examining our favorite research topic: Snowfall extremes at Alta. The paper was just published in what is known as early online release and is available at https://journals.ametsoc.org/view/journals/mwre/aop/MWR-D-23-0206.1/MWR-D-23-0206.1.xml. 

It may be paywalled if you don't work at an institution with a license for American Meteorological Society journals or have an American Meteorological Society membership (apologies).  However, we'll hit some of the highlights in this special two part Wasatch Weather Weenies blogstorm, beginning here with Part I.  

The inspiration for this work was a paper by Larry Dunn, a long-time meteorologist and avid skier, that was published as a NOAA Technical Memorandum in 1983. 

Many of the relationships between flow (especially wind direction) and local snowfall enhancement at various locations in the Wasatch Range that are used today are based on this paper.  

Michael and I thought we would do an update, focused on Alta, and using modern atmospheric analyses, diagnostics, and observations. Using data collected and generously provided by the Alta Ski Patrol, we focused on large 12-hour snowfall events, defined in two ways.  The first was based on snowfall amount.  The second based on the liquid precipitation equivalent (LPE) of snowfall.  Large here means in the top 5% (known as the 95th percentile) with at least 2.54 cm (1 inch) of snow for snowfall or 2.54 mm (0.1 inch) of water for LPE. This 95th percentile works out to be 30.5 cm (12 inches) for snowfall amount and 27.9 mm (1.1 inches) of water for LPE.  

Source: Wasserstein and Steenburgh (2024)

Below is a classic figure from Dunn (1983) illustrating the 700-mb (about 10,000 ft) wind speed and direction during heavy precipitation events at Alta. He used 1" of water equivalent in 24 hours for the threshold for heavy events and presented his results in a hand-drawn diagram that meteorologists sometimes refer to as a wind rose.  This figure identified the high frequency of heavy events in northwesterly flow.  Note also that events peak at a speed of around 10 m/s (about 20 knots)

Source: Dunn (1983)

Below is our updated take using our 1.1" in 12-h threshold and enabling comparisons with climatology (all periods during the cool season), all snow events, all LPE events, and all of the extremes.  In figure d, you can see the high frequency of 700-mb flow from the WNW and NW during snow amount extremes peaking at near 10 m/s. If you look carefully though, you'll see a secondary maximum for flow from the SSW.  For LPE, there is are two clear maxima centered on the SSW and WNW.  You will also notice that snow and LPE extremes have occurred across a wide range of flow directions from SE to WNW (or even N for the former).  

Source: Wasserstein and Steenburgh (2024)

A few key points to take from this figure.  First, snowfall extremes can happen for a wide range of flow directions and wind speeds.  This is why I like to emphasize the great diversity of storms that affect Alta and not discount storms just because they don't have NW flow.  If they dynamics are right, Alta can get it.  Nevertheless, a high frequency of large LPE events occurs for flow form the SW and WNW.  This is known as a bimodal distribution.  For snow though, the WNW maximum is much stronger.  This reflects the influence of snow-to-liquid ratio.  The WNW flow storms tend to be colder, and produce more snow per unit of LPE, so that flow direction lights up more frequently for snowfall amount extremes.  

We also identified seven major synoptic storm types contributing to snowfall extremes at Alta.  Four were associated with enhanced integrated vapor transport (IVT) penetrating inland from the Pacific coast from the south (SIVT), southwest (SWIVT), west (WIVT), or northwest (NWIVT).  

Source: Wasserstein and Steenburgh (2024)

IVT is often used to identify atmospheric rivers, but this turns out to be a complex matter that we will talk about in greater depth in Part II.  

The other three patterns were northwesterly post frontal flow events (of course), frontal (associated with a stationary or cold front), and cold-core lows with southwesterly flow.  

Source: Wasserstein and Steenburgh (2024)

We did not distinguish between northwesterly post frontal flow events with and without lake effect as that would have been another major effort.  We'll let you sort through the radar data to do that!

That's enough for now.  We'll dig into this further in a forthcoming Part II post.