WSDOT - Only 80" of Snow at Snoq Pass this Winter

One of the latest peer reviewed articles I could find on this subject, written by the most outspoken critics of the earlier "1950 - present" analysis by Mote et al (that Amar also heavily criticizes):

Mark T. Stoelinga, Mark D. Albright, and Clifford F. Mass, 2010: A New Look at Snowpack Trends in the Cascade Mountains. J. Climate, 23, 2473–2491.

journals.ametsoc.org/doi/abs/10.1175/2009JCLI2911.1

The abstract from the paper is quoted below.  The full paper is also pretty decent if you want to dive deeper.  The most interesting thing to me in this paper is that they removed the effect of Pacific variability and found that snowpacks have been declining about 2% / decade since 1930.  To my mind, this is the central question in whether our snowpack has been declining long term, much more so then looking at any particular year.

This study examines the changes in Cascade Mountain spring snowpack since 1930. Three new time series facilitate this analysis: a water-balance estimate of Cascade snowpack from 1930 to 2007 that extends the observational record 20 years earlier than standard snowpack measurements; a radiosonde-based time series of lower-tropospheric temperature during onshore flow, to which Cascade snowpack is well correlated; and a new index of the North Pacific sea level pressure pattern that encapsulates modes of variability to which Cascade spring snowpack is particularly sensitive.

Cascade spring snowpack declined 23% during 1930–2007. This loss is nearly statistically significant at the 5% level. The snowpack increased 19% during the recent period of most rapid global warming (1976–2007), though this change is not statistically significant because of large annual variability. From 1950 to 1997, a large and statistically significant decline of 48% occurred. However, 80% of this decline is connected to changes in the circulation patterns over the North Pacific Ocean that vary naturally on annual to interdecadal time scales. The residual time series of Cascade snowpack after Pacific variability is removed displays a relatively steady loss rate of 2.0% decade−1, yielding a loss of 16% from 1930 to 2007. This loss is very nearly statistically significant and includes the possible impacts of anthropogenic global warming.

The dates of maximum snowpack and 90% melt out have shifted 5 days earlier since 1930. Both shifts are statistically insignificant. A new estimate of the sensitivity of Cascade spring snowpack to temperature of −11% per °C, when combined with climate model projections of 850-hPa temperatures offshore of the Pacific Northwest, yields a projected 9% loss of Cascade spring snowpack due to anthropogenic global warming between 1985 and 2025.

I have been keeping snow totals for the pass for many years and I found many inaccuracies with the DOT during this time. I would post the totals every snowfall and compare with the ski resort, DOT and my own findings. I find the DOT numerous times was way off, showing more then what actually fell. Especially on the days of freezing rain where no snow falls the DOT page would report sometimes 8-12" of snowfall. Early years our totals were fairly close with end of season totals sometimes within 10" or so, but starting in 2008/09 that all changed. Since then the DOT records snow depths far above what I show, 50-100" of difference. This made me think they might get snow removal funding based on the amount of snow and they might beef up the numbers?
All I know is something changed in 2008 for their snowfall recording and now lack of accuracy.

Another snow measuring question - NWAC shows 23" at the base of Alpental (3100') and 14" at Hurricane Ridge (5250'). Webcams for both show completely bare ground, so where are the NWAC stations getting these measurements?

Since I'm sitting at home looking at the computer, I'm just curious. If I had driven somewhere expecting 2' of snow on the ground and it was dirt, I might be pissed.

Gamma distribution!

nerd

I have been keeping snow totals for the pass for many years and I found many inaccuracies with the DOT during this time. I would post the totals every snowfall and compare with the ski resort, DOT and my own findings. I find the DOT numerous times was way off, showing more then what actually fell. Especially on the days of freezing rain where no snow falls the DOT page would report sometimes 8-12" of snowfall. Early years our totals were fairly close with end of season totals sometimes within 10" or so, but starting in 2008/09 that all changed. Since then the DOT records snow depths far above what I show, 50-100" of difference. This made me think they might get snow removal funding based on the amount of snow and they might beef up the numbers?
All I know is something changed in 2008 for their snowfall recording and now lack of accuracy.

And therein lies the problem with analyzing and drawing conclusions from long term data sets for which meta-data or changes in data collection methods are not noted.

Not directly related to Snoqualmie Pass, but my favorite benchmark to judge the decline of the Cascade snowpack is to look at the glaciers.

Statistics may lie, but glaciers don't.

Another snow measuring question - NWAC shows 23" at the base of Alpental (3100') and 14" at Hurricane Ridge (5250').  Webcams for both show completely bare ground, so where are the NWAC stations getting these measurements?

Many of the stations are located in sheltered sites on generally flat ground. These locations improve accuracy of snowfall/precipitation measurements, though they often hold snow a bit longer than exposed sites. The Alpental station is located behind the main lodge. If the site hasn't been recently maintained then we may be seeing interference from brush