Thursday, April 30, 2015

// // 3 comments

Climate Data Mining in the Sierra Nevada

California and the Sierra Nevada are well know for the Gold Rush years of the mid 1800s, but for weather weenies like myself (and you likely fall into this classification if you are reading this blog) there is a wealth of beautiful long term climate records for the region just waiting to mined. As a weather and climate researcher data mining is a regular part of my job, and the long term observations really fascinate me.

Dr. James E. Church, who originally came to Reno in the late 1800s to teach foreign languages at UNR, developed the Mount Rose snow sampler in 1906, and the Mount Rose site now has one of the longest snow water equivalent records in the country.


Figure: Dr. James E. Church,  http://www.wcc.nrcs.usda.gov/factpub/wc_ss.html

Donner Summit and Mount Rose both have snow course records archived back to 1910, and can be found at the same pages as the Natural Resources Conservation Service SNOTEL data (http://www.wcc.nrcs.usda.gov/snow/). Below I have plotted up both sites showing the measured snow water equivalent on April 1st, which is considered about the time of maximum snow pack in the Sierra.

Some interesting things stand out when comparing the two sites. First, when considering the current drought 2014 and 2015 are the two lowest values in the Donner record, but at Mount Rose 2015 is tied for third lowest rank with 1948. The lowest value at Mount Rose was in 1924 followed by 1931. There are likely two main factors contributing to these difference: (1) orographic effect (Donner situated just west of the Crest, and Mount Rose in the Carson Range) and (2) elevation. From a skiers perspective, we know that during many storms, Donner and Rose can get similar snowfall or often times Rose is shadowed and Donner gets more, and rarely (during upslope events) Rose can be the big winner. The high freezing levels (ridiculously warm temperatures) during the past two winters is likely the main reason for Donner snow water equivalent being so much lower than Rose (~2000 feet elevation difference), and has played big role in the "snow drought" (here and in the Pacific Northwest).


Streamflow records can tell you quite a bit about drought conditions as well, particularly when you accumulate the streamflow throughout the course of the water year. The Merced River at Happy Isles in Yosemite has one of the longest records in the Sierra, dating back to the early 1900s. 



Shrad Shukla and Andy Wood developed the Standardized Runoff Index (SRI; http://www.hydro.washington.edu/forecast/monitor/info/Shukla_Wood_SRI_GRL08.pdf) to quantify streamflow drought based on the historical distribution. Below is the water year SRI from the Merced River, which is showing the accumulated streamflow throughout each water year. For the SRI, negative values mean drought, and positive values mean wet. This one only goes through 2014, but tells a similar story as the Mount Rose snow course: 1924 and 1931 were worse single year drought than 2014. Based on this SRI 1977 was actually the worst single year drought. Looking at the abysmal snow pack for this year, the 2015 water year SRI may actually be the new worst. 



There are many things to consider when looking at the current drought, and depending elevation, orographic positioning, and variable being considered you will find different results. I've only considered single year droughts here, and this multi-year event (four years now) is certainly one of the most severe in the past 100 years.

Well, I'm droughted out! Hopefully in the near future when can discuss extremes on the opposite end of the spectrum!

3 comments:

  1. Interesting !
    The lowest value at Mount Rose seems to be in 1926 (9" ) not 1924 (17 " )
    mistakes in numbers of 1963 Donner summit (or incredible ... ) :
    feb 1°: 0" march 1° : 0" march 25 ° : 6,3" april 24 : 27.8"

    ReplyDelete
  2. Thanks for pointing this out Jojoski 2! There are definitely some questionable values in both the Mount Rose and Donner data. For instance, the 1983 value for Rose seems waaaay too low, and the Rose SNOTEL has a much different value. It's always good to question the data and cross validate!

    ReplyDelete
  3. It's easier to cross different data from US than thoses for the french or swiss alps (not so much and a large part without free access)
    PS if some snow stats geek like me comes here and want's to improve his french -> can read some pages and follow links of this : http://www.skipass.com/forums/enmontagne/meteo/sujet-129850-300.html

    ReplyDelete