Sierra Nature Notes, Volume 2, February 2002

Continued from:
Monitoring snow from the beach in San Diego:
Automatic snow sensors in the Sierra
Jessica Lundquist
Ph.D. Candidate
Hydroclimatology Group
Scripps Institution of Oceanography
University of California, San Diego

Tweaking the Cosmic Snow Sensors
The next day I met my first cosmic snow sensor. No, a cosmic snow sensor is not as glamorous as the name implies. It consists of a cylinder about the size of a bread box and weighs considerably less than 50 gallons of antifreeze and 4 stainless steel snow pillows! Cosmic radiation, consisting of gamma rays from the sun and distant stars, passes continually through our atmosphere and is relatively harmless. Cosmic snow sensors make use of the fact that gamma radiation is absorbed by water. By placing one sensor on a pole above the snow and burying one in the ground below the snow, we can look at the different amounts of measured radiation. Knowing how much radiation a cubic centimeter of water absorbs, we can then calculate the amount of water in the snowpack.

This works very well in principle. Unfortunately, the electronics that measure cosmic radiation behave differently at different temperatures. The sensor that is buried by snow will be insulated at a temperature near 0°C while the sensor on the pole may be exposed to much colder temperatures. So, when we compare the two measurements of radiation, we need to know how much of a difference the temperature makes before we can calculate how much radiation was absorbed by water. In addition to digging, solving this problem became my task for the summer.

Figure 1 A comparison of snow water depth at Crabtree Meadows as measured by both a snow pillow (red line) and a cosmic snow sensor (blue line). The match is not exact but shows that cosmic snow sensors can give an estimation of snow water within ± 1 inch. (click graph to enlarge)

Most scientific instruments are calibrated in a controlled laboratory, where the instrument's response to a known set of inputs is carefully measured. When this is not possible or practical, most instruments are calibrated against a "standard," an instrument that has already been calibrated and is believed to give accurate responses. Unfortunately, the cosmic snow sensor is a relatively new instrument, and no standards exist. Also, it is hard to recreate ambient high-altitude cosmic radiation in a controlled laboratory (most of which are indoors and at low elevations). Understanding these difficulties, two methods were employed. First, several cosmic snow sensors were deployed next to snow pillows. Because the snow pillows have been functioning for years, their data can become the standard against which to compare the cosmic snow sensors. At higher air temperatures, the cosmic snow sensor shows less snow on the ground than the snow pillow. This can be partially corrected to yield a graph like Figure 1, where the cosmic snow sensor reports snow depth within 1" of liquid water as reported by the snow pillow.

Figure 2 High-altitude field work sometimes calls for creative measures. A cooler filled with an ice bath keeps one cosmic snow sensor at 0°C, while a freezer cools the other sensor to -20°C. The styrofoam container on the right pumps fresh ice water through the system, keeping the temperature in the cooler constant.

Unfortunately, the cosmic snow sensor data is still "noisy" (notice how the line in the graph is not smooth). To attempt further corrections, Frank and I tried to set up "controlled laboratory conditions" near Donner Pass in the northern Sierra. Figure 2 shows the experimental setup. One sensor is kept in a freezer, where the temperature can be adjusted manually. The other sensor is kept in an icebox with ice-water circulating through it so that the temperature is maintained at 0°C. Both instruments are kept at 0°C while the ratio of their response to the incoming cosmic radiation is measured. Then the freezer temperature is lowered to see how the instrument in the freezer changes its measurements while functioning at lower temperatures.

Also unfortunate: the high Sierra doesn't make "controlled conditions"easy! Each time we changed the freezer's temperature, we had to wait 12 hours for it to actually get cold. This required someone, namely me, to stay for a week at the Sierra Snow Lab and periodically feed ice into the cooler while monitoring the actual freezer temperature. While this initially sounds like a relaxing dream job, the week we picked for the experiment coincided with the worst fires of the summer in the surrounding area. Thick smoke filled the air, marring the view. Bees, upset from the smoke, flew about in a random panic, stinging anyone they were unfortunate enough to run into. Helicopters flew overhead every 20 minutes, carrying water to try to douse the nearby fire. Staying alone in these fire-filled mountains without a car, my main consolation was that I didn't have to touch a shovel for an entire week!

Understandably, the experiment results were noisy, but the instrument in the freezer reported relatively higher levels of cosmic radiation when it was colder. This means that during the winter, when the air temperature is significantly below freezing, the cosmic snow sensors will report a larger snow water equivalent than on days when the air temperature is warmer. This relationship is the same as that discovered by comparing the cosmic snow sensors to the snow pillows, where the snowpack was severely underestimated at higher air temperatures.

Still More Data: Back to the Beach!

Figure 3 Frank Gehrke installs the upper portion of the meteorological tower at Tioga Pass while Larry Riddle from Scripps Institution of Oceanography assists. This site will be the first to house a cosmic snow sensor but not a snow pillow.

As such, further work needs to be done to fully understand and calibrate the cosmic snow sensors. Once they are trusted enough to operate independently though, they will revolutionize how we monitor the snowpack. Snow pillows must be placed in open, flat areas, but cosmic snow sensors can operate on steep slopes and under forest canopies, giving a wide range of snow measurements that may be more representative of the snow contained in the watershed as a whole. They are easier to carry and install and can be placed in remote wilderness areas with minimal impact (i.e. less digging and no antifreeze). The first independent cosmic snow sensor site was initiated last summer at Tioga Pass, near 10,000 ft elevation in Yosemite National Park (Figure 3). I still had to dig a 6-foot-deep hole for the meteorological tower which will hold the cosmic sensor; temperature and humidity sensors; a radiometer to measure incoming solar radiation; a solar panel and satellite transmitter to send the data back to Sacramento; and an anemometer to measure the wind. I also got to help haul and mix enough water and concrete (about 40 40-lb bags) to fill the hole

Figure 4 A 30-ft tower with meteorological instrumentation requires a 6-ft-deep hole filled with concrete. For remote mountain sites, the concrete mix and water must all be hauled in. Here, Dan Cayan and Mike Dettinger from SIO mix concrete on a black trash bag while Jessica Lundquist pours water from a gallon jug. Frank Gehrke and Julia Dettinger wait to the left to pat the concrete with sticks once it is poured into the hole.

(Figure 4), which further dashed my hopes of being a lazy armchair scientist. Science is not white coats and a warm quiet lab! On a slope with scattered trees, this will also be the first non-meadow site in the park to send back automated measurements and will be of great use to me as I research snowmelt from the beach in San Diego.

 

 

 

 

 

 

For Further Reading
More information about automatic snow sensors can be found at the following websites:

The National Resource Conservation Service runs Snotel stations in every western state except California. These are very similar to the ones found in the Sierra Nevada.
Pictures and descriptions of a manual snow course survey in Idaho.
California Cooperative Snow Surveys website.
Picture of snow pillow
Article about Frank Gehrke and Dave Hart and snow sensors.
Article about developing cosmic snow sensors.