Better estimates of snow cover

Skiers are not the only people that are interested in accurate information on snow cover. Knowing how much water is stored as snow on mountain slopes is crucial for water resource managers to predicting floods or decide how to allocate resources. Laser and radar-based technologies are two ways to remotely measure snow depth at discrete points, but using these measurements to estimate total snow mass is inherently prone to errors. In a theoretical study published in the journal The Cryosphere, researchers from ENAC’s Laboratory for Cryospheric Science present a methodology to design measurement campaigns so that the error is kept to a minimum.

Whereas in the past snow-depth had to be probed by hand, today’s technologies are able to sample many more points in far less time, and that from a distance. For example using a lidar, which accurately measures distances by means of a laser, it is possible to sample an entire slope at a one-meter resolution from a fixed vantage point across a valley. But because the physical properties of snow vary strongly from place to place, especially in complex terrains or in forests, errors creep in when these point measurements are used to estimate total snow cover.

To improve the accuracy of snow cover estimates, Ernesto Trujillo, the lead author of the study, developed a mathematical model that is capable of reproducing the expected error for a given terrain. Using this model, he explains, snow cover estimates can be improved in two different ways. First, the locations at which snow-depths are remotely assessed can be optimized to minimize the overall estimation error. Alternatively, the model can be used to determine at how many points snow-depth has to be sampled for the error to be reduced to an acceptable level.

Reference: Trujillo, E. and Lehning, M.: Theoretical analysis of errors when estimating snow distribution through point measurements, The Cryosphere, 9, 1249-1264, doi:10.5194/tc-9-1249-2015, 2015.