Stats!

Hey blog fans!
Your web hero and his snow team have been moving forward with data analysis and we've achieved some useful SWE modeling. Yeah!! As would be expected, SWE in the upper elevation snowcourses models fairly close to the snotel sites while the lower elevation sites do not. What methods are we using to model SWE you ask? Well were doing a standard linear regression of SWE to Elevation by month for all snowcourses and snotel sites. Because of the data spread, due to differences in siting and elevation between sites, we also log transformed the data as log SWE/elevation and log SWE/log elevation to get tighter models. We also did a standard regression model of each snowcourse SWE values to each of the three snotel sights to determine which snotel models which snowcourse most accurately. We are currently in the process of evaluating each model by laying all the printouts on a table in the NASA lab to get the best viewpoint. Tune back in later for our forthcoming tech report.