Investigating the Variation in the Annual Progression of Snow Accumulation and Melt in the Sierra Nevada: A Functional Data Analysis Approach

• Activity Number: 195
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Eduardo Montoya*+ and Wendy Meiring and Jeff Dozier
• Companies: California State University, Bakersfield and University of California, Santa Barbara and University of California, Santa Barbara
• Keywords: functional data ; curve registration ; snow ; Sierra Nevada

Abstract

In California, snow from the Sierra Nevada is the source of most of the water, filling reservoirs and recharging aquifers that support downstream habitat, irrigation, urban use, hydropower, and recreation. Management of this resource depends on measurements of snow water equivalent (SWE). Snow pillows provide SWE profiles by obtaining daily measurements throughout each water year of SWE at various spatial locations. The seasonal pattern in a SWE profile at a particular location across water years contains variation both in the magnitude of snowpack events and in the timing of the events (both magnitude and phase variability). This work proposes using landmark curve registration with monotonically constrained B-splines to study the variability of the timing of events in the annual progression of snow accumulation and melt in the Sierra Nevada, thereby providing a new approach to quantify the variation of snow accumulation and snow melt processes. We also discuss an ongoing project to develop a functional linear model to study the association of the phase variability with changes in climate indices and spatial location.