Combination of MODIS Snow, Cloud and Land Spatial Coverage Data with SNOTEL to Generate Inter-Annual and Within Season Snow Depletion Curves and Maps Conference Paper uri icon

Overview

abstract

  • Quantification of spatial coverage of snow and its temporal decline throughout the snowmelt season is an important input for snowmelt runoff models in the prediction of runoff and simulation of streamflow. Several remote sensing methods of snow cover mapping exist today with differing spatial and temporal resolutions that can be used in determining the progressive reduction of snow cover during snowmelt. The Moderate-Resolution Imaging Spectroradiometer (MODIS) satellite sensor has provided global coverage with daily temporal resolution and a spatial resolution on the order of 1 km2 since the year 2000. Several challenges hinder the development and routine use of MODIS snow covered area products, which include cloud and forest obscuration of snow-covered surfaces. Furthermore, the relatively recent launch of the MODIS sensor prevents its direct use in historical or future (e.g., climate change scenario) studies. Based on numerous observations in the literature that snowmelt occurs in a repeating spatial pattern, albeit shifted and/or accelerated or decelerated in time, from one year to the next, a method is presented that makes use of remotely sensed MODIS snow, land and cloud data and concurrent ground-based Snow Telemetry (SNOTEL) data to construct a single, inter-annually applicable and temporally dimensionless snow depletion curve and a corresponding snow depletion map. MODIS Cloud data and ground-based SNOTEL snowfall data during the snowmelt season are used together with the inter-annual snowmelt curve/map to generate within-year snowmelt curves that account for the impact of additional melt-season snowfall. The method is successfully applied to the 9000 km2 Upper Snake River Basin in Wyoming, USA as a test case. The method may be used with historical SNOTEL data to reconstruct snow depletion curves or maps for base periods preceding the availability of current satellite remote sensing, and with synthesized weather datasets for future periods associated with altered climate scenarios; the method may also be used in any current snowmelt season as model input to forecast snowmelt and improve streamflow forecasts.

publication date

  • December 2015