Early findings in comparison of AMSR-E/Aqua L3 global snow water equivalent EASE-grids data with in situ observations for Eastern Turkey

Microwave remote sensing (RS) enables the direct determination of snow water equivalent (SWE), which is an important snow parameter for water resources management. The accuracy of remotely sensed SWE values has always been a concern. Previous studies evaluated global SWE monitoring. However, regional effects such as vegetation, snow grain size, snow density and local meteorological conditions may lead to uncertainties. Thus, regional validation studies that quantify and help to understand these uncertainties and possible error sources are important both for algorithm development and accurate SWE computation.In this study, data of Advanced Microwave Scanning Radiometer (AMSR-E)/Aqua level 3 global SWE Equal Area Scalable Earth (EASE) Grids are compared with ground measurements for 2002-2003 winter period for Eastern Turkey, which includes the headwaters of the Euphrates and Tigris rivers and is fed largely from snowmelt. Thus, accurate determination of SWE is important in optimum resource management for both Turkey and downstream nations. Analyses indicated that AMSR-E generally overestimated SWE in early season. As winter progressed, higher in situ SWE values with respect to AMSR-E were observed which led to underestimation by AMSR-E. The differences between AMSR-E and in situ SWE varied between - 218 and 93 mm.Use of in situ snow densities lead the correlation coefficient between AMSR-E and in situ SWE to increase from 0·10 to 0·32. Underestimation of SWE by AMSR-E occurs after some warm periods, while overestimations occur following refreezing. On rainy days or some days after precipitation within the warm periods, zero AMSR-E SWE values are observed. Copyright © 2008 John Wiley & Sons, Ltd.