Intercomparison of four global precipitation data sets and their correlation with increased Eurasian river discharge to the Arctic Ocean

Recent increases in Eurasian river discharge to the Arctic Ocean have attracted considerable scientific attention but remain poorly understood. Previous studies have examined fire frequency, permafrost thaw, and dam construction as potential mechanisms. Here we focus on precipitation as a driver, us...

Full description

Bibliographic Details
Main Authors: Pavelsky, T.M., Smith, L.C.
Other Authors: College of Arts and Sciences, Department of Geological Sciences
Format: Article in Journal/Newspaper
Language:English
Published: Blackwell Publishing Ltd 2006
Subjects:
Online Access:https://doi.org/10.17615/atsz-7h48
https://cdr.lib.unc.edu/downloads/8623j7154?file=thumbnail
https://cdr.lib.unc.edu/downloads/8623j7154
Description
Summary:Recent increases in Eurasian river discharge to the Arctic Ocean have attracted considerable scientific attention but remain poorly understood. Previous studies have examined fire frequency, permafrost thaw, and dam construction as potential mechanisms. Here we focus on precipitation as a driver, using 198 dam-free Eurasian river basins ranging from 151 to 897,000 km2. Using R-ArcticNet monthly discharge data and four observational and reanalysis precipitation products from the University of Delaware (UDel), University of Washington (UW), NCEP/NCAR (NCEP), and ECMWF (ERA-40), we (1) assess which precipitation data sets best capture spatially realistic patterns as inferred from agreement with river discharge (198 basins; 1958-1989); and (2) determine to what extent observed discharge trends follow Udel precipitation changes (66 basins; 1936-1999). Results from the precipitation intercomparison show for the 74 (of 198) basins displaying statistically significant discharge trends (24 positive, 50 negative; -74% to +89%, mean = -1%), interpolated precipitation products significantly outperform reanalysis data sets, perhaps owing to the fine-scale resolutions examined here. Agreement between discharge and precipitation is 42-86% and 42-97% for UDel and UW, respectively, but approaches zero for NCEP and ERA-40. Comparison of precipitation and discharge trends suggests that precipitation increases play a significant role in observed long-term discharge increases. For the 40 (of 66) basins displaying statistically significant trends in discharge (32 positive, 8 negative; -23% to +50%, mean = +11%), 29 display corresponding trends in precipitation with 35-62% agreement between discharge and precipitation trend. Comparison of discharge trends with basin permafrost properties indicates a possible, but not strong role for permafrost thaw in the observed increases.