Cold region river discharge uncertainty -: estimates from large Russian rivers

We develop an error model to understand the reliability and accuracy of river discharge datasets that are now being used for a variety of important global change questions. The developed error model for cold region river discharge uses standard hydrometric data along with information on the frequency and precision of measurements, characteristics of river channel capacity, and method of discharge computation. The uncertainties of daily, monthly and annual discharge data for the downstream gauges of the six largest Eurasian rivers (Severnaya Dvina, Pechora, Ob', Yenisei, Lena and Kolyma) in the panArctic drainage along with uncertainty of aggregated annual time series are evaluated using the suggested methodology. The study shows that uncertainties associated with discharge determination significantly change from year to year and strongly depend on the computational methods used and frequency of discharge measurements. Recent work by Peterson et al. (2002) has shown increases in river discharge to the Arctic Ocean of the six largest Eurasian rivers of 7% from 1936 to 1999. This paper focuses on determination of reliability in the discharge data which provided such conclusion. The obtained results further confirm the findings of Peterson et al. (2002) concerning the rise in river discharge. We found that errors of the total annual discharge for the six rivers over the period 1950-2000 are in the range 1.5-3.5%. The long-term trend of the observed discharge from these six rivers into the Arctic Ocean for 1936-2000, along with uncertainty associated with discharge data, is 2.0 +/- 0.4 km(3)/year. (c) 2005 Elsevier B.V. All rights reserved.