Using Surrogate Meteorological Data to Predict the Hydrology of a Water Balance Cover

Abstract

The objective of this study was to evaluate strategies for addressing missing meteorological (MET) data when predicting the hydrology of a water balance cover for a waste-containment system using a variably-saturated flow code. Predicting the hydrology of water balance covers typically requires site-specific daily MET data, which may be only partially available (e.g., dew point temperature (Tdew), solar radiation (Rs), wind speed, and cloud cover frequently are only partly available). Thus, some of the input data may need to be estimated or surrogate data employed for hydrologic modeling. The influence of replacing missing MET data with estimates on hydrologic predictions was evaluated for a water balance cover in a semiarid climate. Substitution of single or multiple MET variables with long-term averages led to statistically similar predictions of annual percolation relative to percolation predicted using actual data. Replacing all MET variables (Tdew, Rs, wind speed, and cloud cover) with long-term averages underpredicted percolation by 25% (0.58 mm/year), on average, relative to percolation predicted with actual data. A strategy for estimating MET data via empirical techniques is described that includes estimating (1) Tdew = daily minimum temperature, (2) Rs with the Hargreaves and Samini model, (3) daily wind speed set equal to monthly averages, and (4) cloud cover estimated as a function of solar radiation. This surrogate MET data technique yielded modest overpredictions of annual percolation of 3 and 46% (0.07 and 1.08 mm/year) that were statistically similar to percolation predicted using actual data.