Model for prediction of seepage from small unlined water impoundments

Abstract

A physically based hydrologic model is presented which predicts seepage losses from small water impoundments. Model development included: 1) estimation of historical seepage losses from three stock ponds located at Walnut Gulch Experimental Watershed, Tombstone, Arizona, 2) evaluation of factors controlling seepage losses from these ponds, 3) development of a seepage prediction model based upon the observed factors controlling seepage losses, and 4) calibration and testing of the seepage Prediction model. Results of this study indicate that seepage rates from the stock ponds studied are dynamic, controlled largely by the volumes and frequencies of inflows and the permeabilities of the materials underlying the ponds. Required input to the model includes an inflow record, pan evaporation data, a stage-surface area relationship, and estimates of the model parameters of minimum and maximum hydraulic conductivity. Comparison of model output to actual data indicates the program successfully predicts seepage losses from small unlined water impoundments.