Methods for estimating loads transported by rivers

Ten methods for estimating the loads of constituents in a river were tested using data from the River Don in North-East Scotland. By treating loads de rived from flow and concentration data collected every 2 days as a truth to be predicted, the ten methods were assessed for use when concentration dat a are collected fortnightly or monthly by sub-sampling from the original da ta. Estimates of coefficients of variation, bias and mean squared errors of the methods were compared; no method consistently outperformed all others and different methods were appropriate for different constituents. The wide ly used interpolation methods can be improved upon substantially by modelli ng the relationship of concentration with flow or seasonality but only if t hese relationships are strong enough.