Use of ambient water quality data to refine emission estimates in the Danube Basin

In the frame of an EU/Phare project nutrient balances of the Danube countries were studied. As an element of the procedure, N and P emissions to surface waters were estimated on various scales. The "immission" loads computed from water qualify and flow data were used to refine these estimates. The evaluation indicated the significant role of uncertainties related to scarce observations typical for many countries in the Danube Basin. To obtain the error of annual nutrient load estimates statistical analyses of a number of Hungarian rivers (in the 2 m(3)/s - 2000 m(3)/s mean flow range) were performed. Both, analytical expressions of sampling theory and Monte Carlo simulations were used. The ratio estimate and an analysis of the relation of errors of the annual mean flow and the load, resp. were used to rake advantage of the additional information of flow data in comparison to water quality. An empirical error Function for the range of rivers considered was derived. The effort led to useful results in relation to the selection of the estimation method, the role of sampling frequency and size of the catchment. Conclusions were also drawn on estimation errors of loads of first order tributaries of the Danube and their use in the frame of a nutrient balance study. (C) 1999 Published by Elsevier Science Ltd on behalf of the IAWQ. All rights reserved.