Abstract
The application of dilution gauging techniques to rivers and streams and other variable flows incurs a systematic error caused by the change in discharge with time. The discharge calculated from dilution gauging results may be considered as an estimate of the stream discharge at the injection section or the sampling section, and the magnitude and sense of the error vary according to this choice. On the recession limb of a hydrograph and in other cases where the discharge may be expected to change smoothly and gradually, the equations derived from the residence time model previously proposed by the author may be linearized, and formulae of general applicability developed. In this paper formulae are presented for the error arising in the case of a constant rate tracer injection, and examples are given of their application to results from a number of different flow systems. It is shown that the error may be significant in two situations: where the rate of change of discharge is high and where poor mixing necessitates long gauging reaches and hence long tracer injections.