By means of the Bayesian decision approach the economic value of low flow data is considered with special reference to the design of treatment plants for typical Danish conditions. For a specific case the worth of primary data in form of direct observations of low streamflows has been investigated in terms of an expected opportunity loss (EOL). Indirect information concerning the low flow properties of interest may be obtained by means of a secondary data set in combination with a regression model. In this study is suggested a heuristic method for the economic evalution of a secondary set of data taking into account the uncertainty embedded in the regression model.