How long should a long-term river study be?

I developed a quantitative method to determine appropriate time lengths for long-term river studies. A literature review identified three critical components for evaluating length of long-term studies - the length of relevant time spans, sufficient statistical power to see an effect, and the variability of the system. The first component (relevant time spans) does not require data and, therefore, can provide an a priori estimation of the required study length. The remaining components can be used with historical data or with preliminary data to provide a detailed characterization of study length requirements. The method was tested using fish community and water quality data from a long-term multi-river study and successfully characterized the unique time requirements for each river.