Effects of intersite dependence of nested catchment structures on probabilistic regional envelope curves

Regional flood quantile estimates are affected by intersite correlation between flood sequences observed at different discharge gauges. This study analyses the intersite dependence of nested catchment structures and investigates the possibility of improving the accuracy of regional flood quantiles, by modelling cross-correlations for pairs of nested and unnested catchments separately. Probabilistic Regional Envelope Curves are utilised to derive regional flood quantiles for 89 catchments belonging to Saxony, in the Southeast of Germany. The study area has a nested structure and a definitely stronger intersite correlation for nested pairs of catchments than for unnested ones. Probabilistic Regional Envelope Curves are constructed on the basis of flood flows observed within pooling groups of sites (regions). Their recurrence intervals are based on the number of effective sample-years of data (i.e., equivalent number of uncorrelated data). The evaluation of the effective sample-years of data required the modelling of intersite dependence, which we performed globally, using a cross-correlation formula identified for the whole study area, and by using two different cross-correlation formulas, one for nested pairs and another for unnested pairs. These two modelling approaches returned significantly different effective sample-years of data estimates, and therefore also recurrence intervals, in the majority of the cases. The differences result from various assumptions of the size and homogeneity degree of the pooling group. The reduction of the recurrence interval, when using two different cross-correlation functions, is larger for higher recurrence intervals and for a higher fraction of nested catchment within the pooling group. A separation into nested and unnested pairs of catchments gives a more realistic representation of the characteristic river network structure and improves the accuracy of the estimation of regional information content. Hence, applying two different cross-correlation functions is recommended. http://www.hydrol-earth-syst-sci-discuss.net/6/2845/2009/hessd-6-2845-2009.pdf