Comparative assessment of predictions in ungauged basins – Part 3: Runoff signatures in Austria

摘要

This is the third of a three-part paper series through which we assess theperformance of runoff predictions in ungauged basins in a comparative way.Whereas the two previous papers by Parajka et al. (2013) and Salinaset al. (2013) assess the regionalisation performance ofhydrographs and hydrological extremes on the basis of a comprehensiveliterature review of thousands of case studies around the world, in thispaper we jointly assess prediction performance of a range of runoffsignatures for a consistent and rich dataset. Daily runoff time series arepredicted for 213 catchments in Austria by a regionalised rainfall–runoffmodel and by Top-kriging, a geostatistical estimation method that accountsfor the river network hierarchy. From the runoff time-series, six runoffsignatures are extracted: annual runoff, seasonal runoff, flow durationcurves, low flows, high flows and runoff hydrographs. The predictiveperformance is assessed in terms of the bias, error spread and proportion ofunexplained spatial variance of statistical measures of these signatures incross-validation (blind testing) mode. Results of the comparative assessmentshow that, in Austria, the predictive performance increases with catchmentarea for both methods and for most signatures, it tends to increase withelevation for the regionalised rainfall–runoff model, while the dependenceon climate characteristics is weaker. Annual and seasonal runoff can bepredicted more accurately than all other signatures. The spatial variabilityof high flows in ungauged basins is the most difficult to estimate followedby the low flows. It also turns out that in this data-rich study in Austria,the geostatistical approach (Top-kriging) generally outperforms theregionalised rainfall–runoff model.