Regional flood frequency analysis is currently one of the most often applied methodologies for flood estimation at ungauged sites worldwide. The performance of any regional analysis is very dependent on regionalization - which is its most important step. Currently, prevailing methods of regionalization usually base the formation of flood regions on the similarity of river catchments in physiographic characteristics. This approach is derived from the concept that catchments similar in physiographic characteristics have consequently similar flood response. This concept cannot be guaranteed in general, particularly if the spatial variability of catchment characteristics is large. A usually significant variance of sites’ L-moment ratios L-Cs and L-Ck in a region derived from similarity in physiographic characteristics tends to be explained by their sampling variability. Naturally, one can ask, to what extent this variance may be regarded as sampling variability? If the variance is not caused by sampling variability, then such a region is heterogeneous, having significant differences between sites’ probability distributions.
Weitere Kapitel dieses Buchs durch Wischen aufrufen
- Flood Flow Regionalization Based on L-Moments and Its Use with the Index Flood Method
- Springer Netherlands