Understanding Hydrologic Variability across Europe through Catchment Classification
Anna Kuentz1, Berit Arheimer1, Yeshewatesfa Hundecha1, and Thorsten Wagener2,31Swedish Meteorological and Hydrological Institute, 601 76 Norrköping, Sweden 2Department of Civil Engineering, University of Bristol, BS8 1TR, Bristol, UK 3Cabot Institute, University of Bristol, UK
Received: 21 Aug 2016 – Accepted for review: 22 Aug 2016 – Discussion started: 29 Aug 2016
Abstract. This study contributes to better understanding the physical controls on spatial patterns of pan-European flow signatures – taking advantage of large open datasets for catchment classification and comparative hydrology. We explored similarities in 16 flow signatures and 35 catchment descriptors across entire Europe. A database of catchment descriptors and selected flow signatures was compiled for 35 215 catchments and 1366 river gauges across Europe. Correlation analyses and stepwise regressions were used to identify the best explanatory variables for each signature resulting in a total of 480 regression models to predict signatures for similar catchments. Catchments were clustered and analyzed for similarities in flow signature values, physiography and for the combination of the two. From the statistical analysis, we found: (i) about 400 statistically significant correlations between flow signatures and physiography; (ii) a 15 to 33 % (depending on the classification used) improvement in regression model skills using catchment classification vs. the full domain; and (iii) 12 out of 16 flow signatures to be mainly controlled by climatic characteristics, while topography was the main control for flashiness of flow and low flow magnitude, and geology for the flashiness of flow.
Classifying catchments based on flow signatures or on physiographic characteristics led to very different spatial patterns, but a classification and regression tree (CART) allowed us to predict flow signatures-based classes according to catchment physiographic characteristics with an average percentage of 60 % of correctly classified catchments in each class. As a result, we show that Europe can be divided into ten classes with both similar flow signatures and physiography. We noted the importance of separating energy-limited catchments from moisture-limited catchment to understand catchment behavior. For improved understanding, we interpreted characteristics in hydrographs, flow signatures, physiography and geographical location to define dominant flow-generating processes. We found that rainfall response, snow-melt, evapotranspiration, damping, storage capacity, and human alterations could explain the hydrologic variability across Europe. Finally, we discuss the relevance of these empirical results for predictions in ungauged basins across Europe and for dynamic modelling at the continental scale.
Kuentz, A., Arheimer, B., Hundecha, Y., and Wagener, T.: Understanding Hydrologic Variability across Europe through Catchment Classification, Hydrol. Earth Syst. Sci. Discuss., doi:10.5194/hess-2016-428, in review, 2016.