Classification of worldwide drainage basins through the multivariate analysis of variables controlling their hydrosedimentary response

Quality and amount of waters and sediments conveyed within large drainage basins are crucial for human societies and biodiversity concerns. This work aims to determine the factors controlling the hydrosedimentary response (water discharge and sediment load) of 24 worldwide large drainage basins. In this respect, eleven geomorphologic and climatic variables routinely used in the literature were considered and others as fractal dimension, elongation and mean channel slope are novel for such an issue. In addition, two variables, land cover and lithology indexes, somewhat different from the literature in terms of calculation principles, were also included. All these variables were then subjected to multivariate statistical analyses (CA and PCA) and confronted in a matrix correlation. On the whole, our results display that water discharge is controlled by runoff, precipitation, basin area, elongation and fractal dimension while sediment load is governed by runoff, precipitation and maximum elevation. Mean channel slope and land-use have a minor role while other parameters (hypsometry, lithology, length, slope, mean elevation and temperature) do not play a significant role in the hydrosedimentary response. Such statistical analyses also bring out a classification of these drainage basins, comprising five to six main clusters which are ranged according to the main variables ruling their hydrosedimentary response. Two clusters are essentially governed by geomorphometric parameters (area, elongation, fractal dimension, mean elevation and hypsometry) while one cluster is rather controlled by transfer processes (runoff) and by active tectonic (maximum elevation). Hydrosedimentary response of arctic and continental rivers is controlled by low temperature while two drainage basins show any trend. A comparison of our results with other previous works dealing with this same issue points to some significant disagreements essentially based on the number of drainage basins considered, the number of nature of variables used and the analytical methods carried out. Despite their obvious influence, anthropogenic impacts are not considered in this study. In foreseeable works dealing with such issue, anthropogenic parameters would be required for a better description of hydrosedimentary responses of these large drainage basins.