Trends in Global and Basin-Scale Runoff over the Late Twentieth Century: Methodological Issues and Sources of Uncertainty

While human influence has been detected in global and regional surface air temperature, detection-attribution studies of direct (i.e., land use and water management) and indirect (i.e., climate related) effects of human activities on land surface hydrology remain a crucial and controversial issue. In the present study, a set of global offline hydrological simulations is performed during the 1960-94 period using the Interactions between Soil, Biosphere, and Atmosphere-Total Runoff Integrating Pathways (ISBA-TRIP) modeling system. In contrast to previous numerical sensitivity studies, the model captures the observed trend in river runoff over most continents without including land use changes and/or biophysical CO2 effects, at least when the comparison is made over 154 large rivers with a minimum amount of missing data. The main exception is northern Asia, where the simulated runoff trend is negative, in line with the prescribed precipitation forcing but in contrast with the observed runoff trend. The authors hypothesize that the observed surface warming and the associated decline of permafrost and glaciers, not yet included in most land surface models, could have contributed to the increased runoff at high latitudes. They also emphasize that the runoff trend is a regional-scale issue, if not basin dependent. In line with recent observational studies, their results suggest that CO2 stomatal conductance effects and land use changes are not the primary drivers of the multidecadal runoff variability at continental scales. However, the authors do not rule out a human influence on land runoff, at least through the high-latitude surface warming observed over recent decades.