Reply on RC3

Abstract

Streamflow drought hazard indicators (SDHI) are mostly lacking in large-scale drought early warning systems (DEWS). This paper presents a new systematic approach for selecting and computing SDHI for monitoring drought risk for human water supply from surface water and for river ecosystems that is also relevant for meteorological or soil moisture drought. We rec-ommend considering the habituation of people and ecosystems to the streamflow regime (e.g., a certain interannual variability or relative reduction of streamflow) when selecting indicators. Distinguishing four indicator types, we classify indicators of drought magnitude (water deficit during a pre-defined period) and severity (cumulated magnitude since onset of the drought event). We quantify eight existing and three new SDHI globally using the global hydrological model WaterGAP2.2d. We recommend streamflow hazard indicators that should be included in large-scale DEWS as they are suitable for risk systems that are differently adapted to low water availability and characterized by either perennial or intermittent streamflow regime and the existence or not of large reservoirs. Drought magnitude is best quantified by return period or relative deviation from mean, and severity by return period or water volume below a threshold relative to mean annual streamflow. Both anomaly and deficit indicators should be provided.