Natural and Human Influences on the Link Between Meteorological and Hydrological Drought Indices for a Large Set of Catchments in the Contiguous United States

Abstract

Precipitation-based drought indices are most commonly used in drought monitoring and early warning systems whereas impacts of drought are often related to other domains of the hydrological cycle such as streamflow. Precipitation droughts do not always coincide with streamflow droughts, as the propagation from precipitation to streamflow is affected by climate, catchment properties, and human influences. For monitoring in ungauged catchments it is the question to what extent drought indices solely based on precipitation or other (more recently developed) meteorological drought indices that include evaporation or snowmelt, have a stronger correlation with streamflow, and whether this correlation is weaker in catchments where streamflow is altered by human influences. Results of a correlation exercise between various meteorological drought indices and streamflow showed that the strongest correlation was often found for meteorological drought indices that include evaporation (especially in drier climates) or snow processes (especially in colder climates). Most catchments with an indicated presence of human influences showed a maximum correlation between meteorological drought indices and streamflow that was comparable in strength to the same correlation for catchments with near-natural flow. However, up to 15% of catchments with an indicated presence of human influences show weaker correlations. Drought indices derived from these influenced records with a weaker correlation do not necessarily correspond to reported drought impacts. In conclusion, knowing which meteorological drought index has the strongest correlation with streamflow in different climate zones has the potential of improving large-scale drought monitoring and early warning systems in ungauged areas or regions that lack real-time streamflow availability.