Drought in a human-modified world: reframing drought definitions, understanding, and analysis approaches
Abstract. In the current human-modified world, or Anthropocene, the state of water stores and fluxes has become dependent on human as well as natural processes. Water deficits (or droughts) are the result of a complex interaction between meteorological anomalies, land surface processes, and human inflows, outflows, and storage changes. Our current inability to adequately analyse and manage drought in many places points to gaps in our understanding and to inadequate data and tools. The Anthropocene requires a new framework for drought definitions and research. Drought definitions need to be revisited to explicitly include human processes driving and modifying soil moisture drought and hydrological drought development. We give recommendations for robust drought definitions to clarify timescales of drought and prevent confusion with related terms such as water scarcity and overexploitation. Additionally, our understanding and analysis of drought need to move from single driver to multiple drivers and from uni-directional to multi-directional. We identify research gaps and propose analysis approaches on (1) drivers, (2) modifiers, (3) impacts, (4) feedbacks, and (5) changing the baseline of drought in the Anthropocene. The most pressing research questions are related to the attribution of drought to its causes, to linking drought impacts to drought characteristics, and to societal adaptation and responses to drought. Example questions include
- (i) What are the dominant drivers of drought in different parts of the world?
(ii) How do human modifications of drought enhance or alleviate drought severity?
(iii) How do impacts of drought depend on the physical characteristics of drought vs. the vulnerability of people or the environment?
(iv) To what extent are physical and human drought processes coupled, and can feedback loops be identified and altered to lessen or mitigate drought?
(v) How should we adapt our drought analysis to accommodate changes in the normal situation (i.e. what are considered normal or reference conditions) over time?