Drought in the anthropocene: what/who causes abnormally dry conditions?

Deforestation for agriculture, reservoir construction for hydropower, groundwater abstraction for irrigation, river diversion for navigation. These are only some examples of human interventions in river basins. The consequences of these interventions can be far-reaching, but are often difficult to distinguish from natural influences on the water system, such as meteorological droughts. River basin managers in water-stressed regions need to quantify both human and natural effects on the water system to adapt their water management accordingly. ‘Drought’ is a natural hazard, which is caused by climatic processes and their intrinsic variability, and cannot be prevented by short-term, local water management. ‘Water scarcity’ refers to the long-term unsustainable use of water resources and is a process that water managers and policy makers can influence. Water scarcity and drought are keywords for river basin managers in water-stressed regions, like Australia, California, China and the Mediterranean Basin. The interrelationship between drought and water scarcity, however, is complex. In regions with low water availability and high human pressures, water scarcity situations are common and can be exacerbated by drought events. The worst situation is a multi-year drought in a (semi )arid region with high demand for water. In monitoring the hydrological system for water management purposes, it is difficult (but essential) to determine which part of the temporal variation in a hydrological variable is caused by water scarcity (human induced) and which part by drought (natural). So the urgent question of many water managers is: how to distinguish between water scarcity and drought? Here, we present a new quantitative approach to distinguish, namely the observation-modelling framework proposed by Van Loon and Van Lanen (2013) to separate natural (drought) and human (water scarcity) effects on the hydrological system. The basis of the framework is simulation of the situation that would have occurred without human influence, i.e. the ‘naturalised’ situation, using a hydrological model. The resulting time series of naturalised state variables and fluxes can then be compared to observed time series. Additionally, anomalies (i.e. deviations from a threshold) are determined from both time series and compared. This analysis allows for quantification of the relative effect of drought and water scarcity. To show the general applicability of the framework, we investigated case study areas with contrasting climate and catchment properties in Spain, Czech Republic and the Netherlands. Using these case study areas we could analyse the effect of groundwater abstraction and water transfer on groundwater levels and streamflow. The proposed observation-modelling framework is rather generic. We demonstrate the range of methods that can be used and the range of human influences the framework can be applied to. The observation-modelling framework can help water managers, policy makers and stakeholders in water-stressed regions to combat water scarcity, and to better adapt to drought by decreasing their vulnerability. A clear distinction between drought and water scarcity is needed in the anthropocene.