Search

Abstract Hydrological drought forecasts outperform meteorological ones, which is anticipated coming from catchment memory. Yet, the importance of catchment memory in explaining hydrological drought forecast performance has not been studied. Here, we use the Baseflow Index (BFI) and the groundwater Recession Coefficient (gRC), which through the streamflow, give information on the catchment memory. Performance of streamflow drought forecasts was evaluated using the Brier Score (BS) for rivers across Europe. We found that BS is negatively correlated with BFI, meaning that rivers with high BFI (large memory) yield better drought prediction (low BS). A significant positive correlation between gRC and BS demonstrates that catchments slowly releasing groundwater to streams (low gRC), i.e. large memory, generates higher drought forecast performance. The higher performance of hydrological drought forecasts in catchments with relatively large memory (high BFI and low gRC) implies that Drought Early Warning Systems have more potential to be implemented there and will appear to be more useful.

There still lacks a forecast system that inform end-users regarding the drought impacts, which will be however important for drought management. Here the authors assess the feasibility of forecasting drought impacts using machine-learning and confirm that models, which were built with sufficient amount of reported drought impacts in a certain sector, are able to forecast drought impacts a few months ahead.

Human activities both aggravate and alleviate streamflow drought. Here we show that aggravation is dominant in contrasting cases around the world analysed with a consistent methodology. Our 28 cases included different combinations of human-water interactions. We found that water abstraction aggravated all drought characteristics, with increases of 20%–305% in total time in drought found across the case studies, and increases in total deficit of up to almost 3000%. Water transfers reduced drought time and deficit by up to 97%. In cases with both abstraction and water transfers into the catchment or augmenting streamflow from groundwater, the water inputs could not compensate for the aggravation of droughts due to abstraction and only shift the effects in space or time. Reservoir releases for downstream water use alleviated droughts in the dry season, but also led to deficits in the wet season by changing flow seasonality. This led to minor changes in average drought duration (−26 to +38%) and moderate changes in average drought deficit (−86 to +369%). Land use showed a smaller impact on streamflow drought, also with both increases and decreases observed (−48 to +98%). Sewage return flows and pipe leakage possibly counteracted the effects of increased imperviousness in urban areas; however, untangling the effects of land use change on streamflow drought is challenging. This synthesis of diverse global cases highlights the complexity of the human influence on streamflow drought and the added value of empirical comparative studies. Results indicate both intended and unintended consequences of water management and infrastructure on downstream society and ecosystems.

One of the most effective strategies to reduce the impacts of drought is by issuing a timely and targeted warning from month to seasons ahead to end users. Yet to accurately forecast the drought hazard on a sub-seasonal to seasonal time scale remains a challenge, and usually, meteorological drought is forecasted instead of hydrological drought, although the latter is more relevant for several impacted sectors. Therefore, we evaluate the hydro-meteorological drought forecast skill for the pan-European region using categorical drought classification method. The results show that the hydrological drought forecasts outperform the meteorological drought forecasts. Hydrological drought forecasts even show predictive power (area with perfect prediction > 50%) beyond two months ahead. Our study also concludes that dynamical forecasts, derived from seasonal forecasts, have higher predictability than ensemble streamflow predictions. The results suggest that further development of seasonal hydrological drought forecasting systems are beneficial, particularly important in the context of global warming, where drought hazard will become more frequent and severe in multiple regions in the world.

Streamflow drought forecasting is a key element of contemporary Drought Early Warning Systems (DEWS). The term streamflow drought forecasting, rather than streamflow forecasting, however, has created confusion within the scientific hydro-meteorological community, as well as in operational weather and water management services. The way, how streamflow drought is defined, is the main reason for this misperception. The purpose of this study, therefore, is to provide a comprehensive overview of the differences within streamflow droughts using different identification approaches for European rivers, including an analysis of both historical drought and implications of forecasting of these extreme events. Streamflow data were obtained from a LISFLOOD hydrological model forced with gridded meteorological observed (known as LISFLOOD-Simulation Forced with Observed, SFO). The same model fed with seasonal meteorological forecasts of the European Centre for Medium-range Weather Forecasts system 5 (ECMWF SEAS 5) was used to obtain the forecasted streamflow. Streamflow droughts were analyzed using the Variable Threshold (VT), Fixed Threshold (FT), and the Standardized Streamflow Index (SSI). Our results clearly show that streamflow droughts derived from different approaches deviate from each other both in occurrence and timing, associated with different climate regions across Europe. The occurrence of FT drought is higher than droughts based upon VT and SSI, which highlights the importance of seasonality. FT drought happens earlier in the year than droughts obtained from VT and SSI. The use of aggregating daily streamflow data into monthly time windows for forecasting drought, such as the application of 30-day Moving Average (30DMA), is recommended to identify the VT and FT droughts. This approach will eliminate the undesired minor drought events, which are identified when using non-aggregated daily flow data. There is no unique hydrological drought definition that fits all purposes, hence developers of DEWS and end-users should clearly agree among themselves upon a sharp definition on which type of streamflow drought is required to be forecasted for a specific application. [ABSTRACT FROM AUTHOR]