The catchment studies also showed that the ensemble mean of the large-scale models (running with the WATCH Forcing Data) reproduced drought characteristics similar to more detailed hydrological models that were using local forcing data.
The success of the large and small scale studies gives us confidence that the hydrological models will produce meaningful extremes of runoff when they are applied to the 21st century using the WATCH Driving Data. Comparing the outputs with the atlases will then reveal how the location, timing and severity of extremes will change in the future.
The detailed study of droughts such as those in 1976 and 2003 helps us to understand how they developed over an area, over time, and in intensity. Droughts are highly complex, but tools are being developed that will recognise historical patterns. This will allow us to forecast how large-scale droughts might spread in the future. Given the slow rate at which droughts develop, such tools would have operational applications. The tools will also allow us to anticipate the occurrence of simultaneous droughts across the world. Such occurrences, for example in the world’s main food producing regions, would exert considerable strain on world food security.
WATCH has also identified the best methods for drought characterisation on a global scale and made recommendations for standardisation to allow inter-comparisons of drought characteristics across the world.
WATCH has made significant progress in understanding and recording hydrological extremes in the 20th century and has provided the clearest evidence yet that it is possible to model these both on a European and global scale. However, models need to be improved further, and WATCH has demonstrated that such improvements are accelerated by the availability of comprehensive and up-to-date observed data. Europe has a dense network of river flow measuring sites, but needs the data from these sites to be made more readily available, preferably through a single outlet. A system is already in place for flood warning at JRC in Ispra. This network could be expanded – an evenly spread network of 500 stations would be a starting point – to include sites that are suitable for drought studies. If the momentum of WATCH is to be maintained, then this matter needs to be addressed swiftly.