eWaterCycle

If you would know that ten days from now a flood will be at your doorstep, what would you do? You would have time to prepare emergency measures but also move your belongings to higher grounds. It might even be possible to sacrifice inexpensive farmland to save a central business district. Ten days can make a major difference. This is exactly the aim of the eWaterCycle project: to predict flood and drought events 10 days in advance, worldwide and in high resolution. The featured movie shows a setting in which the city of Mandalay is saved from a hypothetical flood. The model can act as an early warning system and enable companies, people and governments to prepare for the rising water.

With climate change, growing population and increasing pressures on land usage, water management is quickly becoming one of the world’s major problems. Increased urbanization of delta areas in particular is making more and more people vulnerable to flooding.

The Dutch have historical expertise and interest in water management. In the eWaterCycle project, researchers from Delft University of Technology, Utrecht University, and the Netherlands eScience Center are cooperating to build a high resolution, realistic model of the World’s supply of fresh water. Using this model it will not only be possible to build a flood early warning system, but also predict the effect of unsustainable water supply usage, provide support to local governments in making decisions on water protection measures, and provide other information vital to the World’s population.

The development of a high resolution global hydrological model has recently been put forward as the Grand Challenge for the hydrological community. So far, the hydrological community has not made full use of the possibilities recent trends in the availability of computational power offer. Current global hydrological models just lack the resolution required to make accurate predictions and adequately support decision making.

Even more challenging than the refinement of the grid will be the assimilation of the massive amounts of available satellite data. In order to ensure proper parameterization of such a model, massive assimilation of remotely sensed data is needed. All of this data needs to be incorporated into a very high resolution simulation, resulting in a model which can only be run on large supercomputers. Updating a global hydrological model with Earth observations will be a major computational challenge that demands close cooperation between experts from the domains of ICT and hydrology. This project aims to exploit high performance computing infrastructures to run these high resolution simulations. Ultimately, these models will help in making the world a safer place, saving both lives and money.