As the severity of floods increases, possibly due to climate and landuse change, there is an urgent need for more effective and reliable warning systems. This demand is particularly high amongst private companies and governmental agencies who are engaged in the flood protection business.

Conventional flood warning systems are no more than a dedicated user interface built upon hydrological and hydraulic models. They often fail to account for inherent uncertainties in the weather forecast. They can provide flood forecasts for a few hours in advance but are not capable of producing reliable forecasts of a few days (2-10 days) in advance. A sufficiently long lead-time is critical for evacuating people and protecting valuable assets including vital infrastructure. In countries like China with a high development growth rate and an increasing flood threat, failing to satisfy this need has grave repercussions. This unsatisfied need has also led to a market for novel climate and hydrosystems modelling approaches and tools.

Operational medium range flood forecasting systems are increasingly moving towards the adoption of ensembles of numerical weather predictions (NWP), known as ensemble prediction systems (EPS), to drive their predictions. Cloke and Pappenberger (2009) [PDF] reviewed (1) the scientific drivers of this shift towards such 'ensemble flood forecasting'; (2) the literature evidence of the 'added value' of flood forecasts based on EPS. They also discussed the questions surrounding best practice in using EPS in flood forecasting systems; and pointed to remaining key challenges in using EPS successfully. They listed key case studies (hindcasts) evaluating ensemble flood forecasting up to early 2009 (see Table).

Successes in early flood warning have been demonstrated by the NEWS KCL Team at King's College London using multiple numerical weather/climate prediction systems in conjunction with spatially distributed catchment models. The best available technology in the market now is the European Flood Alert System (EFAS) developed by the European Joint Research Centre. In comparison to the EFAS, the proposed NEWS technology has clear advantages in that it can (1) provide medium-range flood simulations with a lead-time of 3 to 10 days and (2) assess risk and uncertainties based on a grand ensemble system. NEWS will be built upon twelve global ensemble prediction systems through the 'THORPEX Interactive Grand Global Ensemble' (TIGGE) system. The grand ensemble approach makes it possible to account for uncertainties originating from initial conditions and stochastic physics, numerical implementations and data assimilation.

NEWS is capable of producing reliable early warning which represents the very best of current technology. A research prototype has been successfully demonstrated. It was used to predict a flood event in the Upper Severn catchment in January 2008. An atmospheric-hydrologic-hydraulic cascade probabilistic flood warning framework has also been established. The work was carried out by the applicant at King's College London within a UK NERC funded project that has led to a high impact journal article in Meteorological Applications published in January 2009 [He et. al. 2009]. The same framework was cloned, validated and applied in the Upper Huai Catchment in China during the ICUK Partnership Grant (PG) phase and the Proof of Concept phase [He et. al. 2010].

NEWS is also expected to contribute in the bid to develop environmental management tools to adapt to climate change in both the UK and China. As climate change is expected to herald in a new wave of extreme events, we are beginning to see emphasis on effective long-term flood prediction with quantifiable uncertainties and minimum tolerance for error. The situation sets the stage for future joint partnership especially with respect to the development of "climate sensitive" long-term flood prediction systems. NEWS will be developed into a seamless flood prediction system covering a full spectrum of lead times ranging from hours to decades.

Climate scientists in race to predict where natural disaster will strike next. Conference in Boulder will step up world's efforts to establish an early warning system for extreme weather events. Guardian observer reported the meeting to take place in Boulder, Colorado amid the devastating floods in Pakistan, the killing landslides in Gansu, China, the heat waves in Russia and many other extreme events that took place in the 2010 summer. The world's leading climate scientists will gather in the United States to hammer out plans to set up an early warning system that would predict future meteorological disasters caused by global warming.