Climate, water policy and management Since about 1980 there has been a distinct change in our understanding of the nature and origin of the statistics of hydrological variables, as measured in an individual catchment or region. Previously the assumption was that these statistics are entirely haphazard in nature and indeterminate in origin, and do not change with time. Thus the most important hydrological variables (such as precipitation, runoff and potential evaporation) are sampled over a calibration period (of perhaps only a few decades), and the statistics observed within that period are then used as the basis for hydrological design and water resources management. Now, however, there is increasing realisation that the nature of the locally observed statistics of hydrological variables is not stationary and may contain long-term trends caused by global-scale phenomena.At the seasonal to interannual timescale, the influence of El Niño and La Niña on hydrological statistics (and the occurrence of extreme hydrological events such as floods and droughts) is now well recognised – even catchments remote from the Pacific may be affected. There is also observational evidence of a relationship between the strength of the Asia-Australian monsoon and El Niño, and indications that these phenomena are together related to seasonal variations in Siberian snow cover. Similarly, recent studies suggest an association between the North Pacific Oscillation and precipitation in Europe and the Middle East. These relationships (and others yet to be identified) can generate seasonal distortions in the statistics of hydrological variables, thus threatening the validity of the operational rules applied to water management systems.There are indications that the strength of important fluctuations in the global climate (such as those associated with El Niño and the Asian-Australian monsoon) may themselves vary at the decadal timescale, which brings into question hydrological designs based on observations made over 30 years or less. Moreover, model studies suggest, and observational evidence tends to confirm, that an enhanced hydrological cycle is likely to be an important consequence of global climate change caused by "greenhouse warming". Some developed countries now have the capability to use models and data gathered with advanced technologies (such as remote sensing) to improve the prediction of the impact of multiple stresses present in individual catchments. Such improved management tools are, however, rarely applied in the extensive regions of the world where water-resource issues are most extreme and where their potential benefit for human welfare is greatest. Thus, it is clear that the basic paradigm, that is, the assumption of stationarity that underlies hydrological design and management (e.g. flood management), is open to question, but, in the absence of reliable alternative understanding and methods, current practice is locked in place by professional and legal precedents. There are now huge opportunities to develop hydrological understanding relevant to these policy issues. The past success of the scientific community now involved in the Global Energy and Water Cycle Experiment (GEWEX), the International Geosphere Biosphere Program (IGBP), and the Climate Variability and Predictability programme (CLIVAR) engenders optimism. It is likely that new and beneficial understanding of the Earth's hydrological cycle will emerge in the course of the next decade under the World Climate Research Program. Remote-sensing systems are now better able to provide global observations to monitor fluctuations and change in the Earth's atmosphere, oceans and continents. Field measurements using reliable, unsupervised hardware with remote data capture is now feasible. Meanwhile, the explosive growth of computer technology promises the capability to describe the entire globe with models having a grid scale of just a few tens of kilometres within a few years. Further, it has fostered a revolution in information transfer, bringing the capability to transfer data and knowledge at unprecedented rates.HELP will complement the global data that GEWEX and CLIVAR will provide with simultaneous, in situ hydrological observations in representative research catchments around the world. A particular focus of attention will be on extreme events (floods and droughts). An education programme is also required, to promulgate the use of modern hydrological monitoring and data transfer techniques and to disseminate the understanding and application of the relationship between global processes and regional hydrology.

The hydrological science contribution The overarching question that motivates research into water and climate is:

how can knowledge, understanding, and predictive modelling of the influence of global variability and change on hydrological variables and remotely sensed data be used to improve the management and design of water resource, agro-hydrological and eco-hydrological systems?

• How significant is the relationship between the statistics of hydrological variables and observable global phenomena, and how does this change with location?
• How can remote data capture, and advanced information transfer technologies best be applied to improve the management and design of water systems?
• How can predictions of seasonal-to-interannual variations be used to improve the management of water, including for disaster prevention (floods and droughts)?
• How significant are multi-decadal fluctuations in climate, and how can knowledge of such fluctuations be used to improve the design of water systems?
• What is the hydrological significance of potential anthropogenic climate change, and how can predictions of such change best be used to improve design of water systems?