Earth is a unique, living planet due to its abundance of water and the vigorous cycling and replenishing of water throughout our global environment. Water is indeed abundant in our environment, but it is neither evenly distributed over the globe nor always in a form amenable for human use. The increase of human settlements in flood-prone areas amplifies the hazards associated with intense precipitation events; population growth and rising living standards increase the risk of severe water shortages. Indeed, the public list concerns about water availability and quality above all other environmental issues.

Because the water cycle is central to the process and consequence of climate change, we must seek to advance water-cycle observation, scientific understanding, and ultimately its prediction to enable society to cope with future climate adversities. We must determine the predictable changes in the water cycle, particularly precipitation and replenishment of water resources that are associated with observed and projected global climate changes. To achieve this objective, we must integrate and interpret past, current and future space-based and in-situ observations that are global in scope and instill improved understanding and information into global prediction systems.

The first-order requirement of water-cycle science is an ability to quantify the rate of the water cycling throughout our global environment. To do this, measurements of global changes in the vertical fluxes of water (precipitation and evaporation), the amount of water stored in various land reservoirs (soil moisture, inland water bodies, possibly total ground water mass), and lateral fluxes of water over land (river flow) are needed. These observations will provide the basis for: diagnostic studies of trends, transient variability and predictability; model validation; and initial-value data for numerical prediction.

The ability to produce believable probabilistic forecasts of rainfall and snowfall at relevant time and space scales is the linchpin of all potential applications of climate change science for the protection of human health and assets, long-term water system management, and infrastructure planning. Yet, predicting the variability in the amount and distribution of precipitation beyond the range of deterministic weather forecasts still represents a formidable challenge for climate models.

Therefore, the next requirement of water-cycle science would be to advance climate models so that they faithfully represent processes which control the replenishment of atmospheric water vapor – the surface energy budget and the resulting evaporation from the surface of the oceans and land – and processes which govern precipitation - weather systems, condensation, atmospheric radiation and cloud dynamics. Meeting this goal will require major model improvements beyond the current state-of-the-art, an equally major investment in supporting computer facilities, and more importantly, global-wide measurements and analyses of the atmospheric, land, and ocean processes that control the global water and energy cycles. The principal challenge is the vast difference in spatial scales between atmospheric phenomena that control the planetary radiant energy budget and those that govern moist processes.