Background
The author writes that "despite their obvious environmental, societal and economic importance, our understanding of the causes and magnitude of the variations in the global water cycle is still unsatisfactory," noting that "uncertainties in hydrological predictions from the current generation of models pose a serious challenge to the reliability of forecasts and projections across time and space scales."

What was done
Hoping to illuminate the way forward in this regard, Martin states that the purpose of his paper is to provide "an overview of the current issues and challenges in modelling various aspects of the Earth's hydrological cycle." And he indicates, in this regard, that "the challenge to quantify and reduce uncertainty in the large-scale response of the global water cycle is immense, not least because the hydrological cycle involves almost every component of the climate system," which he subsequently goes on to identify and describe in more detail.

What was learned
In the words of the UK's Met Office Hadley Centre researcher, he identifies the essential components of the way forward as including "the global water budget and water conservation, the role of model resolution and parameterization of precipitation-generating processes on the representation of the global and regional hydrological cycle, representation of clouds and microphysical processes, rainfall variability, the influence of land-atmosphere coupling on rainfall patterns and their variability, monsoon processes and teleconnections, and ocean and cryosphere modelling."

What it means
In light of the many aspects of this broad undertaking, Martin concludes that "continued collaborative activity in the areas of model development across timescales, process studies and climate change studies will provide better understanding of how and why the hydrological cycle may change, and better estimation of uncertainty in model projections of changes in the global water cycle," all of which elements of the gigantic task suggest that we still have a long, long way to go before we will be capable of producing projections of the global water cycle that have even a modest degree of reliability. So until that time, somewhere in the future, we need to take current model projections of the world's water cycle with a good bit of scepticism.