Precipitation extremes with climate variability and change

Significant gaps exist in our understanding of hydro-meteorological processes in the context of climate variability or change. However, despite the uncertainties, developing relatively credible insights for precipitation extremes at scales relevant for hydrology is necessary and may be possible. Statistical analyses of observed and model-simulated precipitation data, particularly methods based on extreme value theory, have demonstrated the potential to yield new insights. Specifically, a delineation of the impacts of global climate change versus regional changes in land use or urbanization may be possible and could be important for policy-makers. Precipitation extremes have known dependence on variables like sea surface temperatures, atmospheric temperature profiles and wind velocities, some of which may be better predicted than precipitation from models, exhibit less variability in observations and may not be as subject to thresholds and intermittences in either models or observations. Thus, leveraging the information content in these auxiliary variables through data mining or network science based approaches, especially if the techniques are informed by process understanding at multiple scales, may help improve regional projections of precipitation and corresponding extremes. Enhanced regional projections of precipitation and their extremes can help drive models of hydrology and hence better inform water managers, especially at scales that matter for water resources planning or managing hydraulic infrastructures. A combination of physics-based models, data-guided mathematical approaches, and quantitative techniques informed by conceptual process understanding, may be a way forward to understand the possible consequences of climate variability and global or regional change on precipitation extremes. Examples and case studies are presented from the published literature and from ongoing research.