Using paleohydrology to inform water management and constrain climate models

Dr. David Sauchyn

Wednesday, August 21, 8:30-9:15

Dave Sauchyn is Professor of Geography and Environmental Studies at the U of R and Director of the Prairie Adaptation Research Collaborative. His research interests are variability in prairie hydroclimate over the Common Era, and planned adaption to minimize the adverse impacts of climate change. Dave has given more than 500 invited talks on aspects of climate variability and change. He can trace his prairie roots back to the 1910s when his grandparents arrived from eastern Canada and from Ukraine.

Abstract: The use of instrumental hydrometric data to estimate water supplies and the probability of extreme conditions relies on the critical assumption that observations are stationary over time. The prominent hydrologist Vit Klemes wrote in 1989 “It is of course well known that this assumption is not true in general and that it diverges from reality with the length of the period considered.” Research in paleohydrology highlights this disconnect between reality and stationarity. It provides context for the interpretation of reference hydrology and climate model projections and for understanding of the links between atmosphere-ocean dynamics and regional hydroclimatic variability. The value added by a longer perspective, and the fundamental necessity of water for life, gives all research in paleohydrology societal relevance; however, relatively few paleo-hydrologists have collaborated directly with water managers to influence water resource policies or management practices.

Ice cores, banded corals, laminated speleothems, tree rings and varved sediments preserve annual hydroclimatic signals. Among these proxies, tree-rings have the advantage of absolute annual resolution, wide distribution and massive replication. Water balance variables are inferred from radial tree growth at dry sites, where it is limited by the availability of soil moisture. Continuous time series of annual/sub-annual tree-ring data can be processed using the methods employed by engineering hydrologists, including uncertainty analysis and disaggregation of proxy data in space and time. Much of this research has shown that existing water policies and management practices are based on an underestimation of the severity of excessively wet and dry conditions. Recent studies, however, have reached a different conclusion; that extreme events in early 21^st century are unprecedented in the paleo record. This recent shift in the outcomes and interpretation of dendro-hydroclimatology suggests an emerging signal of global climate change and the anticipated hydrologic response to greenhouse gas warming.

When climate proxy and model data are compared, new insights arise about the modes and drivers of climate variability and whether they are adequately simulated by numerical climate models. The models tend to be more energetic at high versus low frequencies relative to instrumental and proxy hydroclimatic data, implying that climate projections underestimate the future risk from persistent droughts and pluvials. Research in paleo-hydroclimatology also addresses an important distinction between cyclical variability and incremental long-term climate change and key practical questions about whether observed and projected climate changes are beyond the range of natural variability. Trends in short instrumental records (i.e., non-stationarity in mean values) can be falsely perceived as climate change if the signal of incremental change is relatively weak. Natural variability masks the signal of climate change and contributes to substantial uncertainty in model simulations especially at mid- and high-latitudes and for precipitation-related variables.