Seasonal drought prediction for North America is currently limited to the impacts of tropical Pacific sea surface temperature (SST) anomalies in the winter half year. Predictability is small in winters without strong El Ni˜nos or La Ni˜nas and in the summer half year. However, droughts are year round events and summer droughts can be extremely costly via impacts on agriculture. Does this situation represent the limits of predictability or could predictability be improved based on atmospheric responses to other tropical SST anomalies? Further, could prediction on the basis of tropical Pacific SST anomalies be refined by taking account of the differing patterns of anomalies and the potential different patterns of atmospheric and precipitation response? These questions will be addressed in this proposal. The work proposed combines two thrusts. The first is an extensive set of numerical experiments with comprehensive atmosphere models and diagnostic models. 100-member super-ensembles of climate model simulations will be generated in which various realistic Pacific and Indian Ocean SST anomalies are switched on, creating dis-equilibrium, and the day-by-day transient adjustment of the atmospheric circulation and precipitation back to statistical equilibrium is analyzed. This modeling approach allows establishment of cause and effect in the response to SST anomalies in a way that is impossible by analyzing observations, or models forced with realistically evolving SSTs, where the atmosphere is always in statistical equilibrium with the SSTs. At least two different climate models will be used. The transient adjustment in the climate model simulations will be fully diagnosed using nonlinear storm track models, a linear stationary wave model and a linear quasi-geostrophic model of transient eddy propagation. In this way the transient eddy-mean flow interactions that connect tropical SST anomalies and the precipitation over North America will be fully understood. Imposed SST anomaly patterns will include eastern Pacific centered El Ni˜nos and La Ni˜nas, El Ni˜no-Modoki patterns centered in the central Pacific, and Indian Ocean anomalies. Ground truth will be analyses of observed relations between the SST anomalies and reanalyzed circulation and instrumental precipitation fields. The work will allow a more comprehensive understanding than available to date of how tropical Pacific and Indian Ocean SST anomalies can impact North American precipitation and drought, what matters in the SST anomalies, what the strength of the relations are, how this depends on season, and, most importantly, what the physical mechanisms are that couple the mean and transient atmospheric circulation and the moisture budget. The second thrust will apply the lessons learned to analysis of two specific great North American droughts: the post 1998 drought that still goes on but with a focus on the 1998 to 2004 period, and the early to mid 1890s drought which was a turning point in the economic and agricultural development of the American West.
Relevance to NOAA’s goals: The proposal will help fulfill aims of the MAPP program on “Research to Advance Understanding, Monitoring and Prediction of Drought” with stated goal of “understanding predictability of past droughts over North America” via an intensive modeling and observational study of general predictability offered by tropical SST anomalies and detailed analysis of two historical North American droughts and will help advance NOAA’s goal to provide useful support to the social challenge of dealing with “climate impacts on water resources”.
Relevance to society as a whole: Improved drought prediction could potentially be used to guide adaptation strategies and minimize the tremendous costs to individuals, organizations and the entire nation that currently occur when drought strikes and water becomes limited for agricultural, municipal and other uses. Such advances require a clear assessment of what the limits of predictability are and what cannot be predicted in advance and must remain as inevitable surprises.