Drought in the 21st Century


The central science question to be addressed by this proposal is: What is the susceptibility of the continental U.S. to drought over the next century, and what role is anthropogenic warming likely to play in U.S. drought susceptibility? Among U.S. natural hazards, drought is the most costly. The 1988 drought alone, in 2002 dollars, cost almost $62B, making it by far the most costly natural disaster. A number of studies over the last decade have suggested that, as a result of greenhouse warming, the interior of the northern hemisphere continents will become more susceptible to droughts over the next century. Given economic sensitivities to drought, understanding the nature of drought risk sensitivity to climate change is a question that we believe is a key national concern.

A key shortcoming of past studies of future climate and drought is that the land surface representations used in climate models, in general, have not been able to produce realistic land surface hydrologic conditions. Furthermore, past studies that have evaluated the potential for future drought have not considered the possible role of vegetation change. We propose to use the Community Climate System Model (CCSM), in conjunction with a modified version of the Community Land Model (CLM) that will incorporate a more realistic representation of land surface hydrology, to evaluate the susceptibility of the U.S. to drought over the next century. In so doing, we will address the following subsidiary questions:

a) What have been the space-time signatures of 20th century drought on precipitation, soil moisture, and streamflow, and how might those change in the 21st century?

b) What is the role of climate-vegetation feedbacks in exacerbating or ameliorating North American drought severity and intensity?

c) Can useful estimates of drought recovery probabilities be developed based on off-line and or coupled ensemble climate prediction methods?


The tasks we propose to undertake are:

Task 1: Improve land surface hydrology representation in CLM. We intend to implement the runoff generation parameterizations, the soil column representation, and possibly the snow model from the Variable Infiltration Capacity (VIC) model into CLM, which are expected to improve its ability to represent land surface hydrologic conditions such as soil moisture and streamflow.

Task 2: Perform offline tests of upgraded version of CLM. We will test the updated CLM in off-line tests using hydrologic and energy flux data from the Arkansas-Red River basin, and the Southern Great Plains CART-ARM facility within it, as well as other surface hydrologic data.

Task 3: Evaluate performance of upgraded model in climate simulations. This task will focus on the period 1950-present, for which CCSM can be run with observed SSTs and for which the University of Washington has developed estimates of the space-time characteristics of U.S. droughts.

Task 4: Simulate and evaluate projected drought characteristics. Transient climate simulations for the period 1870-2100 will be performed with the fully coupled CCSM. The simulations will be evaluated for their ability to simulate historical droughts in the U.S., which will provide the basis to evaluate future changes in U.S. drought risk in the 21st Century.

Task 5: Drought recovery evaluation. We will explore the possibility of using coupled and/or off-line simulations for future conditions starting with initial conditions representative of past or simulated future droughts, to evaluate the potential for alternate methods of estimating drought recovery.

Task 6: High resolution simulations for the U.S. We will develop a prototype high-resolution version of the CLM that runs on a finer spatial grid than the current CLM. We expect that this high-resolution version of the model, implemented on a 1/8 degree grid for the U.S., will better resolve spatial variability in surface properties and lead to better simulation of U.S. drought in the CCSM.