Integrating Climate Models and Impacts
- Continuing Projects
- New Projects Under Way during FY98
Climate Change Scenario Development over the Great Plains with an Application to Crop Models
Mearns, along with McDaniel, Tsvetsinskaya, and Theodoros Mavromatis (University of East Anglia, UK) completed work on the National Institute for Global Environmental Change (NIGEC) project, "Development of a Nested Regional Climate Change Scenario with an Application to Crop Climate Models." The project involved regional modeling with RegCM2 by Giorgi and Christine Shields of CGD and detailed model evaluation and application to crop models by Mearns and colleagues at U Nebraska. One goal of the project was to appropriately integrate climate modeling work, model analysis, climate change scenario formation, and application to impacts models.
Back to Top
Regional Modeling Work
- Provision of Climate Change Scenarios. The NIGEC-funded project has provided climate change scenarios for other researchers nationally. Several researchers have been provided with mean change climate scenarios for the entire domain. The runs are now being used as additional climate change scenarios for the US National Assessment for the Basin and Range Region and the Southwest Region.
- A comparison of regional climate model results with a semi-empirical statistical downscaling technique (developed by Bogardi and colleagues at U Nebraska) for the Central Great Plains has been completed. Results indicate that climate changes in the Regional Model are more pronounced than those calculated via empirical downscaling of the GCM results. Different directions of change of climate, particularly for higher-order moments, are found in the two models. A paper describing these results will appear in a special issue of the Journal of Geophysical Research on the Regional Climate Model, RegCM.
Back to Top
Crop Modeling Work
- Determining the effect of the spatial scale of climate change scenarios on simulation of agricultural impacts of climatic change. They developed a coarse-scale climate change scenario from the General Circulation Model (GCM) used to drive the regional climate model and a high-resolution scenario generated from control and 2xCO2 results of the Regional Climate Model (RegCM2). They then applied the climate change scenarios to the EPIC corn, wheat, and soybean crop models, for GCM grid boxes in the central Great Plains. They found that different scale scenarios produced substantial differences in the impacts of climate change on some of these agricultural crops.
- They also investigated uncertainty in agricultural impacts assessment based on using different impacts models. They ran the CERES-wheat and corn models for the same coarse and fine scenarios described above and found that the CERES models produced substantial differences from those produced by the EPIC model. Detailed analyses of why these differences occurred were performed. A paper describing these results is in press in a special issue of JGR (Journal of Geophysical Research) on uses of the Regional Model RegCM.
Back to Top
Climate Variability and Agriculture in the Southeast US
Work on two overlapping three-year projects (NASA/USEPA/USDA)has continued by Mearns, Katz, McDaniel, Tsvetsinskaya, Mavromatis, Gregory Carbone (U South Carolina), Bette Walker-Shea (U Nebraska), and William Easterling (Pennsylvania State U). Regional climate modeling, conditioned stochastic modeling, and transient coupled GCM runs form the basis of three different types of climate change scenarios. Remote sensing, crop and economic modeling, and spatial scaling analysis make up the other elements of the projects.
Major accomplishments in the projects this year have included: (i) Development of a seasonal observed daily temperature and precipitation data set for the entire Southeastern US, gridded on a .5-degree grid; (ii) Development of two different soils data bases; (iii) Detailed site validation of wheat, soybean, sorghum, and corn CERES models; (iv) Advanced work on remotely sensed AVHRR and SPOT data; (v) Analysis of observed temperature and precipitation data sets in the Southeast US for development of stochastic models conditioned on ENSO phases (see section on stochastic modeling for further details); (vi) Generation of the second set of five-year control and doubled CO2 runs over the Southeast US with the Regional Climate Model; (vii) Preliminary multi-year runs of most crop models over the full domain, and comparison with county yields. The interannual variability of the spatial distribution of observed yields is well-reproduced by the models, especially for adjacent years when the variability is high.
|Relative change (fraction) in winter wheat yields, 1988 compared to 1987, for the southeastern US.|
A. Observed county yields;
B. CERES-Wheat simulated yields;
Back to Top
The Yangtze Delta of China as an Evolving Metro-Agro-Plex (China MAP)
A three-year project funded by NASA on this subject was pursued by Mearns, McDaniel, Giorgi, and Wei Gao (Purdue U), in collaboration with Bill Chameides (Georgia Institute of Technology). This is an international, multidisciplinary research project focusing on the effects of regional environmental change on agriculture in China, the most populous and rapidly developing nation in the world. The project includes the assessment of major pollutants (ground level ozone, nitrogen oxides, gaseous sulfur oxides) and their effects on present-day and future agricultural yields of crops, as well as the effects of particulate emissions and land-use changes on the regional climate in China and their concomitant impact on future agricultural yields. Mearns and colleagues are modeling primarily wheat and rice crops for the region, using CERES and UCLA-YIELD crop models. The CERES Wheat and Rice models have been tested for locations within the Yangtze River Delta, for a series of years. Sensitivity analyses of the crop model responses to decreased solar radiation have been performed (see figure). Decreases in solar radiation occur in the region due to heavy sulfate emissions.
Back to Top
Other Changed Climate and Yield Projects:
- Mearns and McDaniel continued the study of The Effects of Climate Variability on Forest Dieback in the Northeast US (NOAA). They are devising a model that can be used to identify the effects of changes in the frequency of ENSO and NAO (North Atlantic Oscillation) events on forest dieback. This is an integrated assessment project being coordinated with colleagues at Science and Policy Associates and Resources for the Future.
- Mearns continued the study of The Effect of Climate Change on Wheat Yields in Italy with Carlo Pona (Agency for New Technologies, Energies, and Environment, Italy [ENEA]). Climate change scenarios for Italy have been generated from output of the RegCM2 European runs. Numerous sensitivity analyses with CERES-wheat have been performed for locations in Italy; climate change crop model runs have been made; and the effects of changes in daily climate variability are being examined.
Back to Top
Evaluation of the Atmosphere/Ocean (A/O) GCMs for the US National Assessment
In a project funded by the NIGEC National Office, Mearns, with Ruth Doherty (U Wisconsin) evaluated the control runs of the two major A/O GCMs being used to provide climate change scenarios for impacts used in the US National Assessment: the Canadian CGCMI and the British HADCM2. Both the control runs and future runs were evaluated. A report has been prepared, and the results will be displayed at the National Assessment Scenarios Web site.
Back to Top
Development of Transient Runs of the RegCM2 over the Continental US
With support from CMAP, Mearns, with Eric Small (New Mexico Institute of Mining and Technology) and Filippo Giorgi (CGD/Trieste Institute of Physics), have developed a project to produce multi-year runs of the RegCM2 over the continental US at a grid point spacing of 50 km. Initial and lateral boundary conditions are being provided by Climate Systems Model (CSM) runs, including greenhouse gases and sulfate aerosols. Approximately 20 years of current climate are being produced, and 20 years along two different points in the run that extends to 2100. Currently, improvements in the RegCM are being made in preparation for these runs. These runs will form the basis of a multi-agency-supported, nationwide program focused on furthering understanding of regional climate dynamics in the US and on detailed regional impacts of climate change in the US.
Back to Top