The agricultural community and decision makers require tools to reliably predict crop yield, to assess optimal management practices and economic impacts. UVMRP is currently developing the Climate-Agroecosystem-UV Interactions and Economic (CAIE) system with collaborators from the University of Maryland and Colorado State University. CAIE system that will couple an advanced regional climate model with models of crops and ecosystems and economics, integrating the data from the monitoring network, the results of the effects studies, and satellite observations. This system will allow studies on how climate and crop production interact and how the interaction impacts economics and management practices.
CAIE is an integrated assessment system that simulates climate, UV radiation, crop growth, and economics. It will be able to predict how crop yield and quality will respond to changes in environmental including temperature, moisture (drought), nutrients, UV-B radiation, CO2 concentration, aerosols and other air pollutants. It will be capable of simulating agriculturally important crops (such as cotton, corn, soybean, wheat, and rice), forests, and rangelands. CAIE will include an economic assessment model that predicts economic growth of crop agriculture (total factor productivity change) based on climate. The system will have the ability analyze policy, land use, and management practices. Ultimately, the system will provide the science support for U.S. policy makers to not only establish necessary incentives and safety nets for producers, but also to assess potential risks, determine optimal practices, design effective policies, and identify mitigation and adaptation strategies to achieve sustainable development of agriculture.
CAIE has the following components:
- CWRF (Climate-Weather Research and Forecasting model) is a state of the art model that comprehensively simulates the processes behind regional climate and weather. CWRF runs on horizontal grids at U.S. nationwide scale and is capable of simulating terrestrial hydrology, precipitation, and radiation, all of which are critical for crop growth.
- CSSP (Conjunctive Surface-Subsurface Process model) simulates canopy effects, soil temperature/moisture distributions, terrestrial hydrology variations, and land-atmosphere exchanges of water, heat, and moment fluxes.
- LDAS (NASA Land Data Assimilation System) assimilates best available observations to produce spatially and temporally consistent land-surface model datasets, intended to reduce the errors in key variables of climate and weather models.
- DSSAT (Decision Support System for Agrotechnology Transfer) is a crop model capable of simulating many species individually, based on the genetic characteristics of each species.
- GOSSYM (a shortened scientific name for the genus of cotton, Gossypium) is a mechanistic model developed by USDA to simulate cotton growth given soil, weather, and management practices.
- DayCent-UV (Daily version of Century model with UV module) is a modified version of a widely used terrestrial ecosystem biogeochemistry model DayCent, which simulates photosynthesis, plant production, carbon allocation, autotrophic and heterotrophic respiration, decomposition, evaporation, transpiration, phenology, disturbances such as fire and grazing, and management practices such as fertilizer use and irrigation.
- TUV (Tropospheric Ultraviolet and Visible radiation model), is a well-tested radiation transfer model developed by National Center for Atmospheric Research.
- UV-Canopy (3-D Canopy UV radiation transfer model) is a canopy level radiation transfer model that predicts UV-B within and below a canopy.
- FASOMGHG (Forest and Agricultural Sector Optimization Model – GreenHouse Gasses version) is an economic model used by EPA that simulates land allocation and the economic impacts of changing land allocation and production practices.
For more information on ongoing research, select an interest below.
CWRF
The heart of CAIE system is a climate-crop model that simulates crop growth and yield based on weather, soil, and crop physiology. The climate-crop model is centered on a state of the art Climate-Weather Research and Forecasting model (CWRF) that simulates the processes behind regional climate and weather at a national scale (Liang et al 2012a). It is capable of simulating terrestrial hydrology, precipitation, and solar radiation that are all critical for crop growth. WRF has Conjunctive Surface-Subsurface Process model (CSSP) that simulates plant canopy properties, soil temperature/moisture distributions, terrestrial hydrology variations, and land-atmosphere exchanges of water, heat, and moment fluxes. In the climate-crop model, CWRF is coupled multiple models of crops by passing information through CSSP. One of the crop model is GOSSYM (a shortened scientific name for the genus of cotton, Gossypium), a mechanistic model for cotton developed by USDA to simulate cotton growth given soil, weather, and management practices.
UVMRP has coupled the climate model CWRF with the cotton growth model GOSSYM, and compared the results against reported yield (Liang et al 2012b, Liang et al 2012c). The results matched the reported yield well, especially how the yield varied in time and space. GOSSYM was modified to incorporate our results from studies on effects of UV-B radiation on cotton growth. The coupled model has the ability to simulate cotton growth and productivity under changing environmental stress factors such as extreme temperatures, drought, CO2, and solar UV-B radiation.
UVMRP is in a process of coupling CWRF with another crop model DSSAT-corn. This requires extensive modifications and recoding of DSSAT-corn, and we are in the process of doing so. Once coupled, they will be tested by retrospectively simulating crop yields over various U.S. regions with historical climate data. We will compare the simulations with both observational data and model simulations from other studies to attribute crop yields to key environmental factors and stressors, including temperature, precipitation, soil moisture, UV-B radiation, and CO2 fertilization.
The climate-crop model must accurately simulate UV-B radiation that the plants experience, and UVMRP will incorporate the models that simulate UV-B radiation passes through the atmosphere (TUV) and through plant canopies (CUV). TUV (Tropospheric Ultraviolet and Visible radiation model) is a well-tested model developed by National Center for Atmospheric Research, and CUV (3-D Canopy UV radiation transfer model) is a model that predicts UV-B radiation within and below a plant canopy. Once integrated into CWRF, TUV and CUV will be able integrate near-real time meteorological conditions and NASA satellite assimilated data to retrieve UV-B radiation covering the entire U.S. from 1979 onward. We will test them against UVMRP data.
Liang, X.Z. M. Xu, X. Yuan, T. Ling, H.I. Choi, F. Zhang, L. Chen, S. Liu, S. Su, F. Qiao, Y.X. He, J.X.L. Wang, K.E. Kunkel, W. Gao, E. Joseph, V. Morris, T.-W. Yu, J. Dudhia, and J. Michalakes. 2012a. Regional Climate–Weather Research and Forecasting Model. Bulletin of the American Meteorological Society, 93, 1363–1387, doi:10.1175/BAMS-D-11-00180.1
Liang, X., M. Xu, W. Gao, K. R. Reddy, K. Kunkel, D. L. Schmoldt, and A. N. Samel. 2012b. A Distributed Cotton Growth Model Developed from GOSSYM and Its Parameter Determination. Agronomy Journal, 104(3), 661-674. doi: 10.2134/agronj2011.0250
Liang, X., M. Xu, W. Gao, K. R. Reddy, K. Kunkel, D. L. Schmoldt, and A. N. Samel. 2012c. Physical Modeling of U.S. Cotton Yields and Climate Stresses during 1979 to 2005. Agronomy Journal, 104(3), 675-683. doi: 10.2134/agronj2011.0251
Wang, X.L., W. Gao, J. Slusser, K.R. Reddy, Z.Q. Gao, and M. Xu, 2006, Preliminary results of a UV-B effect incorporated GOSSYM Model. In Remote Sensing and Modeling of Ecosystems for Sustainability III, 62980O-1—62980O-10, Published by SPIE, Bellingham, WA, USA
DayCent-UV
UVMRP will develop an UV version of DayCent. DayCent is a well-tested model that simulates variety of ecosystems including grasslands of western U.S. that receives high doses of UV-B radiation. In such places UV-B radiation and visible light can accelerate litter decay in a process called photodegradation, possibly releasing more nutrients from the litter to be taken up by plants or be lost from the system through runoff. We will develop an UV-B module in the DayCent model to incorporate photodegradation. The DayCent-UV model will be calibrated and validated against the experimental data from publications, 10-year litter decay data, and other observed ecosystem variables at multiple western U.S. sites.
UVMRP will couple DayCent-UV with CWRF to examine the effects of UV-B on ecosystems and how they in turn impact climate. The procedure here will be similar to those for DSSAT. We will extensively modify and recode DayCent-UV to be compatible with CWRF. We will also develop a method to incorporate management strategies from national inventories.
Economic Model
Policy makers and decision makers in agriculture must evaluate the economic consequence of agricultural production under climate change at the regional or national level. UVMRP will thus incorporate into CAIE an economic model FASOMGHG (Forest and Agricultural Sector Optimization Model – Green House Gasses version) used by EPA to simulate land allocation and the economic impacts of changing land allocation and production practices. FASOMGHG simulates the biophysical and economic processes that determine technical, economic, and environmental implications of bioenergy production, climate change and policy intervention. We will expand it to reflect municipal and energy sector demands for water, considering aquifer depth, pumping cost, wind energy, and solar power.
The economic dynamics also reflect the changes in resource allocation and energy and land use at relatively longer time and spatial scales than represented in a regional model. We will thus develop a feasible approach that links CWRF-CROP predicted climate and crop yield distributions at the county level to the statistical economy model of aggregate agricultural productivity at the national level. We will look for the correlation between total factor productivity change (TFPC) and key climate indices and gain the physical understanding of their relationships. We will then develop a multivariate model for TFPC and climate indices to predict TFP growth, and apply this regression model to regional climate changes projected by the CWRF-CROP model to predict the potential trends of future U.S. agricultural TFP. The outcome of this research will lead to an interdisciplinary approach for developing an integrated system model infrastructure CAIE to achieve a credible and quantitative assessment of key stress factors, climate feedbacks and economic impacts for U.S. agriculture, and consequently predict the likely changes in agricultural productivity in a changing climate.