Introduction
Crop growth modeling is an efficient and feasible technique to investigate the effects of enhanced UV-B radiation combined with other changing abiotic and biotic factors such as high temperature, water stress, and increased CO2 concentration as predicted in global warming. Coupling crop growth models with climate models makes it possible to comprehensively assess the risks of agriculture production due to climate change and UV-B enhancement. While crop modeling is helpful for agricultural management and assessment at a specific site, remote sensing techniques and data are especially useful to extend these studies and applications to a regional and global scale or to where ground observations are not available. Our recent research goal is to integrate the existing crop growth models (e.g., GOSSYM, DSSAT), climate models (e.g., CWRF), radiation transfer models (e.g., TUV), ecological models (e.g., GeoPro), soil process models, and hydrological models along with the application of remote sensing techniques and datasets to develop a comprehensive crop impact assessment system.
Development of an Integrated Crop Impact Assessment System
Understanding agricultural response to climate and environmental changes is privotal for providing decision support to stakeholders, such as agricultural producers, land managers, and policy makers trying to meet the food and fiber needs of a growing population. A fully integrated impact assessment system is needed to predict climate-crop interactions and manage sustainable agriculture production for high profitability and environmental quality. Such a system must be sufficiently comprehensive to include crop growth models, ultraviolet-visible solar radiation, Earth's climate, air and water quality models, and satellite and in-situ observations. Our goal is to develop an integrated crop impact assessment system capable of achieving credible and quantitative assessments of key stress factors and evaluating alternative cultural practives for systainable agriculture production.
The primary objectives of the assessment system are to:
- Provide a state-of-the-art prediction and assessment tool tailored to the agricultural community and deliver credible products that have spatial and temporal resolution better than currently available, with the ultimate target at the farm level;
- Integrate, as a fully coupled system, existing models that are currently being implemente independently;
- Incorporate more fully remotely sensed data from satellites together with ground measurements to improve the system predictability and scale;
- Include direct and indirect impacts of air quality on regional climate and crop growth variations--an impact that has not been properly addressed to date;
- Couple comprehensive terrestrial hydrology and water quality modeling capabilities to assess water quantity and quality regulation in response to emerging agricultural management strategies; and
- Develop extensions to economics and policy so that there is a strong feedback loop between the users of the assessment information and scientists developing the assessment system.
Remote Sensing Application
Works have been conducted or have been ongoing on various applications of remote sensing techniques and datasets, including:
- Computation of net primary productivity using remote sensing techniques and datasets.
- Comparison of ground measurements of erythema-weighted UV radiation with TOMS data retrievals.
- Spatial distribution of UV radiation, PAR, and temperature in the United States.
- Grid values of input data for climate models.
GeoPro Model
GeoPro model (Geo_Process model based on the theory of ecosystem and photosynthesis) is a terrestrial ecosystem process model developed from the existing models of Biome_BGC, BEPS, Century, CLM, etc. Biogeochemical mechanism is employed in this model. Using the chemical, physiological and physical approaches, along with the ecological, biogeochemical, phonological, and hydrological processes, the model simulates the photosynthesis, autotrophic respiration (RA), allocation of carbon, evaporation (E), transpiration, radiation balance.
The GeoPro model includes the radiation transferring module, carbon module, water balance module, and photosynthesis- respiration module.
The output of the GeoPro model includes net primary productivity (NPP), evaportranspiration (ET), gross primary productivity (GPP), soil carbon(SC), vegetation carbon(VEGC), and radiation.
Modeling UV-B Effects on Crops
Crop modeling provides a feasible approach to study the comprehensive effects of multiple factors. Various crop models have been developed to simulate the process-level responses of crop growth, development, and yield to governing environmental variables, including temperature, solar radiation (visible to infrared), water stress, and nitrogen supply. Algorithms have been developed for UV-B indices of cotton growth and have been incorporated into the GOSSYM cotton growth model. The model evaluation results showed a satisfactory agreement between measured and simulated results. The verified model can then be used to simulate the impacts of UV-B radiation on cotton yield under multi-stress conditions in the field. Taking the data at Stoneville, Mississippi, USA from 1964-1993 as an example, the results showed a significant reduction in cotton yield when UV-B irradiance exceeded 12 kJ m-2. The cross effect of UV-B and water stress on cotton yield is evedent. Cotton yields declined by 20% at 12 kJ m-2 of UV-B radiation under irrigated conditions. Similar reduction occurred at a lower UV-B radiation level (11 kJ m-2) under rain fed conditions. On an average using both management conditions, a 50% reduction in cotton lint yield would occur at UV-B irradiance of 14 kJ m-2 which is projected to occur when stratospheric ozone declines by 30%
Development of CWRF Climate Model
With joint support from the USDA UV-B Monitoring and Research Program, Dr. Xin-Zhong Liang's group at the Illinois University at Urbana-Champion has developed the CWRF (Climate Weather Research and Forecast) model for regional climate studies and forecasting. It is a climatic extension of WRF (Weather Research and Forecast) model. This extension incorporates all WRF functionalities for NWP while enhancing the capability for climate applications. Hence the CWRF can be applied for both weather forecasts and climate predictions. The CWRF model will be the primary climate model in our Integrated Crop Impact Assessment System as describe in a previous section. Crop models, including GOSSYM, cotton2k, and DSSAT, as well as radiative transfer models, such as TUV, will be coupled with the CWRF climate model. Hydrological models, soil process models, and ecosystem models will also be integrated under the framework of CWRF model. Further developments of CWRF model to improve the simulation results are under going.
Modeling UV-B Radiation in Row-Crop Canopies
To determine the physiological effects on plants of any increases in UV-B radiation, the irradiances at the potential sensitive plant surface need to be known. A number of radiative transfer models for plant canopies but because of the importance of sky diffuse radiation to the global UV-B irradianc, models designed to estimate photosynthetically active radiation or total solar radiation may not accurately model the UV-B radiation. We compared spatially and temporally averaged measurements of the UV-B canopy transmittance of a relatively dense maize canopy (sky view: 0.27°) to the estimations of two one-dimensional models differing mainly in the handling of sky radiance. The model that considered the distribution of sky radiance tended to underestimate the canopy transmittance, the model that assumed an isotropic sky radiance distribution accounted for only about 0.01 of the model error. Consequently, the sky radiance distribution is probably not important in modelin such dense crop canopies. The model that overestimated transmittance and had the generally larger errors, a modified Meyers model, used the assumption of uniform leaf angle distribution, whereas in the other model, designated the UVRT model, leaf angle distributions were estimated by sample measurements. Generally this model would be satisfactory in describing the statistically average UV-B irradiance conditions in the canopy. This model may also be applied to other dense plant canopies including forests.
This model, developed for describing the vertical distribution of UV-B radiation in dense row-crop canopies, will be integrated into our Crop Impact Assessment System to evaluate the physiological effects of UV-B on plants. The UV-B irradiances at different levels in the canopy can be calculated with an acceptable accuracy using this model with the input of UV-B irradiance at the top of the canopy and the structure of the canopy.
Daily Collumn Ozone Retrieval
Studies have shown significant effects of ozone on plants and dominant absorption of UV-B by the atmospheric ozone.
A model for direct-Sun collumn ozone retrieval using the ultraviolet multifilter rotating shadow-band radiometer (UV-MFRSR) has been developed. Compared with the measurements of collocated Dobson and Brewer, the total uncertainty of ozone retrievals is 2.0% at the three observation sites, Mauna Loa, Hawaii, USA (19.539°N, 155.578°W), Toronto, Ontaio, Canada (43.78°N, 79.47°W), and Regina, Saskatchewan (50.197°N, 104.7°W) during four months in 1999. Software has been developed to compute total column ozone routinely. The advantages of total vertical column ozone retrieval using UV-MFRSR include relatively low cost, computer-controlled operation, automated calibration stability checks, and minimal maintenance. It allows for the real-time measurement of total vertical column ozone. The UV-MFRSR is being used at 35 sites across the Unitied States, two sites in Canada, and one site in New Zealand that form the USDA UV-B Monitoring and Research Program. This constitutes a unique network of total vertical column ozone measurements.
Home | Mission | Research | Publications | Collaboration | Outreach 