Study On Maize Yield Cstimation Using Remote Sensing Technology Integrated With Coupled WOFOST And HYDRUS Models

Study on interactions and feedbacks between crop growth and water cycle in agro-ecosystems can provide help for increasing water efficiency of irrigated oasis, optimizing irrigation scheme and alleviating the competition between agricultural water and ecological water. Modeling interactions and feedbacks of relevant processes based on model coupling can help us to properly understand agro-eco-hydrological processes of irrigated oasis and to provide reference for water management and water saving. Realization of crop yield estimation based on coupled crop growth and hydrologic model can provide help for decision makers to ensure regional or national food security.The middle basin of Heihe River, located in arid region of northwest China, is the most developed area of Heihe river basin. However, with economic development and an increasing population, there is an increasing competition between the limited water resources and the increasing demand for crop irrigation. An increase in water efficiency of agro-ecosystems, especially irrigated agro-ecosystems in arid regions, is an urgent task. Accurate knowledge of water demand and transpiration during crop growth is critical for sustainable water management and water saving. To predict crop yield under different environmental conditions and climatic conditions, the crop growth model, WOFOST, and the vadose zone hydrologic model, HYDRUS-1D, are coupled to quantify water demand and transpiration during crop growth. FSEOPT program and SCE-UA algorithm are used to calibrate crop growth parameters and soil hydraulic parameters. The results show that the simulations agree well with the observations. The related parameter values are reasonable for local crop (maize) characteristics and soil properties in the study field. The calibrated model is then used to evaluate the water balance and to search for a potential, water-saving scheme. The ratio between actual root uptake and potential transpiration is used as an indicative factor to guide irrigation. The simulated results indicate decreasing in deep percolation can save irrigation water. Based on guided irrigation scheme and the coupled model, global sensitivity analysis (SA) method is further applied to study the effect of the coupled model parameters, climatic and environmental conditions on maize yield. The SA analysis results show that8out of33parameters (HYDRUS parameters, ZIT, SLATB1, IDSOW, EFFTB, RDMCR, KDIFIB, CFET) have great effect on output of the coupled model. The SA analysis results also suggest the coupled model has not over-parameterization. An uncertainty analysis (UA) method is used to predict probability of maize yield in uncertain environmental conditions. The UA analysis results indicate that the uncertainty analysis using Monte Carlo method can reveal the risk of a possible loss of maize yield with irrigation decrease and provide the probability of yield in the uncertainty range of crop parameters and environment parameters. The coupled model integrating with various groundwater depth and irrigation schemes can be used to evaluate the uncertainty influence of groundwater depth on maize yield. The results indicate that the irrigation demand of maize increases when groundwater depth more than2.0m. A small amount of irrigation can guarantee maize growth When groundwater depth less than2.0m. An object-oriented classification method integrated with NDVI time series data and crop phenological information is used to extract the planting distribution data of maize in middle basin of Heihe river. To apply the coupled model at the regional scale, the Ensemble Kalman filter (EnKF) is used to assimilate the leaf area index (LAI) data extracted from Lansat-ETM+imagery. As the result of assimilation, region-wide spatial distribution of the maize yield in Zhang Ye Oasis was obtained.Synthetically, the method of integrating the coupled WOFOST and HYDRUS models with uncertainty analysis and sensitivity analysis can be used for guiding agricultural irrigation, saving water resources, predicting agricultural production and researching effects of the climatic and environmental change on agricultural production. The method of integrating the object-oriented classification method with NDVI time series data and crop phenological information can be used to differentiate varieties of crops. The method of integrating the EnKF with remote sensing information can be used as a useful tool to apply the coupled WOFOST and HYDRUS models at regional scale.