Search

Abstract: Crop growth models have been applied successfully in forecasting crop yield at a local scale, while satellite remote sensing has the advantage of retrieving regional crop parameters. The new assimilation method of integrating the crop growth model with remote sensing has presented great potential in regional crop yield assessment. In this study, the Moderate Resolution Imaging Spectrometer (MODIS) leaf area index (LAI) data product was assimilated into the World Food Studies (WOFOST) crop growth model. Using the Extended Fourier Amplitude Sensitivity Test (EFAST) global sensitivity analysis approach, several local and regional crop parameters were identified to be recalibrated. The Shuffled Complex Evolution (SCE) optimization algorithm was used to estimate the emergence date, initial biomass and initial available soil water by minimizing the differences between the corrected MODIS-LAI and simulated LAI. Results indicated that the accuracy of water-limited crop yield was improved significantly after the assimilation. The root mean square error (RMSE) reduced from 983 kg/ha to 474 kg/ha and 667 kg/ha respectively in two different optimization schemes. [Copyright &y& Elsevier]

Abstract: Based on the survey result of the nitrate content of the groundwater samples, this paper studies the variability of nitrate content of the groundwater in the Shandong province by using an optimised sampling model. Meanwhile, it also calculates the sample size required by the survey based on precision (relative error) and guarantying degree of probability. The calculation indicates that the coefficient of variation of nitrate content of groundwater in the Shandong province is 1.12, with the set survey relative error of 10%, confidence probability of 95%, and 482 required survey samples. By using pseudo-random numbers, the residential areas are selected from county level units as the survey sites, and the groundwater sources of the residential sites of villages and townships are taken as the survey targets. The first-order spatial distribution test on the residential sites is conducted by using the nearest neighbor index method, to analyse the spatial relation between the nearest neighbor sites. For the sampling sites which accord with the first-order spatial distribution, the multiple order spatial distribution test is conducted by using Ripley’s function, so as to meet the requirement of spatial random distribution of sample sites. [Copyright &y& Elsevier]