中文摘要：

【目的】基于小麦籽粒品质预测模型和GIS技术,探索主要籽粒品质指标区域模拟与空间变异分析的方法。【方法】首先利用文献资料对已有小麦籽粒品质预测模型进行验证和评价,并基于江苏省40个生态点2000-2003年逐日气象数据和5个生态点、6个品种类型的小麦田间试验数据,对籽粒品质预测模型进行先计算后插值（calculate first,interpolate later,CI）和先插值后计算（interpolate first,calculate later,IC）两种升尺度方法的研究与评价;在此基础上进行小麦籽粒蛋白质含量、湿面筋含量和沉降值3项籽粒品质指标的区域模拟;最后,运用GIS和地统计学方法,分析江苏省3项籽粒品质指标的空间变异,获取3项籽粒品质指标的空间分布栅格图。【结果】IC方法的模拟精度较高,3项籽粒品质指标在不同地点、不同品种上的模拟值与实测值的RMSE基本小于20%;江苏省3项小麦籽粒品质指标在最大变程7.16km范围内显著相关,表现为东西经向和南北纬向变异较大的各向异性分布;空间栅格图能直观显示3项籽粒品质指标的空间分布以及纬度和经度方向的变异趋势。【结论】利用基于IC的小麦籽粒品质模型升尺度方法进行区域籽粒品质模拟和空间变异分析是可行的，为小麦籽粒品质的生态变异研究提供了参考。

英文摘要：

Objective The aim of this study is to explore the method of simulating the grain quality index and analyzing the spatial variation characteristics based on GIS and wheat quality index prediction model.Method Firstly,literature data with five varieties at five eco-sites were used for the evaluation of grain quality model at site scale.Secondly,based on the weather data sets of 2000-2003 at 40 eco-sites in Jiangsu province and experimental data from five eco-sites and six wheat cultivars,three main wheat grain quality indices at regional scale were calculated with two methods such as ＇calculate first, interpolate later＇ （CI） and ＇interpolate first, calculate later＇ （IC）. Finally, the spatial variation characteristics of three wheat grain quality indices in Jiangsu province were analyzed, and three spatial distribution raster maps for grain protein content, wet gluten content and sedimentation were made based on the geo-statistics and GIS. [Result] The IC was suggested to be preferable for up scaling the grain quality model, with the RMSE less than 20% between simulated and observed values for three quality indices. The spatial autocorrelation of grain protein content, wet gluten content and sedimentation under research region was significant within the 7.16 km variation range, and the anisotropic structure varies more evidently in longitudinal and latitudinal directions. The spatial raster maps could show the distribution and the variation trend of regional grain quality effectively. [Conclusion] The result indicated that the IC based simulation on regional spatial variation of wheat grain quality is feasible. This study might provide reference for analysis of ecological variation on crop quality.