Abstract: Based on the statistical data of 126 counties in Henan, using mathematical statistics method and spatial autocorrelation analysis, spatial changes and variation character of grain yield per hectare at county level in Henan Province during 1980?2012 are discussed by SPSS, ArcGIS and GeoDa software, and the increase potential of grain yield per hectare and total yield at regional scale are calculated. The results show that:1) During 1980?2012, grain yield per hectare and its absolute difference at county level take on increasing trend by and large, while the relative difference increases first and then decreases. The distributional pattern of grain yield per hectare changes from positive skewness to negative skewness, and the agglomeration degree of grain yield per hectare increases first and then decreases. 2) Grain yield per hectare at county level in Henan has a strong spatial autocorrelation with evident pattern “high in the eastern while low in the western”. The increment of grain yield per hectare descends from East to West at county level, and counties with higher increment of grain yield per hectare gather in the eastern plain area of Henan. 3) The changes of grain yield per hectare at county level show a positive spatial autocorrelation and the characteristic of spatial clustering of regional high value and low value is significant. The counties with “High-High” and “Low-Low” correlation are the majority. The counties with significant “High-High” correlation are mostly located in the eastern of Huang-Huai plain area and northwestern of the Hebei-Shandong-Henan low-lying plain area, while the counties with significant “Low-Low” correlation are mainly distributed in western Henan mountainous-hilly area and western of Huang-Huai plain area. 4) Henan Province is divided into 5 first-grade regions and 11 second-grade regions according to the coupled conditions of farming system regionalization and LISA cluster map. The potential output per hectare of the 5 first-grade regions could be sorted in descending order as hilly-mountainous region of Hubei-Henan-Anhui, piedmont region of Taihang mountain and Yanshan mountain, western Henan mountainous-hilly region, Huang-Huai plain region, and Hebei-Shandong-Henan low-lying plain region. The total potential output of Henan Province is 1.27×10⁷ tons, and the 5 regions could be sorted as Huang-Huai plain area, hilly-mountainous region of Hubei-Henan-Anhui, piedmont region of Taihang mountain and Yanshan mountain, western Henan mountainous-hilly region, and Hebei-Shandong-Henan low-lying plain region.

Key words: grain yield per hectare, potential output, mathematical statistics method, spatial autocorrelation analysis method, Henan Province

摘要： 基于126个县域数据, 采用描述性统计指标和空间自相关分析法, 揭示1980?2012年河南省县域粮食单产空间格局及其变化, 并结合全国耕作制度分区测算粮食增产潜力。结果表明: 1) 县域粮食单产波动性增加, 绝对差异扩大, 相对差异先增大后缩小, 县域粮食单产分布由正偏态转为负偏态, 集聚程度先增强后减弱; 2) 县域粮食单产空间自相关显著, 粮食单产“东高西低”的分异格局明显, 粮食单产增加量自东向西阶梯性递减, 东部平原县域的粮食单产增加显著; 3) 1980?2012年县域粮食单产变化的空间集聚特征明显, 局部分异显著, 显著高高集聚(HH)类型区集中在黄淮平原东部和冀鲁豫低洼平原东北部, 显著低低集聚(LL)类型区主要分布在豫西山地丘陵区和黄海平原区西部; 4) 将河南省划分为5个一级、11个二级增产类型区, 鄂豫皖丘陵山地区、燕山太行山山前平原区的粮食单产增产潜力大; 河南省总增产潜力约为1.27×10⁷t, 黄淮平原区>鄂豫皖丘陵山地区>燕山太行山山前平原区>豫西山地丘陵区>冀鲁豫低洼平原区。

关键词: 粮食单产, 增产潜力, 数理统计方法, 空间自相关模型, 河南省