基于主成分分析和聚类分析的山东省农业节水分区

Agricultural Water-saving Regionalization in Shandong Province Based on Principal Component Analysis and Cluster Analysis

基于压力-状态-响应模型,选取干燥度、作物需水量、节水灌溉比例等共18个较全面、易获得、概括性强的指标,构建山东省农业节水分区指标体系;利用主成分分析法对分区指标进行降维处理,运用聚类分析法对山东省17个城市进行农业分区处理。结果表明,山东省可划分为黄泛平原区,鲁中山区,鲁南山区,鲁东丘陵区,鲁西南平原区,鲁中、鲁东平原区共6个农业节水类型区,各类型区缺水程度、主要作物类型、社会经济发展状况等均存在差异,应从农业种植结构调整、田间节水技术应用等方面提出适宜各类型区发展的农业节水模式。

Based on pressure-state-response model,18 more comprehensive,easily available,and generalizable indexes such as aridity,crop water requirement,proportion of water-saving irrigation and so on were selected to construct the index system of agricultural water-saving zone in Shandong province.Principal component analysis method was used to reduce dimensions of the zoning indexes,and cluster analysis method was applied for zoning of 17 cities in Shandong province.The results show that Shandong province can be divided into six water saving agricultural zones including the Yellow River flood plain area,central Shandong mountain area,southern Shandong mountain area,eastern Shandong hilly area,southwest Shandong plain area,as well as central Shandong and eastern Shandong plain area.There are differences in water shortage degree,main crop types and social economy development status in different types of regions,the agricultural water-saving modes which are suitable for the development of various types of regions from the aspects of agricultural planting structure adjustment and field water-saving technology application are suggested.