利用整群抽样进行地类监测

STUDIES ON THE SURVEYING LANDS AREAS BY MEANS OF MASS SAMPLING TECHNIQUE

引言 根据山西省农业区划委员会关于建立农业资源动态监测体系的研究课题,我们从1981年开始经过三年时间对晋南的永济、运城、临猗、闻喜、夏县等五个县(市)的试点研究,基本建成了以进行土地类型面积和农作物产量调查为主要内容的动态监测体系。这是数理统计中抽样推断理论在农业上的应用,它还可以对大范围多内容的农业资源的变化进行动态监测。因此,“农业资源动态监测”被正式列入国家的“七五”计划。

In this paper, the maternal detecting system, which main functions were to monitor the variations of lands areas, was rebuilt by means of the mass sampling technique. This new rebuilt system may raise the accuracy of the surveyed land areas. The surveyed land area increase from one land area to four. The system may reduce the least surveyed land area from 30 thousand hectare to 7 thousand hectare within a county. These results were proved by the simulating trial information of masses. By experiments of the 10 distances on the 50 detected wheat points, 111.11 metres is selected to be the best distance.This new system made it possible to detect variations of the more agricultural resources, for example land areas, crops yields, soil water etc.