测绘通报 ›› 2017, Vol. 0 ›› Issue (3): 46-51,90.doi: 10.13474/j.cnki.11-2246.2017.0082

• 学术研究 • 上一篇    下一篇

顾及空间自相关的地理国情信息统计格网尺度选择——以植被覆盖信息统计为例

连世忠, 丁霖, 陈江平   

  1. 武汉大学遥感信息工程学院, 湖北 武汉 430077
  • 收稿日期:2016-04-12 出版日期:2017-03-25 发布日期:2017-03-31
  • 通讯作者: 丁霖。E-mail:dinglin@whu.edu.cn E-mail:dinglin@whu.edu.cn
  • 作者简介:连世忠(1990-),男,硕士生,工程师,主要从事地理国情监测数据分析与评价方面的工作。E-mail:371664651@qq.com
  • 基金资助:
    国家自然科学基金重点项目(41331175)

Scale Selecting of Geographic National Conditions Information Statistical Grids with Spatial Autocorrelation——A Case Study of Vegetation Cover Information Statistics

LIAN Shizhong, DING Lin, CHEN Jiangping   

  1. School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430077, China
  • Received:2016-04-12 Online:2017-03-25 Published:2017-03-31

摘要: 统计格网尺度的不同会带来统计结果的差异,如何选择统计格网是地理国情信息统计的重要工作。本文提出了一种顾及空间自相关的地理国情信息统计格网尺度选择方法。采用地理国情普查数据,在50 m、60 m、70 m、80 m、90 m、100 m、250 m、500 m和1000 m几个尺度下,以植被覆盖信息统计为例,利用面积占优法和中心点归属法两种方法分别进行格网化,得到了不同尺度的植被格网数据;计算植被覆盖面积统计误差,分析不同尺度下植被覆盖信息的空间自相关的变化特征,并利用Moran’s I系数差值进行尺度选择,得到了植被覆盖信息统计格网的适宜尺度。以龙沙区和清涧县作为研究区域,结果表明,在地理国情植被覆盖信息统计时,不同地区的格网统计适宜尺度是不一样的,植被覆盖度中低的龙沙区的适宜尺度为100 m,而植被覆盖度高的清涧县的适宜尺度为250 m。

关键词: 空间自相关, 尺度选择, 统计格网, 植被覆盖, 地理国情

Abstract: The difference in statistical grid scales will bring different statistical results, so it is important to select statistical grids for the geographical conditions information statistics. A method of choosing statistical grids scale of geographical conditions information with spatial autocorrelation being taken into account is proposed. By using geographic conditions census data and taking vegetation cover information statistics as an example, the study gets vegetation cover girds data, under the rule of maximum area and the rule of centric cell in scales of 50 m, 60 m, 70 m, 80 m, 90 m, 100 m, 250 m, 500 m and 1000 m, and meanwhile calculates vegetation cover statistical errors, analyzes changes of spatial autocorrelation of vegetation cover gird data at different scales to make scale selection, and then uses vegetation cover statistical errors to obtain an appropriate statistical grid scale of vegetation cover information statistics. The results show that for vegetation cover information statistics of geographic national conditions, the suitable scale is 250 m in areas with high degree of vegetation coverage, and 100 m in areas with low degree of vegetation coverage.

Key words: Spatial autocorrelation, scale selecting, statistical grids, vegetation cover, geographic national conditions

中图分类号: