地理研究 ›› 2017, Vol. 36 ›› Issue (5): 872-886.doi: 10.11821/dlyj201705006

• 研究论文 • 上一篇    

空间权重对分析地理要素时空关联格局的影响——基于中国大陆省域水资源消耗强度的实证

臧正1(), 邹欣庆1,2(), 宋翘楚1   

  1. 1. 南京大学地理与海洋科学学院,南京 210023
    2. 中国南海研究协同创新中心,南京 210093
  • 收稿日期:2016-08-15 修回日期:2016-12-23 出版日期:2017-05-20 发布日期:2017-05-31
  • 作者简介:

    作者简介:臧正(1978- ),男,吉林白城人,博士研究生,助理工程师,主要从事自然资源管理及区域可持续发展研究。E-mail:zangzheng@126.com

  • 基金资助:
    国家基础科学人才培养基金项目(J1103408)

Impacts of spatial weight on the analysis of spatial-temporal patterns of geographic factors: An empirical study on the intensity of water resource consumption in provinces of Chinese Mainland

Zheng ZANG1(), Xinqing ZOU1,2(), Qiaochu SONG1   

  1. 1. School of Geographic and Oceanographic Sciences, Nanjing University, Nanjing 210023, China
    2. Collaborative Innovation Center of South China Sea Studies, Nanjing 210093, China
  • Received:2016-08-15 Revised:2016-12-23 Online:2017-05-20 Published:2017-05-31

摘要:

基于区域经济发展水平、资源禀赋等地理要素对水资源利用效率的耦合影响,以2003-2013年中国大陆省区为例,应用多种空间权重方案对省际水资源消耗强度的时空格局进行比较分析,针对评价结果的不确定性,提出了基于GDP-距离倒数的改进空间权重方案,并进行实证研究。结果表明:① 应用对称性空间赋权方法,得出的省际水资源强度的空间自相关特征检验结果具有局限性,依据基于邻接或距离原则构建的空间权重,评价结果不够稳健;② 应用GDP-距离倒数的组合空间权重方案,能够充分体现经济—地理要素对水资源利用效率的非对称性影响,所得局部空间自相关评价结果的可靠性提高;③ 在研究时段内,中国大陆省际居民生活用水强度的时空异质性最小,生态环境用水强度异质性最大,工业和农业用水强度的时空异质性相对较小并且具有显著的全局空间自相关特征;④ 省际工业用水强度具有南北分异特征,局部低—低集聚区位于黄河以北并且相对稳定,高—高集聚区位于长江以南并趋于弱化;⑤ 省际农业用水强度的局部高—高集聚、高—低集聚区位于中国大陆西部,且研究时段内表现相对稳定。基于GDP-距离倒数的组合空间权重方案,能够同时体现省际经济水平及空间距离的临近度,因此可在一定程度上提高相关研究中评价结果的可靠性。

关键词: 组合权重, 探索性空间数据分析方法, 水资源消耗强度, 时空格局, 中国大陆

Abstract:

In order to evaluate the combined impacts of regional economic development, resource endowment and other geographical factors on the efficiency of water resource utilization, we applied various spatial weight schemes to analyze the inter-provincial spatial patterns of intensity of water resource consumption in Chinese Mainland from 2003 to 2013. Due to the uncertainty of the results, we proposed a modified spatial weight scheme based on GDP-distance reciprocal. The results showed the limitation of the symmetry spatial weight method in examining the spatial autocorrelation of the inter-provincial water resources, hence the evaluation could be inadequate in terms of the adjacent or distant principles. The modified spatial weight scheme based on GDP-distance reciprocal properly demonstrated the influences of economic geographical factors on the efficiency of water resource utilization, therefore it improved the reliability of the evaluation of local spatial autocorrelation. Within the time window of our study, we found that the spatial-temporal heterogeneity was the lowest in the intensity of residential water utility, whereas it was the highest in that of eco-environmental water utility. The spatial-temporal heterogeneities of the intensities of industrial and agricultural water utility were low and displayed an obvious spatial autocorrelation. Inter-provincial intensity of industrial water utility was distinct between north and south China. The low-low concentrated areas, distributed to the north of the Yellow River were relatively stable, while the high-high concentrated areas located to the south of the Yangtze River were declining. As for the inter-provincial intensity of agricultural water utility, both high-high and high-low concentrated areas were in western China and relatively stable during 2003 to 2013. Our study suggested the spatial weight scheme based on GDP-distance reciprocal could reveal the proximity of both inter-provincial economic levels and the spatial distances, thus may improve the reliability of the evaluation.

Key words: combination weighting, ESDA, water resource consumption intensity, spatial and temporal pattern, Chinese Mainland