An integrated ESDA-based simulation on spatial distribution of rural land conflicts at provincial scale

Abstract

An integrated ESDA-based analysis system (IEAS) for supporting conflict management of rural land at provincial scale is developed in this paper. The IEAS consists of modules of classification for rural land conflict, and distribution assessment. Self-organizing map (SOM) and spatial autocorrelation technique (SAT) are used to classify the rural land conflicts and to assess the spatial distribution of them, respectively. A case study implementing the system is performed on Jiangxi province in China. After four categories are classified by SOM, the global and local spatial autocorrelation are investigated by SAT. The results reveal that there are different spatial characteristics on the rural land conflicts in Jiangxi province, China. Specifically, the location of significant Gi* identified areas where the differences in LAI and stand volume occur and are spatially clustered. However, the global index Moran,s I is -0.0013 (z i =0.5247), which shows that there is not evidence of spatial autocorrelation for rural land conflicts at provincial scale. This paper shows SOM can be used to achieve these goals via kernels derived, especially when it is used in conjunction with more classical methods. Meanwhile, analyzing the global and local spatial autocorrelation of the differences identified those areas that have systematic sensitivity to specific model inputs. This information may then be used to aid in the interpretation of spatial distribution of rural land conflicts. Obviously, the technique may offer a fresh perspective on such conflict management issues, and potentially also serve advantages over existing approaches.8