Land use zoning using a coupled gridding-self-organizing feature maps method: A case study in China

Abstract

Huangshui River Valley (HRV) is a typical valley area located in the northeast of the Qinghai-Tibetan plateau. Its long and narrow space, large vertical gradient and diverse land use types bring great challenges to the land use zoning in this region. The paper aims to propose a gridding-self-organizing feature maps (SOFM) coupled method based on grid cell, which will be more reliable and refined, if compared to the method of commonly used administrative regions as the basic analytic units. Firstly the focused study area is divided into 500 m × 500 m grid cells. Six indexes, including grassland, cultivated land, forest land, construction land, population density and per capita income are both spatialized and quantified in every grid. Meanwhile, that SOFM neural network model is at the usage of classifying the land use into six zoning areas can be certain. And urban area, urban-rural transition area, agriculture-forestry transition area, typical agricultural area, agriculture-pastoral transition area, and typical pastoral area are sure to belong to them. Its results can vividly indicate the actual situation where land use pattern is in constant changes in Huangshui River Valley (HRV). The research will provide a specific method for the substantial utilization of land resources.