丘陵山区耕地功能恢复潜力评价及分区研究以江西省赣州市南康区为例

EVALUATION AND ZONING STUDY OF THE POTENTIAL FOR RESTORING THE FUNCTION OF CULTIVATED LAND IN HILLY AND MOUNTAINOUS AREAS*——TAKING NANKANG DISTRICT， GANZHOU CITY, JIANGXI PROVINCE AS AN EXAMPLE

关键词:  耕地功能恢复  潜力评价  CRITIC-TOPSIS模型  K-means聚类  整治分区

Key words:  cultivated land functional restoration  potential evaluation  CRITIC-TOPSIS model  K-means clustering  remediation zoning

摘要:

目的 开展耕地功能恢复潜力评价，科学确定耕地功能恢复适宜性等级及重点区域，对于保障国家粮食安全具有重要意义。方法 文章以江西省赣州市南康区为研究区，遵循“内涵界定—潜力识别—分区调控”的研究思路，从拟恢复耕地地块的自然资源禀赋、区位条件、工程条件、经济因素4个维度选取15项评价指标，采用CRITIC-TOPSIS模型，定量识别耕地功能恢复重点区域及其空间分布格局。结果 （1）南康区耕地恢复属性地类总面积4 041.02 hm²，主要以林地、园地、坑塘水面和草地为主，面积分别为1 871.18、1 237.63、892.47和39.73 hm²，占比分别为46.30%、30.63%、22.09%和0.98%，恢复性地类分布呈“中部高、南北低”的空间格局。（2）基于CRITIC-TOPSIS模型，南康区耕地功能恢复潜力指数为［34.2~92.4］，从高至低分为5个恢复潜力等级，面积分别为993.86、1 666.58、1 040.09、309.34和31.15 hm²，占恢复属性地类面积的24.59%、41.24%、25.74%、7.65%和0.77%，其中1~3等恢复潜力土地居多，占91.57%。（3）基于带轮廓系数的K-means聚类算法，将耕地整治潜力划分为容易恢复区、一般容易恢复区、较难恢复区，面积分别为1 782.56、1 101.43和1 157.04 hm²，占比分别为44.11%、27.26%和28.63%，恢复潜力较大区域主要分布在研究区中部、南部与北部部分地区。结论 在规划层面划定整治分区，分类施策实施耕地恢复，可为耕地功能恢复整治项目筛选、耕地进出平衡方案编制与实施提供科学依据，对促进耕地可持续利用与国土空间格局优化具有重要意义。

Abstract:

In order to ensure national food security, it is of great importance to evaluate the functional restoration potential of farmland and to scientifically identify the key areas of the functional restoration potential of farmland. Taking Nankang district, Ganzhou city, Jiangxi province as the study area, and following the research idea of "connotation definition—potential identification—area regulation", 15 evaluation indicators were selected from the dimensions of natural resource endowment, location conditions, engineering conditions and economic factors of arable land to be restored, and the TOPSIS model was used to quantitatively identify the key areas of arable land functional restoration potential and their spatial distribution patterns. The results showed that (1) The total area of restored attribute land types in Nankang district was 4 041.02 hm², mainly dominated by forest land, garden land, water surface of pits and ponds and grassland, with the areas of 1 871.18, 1 237.63, 892.47 and 39.73 hm², respectively, accounting for 46. 30%, 30.63%, 22.09% and 0.98%, respectively, and the distribution of restored land types showed a spatial pattern of "high in the middle and, low in the north and south". (2) Based on the CRITICAL-TOPSIS model, the potential index of arable land function restoration in Nankang District was [34.2 ~ 92.4], and it was divided into five potential restoration classes from high to low, with the areas of 993. 86, 1 666.58, 1 040.09, 309.34, 31.15, 1 040.09, 309.34, and 31.15 hm² , accounting for 24.59%, 41.24%, 25.74%, 7.65%, 0.77% of the area of restoration attribute land classes, of which 1-3 potential restoration class land was predominant, accounting for 91.57%. (3) Based on the k-means clustering algorithm with contour coefficients, the restoration potential of arable land was classified into easy to restore, easier to restore and more difficult to restore areas, with areas of 1 782.56, 1 101.43 and 1 157.04 hm², accounting for 44.11%, 27.26% and 28.63% respectively, and the areas with higher restoration potential were mainly located in the central, southern and northern parts of the study area. In summary, the delineation of arable land concentration areas or arable land protection cluster areas at the planning level and the implementation of arable land restoration by classified measures can provide a scientific basis for the selection of arable land functional restoration and restoration projects, the preparation and implementation of arable land in/out balance programmes, and is of great significance for the promotion of sustainable use of arable land and the optimisation of the spatial pattern of land.