| Farmland is the land resource basis for food production and has an irreplaceable special status in China’s economic and social development and ecological civilization construction.Farmland protection has always been a global and strategic issue in China’s modernization process.Since the reform and opening up policy was introduced,the industrialization and urbanization process at the expense of sacrificing farmland resources had made it difficult to reverse the trend of non-agriculturalization of farmland in some regions.In recent years,due to the continuous reduction of farmland area,the potential for increasing grain production has been limited,which seriously threatens the country’s food security.Basic farmland is the essence of farmland.Delineating basic farmland and implementing special protection on it is the most important task of farmland protection,as well as the national will and rigid constraint.Meanwhile,it is also an important measure to stabilize grain production capacity and ensure regional food security,which has great significance on regional development.At present,most of the researches on basic farmland delineation are based on characteristics evaluation of farmland,and lack coupling research on multi-factors such as spatial connectivity and scale constraint of farmland.Thus,the research is not systematic enough on the whole.At the same time,a series of technical problems have emerged in the actual process of basic farmland delineation.For example,the standard for selecting farmland into basic farmland is ambiguous,the technical methods adopted in basic farmland delineation are not rich enough,and the spatial pattern of delineated basic farmland is fragmented.Under these circumstances,solving these problems will help improve the rationality and scientificity of basic farmland delineation.To this end,this study took Changsha as research area,coupled the spatial connectivity and comprehensive quality of farmland into basic farmland delineation,designed and realized the technical framework of "connectivity for priority,quality for filtering and scale for constraining" for basic farmland delineation.The process is based on multi-source data such as land use data,spatial planning data,economic and social statistics and physical geographic attributes.A variety of technical methods such as buffer analysis,technique for order preference by similarity to ideal solution(TOPSIS)and spatial autocorrelation analysis etc were also used to achieve the research goals.Specifically,the main contents and conclusions of this study include the following three parts:(1)Analysis of the spatial connectivity of farmland.A total of 2385 contiguous farmland were identified in the study.Spatially,they were mainly distributed in the western and northern part of the city and peripheral area of the central city.Depending on the size of each contiguous farmland,they were divided into 6 classes.The first-class contiguous farmland with a scale over 666.67 hectares were distributed in 6 counties(cities or districts)other than Furong District,Tianxin District and Yuhua District,in which Ningxiang City was the one with the largest distribution.Based on the identification results of contiguous farmland,the connectivity network of farmland was constructed and the connectivity indexes were calculated.It was found that neither the overall connectivity nor the centralization degree of the network was high enough,but the potential for forming contiguous farmland was relatively large.Patches with high-level connectivity were mainly distributed in Ningxiang City and Changsha County,while low-level connectivity patches were widely distributed in other counties(cities or districts).Furthermore,based on morphological characteristics,the farmland was divided into three landscape types,among which the matrix farmland and edge farmland together constituted connected farmland and accounted for 91.92%of the total farmland,while the island farmland represented fragmented farmland without spatial connectivity and accounted for 8.08%.Connected farmland was mainly distributed in plains and some valleys along rivers.Fragmented farmland was scattered in western,northeastern and southeastern mountains,hills and terraces and the edge of central city.(2)Evaluation of comprehensive quality of farmland.Firstly,a comprehensive quality evaluation system for farmland was constructed from the aspects of natural quality,site conditions and ecological value of farmland,in order to carry out scientific quantitative measurement of farmland quality.Secondly,taking each farmland patch as an evaluation unit,the method of TOPSIS was used to calculate the relative proximity Ci of comprehensive quality of each unit,and the value from high to low corresponded to the first-class to fourth-class farmland.The first-class and second-class farmland was mainly distributed in Wangcheng District,Ningxiang City and Changsha County.The third and fourth class farmland was mainly distributed in Ningxiang City.For farmland distributed in the central urban area,both the scale of farmland and the level of farmland comprehensive quality was low.Thirdly,a spatial autocorrelation analysis of comprehensive quality of farmland was conducted.The results showed that there existed a significant spatial positive correlation among the comprehensive quality of farmland in Changsha.Then the farmland was divided into five types:H-H type,H-L type,L-H type,L-L type and non-significant type.Based on the results of overlay analysis between comprehensive quality classification and spatial autocorrelation classification of farmland,it was found that the local spatial autocorrelation results were basically consistent with the results of farmland quality evaluation.According to this,the farmland was divided into four zones in terms of comprehensive quality.(3)Delineation of basic farmland.Taking the year of 2035 as target year,the basic farmland scale was predicted by the index method and the grain demand method,and the threshold of the basic farmland scale in Changsha in 2035 was determined.On this basis,according to the principle of "connectivity for priority,quality for filtering and scale for constraining",this study delineated 231407.01 hectares of basic farmland in total.The basic farmland was mainly distributed outside the central urban area.Specifically,Ningxiang city,Liuyang city and Changsha county were the key areas for basic farmland delineation.In terms of landscape pattern characteristics,the spatial distribution of basic farmland patches was relatively concentrated.The overall degree of cohesion was good,the degree of fragmentation was low,and the shape of patches was relatively regular.Comparing the results of this study with the results of the original land use plan,it was found that the basic farmland delineated in this study had several advantages in terms of scale,quality and spatial layout,and was more in line with the requirements of basic farmland delineation.In summary,the study analyzed the spatial connectivity of farmland,evaluated the comprehensive quality of farmland,predicted the scale of basic farmland,and finally delineated the basic farmland.The research results achieved the trinity of spatial connectivity,quality and scale in basic farmland delineation,and thus provided a theoretical reference for guiding regional basic farmland delineation. |