工业化区域撂荒耕地空间格局演变及影响因素分析

Spatial pattern evolution of abandoned arable land and its influencing factor in industrialized region

该文以箱包产业发展迅速的河北省高碑店市为研究区,基于1999-2001年、2007-2009年、2015-2017年间18期Landsat TM/OLI数据,采用CART决策树分类方法提取出研究区的撂荒耕地范围并分析其空间格局变化及影响因素。研究表明:1)对CART决策树分类结果进行精度验证,18期影像的分类精度介于87.5%~96.4%之间,可以满足本研究的精度要求。2)高碑店市耕地撂荒类型以季节性(春季)撂荒为主,并且季节性撂荒和常年撂荒的耕地面积均呈现出逐渐减少的趋势。3)耕地撂荒的主要形式由大规模集中式撂荒向小规模分散式撂荒转变。4)农村工业发展是导致耕地撂荒的主要驱动因素,距离产业中心越近的地区其耕地撂荒率越高;交通条件及耕作半径也在一定程度上影响耕地撂荒,但其影响程度逐年减弱;作物收益水平差距导致耕地撂荒呈现出季节性差异,而耕地流转能有效抑制当地耕地撂荒现象。该研究结果能为全国其他类似地区的撂荒耕地研究提供参考,并对制定保障国家粮食安全以及促进区域可持续发展的相关政策提供依据。

This paper takes Gaobeidian City of Hebei Province as the research area.Based on the Landsat TM/OLI data from 1999 to 2001,2007-2009 and 2015-2017,the CART decision tree classification method is used to extract the distribution of abandoned arable land in the study area.Finally,we analyze its spatial pattern change characteristics and influencing factors.The study draws the following conclusions:1)Using CART decision tree classification method to interpret remote sensing images in Gaobeidian City and verify the accuracy.The result shows that:1)the classification accuracy of 18-stage images are between 87.5%and 96.4%,which can meet the accuracy requirements of this study;2)The type of abandoned arable land in Gaobeidian City is mainly seasonal abandonment.The area of abandoned arable land reached 21 888.42 hm2 in the spring of 2001,and the area of seasonal abandoned arable land and perennial abandoned arable land are gradually decreasing;3)The analysis of landscape indicators including plaque number(NP),average plaque area(MPS),median plaque area(PSMD),plaque area standard deviation(PSSD),and average shape index(MSI)of the abandoned arable land shows that the main form of the abandonment of arable land has changed from large-scale centralized abandonment to small-scale decentralized abandonment;4)The development of rural industry is the main driving factor leading to the abandonment of arable land.The result of the buffer analysis shows that the closer the industrial center is,the higher the comprehensive abandonment rate;The traffic conditions and farming radius also affect the abandonment of arable land to a certain extent,but in the flat plains region,its impact gradually weakened;5)The gap in crop yields leads to seasonal differences in the cultivated land reclamation in Gaobeidian City.The long-term low net income per unit area of wheat is the main factor leading to the large-scale spring abandonment of arable land in Gaobeidian City,and the arable land transfer can effectively inhibit the abandonment of arable land.The results of rural survey show that the arable land transfer rate in the six surveyed villages shows a significant negative correlation with the change of the arable land abandonment rate.Due to the low resolution of remote sensing images used in this study,the interpretation accuracy may be affected.In addition,the selection time of remote sensing images is mainly based on the growth of spring wheat and summer maize,which do not take into account the late planting of summer stubble carrot and other crops in different growth periods.Although the sowing area of such crops is small,it may still cause errors in the interpretation of abandoned arable land.In order to solve the above problems,it is necessary to use Google high-resolution image for manual visual interpretation and correction,but this method will cost a lot of manpower and time.In future research,other remote sensing data sources with higher spatial resolution and richer spectral information can be considered for interpretation in order to solve these problems.The research results can provide reference for the study of abandoned arable land in other similar areas in China,and provide basis for the formulation of national food security and regional sustainable development policies.