Abstract：[Background] Dengkou county is a typical arid and semi-arid area with obviously serious desertification. Ecological environment protection and treatment needs to be solved urgently. In the process of urbanization, balancing the three types of land for construction land, sandy land and ecological land is particularly important. Based on the GD_SVM_CA-Markov model, this paper aims to analyze the change of the dynamic distribution of the landscape in Dengkou county from two dimensions of time and space, to explore its change pattern, and carry on the simulation prediction, so as to provide a certain decision-making reference for the local urban development planning, the desertification control and the ecological environment protection.[Methods] Based on 10 driving factors (DEM, slope, aspect, NDVI, groundwater depth, evapotranspiration, population density, the nearest distance to water area, the nearest distance to settlement, the nearest distance to road), the land use suitability atlas was created by using GeoDetector to explore the relationship between land use change and 10 driving factors and MCE module provided by IDRISI software; Through SVM to define the transformation rules of the cell, thus the improvement of CA model was achieved; Based on the land use data of the two periods of 2006 and 2011, the Markov model was used to generate the land use transfer matrix. The landscape pattern simulation of study area in 2016 based on the GD_SVM_CA-Markov model was implemented with the above process integrated. In order to test the simulation accuracy, the Kappa coefficient was used for the test[Results] From 2006 to 2016, the landscape area of construction land in Dengkou county increased from 5 785.55 hm2 to 8 952.67 hm2, the landscape area of sandy land decreased from 76 616.15 hm2 to 56 460.50 hm2, the landscape area of water area increased from 23 859.88 hm2 to 24 679.10 hm2, the landscape area of woodland and grassland increased from 117 452.37 hm2 to 128 120.87 hm2. For construction land, there was 15.64% probability of conversion to arable land. In the case of water area, there was 11.56% probability of turning into arable land. In terms of sandy land, there was 18.37% probability of turning into woodland and grassland. The influence degree of the 10 driving factors on the landscape type change in Dengkou county was 0.248 816, 0.048 784, 0.134 342, 0.951 212, 0.975 924, 0.873 667, 0.520 317, 0.256 226, 0.413 550, 0.178 658 respectively according to the above order. The Kappa coefficient of the CA-Markov model simulation results of 2016 was 0.862 8, the Kappa coefficient of the GD_SVM_CA-Markov model simulation result of 2016 was 0.925 0. Based on the land use data of 2016 and the land use transfer data of 2011-2016, the GD_SVM_CA-Markov model was used to simulate and predict the spatial distribution pattern of landscape in 2021. During 2016-2021, the landscape area of construction land increased from 8 952.67 hm2 to 11 610.21 hm2, the landscape area of sandy land increased from 56 460.50 hm2 to 67 235.11 hm2, and the landscape area of ecological land such as water area and woodland and grassland decreased from 152 799.97 hm2 to 143 670.04 hm2.[Conclusions] Hydrological conditions, vegetation cover and population factor are the decisive factors that determine the temporal and spatial changes of local landscape types. Thus, at the same time as urban development, it is necessary to pay attention to ecological and environmental protection. The simulation result of 2016 based on GD_SVM_CA-Markov model has higher overall simulation accuracy and is better than the simulation result of 2016 based on CA-Markov model. Therefore, it is feasible to use GD_SVM_CA-Markov model to simulate and predict the spatial distribution pattern of landscape in Dengkou county.
王宇航, 于强, 岳德鹏, 张启斌, 马欢. 基于GD_SVM_CA-Markov模型的县域景观格局模拟[J]. 中国水土保持科学, 2018, 16(3): 134-141.
WANG Yuhang, YU Qiang, YUE Depeng, ZHANG Qibin, MA Huan. Landscape pattern simulation within a county based on GD_SVM_CA-Markov model. SSWC, 2018, 16(3): 134-141.
廖李红,戴文远,陈娟,等. 平潭岛快速城市化进程中三生空间冲突分析[J]. 资源科学,2017, 39(10):1823. LIAO Lihong, DAI Wenyuan, CHEN Juan, et al. Spatial conflict between ecological-production-living spaces on Pingtan Island during rapid urbanization[J]. Resources Science, 2017, 39(10):1823.
[2]
乔琨,朱文泉,胡德勇,等. 北京市不同功能区不透水地表时空变化差异[J]. 地理学报,2017,72(11):2018. QIAO Kun, ZHU Wenquan, HU Deyong, et al. Examining the distribution and dynamics of impervious surface in different functional zones of Beijing[J]. Acta Geographica Sinica, 2017, 72(11):2018.
[3]
郑慧,贾珊,赵昕. 新型城镇化背景下中国区域生态效率分析[J]. 资源科学,2017,39(7):1314. ZHENG Hui, JIA Shan, ZHAO Xin. An analysis of regional eco-efficiency in China under the background of new-type urbanization[J]. Resources Science, 2017, 39(7):1314.
[4]
王新越,刘二恋,候娟娟. 山东省旅游城镇化响应的时空分异特征与类型研究[J]. 地理科学,2017,37(7):1087. WANG Xinyue, LIU Erlian, HOU Juanjuan. Characteristics and types of spatial-temporal differentiation of tourism urbanization response in Shandong province[J]. Scientia Geographica Sinica, 2017, 37(7):1087.
[5]
李志,刘文兆,郑粉莉. 基于CA-Markov模型的黄土塬区黑河流域土地利用变化[J]. 农业工程学报,2010,26(1):346. LI Zhi, LIU Wenzhao, ZHENG Fenli. Land use change in Heihe catchment on loess tableland based on CA-Markov model[J].Transactions of the CSAE, 2010, 26(1):346.
[6]
王友生,余新晓,贺康宁,等. 基于CA-Markov模型的藉河流域土地利用变化动态模拟[J]. 农业工程学报,2011,27(12):330. WANG Yousheng, YU Xinxiao, HE Kangning, et al. Dynamic simulation of land use change in Jihe watershed based on CA-Markov model[J]. Transactions of the CSAE, 2011, 27(12):330.
[7]
李俊,董锁成,李宇,等. 宁蒙沿黄地带城镇用地扩展驱动力分析与情景模拟[J]. 自然资源学报,2015,30(9):1472. LI Jun, DONG Suocheng, LI Yu, et al. Driving force analysis and scenario simulation of urban land expansion in Ningxia-Inner Mongolia area along the Yellow River[J]. Journal of Natural Resources, 2015, 30(9):1472.
[8]
张晓娟,周启刚,王兆林,等. 基于MCE_CA-Markov的三峡库区土地利用演变模拟及预测[J].农业工程学报,2017,33(19):268. ZHANG Xiaojuan, ZHOU Qigang, WANG Zhaolin, et al. Simulation and prediction of land use change in Three Gorges Reservoir Area based on MCE_CA-Markov[J]. Transactions of the CSAE, 2017, 33(19):268.
[9]
王劲峰,徐成东. 地理探测器:原理与展望[J]. 地理学报,2017,72(1):116. WANG Jinfeng, XU Chengdong. Geodetector:Principle and prospective[J]. Acta Geographica Sinica, 2017, 72(1):116.
[10]
张兰影,庞博,徐宗学,等. 基于支持向量机的石羊河流域径流模拟适用性评价[J].干旱区资源与环境,2013,07:113. ZHANG Lanying, PANG Bo, XU Zongxue, et al. Assessment on the applicability of support vector machine-based models for runoff simulation in Shiyang river basin[J]. Journal of Arid Land Resources and Environment, 2013, 07:113.
[11]
马欢,于强,岳德鹏,等. 基于MAS-LCM的沙漠化空间模拟方法研究[J].农业机械学报,2017,48(10):134. MA Huan, YU Qiang, YUE Depeng, et al. Spatial simulation method of desertification based on MAS-LCM model[J].Transactions of the Chinese Society for Agricultural Machinery,2017,48(10):134.
[12]
YU Q, YUE D P, WANG J P, et al. The optimization of urban ecological infrastructure network based on the changes of county landscape patterns:A typical case study of ecological fragile zone located at Dengkou (Inner Mongolia)[J]. Journal of Cleaner Production, 2017, 163, S54.
[13]
周亮,周成虎,杨帆,等. 2000-2011年中国PM2.5时空演化特征及驱动因素解析[J]. 地理学报,2017,72(11):2079. ZHOU Liang, ZHOU Chenghu, YANG Fan, et al. Spatio-temporal evolution and the influencing factors of PM2.5 in China between 2000 and 2011[J]. Acta Geographica Sinica, 2017, 72(11):2079.
[14]
林志垒,晏路明. 高光谱影像的BDT-SVM地物分类算法与应用[J]. 遥感技术与应用,2016,31(1):177. LIN Zhilei, YAN Luming. The object classification algorithm and application for hyperspectral imagery based on BDT_SVM[J]. Remote Sensing Technology and Application, 2016, 31(1):177.
[15]
宋荣杰,宁纪锋,刘秀英,等.基于纹理特征和SVM的QuickBird影像苹果园提取[J].农业机械学报,2017,48(3):188. SONG Rongjie, NING Jifeng, LIU Xiuying, et al. Apple orchard extraction with quickbird imagery based on texture features and support vector machine[J].Transactions of the Chinese Society for Agricultural Machinery,2017,48(3):188.
[16]
陈宝芬,张耀民,江东. 基于CA-ABM模型的福州城市用地扩张研究[J]. 地理科学进展,2017,36(5):626. CHEN Baofen, ZHANG Yaomin, JIANG Dong. Urban land expansion in Fuzhou City based on coupled cellular automata and agent-based models (CA-ABM)[J].Progress in Geography, 2017,36(5):626.
[17]
何丹,周璟,高伟,等. 基于CA-Markov模型的滇池流域土地利用变化动态模拟研究[J]. 北京大学学报(自然科学版),2014,50(6):1095. HE Dan, ZHOU Jing, GAO Wei, et al. An integrated CA-Markov model for dynamic simulation of land use change in Lake Dianchi watershed[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2014, 50(6):1095.
[18]
赵冬玲,杜萌,杨建宇,等. 基于CA-Markov模型的土地利用演化模拟预测研究[J].农业机械学报,2016,47(3):278. ZHAO Dongling, DU Meng, YANG Jianyu, et al. Simulation and forecast study of land use change based on CA_Markov model[J].Transactions of the Chinese Society for Agricultural Machinery,2016,47(3):278.
[19]
布仁仓,常禹,胡远满,等. 基于Kappa系数的景观变化测度:以辽宁省中部城市群为例[J]. 生态学报,2005,25(4):778. BU Rencang, CHANG Yu, HU Yuanman, et al. Measuring spatial information changes using Kappa coefficients:A case study of the city groups in central Liaoning province[J]. Acta Ecologica Sinica, 2005, 25(4):778.
2018, Vol. 16 
