%0 Journal Article %T 海河上游清水河流域土地利用变化特征 %T Characteristics of Land Use Change in Qingshui River Watershed in Upper Reaches of Haihe River %A 刘孟竹,张红娟,杨佳,裴宏伟 %A Liu Mengzhu %A Zhang Hongjuan %A Yang Jia %A Pei Hongwei %J 水土保持通报 %J 水土保持通报英文版 %@ 1000-288X %V 40 %N 3 %D 2020 %P 147-154 %K 清水河流域;遥感;土地利用;Logistic-CA-Markov模型;预测模拟 %K Qingshui River watershed;remote sensing;land use;logistic-CA-Markov model;prediction simulation %X [目的] 海河上游清水河流域处于半湿润向半干旱过渡的生态环境脆弱区,研究该地区土地利用时空变化规律,旨在为当地生态环境建设提供一定的决策依据。[方法] 基于清水河流域1990—2018年4期Landsat遥感影像解译的土地利用数据,在优选的最佳空间尺度下通过Logistic-CA-Markov模型对清水河流域土地利用变化进行分析和预测。[结果] ①近30 a来清水河流域土地结构主要以耕地、林地、草地为主,三者面积之和比例超过总量的96%;其中,建设用地面积持续增加且变化程度最剧烈。②通过对受试者工作特征(ROC)曲线进行拟合优度评价,在100 m×100 m空间尺度下,各个地类ROC值达到最优水平。③2018—2027年间,草地和水域面积预计分别减少42.44,5.43 km2,耕地、林地和建设用地面积分别增加16.51,16.89,14.67 km2。[结论] ①100 m×100 m是针对具有典型小流域特征的清水河流域进行预测模拟研究的最合适尺度;②因2022冬奥会的筹备促使建设用地的开发力度加大将成为未来清水河流域土地格局变化的主要因素。 %X [Objective] The Qingshui River watershed in the upper reaches of the Haihe River is located in a fragile ecological region where climate varies from arid to semiarid. The spatiotemporal variation of land use in this region was studied for local decision making on ecological construction.[Methods] Based on land use data interpreted from Landsat remote-sensing images at four periods during 1990-2018, the land use changes of the Qingshui River watershed were analyzed and predicted by using the logistic-CA-Markov model under an optimal space scale.[Results] ① During the past 30 years, the land use of the Qingshui River mainly consisted of cultivated land, forest land, and grassland, and the total area of these three land use types accounted for more than 96% of the study area. The area of construction land increased, and its change intensity was the maximum. ② The receiver operating characteristic (ROC) value of each land type reached the best level at a scale of 100 m×100 m, which was obtained based on the goodness-of-fit tests for ROC curves. ③ The area of grassland and water area would decrease by 42.44 and 5.43 km2, respectively, from 2018 to 2027. The areas of cultivated land, forest land, and construction land would increase by 16.51, 16.89, and 14.67 km2, respectively.[Conclusions] ① The scale of 100 m×100 m was the suitable space scale for the prediction simulation on a typical small watershed, such as the Qingshui River watershed. ② The preparations for the 2022 Winter Olympics drastically accelerated the exploitation intensity of construction land, and they were the dominant driver of the land structure change in the Qingshui River watershed. %R 10.13961/j.cnki.stbctb.2020.03.021 %U http://stbctb.alljournal.com.cn/stbctb/home %1 JIS Version 3.0.0