【作者】 王彦阁；

【导师】 慈龙骏；

【作者基本信息】 中国林业科学研究院 ， 水土保持与荒漠化防治， 2010， 博士

【摘要】 目前密云水库是北京惟一地表饮用水水源地,北京城镇地表水供水量50%以上来自密云水库。为了保护密云水库水质、水量及周边的生态环境,多年来国家和北京、河北两地政府在密云水库流域节水灌溉、治污、水源涵养、水土保持等方面做了大量工作,流域内的植被恢复和重建工作取得了很大的成绩,也导致了流域土地利用/覆被及景观格局发生了大幅度的变化。为此,在广泛收集密云水库流域现有资料的基础上,以TM和MSS遥感卫星影像为基础,采用遥感技术和地理信息系统技术建立了密云水库流域的1978、1988、1998年和2008年四期土地利用空间信息数据库,进行流域的土地利用/覆盖变化研究。建立基于CA、Markov的复合模型来模拟和预测流域土地利用的变化,进一步探索土地利用变化背后的驱动机制以及预测未来土地利用格局。将定性分析和定量计算相结合,采用模拟结果和现实状况相比较、DCCA模型等方法研究了政策因素对土地利用变化驱动作用,揭示人类社会影响下区域土地利用变化的过程、机理等,以期帮助土地管理者分析不同宏观政策驱动下的土地利用变化趋势,为合理利用土地规划和决策提供支持。同时,为了研究该区森林景观的结构与功能,诊断森林景观健康状况,平衡区域发展,保护脆弱生态环境,并使得生态补偿标准更加科学化和合理化,本文对密云水库流域土地利用现状进行了基于生计改善的森林景观恢复优先区以及基于生态效益的森林景观保护区进行了划分,以期为调整社会经济发展策略及该流域森林景观恢复和改善提供科学依据。本研究的主要成果如下:1、1978-2008年以来,耕地、草地、水域及未利用土地面积减少,而林地和城乡、工矿、居民用地面积增加。流域土地利用方式以耕地、林地和草地为主,但其所占比例有所变化。1978-2008年有林地面积共增加4 891.75km2,草地资源的大面积退化也是密云水库流域土地利用空间变化的显著特征之一,可能由于高覆盖度草地的大量退化,中、低覆盖草地在某些相应时段内呈增加趋势,总体看来,草地面积在大量缩减,同时草地质量在下降;耕地在1978-2008年间共减少1 350.04km2。2、从密云水库流域土地利用类型空间转移情况看,主要表现为旱地、有林地、灌木林地、疏林地以及高、中、低覆盖度草地的变化。1978-2008年间,旱地的减少面积为1 858.67km2,主要去向为有林地、灌木林地和高覆盖度草地;旱地的增加面积为521.04km2,主要来源是高覆盖度草地。有林地增加面积为4 893.55km2,主要来源于旱地、疏林地。灌木林地的减少面积为752.65km2,主要去向为有林地、高覆盖度草地、中覆盖度草地;灌木林地增加面积为1 689.21km2,主要来源于旱地、疏林地、高覆盖度草地。疏林地的减少面积为3 816.31km2,主要去向为有林地、灌木林地;疏林地增加的面积为383.81km2,主要来源于旱地、灌木林地、高覆盖度草地。高覆盖度草地的减少面积为1 394.31km2,主要去向为有林地、灌木林地;高覆盖度草地的增加面积为769.82km2,主要来源于旱地、低覆盖度草地。中覆盖度草地的减少面积为274.41km2,主要去向为旱地、有林地、高覆盖度草地;中覆盖度草地增加的面积为540.45km2,主要来源于旱地、高覆盖度草地。低覆盖度草地的减少面积为613.28km2,主要去向为有林地、高覆盖度草地。3、1978年到2008年间,水田、滩地、有林地的破碎化程度降低,疏林地、河渠的斑块边界总长度逐渐减少,表明其形状逐渐趋于规则、简单,受人类活动的干扰较强。从景观格局整体变化趋势看,斑块数、斑块密度和斑块平均面积的变化趋势说明密云水库流域北京部分的景观趋于完整化,而河北部分的景观趋于破碎化。也说明了1988年后由于河北部分景观破碎化程度的增加导致了整个密云水库流域的景观破碎化程度在增加。北京部分和河北部分的景观多样性指数和景观均匀度指数都呈逐渐降低的趋势,北京降低的速度快于河北部分。4、密云水库流域河北部分1988-1998年旱地、疏林地和高覆盖度草地减少的速度大于1978-1988年,而有林地、灌木林地和中覆盖度草地增加的速率大于1978-1988年;1998-2008年有林地、疏林地、中覆盖度草地增加的速度小于1988-1998年,而旱地、灌木林地、高覆盖度草地减少的速度小于1988-1998年。密云水库流域北京部分1988-1998年疏林地、高覆盖度草地减少的速度大于1978-1988年,而有林地、灌木林地增加的速度大于1978-1988年,2008年的模拟结果主要表现为疏林地的模拟结果比现实结果要大,而旱地、灌木林地的面积的模拟结果都比现实结果要小。这也说明政策的干扰加剧了1988-1998年间林地的增加以及耕地、草地的减少,而1998-2008年政策的干扰强度较前十年有所降低。DCCA排序结果显示国家生态建设政策引导和调控下的人类活动是导致密云水库流域土地利用变化的最主要原因,是林地面积大幅度增加的直接驱动力。5、2020年预测结果与2008年现状相比较显示,耕地、草地仍呈减少的趋势,林地、水域和建设用地较2008年有所增加,总体看来2020年土地利用的变化趋势和1978-2008这30年间每10年一个时间段的变化趋势大体相同,但变化幅度远远减小。6、基于生计改善的森林景观恢复优先区以及基于生态效益的森林景观保护区划分结果显示:密云水库社会经济发展潜力一级分区主要分布在赤城县、丰宁县和密云县,二级分区主要分布在密云县、延庆县、滦平县和丰宁县,三级分区主要分布在赤城县、丰宁县、怀柔县、延庆县、滦平县和兴隆县,四级分区主要分布在赤城县和丰宁县;属于森林景观恢复一级优先区的村数有7个,多分布在赤城县,属于二级优先区的有377个村,多分布在赤城、丰宁和沽源县,属于复三级优先区的有606个村,多分布在赤城、密云和丰宁县,属于四级优先区的有116个村,多分布在怀柔区、密云县和延庆县;密云水库流域森林碳汇功能第一和第二分区所占面积最大,分别为441 928.95 hm2和318 505.95hm2,占总面积的46.40%和33.44%,第五分区最小,仅占总面积的3.17%;森林生态系统水源涵养功能分区中第二分区和第三分区所占的比例较大,分别为55.45%和28.28%,第五分区所占的比例最小,仅2.34%;生态系统服务功能第二分区和第三分区的面积最大,分别为542 812.69hm2和256 413.25hm2,分别占总面积的56.99%和26.92%,第五分区面积最小,仅占0.30%。更多还原

【Abstract】 Fresh water resource scarcity is a global problem, Beijing as one of the world’s serious water shortage cities, the per capita water resource is less than 300m, which is 4% of the world average value, and this makes Beijing a serious wat er deficient area. Miyun Reservoir is currently the only surface water source in Beijing, supplying more than 50% of Beijing urban surface water resource. In order to protect the water quality, quantity, and surrounding environment of Miyun Reservoir, national and Beijing, Hebei governments did a lot of work about water saving irrigation, pollution control, water and soil conservation in Miyun Reservoir watershed during the last few years, which made great achievements in vegetation restoration and reconstruction, but also led to a significant change of land use / cover and landscape pattern. For this reason, on the basis of available information of Miyun Reservoir watershed, based on the TM and MSS remote sensing satellite images, using remote sensing and GIS technology, we established 4 spatial information database of land use in the Miyun Reservoir watershed in 1978, 1988, 1998 and 2008 to analysis land use/cover change. We established CA, Markov composite model to simulate and predict land use changes to further more explore the driving mechanism and to forecast future land use pattern. We combined the qualitative analysis and quantitative calculation: compare the simulation results and the actual situation, apply DCCA model to analysis the impact of policy factors to land use change, reveal the land use change process and mechanism under the human activities effect, which can help land managers analysis land use trend under different macro-policy, to support rational land use planning and decision-making support. At the same time, in order to study the forest landscape structure and function, diagnosis the health status of forest landscape, balance regional development, protect the fragile ecological environment, and make scientific and reasonable ecological compensation standards, in this paper we zoned different priority areas of forest landscape restoration based on the livelihood improvement and different priority areas of forest protection based on ecological benefits, in order to provide scientific basis for adjusting social and economic development strategies and forest landscape restoration and improvement.The main results of this study are as follows:1. The farmland, grassland, water area and unused land have decreased, while the forest land, urban and rural, mining, residential areas have increased during 1978-2008. The farmland, forest land, and grassland were the main land use types in this watershed, but their proportion changed from 1978 to 2008. The forest land increased 4 891.75km2 from 1978 to 2008, and grassland degradation is also one of remarkable characteristics of spatial change of land use in Miyun Reservoir watershed, may be due to the mass degenerative high coverage grassland, medium and low coverage grassland tended to increase in some time period, in general the area of grassland greatly reduced, while quality of the grassland declined too; there was 1 350.04km2 farmland reduced during 1978-2008.2. As to the transfer situation of land use types in Miyun reservoir watershed, it mainly shown as the change among nonirrigated farmland, forest land, shrub land, high coverage grassland, moderate coverage grassland and low coverage grassland. The nonirrigated farmland reduced 1 858.67km2 during 1978-2008, and mainly changed to forest land, shrub land and high coverage grassland, the increased area of nonirrigated farmland is 521.04km2, which mainly from high coverage grassland. The increased area of forest land is 4 893.55km2, which mainly from nonirrigated farmland and open forest land. The shrub land reduced 752.65km2 during 1978-2008, and mainly changed to forest land, high coverage grassland and moderate coverage grassland, the increased area of shrub land is 1 689.21km2, which mainly from nonirrigated farmland, open forest land and high coverage grassland. The open forest land reduced 3 816.31km2 during 1978-2008, and mainly changed to forest land and shrub land, the increased area of open forest land is 383.81km2, which mainly from nonirrigated farmland, shrub land and high coverage grassland. The high coverage grassland reduced 1 394.31km2 during 1978-2008, and mainly changed to forest land and shrub land, the increased area of high coverage grassland is 769.82km2, which mainly from nonirrigated farmland and low coverage grassland. The moderate coverage grassland reduced 274.41km2 during 1978-2008, and mainly changed to nonirrigated farmland, forest land and high coverage grassland, the increased area of moderate coverage grassland is 540.45km2, which mainly from nonirrigated farmland and high coverage grassland. The low coverage grassland reduced 613.28km2 during 1978-2008, and mainly changed to forest land and high coverage grassland.3. The fragmentation degree of irrigated farmland, beach land and forest land reduced from 1978 to 2008, the total edge of open forest land and river trench gradually decreased, which indicating that their shape becoming more and more regular and simple, strong disturbed by human activities. From the whole developing trend of landscape pattern we can see that the trend of patch number, patch density and mean patch area shows the landscape of Beijing part tend to be integrated, while the Hebei part tend to fragmentation, and the increase in the landscape fragmentation degree of Hebei part after 1988 has led to the increasing of landscape fragmentation of Miyun Reservoir watershed. Both the landscape diversity index and landscape evenness index of Beijing and Hebei part gradually decreased, and the decreasing speed in Beijing part was faster than Hebei. ?4. The decreasing speed of nonirrigated farmland, open forest land and high coverage grassland in Hebei part during 1988-1998 is higher than 1978-1988, while the increasing speed of forest land, shrub land and moderate coverage grassland is higher than 1978-1988. The increasing speed of forest land, open forest land and moderate coverage grassland in Hebei part during 1998-2008 is higher than 1988-1998, while the decreasing speed of nonirrigated farmland, shrub land and high coverage grassland is higher than 1988-1998. The decreasing speed of open forest land and high coverage grassland in Beijing part during 1988-1998 is higher than 1978-1988, while the increasing speed of forest land and shrub land is higher than 1978-1988. The simulation results of land use situation in 2008 in Beijing part mainly show that the simulation result of open forest land is larger than actual result, while the nonirrigated farmland and shrub land are smaller than actual result. This implied the policy disturbance exacerbated the increase of forest land and the decrease of farmland and grassland in 1988-1998, but the policy disturbance intensity decreased during 1998-2008. The results of DCCA analysis show that human activities guided and regulated by national ecological construction projects is the main driving force of land use change in Miyun Reservoir watershed, it is also the direct driving force of large increased area of forest.5. Compared the simulated result in 2020 to the actual situation in 2008 we can see that farmland and grassland continued to show a decreasing trend, forest land, water area and construction area has increased compared with 2008. Overall, the land use change trend during 2008-2020 is similar to the trend of the last 3 time period, but the change speed reduced significantly.6. The zoning of different priority areas of forest landscape restoration based on the livelihood improvement and different priority areas of forest protection based on ecological benefits showed that: the 1st area of socio-economic development potential areas is mainly distributed in Chicheng County, Fengning County and Miyun County, the 2nd area is mainly distributed in Miyun County, Yanqing County and Luanping County, the 3rd area is mainly distributed in Chicheng County, Fengning County, Huairou County, Yanqing County, Luanping County and Xinglong County, the 4th area is mainly distributed in Chicheng County and Fengning County; there are 7 villages belong to the 1st area of forest landscape restoration, and most of them located in Chicheng County, there are 377 villages belong to the 2nd area of forest landscape restoration, and most of them located in Chicheng, Fengning and Guyuan counties, there are 606 villages belong to the 3rd area of forest landscape restoration, and most of them located in Chicheng and Miyun counties, there are 116 villages belong to the 4th area of forest landscape restoration, and most of them located in Huairou, Miyun and Yanqing counties; there are large areas of the 1st and the 2nd areas of forest system carbon sequestration function area, which are 441 928.95 hm2 and 318 505.95hm2 respectively, about 46.40% and 33.44% of the total area, and the 5th area has the smallest area, which is about 3.17% of the total area; there are large areas of the 2nd and the 3rd areas of forest system water conservation function area, which are 55.45% and 28.28% of the total area respectively, and the 5th area has the smallest area, which is about 2.34% of the total area; also there are large areas of the 2nd and the 3rd areas of forest ecosystem function area, which are 542 812.69hm2 and 256 413.25hm2 respectively, about 55.45% and 28.28% of the total area, and the 5th area has the smallest area, which is about 0.30% of the total area.更多还原