题名基于NDVI的延河流域植被格局时空变异的影响因子研究
作者刘洋
学位类别博士
答辩日期2010
授予单位中国科学院研究生院
授予地点北京
导师陈利顶
关键词黄土丘陵沟壑区 NDVI 植被格局
中文摘要
      陕北黄土丘陵沟壑区是我国水土流失最严重的地区之一,造成该地区水土流失严重的最主要原因是植被覆盖度较低。深入研究影响该地区植被生长及分布的主要因子,尤其是研究较大尺度上植被生长及其分布的主要影响因子及其影响机制,可为该地区植被恢复及水土流失治理提供科学依据。本文以黄土丘陵沟壑区的典型流域——延河流域为主要研究对象,以NDVI作为指示植被格局的指标,以遥感影像为主要数据来源,综合应用GIS空间分析方法、相关和回归分析方法以及人际网络等方法,分别探讨了大尺度植被格局的主要影响因子——气候、地形、政策和人类活动对延河流域植被格局的作用特征,并明确了导致该地区植被格局时空变异的主导因子。所得出的主要结论如下: (1)气候对植被年内、年际NDVI影响显著,且不同季节对植被起主导作用的气候因子不同。平均气温是NDVI年内变异的主要影响因子,降水只有在满足一定的气温条件时才会对植被产生作用。影响NDVI年际变异的主要气象因子是平均最高气温,平均最高气温越高,年NDVI值越低。 (2)退耕还林还草政策的实施使延河流域NDVI有了明显提高,即植被盖度得到了较明显的提高。但流域NDVI的增长幅度及增长的显著性具有明显的空间差异,自然条件(主要为土壤水分)和人类活动可能是造成植被恢复效果空间差异的主要原因。退耕还林效果与退耕投入不成比例,尤以流域上游地区最为严重。 (3)坡度、南坡向和东坡向是影响植被空间异质性的主要地形因子。它们与NDVI的相关系数沿河流上游到下游均呈现“高-低-高”的变化趋势。在流域中游地区,受人类活动的干扰,植被对地形的依赖性减弱,导致二者相关性较差。“梁”和“峁”是研究区的主要地貌特征。研究结果认为,可将黄土梁和黄土峁作为研究延河流域地区地形-植被关系的基本单元。 (4)人类活动对植被的干扰以居民点分布集中的河流为源区向周边逐渐减弱,0~500 m的范围是人类活动对植被作用的主要区域,主要表现为这一范围内的NDVI均值较低、标准差较高。人均果园面积、人均秋粮面积和人口密度是干扰植被的主要人类活动类型。 (5)采用人际网络法对气候、政策、地形、人类活动与植被之间关系的研究结果显示,人类活动是“影响因素-植被”这一人际网络中至关重要的一环。它具有最高的出度中心程度,说明在不同影响因素与植被的关系网中,人类活动是需要重点跟踪的对象。
英文摘要      Loess hilly and gullied area in northern Shaanxi Province suffers the most serious soil erosion in China. The most important reason for soil erosion is low vegetation coverage. Thus, detecting factors affecting vegetation, especially that in large scales, is vital for vegetation rehabilitation and soil erosion control in the loess hilly and gullied area. Taking the typical watershed of the loess hilly and gullied area, the Yanhe watershed, as the study area, this study explores the impacts of climate, topography, policy and human activities on NDVI (Normalized Difference Vegetation Index) respectively, and determines the most important factor that accounts for spatio-temporal changes of NDVI. Remote sensing images are the main study materials, and methods employed in this study include ArcGIS spatial analysis, correlation analysis, regression analysis, social network analysis, and etc. The main results are as follows: (1) Meteorological variables had significant impacts on both annual and inter-annual variability of NDVI, and they influenced NDVI variability at different seasons and time lags. Mean temperature was the most important meteorological variable affecting annual variability of NDVI; precipitation influenced NDVI only when temperature reached certain level. The mean lowest temperature affected NDVI variability most significantly in spring and autumn, and precipitation affected NDVI variability most significantly in summer. The most important meteorological variable affecting inter-annual variations of NDVI was the mean highest temperature, and the higher the mean highest temperature was, the lower the NDVI was. (2) NDVI in the Yanhe watershed increased obviously, illustrating that the Grain-for-Green Project (GGP) has gained a great success. However, the increase of NDVI showed a great spatial variance. NDVI increased significantly in the lower reaches of the watershed, but in other areas the increase was insignificant. Soil water availability and human activities were the main causes of the uneven increase of the NDVI. It was demonstrated that the ecological benefit of the GGP could not fully compensate for the economic input especially in the upper reaches of the watershed. (3) Slope, aspect southness, and aspect eastness were the most important topographic attributes affecting NDVI among all the tested topographic indicators. Pearson’s r (absolute value) between slope, southness, eastness and NDVI showed a high-low-high trend along Yanhe River, indicating that the influence of topography on vegetation was much weaker in middle part of the Yanhe watershed. This may be caused by denser population and stronger human activities such as land use changes in middle part of the watershed. Relationships between aspects and NDVI became more prominent when the observation scale approached average width of “liang”, and relationships of aspects, slope and NDVI broke down when the resolution exceeded average size of “mao”, indicating that that a “liang” or a “mao” would be an appropriate unit for observing the topography-vegetation relationships in the Yanhe watershed. (4) People located mainly in the river valley of the Yanhe watershed, and the influence of human activities on vegetation mainly happened within the area of 0~500 m distances to the Yanhe River. In this area, the mean NDVI was lower and the standard deviation of NDVI was higher than those in other areas where distances to the river were further than 500 m. The average per capita orchard, the average per capita area for autumn grain, and population density were main human activities affecting vegetation. (5) In the social work of “influencing factors - vegetation”, human activities had the highest out Degree Centrality, indicating that human activities was the most important factor affecting vegetation and other factors. Thus, it should be paid more attention in the course of vegetation rehabilitation.
内容类型学位论文
源URL[http://ir.rcees.ac.cn/handle/311016/35030]  
专题生态环境研究中心_城市与区域生态国家重点实验室
推荐引用方式
GB/T 7714
刘洋. 基于NDVI的延河流域植被格局时空变异的影响因子研究[D]. 北京. 中国科学院研究生院. 2010.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace