选取北京市饮用水源地——密云水库上游流域为研究区,采用空间分析、冗余分析与多元线性回归等多种方法,识别不同空间尺度及季节变化下景观格局与水质的相关关系.研究结果表明,研究区河流水质与流域景观特征显著相关,景观变量能够解释水质变异的58%.子流域面积对水质的影响最大,其他影响变量依次为居民用地比例、林地比例及聚集度指数.季节性流量变化与人为活动的强度对景观与水质关系有很大的影响;雨季前景观对水质的影响最弱,雨季中景观对水质的解释能力最强,雨季后景观对水质的影响减弱,略大于雨季前.岸边带尺度景观对于水质的解释能力要大于流域尺度.通过比较50m,100m,150m,300m4个尺度岸边带景观对水质的影响,发现100m宽岸边带内景观与水质关系最为显著.研究结果对岸边缓冲带的设计及水污染的控制具有指导意义.
Spatial analysis, redundancy analysis and multiple linear regressions were used to identify the correlation between landscape pattern and water quality at various spatial scales in different seasons within upper watershed of Miyun Reservoir. The results show that there is a significant correlation between water quality and landscape characteristics in the study area. Landscape variables can explain 58% variation of water quality. Drainage area has the most substantial influences on water quality. Other landscape variables with significant influences on water quality include, in sequence, percentage of residential area, percentage of forest, and contagion index. Seasonal fluctuations in flow and the intensity of human activities have substantial influences on the landscape-water quality relationship. Before the rainy season, landscapes have the least influences on water quality. During the rainy season, landscapes can best explain water quality variation. After the rainy season, the influences of landscapes on water quality decrease, but are still slightly stronger than those before the rainy season. Landscapes at the riparian scale can better explain variation of water quality than those at watershed scale. By comparing four riparian widths of 50 m, 100 m, 150 m and 300 m, we find that landscape factors at 100 m riparian have the strongest influence on water quality variation, and this may be useful for establishing standard riparian buffer structure and management systems of water pollution control.