【作者】 卢俊峰；

【导师】 赵英；

【作者基本信息】 哈尔滨工业大学 ， 资源与环境（专业学位）， 2022， 硕士

【摘要】 高分辨率遥感影像和技术在流域的应用弥补了水环境和生态监测中传统站点和采样时空不连续和无法全面表征流域信息等缺陷,可以更为全面的反映人类活动对流域水环境和生态系统的干扰程度,是推进流域水环境保护和景观生态健康发展的重要基础。本研究选取土地利用类型丰富、人类活动频繁的松花江流域肇源至哈尔滨段30km缓冲区高分辨率遥感影像作为研究对象,分析土地利用变化和驱动因素。基于差分自回归移动平均模型(ARIMA)对水环境指标演变进行分析,通过Pearson相关系数分析流域景观格局和水环境关联。构建自然-社会-景观格局三个维度综合指标体系,基于空间主成分分析法(SPCA)进行数据融合,对流域景观生态风险进行评价,为流域生态环境保护和规划提供依据。本研究基于高分辨率遥感影像,实现了非结构化数据到半结构化数据和结构化数据的有效分析,补充了目前环境领域普遍单一结构化数据的应用。本研究从景观格局角度出发分析流域水环境变化可能的关联。相比于传统生态评价方式,本研究将多源数据融合,在高分辨率遥感影像的景观格局基础上,同时考虑了自然因素和社会经济因素对风险等级的影响,评价结果更准确和全面。研究主要结论如下:(1)流域主要土地利用类型是旱田、森林、裸地和建筑,旱田占比最高,平均47.87%,面积是9441.74km~2-12458.99km~2。2005-2018年流域缓冲区内土地利用程度综合指数介于264.60-276.68,土地利用程度仍处于增加发展阶段。(2)流域缓冲区内松花江干流水质较好,阿什河和呼兰河水质偏差。2005-2020年水质得到整体改善,各水质指标离群值大量减少,除溶解氧外的水质指标分布向低水平集中,整体水质分布趋于稳定,流域缓冲区水环境得到提升。(3)建筑斑块密度与各负向水质指标呈现中等偏强程度显著负相关,流域缓冲区内水环境改善与建筑地类破碎度增加有关。森林最大斑块指数与高猛酸盐指数、氨氮、五日生化需氧量等呈现中等程度显著负相关,水环境改善也与缓冲区内自然景观稳定性提升、破碎度下降和空间复杂程度提升有一定显著性正相关。(4)景观生态风险值在流域缓冲区内不是随机分布,而是存在显著的空间异质性和依赖性特征,各个风险等级分布并不均匀,具有一定复杂性。流域内景观生态风险空间分布与景观类型联系密切。高风险区域集中在建筑密集区域,特别是哈尔滨人口密集的市区。基于景观生态风险评价,对不同风险等级的具体生态保护和规划提出建议,促进流域可持续发展。更多还原

【Abstract】 High resolution remote sensing images and the technical application made up for temporal and spatial discontinuity and inability to fully characterize river basin information defects in traditional sites and sampling in water environment and ecological monitoring in river basin,and can reflect the disturbance degree of human activities to the water environment and ecosystem of the basin more comprehensively,is the important basis of healthy development of water environment protection and landscape ecological in river basin.In this study,high-resolution remote sensing images of 30km buffer zone from Zhaoyuan country to Haerbin of Songhua River with abundant land use types and frequent human activities were selected as the research object to analyze land use change and driving factors.The evolution of water environment index was analyzed based on ARIMA,and the correlation between landscape pattern and water environment was analyzed by Pearson correlation coefficient.A comprehensive index system of three dimensions of natural,social and landscape pattern was constructed,and data fusion was carried out based on spatial principal component analysis(SPCA)to evaluate the ecological risk of river basin landscape,providing a basis for river basin ecological and environmental protection and planning.Based on high-resolution remote sensing images,this study achieves effective analysis from unstructured data to semi-structured and structured data,complementing the common application of single structured data in the current environmental field.This study analyzed the possible correlation between water environment change in river basin from the perspective of landscape pattern.Compared with the traditional ecological evaluation method,this study integrates multi-source data and considers the impact of natural factors and social and economic factors on the risk level on the basis of landscape pattern based on high-resolution remote sensing images,so the evaluation results are more accurate and comprehensive.The main conclusions are as follows:(1)The main land use types in the basin are dry land,forest,bare land and buildings,and dry land accounts for the highest proportion with an average of 47.87%,covering an area of 9441.74km~2 to 12458.99km~2.The comprehensive index of land use degree in the river basin buffer zone from 2005 to 2018 ranged from 264.60 to 276.68,and the land use degree was still in the stage of increasing development.(2)In the buffer zone,the water quality of Songhua River is good,but the water quality of Ash River and Hulan River is different.From 2005 to 2020,the overall water quality was improved,the outliers of all water quality indexes decreased greatly,and the distribution of water quality indexes except dissolved oxygen concentrated to a low level.The overall water quality distribution tended to be stable,and the buffer water environment of the basin was improved.(3)There was a significant negative correlation between building patch density and each negative water quality index,and the improvement of water environment in the river basin buffer zone was related to the increase of building land fragmentation.The maximum patch index of the forest was negatively correlated with the hyperbolic acid index,ammonia nitrogen and five-day biochemical oxygen demand,and the improvement of water environment was positively correlated with the improvement of the stability,fragmentation and spatial complexity of the natural landscape in the buffer zone.(4)The landscape ecological risk values in the river basin are not randomly distributed,but have significant spatial heterogeneity and dependence characteristics.The distribution of each risk level is not uniform and has a certain complexity.The spatial distribution of landscape ecological risk is closely related to landscape types.High-risk areas are concentrated in built-up areas,especially in the densely populated urban area of Harbin.Based on landscape ecological risk assessment,suggestions on specific ecological protection and planning of different risk levels were put forward to promote sustainable development of the river basin.更多还原