ABSTRACT

Wetlands are important ecosystems but face a grim future due to the impacts of human activities, such as lost area and function. Their landscape patterns and the environment quickly respond to external disturbance. This paper selected Jiaozhou Bay as the study area and used man-machine interactive remote sensing to extract information from wetlands. Human disturbance indices and landscape transition methods were used to conduct a scenario analysis of the evolution of landscape disturbance in Jiaozhou Bay wetlands from 1973 ~ 2018. The quantitative and spatial changes in landscape disturbance types on the wetlands were also analyzed, the changes were clustered and merged, and the dynamic change types of disturbance in the study area were summarized. Among landscape disturbance types, anthropogenic renewable and non-renewable landscapes increased continuously: in particular, the anthropogenic non-renewable type had the largest area increase, a testament to continued human disturbance. After these types of changes are sorted out, a K-Means clustering algorithm was used to classify the dynamic characteristics of disturbance transition into nine types. Due to their superior physiographic conditions, the wetlands of Jiaozhou Bay have been more vulnerable to disturbance by human activities, national industrial policies, socio-economic development, and other factors have also contributed to intensified disturbance.

摘要

湿地是重要的生态系统，但由于人类活动的影响，如面积和功能丧失，未来前景黯淡。它们的景观格局和环境对外部干扰迅速做出反应。本文以胶州湾为研究区，利用人机交互遥感提取湿地信息。利用人为干扰指数和景观转换方法对胶州湾湿地1973~2018年景观干扰演变进行情景分析，分析湿地景观干扰类型在数量和空间上的变化，对变化进行聚类分析。并总结出研究区扰动的动态变化类型。在景观干扰类型中，人为可再生和不可再生景观不断增加：尤其是人为不可再生类型面积增加最大，证明了人类干扰的持续。在对这些类型的变化进行梳理后，采用K-Means聚类算法将扰动过渡的动态特征分为九类。胶州湾湿地由于其优越的地理条件，更容易受到人类活动的干扰，国家产业政策、社会经济发展等因素也导致了干扰加剧。采用K-Means聚类算法将扰动转移的动态特征分为九类。胶州湾湿地由于其优越的地理条件，更容易受到人类活动的干扰，国家产业政策、社会经济发展等因素也导致了干扰加剧。采用K-Means聚类算法将扰动转移的动态特征分为九类。胶州湾湿地由于其优越的地理条件，更容易受到人类活动的干扰，国家产业政策、社会经济发展等因素也导致了干扰加剧。