Dissolved oxygen is a key parameter to measure water environment quality and ecosystem health. Currently, the problem of hypoxia (low oxygen) is prominent in coastal areas in China, but there is a lack of research on the spatiotemporal characteristics of dissolved oxygen and the control mechanism of hypoxia in the watershed-coastal system. Based on the data of 135 surface water (including estuaries) and 66 coastal water monitoring sites in Fujian Province from 2011 to 2020, this study analyzed the spatiotemporal variation pattern of dissolved oxygen at seasonal and interannual time scales. The data of hypoxia (10% quantile, corresponding to 67% saturation) were selected to study the characteristics and control mechanism of hypoxia in four types of water bodies (i.e., rivers, reservoirs, estuaries, and coastal waters) using mathematical statistics and a random forest model. The results showed that the dissolved oxygen saturation was the highest in the coast[(98.2±10.2)%] and the lowest in the estuary[(79.2±17.9)%]. Compared with that in the 12th Five-Year Plan (2011-2015), the frequency of hypoxia detection in rivers and reservoirs in the 13th Five-Year Plan (2016-2020) was significantly reduced, but the change in estuaries was not significant. Counting the points with hypoxia detection, the multi-year average hypoxia detection frequency of rivers and reservoirs was highest in autumn, and the frequency of estuaries was highest in summer. Hypoxia in reservoirs and estuaries was the most prominent but with different mechanisms. Specifically, hypoxia in reservoir reaches was related to summer runoff carrying large amounts of organic matter input, stratification leading to continuous oxygen depletion in the bottom water, and vertical mixing or discharge through dams in autumn, whereas hypoxia in estuaries was associated with strong pollution inputs and reductive materials. Systematic management and regionalized control mechanisms need to be established to further strengthen watershed-coastal pollution abatement to help mitigate eutrophication and hypoxia problems.
Keywords: control mechanism; dissolved oxygen (DO); eutrophication; hypoxia; random forest model; seasonal variation; spatial distribution.