Environmental Parameters Related to Hypoxia Development and Persistence in Jinhae Bay from 2011 to 2016 and Their Potential for Hypoxia Prediction

Jinhae Bay, a semi-enclosed bay on the southern coast of Korea, is a major aquaculture area that forms a spawning ground and nursery for commercially important fishes. Since the late 1960s, industrial and domestic waste from adjacent cities and industrial complexes has been released into the region, resulting in chronic hypoxia and red tides. As a central site of environmental monitoring efforts for aquaculture and fisheries in southeastern Korea, Jinhae Bay was surveyed every 2 months usually, and every 2–3 weeks during the hypoxia season, with the seawater properties observed at approximately 31–34 stations. The maximum area and duration of hypoxia in Jinhae Bay occurred in 2016 (316 km² and 26 weeks, respectively), with minima of area in 2013 (213 km²) and duration in 2011 (15 weeks). Correlation analyses of the seawater properties, weather parameters, and hypoxia indices showed that the hypoxic area was positively correlated with the surface-water temperature, air temperature, and rainfall; the minimum dissolved oxygen concentrations were negatively correlated with the air and water temperatures and bottom-water nutrient levels; and the water stability was negatively correlated with the surface-water salinity and positively correlated with both the surface- and bottom-water nitrate and silicate concentrations. These findings imply that the air temperature and precipitation may be important factors in the development and persistence of hypoxia in Jinhae Bay via the control of the stratification intensity and eutrophication of the water column. Therefore, we tested these parameters for their potential to predict hypoxia. Based on our results, we propose the following trends of hypoxia in Jinhae Bay: the initial hypoxia development generally depends on the criteria of an air temperature ≥ 19.5 °C for 1 week and total precipitation > 100 mm over 4 weeks, and it becomes more severe (≥50% coverage) under strong eutrophication, mainly due to organic matter discharge following heavy rainfall, based on the logarithmic correlation with the 4-week rainfall (R² = 0.6). Therefore, the hypoxic area index can be predicted using its linear regression relationships with the 1-week air temperature and 4-week precipitation (R² = 0.56). This study tested the prediction of the hypoxic area based on a simple calculation method and weather parameter criteria, and it demonstrated the potential of this method for precisely forecasting hypoxia in combination with biogeochemical models or other mathematical solutions to prevent massive fishery damage.