Author(s): Takanori Nagano; Ryuta Nakada; Masayasu Irie; Satoshi Watanabe
Linked Author(s):
Keywords: Hypoxia; Climate change; ROMS; Osaka Bay; D4PDF
Abstract: Global warming decreases the solubility of oxygen in seawater and increases the biogeochemical oxygen consumption in the water column, leading to further hypoxia. The effects of environmental factors, such as climatic mixing and nutrient loads from rivers on the expansion of hypoxia, vary based on regional characteristics. A previous study based on a deterministic approach reported the earlier occurrence and expansion of hypoxia in Osaka Bay, a hypoxic coastal sea in Japan. However, river discharge and wind conditions contribute to the currents, density distributions, and formation of hypoxia in the bay, but they vary greatly; therefore, deterministic experiments cannot evaluate these future changes and their uncertainties. Thus, we aimed to assess the effects of uncertainties in climate model predictions on the prediction of hypoxia in Osaka Bay. We discussed changes in the spatial distribution of hypoxia and the meteorological factors that contribute to these changes based on three-dimensional hydrodynamic and biogeochemical simulations using multiple forecast meteorological datasets. The simulations used Regional Ocean Modeling System and a nitrogen-based biogeochemical model. To consider climate model uncertainties, 30 ensembles of present and future scenarios were randomly selected from the Database for Policy Decision-Making for Future Climate Change product (Mizuta et al., 2017), which contained downscaling experiments using six models. The simulation results showed a trend of early hypoxia expansion in the warming scenario, including variations in the meteorological field and freshwater inflow. In future ensembles, the average volume of hypoxia was 11% larger than in the present scenario over the time calculation period. The model estimated that the volume of hypoxia increased as an average discharge increased. However, the higher bottom water temperatures were not clearly associated with the magnitude of hypoxia. The simulated oxygen distribution analysis suggested that hypoxia was most developed around August in the present scenario. However, in the warmer future scenario, the volume of hypoxia was projected to occur one month earlier.
Year: 2024