부산대학교 도서관

Global climate models provide useful tools to forecast large‐scale anthropogenic trends and the impacts on ocean physics and marine biology and chemistry. Due to coarse spatial resolution, they typically lack the ability to represent important regional processes while underestimating mesoscale variability and vertical mixing. This means they provide limited value when it comes to regional climate projections. We developed a regional submesoscale‐permitting physical/biogeochemical model to dynamically downscale the output of a CMIP6 Earth System Model for three different Socioeconomic Pathways for the main Hawaiian Islands. We describe the methodology for downscaling the CMIP6 ocean physics and biogeochemistry along with atmospheric conditions in order to offline nest a regional model. We expect the large‐scale spatial and temporal features of the global model to be retained by the regional model, while adding representation of the regional processes that are crucial to understanding climate change on a local scale. We compare the regional model representation against both observed data and a regional reanalysis over the first two decades of the century. We show that the regional model maintains the large‐scale trends and interannual variability provided by the CMIP6 model while well‐representing the regional dynamics that drive the short‐term variability. To better illustrate the benefit of the downscaling, we present preliminary analysis of the downscaled results to examine climate impacts on the island corals that are not resolved by the global models. This analysis reveals that coastal corals are likely to experience unprecedented ocean acidification and substantial warming over the course of the century. Plain Language Summary: Climate models play an important role in projecting the changes in the ocean that may result from human‐induced climate change; however, these models lack the ability to represent critical features important to regional localities. This means that high‐resolution regional models are required to understand the impacts of climate change important to local communities. We describe how to downscale climate model solutions such that they can be used by high‐resolution regional models to project local impacts of climate change. Our application is for the main Hawaiian Islands in the Pacific ocean that are highly dynamic but are missing from climate models. We also provide preliminary results of these projections that show regional impact on the island corals due to climate change. Key Points: The first dynamically downscaled submesoscale permitting future climate projections for the main Hawaiian Islands are presentedA careful treatment of global model output is critical to capture essential local oceanographic characteristicsHawai'i's near‐surface ocean is likely to experience unprecedented acidification and substantial warming over the course of the century [ABSTRACT FROM AUTHOR]