Seasonal forecasting of subsurface marine heat waves

Abstract

Marine heat waves (MHWs) cause devastating damage to ecosystems and ocean services, with effects being identified mostly below the ocean surface. Current forecasting efforts, however, focus only on metrics based on sea surface temperature. To create a more user-relevant detection system, it is necessary to provide forecasts of subsurface events. Here, we demonstrate the feasibility of seasonal forecasting of subsurface MHWs by using ocean heat content, a more relevant indicator than surface temperature for marine stakeholders. We validate summer MHW indicators in a fully-coupled seasonal forecast system against a global ocean reanalysis and satellite data. Our main result is that subsurface summer MHWs are predicted with greater skill than surface MHWs across much of the global ocean. Sub-surface MHWs are typically longer-lasting than surface events, rendering them easier targets for forecasting systems. Despite the long-lasting nature of subsurface MHWs, we also show that the dynamical forecast system used here typically outperforms a MHW-persistence model, indicating the capability for capturing the onset and decay of MHWs. Lastly, we highlight the role of warming oceans in MHW detection skill, by removing linear trends. This work highlights the need for a wider appreciation of subsurface ocean phenomena and the increased uptake of seasonal forecasting indicators by marine stakeholders such as marine protected areas and fisheries.