Subseasonal-to-seasonal (S2S) forecasts have the potential to provide advance information about weather and climate events. The high heat capacity of water means that the subsurface ocean stores and re-releases heat (and other properties) and is an important source of information for S2S forecasts. However, the subsurface ocean is challenging to observe, because it cannot be measured by satellite. Subsurface ocean observing systems relevant for understanding, modeling, and forecasting on S2S timescales will continue to evolve with the improvement in technological capabilities. The community must focus on designing and implementing low-cost, high-value surface and subsurface ocean observations, and developing forecasting system capable to extract their observation potential in forecast applications. S2S forecasts will benefit significantly from higher spatiotemporal resolution data in regions that are sources of predictability on these timescales (coastal, tropical, and polar regions). While ENSO has been a driving force for the design of the current observing system, the subseasonal time scales present new observational requirements. Advanced observation technologies such as autonomous surface and subsurface profiling devices as well as satellites that observe the ocean-atmosphere interface simultaneously can lead to breakthroughs in coupled data assimilation (CDA) and coupled initialization for S2S forecasts. These observational platforms should also be tested and evaluated in ocean observation sensitivity experiments with current and future generation CDA and S2S prediction systems. Investments in the new ocean observations, as well as model and DA system developments, can lead to substantial returns on cost savings from disaster mitigation as well as socio-economic decisions that use S2S forecast information.
KAUST Repository Item: Exported on 2021-03-03
Acknowledgements: The authors would like to thank support by a grant from NOAA Climate Variability and Prediction Program (NA14OAR4310276) and the NSF Earth System Modeling Program (OCE1419306). PMEL contribution number 4888. CD was funded by NA16OAR4310094. SG and DG were funded by NASA awards NNX14AO78G and 80NSSC19K0059. DY was supported by NSFC (91858204, 41720104008, and 41421005).