Ocean and sea ice initialization and configuration
Sub-seasonal to seasonal prediction is now routinely performed with coupled models because coupling of the atmosphere to the ocean is thought to become important for lead times longer than two weeks, contributing, for example to predictability of monsoon variations (e.g. Hendon et al., 2012) and the MJO (e.g. Woolnough et al., 2007). Sub-seasonal prediction of regional SST variations, marine heat waves, and variations in near-surface currents is also of direct interest for a wide range of activities and enterprises including, management of fisheries, offshore mining activities, ocean transportation, and anticipating coastal impacts resulting from wave activity and inundation. Sea ice is also considered as part of the coupled ocean system. Sub-seasonal prediction of sea ice has wide potential application as well but its potential has not yet been widely investigated.
This sub-project will work in coordination with WGSIP, DAOS, and PDEF (Predictability, Dynamics, Ensemble Forecasting) Working Groups to promote improved sub-seasonal predictions though improved initialization of the ocean-sea ice state and depiction of key ocean and sea-ice processes that provide predictability at sub-seasonal timescales. The project will also promote improved understanding and prediction of sub-seasonal variations of the ocean and sea ice, including marine heat waves and sea-ice extremes.
Proposed studies
  • What are the ocean-atmosphere coupled processes that directly drive/influence sub-seasonal variability and how well are these processes depicted in S2S models? What role does ocean model resolution play? As well as assessing hindcast output, some coordinated case study hindcasts can be developed that will provide ocean model output with sufficient temporal and spatial resolution in order to assess how well key processes are depicted such as the diurnal cycle of the mixed layer in the warm pool which is thought to play a key role in evolution of the MJO. An example from Shinoda et al. (2013) of the diurnal cycle of SST in the equatorial Indian Ocean during supressed phase of the MJO is shown in Figure 3. The COAMPS model used in this example has 0.5 m vertical resolution in the mixed layer and so well represents the strong diurnal variations in the suppressed phase. Assessing the depiction and prediction of this kind of variability across the full range of S2S models will provide insight into the resolution required to enhance forecast skill of the MJO.
  • What aspects of the ocean initial state provide predictability for sub-seasonal to seasonal timescales? This activity will encompass intercomparison of ocean initial states so to provide insight into current capability to capture key oceanic features that are providing predictability. The role of model ocean mean-state drift and its impact on depiction and prediction of sub-seasonal variability also should be addressed in order to guide forecast system development.
  • What are the mechanisms and predictability of sub-seasonal marine variability, especially focusing on near–surface variations, ocean fronts, upwelling, and extreme events, such as ocean heat waves, that are relevant of fisheries and coral bleaching? The impact of ocean model resolution for depiction of these events should be addressed in order to guide forecast system development. Case study prediction of some key marine heat wave events that, for instance, resulted in the recent bleaching of the Great Barrier Reef, should be coordinated across several modelling centres in order to assess current capability and to highlight key deficiencies that are acting to limit predictive skill.
  • What is the current capability to make sub-seasonal prediction of sea ice, which can be assessed with the S2S hindcast datasets? What is the sensitivity of forecast skill and predictability to initial state? What are the key processes driving sub-seasonal variations of sea ice and how are these key processes represented in the S2S models?
Figure 3. Observed and simulated near surface temperature variation in the equatorial Indian Ocean during transition from the convective (5-18 April) to the suppressed (20-30 April) phase of the MJO (from Shinoda et al., 2013). The COAMPS model has .5 m resolution in mixed layer and can well depict the diurnal variation of SST during suppressed phase of the MJO.