Incorporating a Semi-Stochastic Model of Ocean-Modulated Westerly Wind Bursts into an ENSO Prediction Model
Access StatusFull text of the requested work is not available in DASH at this time ("restricted access"). For more information on restricted deposits, see our FAQ.
MetadataShow full item record
CitationGebbie, Geoffrey A., and Eli Tziperman. 2009. Incorporating a semi-stochastic model of ocean-modulated westerly wind bursts into an ENSO prediction model. Theoretical and Applied Climatology 97(1-2): 65-73.
AbstractPrediction models of the El Niño-Southern Oscillation (ENSO) phenomenon often represent westerly wind bursts (WWBs), a significant player in ENSO dynamics, as stochastic forcing. A recent paper developed an observationally motivated semi-stochastic statistical model that quantifies the dependence of WWBs on large-scale sea-surface temperature. This WWB model is added here to a hybrid coupled model, thus activating a two-way SST-WWB feedback. The WWB model represents both the deterministic and stochastic elements of WWBs and thus is especially appropriate for ensemble ENSO prediction experiments. An ensemble of retrospective forecasts is performed for the years 1979–2002. Overall statistical measures of predictability are neither degraded nor improved relative to the hybrid, coupled general circulation model, perhaps because of the limitations of the hybrid coupled model and the initialization procedure used. While the present work is meant as a proof-of-concept, it is found that the addition of the WWB model does improve the prediction of the onset and the development of the large 1997 warm event, pointing to the potential for ENSO prediction skill improvement using this approach.
Citable link to this pagehttp://nrs.harvard.edu/urn-3:HUL.InstRepos:3445987
- FAS Scholarly Articles