Marine weather variability, tropical cyclone prediction and impacts in the Southwest Indian Ocean
Loading...
Date
2005
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The southwestern Indian Ocean is characterized by significant inter-annual climate variability and numerous tropical cyclones. Long-and short-term variability and associated oceanic and atmospheric fields are analyzed using model-assimilated data. The evolution of monsoon circulations associated with composite tropical cyclones distinguished by trade is studied. Case study tropical cyclones are investigated and impacts evaluated through comparison of Yea!' and 'model' datasets.
The seasonal variability of intense TC is studied through composite, statistical correlation analysis, cross-modulus wavelet-filter, hovmoller analysis and multi-variate modeling.
The intense TC index is characterized with biennial to decadal cycles that may be related with the QBO and the ocean thermohaline circulation respectively. The decade 1960-69 was the most active while 1980-89 was the least active in terms of intense tropical cyclone days in SWIO. New predictors are uncovered that significantly improve the seasonal prediction of intense tropical cyclones. Te new multivariate models are performing about 42 % better than the previous model of jury et al (1999) in the period 1960-2002. One predictor is the geo-potential height (explaining 31 % TC days variability) in the southeast Pacific. It appears to foretell of downstream oscillations in the sub-tropical jet stream which bifurcates and govern wind shear over the Indian Ocean. A duct pattern seems to promote negative vorticity anomalies of -2.0X1061 and convergence anomalies of +1.2X106s *over SWIO.
An anti-phase association between South America monsoon convection and intense TCs is found through the Atlantic Zonal circulation especially in the ENSO years. A mild La Nina type of mode is favorable for more TCs. A westward propagating cyclonic circulation is coupled with a transient ocean Rossby wave in respect of the year-to-year variability in tropical cyclone days in the SWIO.
The fish resources of the SWIO are negatively associated with intense TC days, e.g. more TC implies lower catch rates. The east South African fish catch also responds to the GPH wave three patterns.
Tire daily variability is investigated through composite, hovmoller analysis, standard empirical tables and equations.
Daily sequences of composite conditions for tropical cyclones moving west, southwest and southward are compared. A link is established with the northern sub-tropical jet stream that may influence the intensity and track of cyclones in the SWIO. The temporal and spatial variability of TC rainfall and radial winds is significant. Westward and southwestward-moving TCs have rain bands of rainfall intensity of 30 mm h1 that affect the equatorial region. Southwestward-moving TC maintains its intensity while southward-moving TC endures rapid kinematic and therrnodynamfc transformation. Westward-moving TC suffers from the blocking effect of Madagascar highlands.
In our case study analysis, it is found that the NCEP model consistently underestimates wind speed by a factor of two within a 300 km radius of tropical cyclones, when compared to QuikSCAT satellite winds. As a consequence swell-driven storm surges are also underestimated. Infrared and microwave satellite rainfall comparisons are done and results are presented.
Description
Thesis submitted in fullfilment of the requirement for the Degree of Master of Science inthe Department of Geography and Environmental Studies in the Faculty of Science at the University of Zululand, South Africa, 2005.
Keywords
Weather forecasting--Indian Ocean, Cyclones--Indian Ocean, Marine meteorology--Indian Ocean