13th Conference on Applied Climatology

4.2

A seasonal climate forecast methodology for ITCZ associated rainfall over Eastern Africa

Ronald P. Lowther, Air Force Institute of Technology, Wright-Patterson AFB, OH; and J. F. Griffiths

A methodology is developed for seasonal climate forecasting of Inter-Tropical Convergence Zone (ITCZ)-associated rainfall over Eastern Africa (a monsoon related system) which shows reasonably dependable predictability. Using this methodology, rainfall probability decision flowcharts are produced which, are simple enough to use in seasonal forecast efforts by a variety of users now. Their purpose is to provide improved predictive guidelines for forecasting either actual rainfall amounts or for an indication of rainfall trends three months in advance.

Explored is whether rainfall trends during the passage of the ITCZ can be determined by examining rainfall trends, which, occur earlier as the ITCZ traverses a region. Eastern Africa is divided into four seasonal rainfall patterns, within four regions, corresponding to the latitudinal displacement of the ITCZ. Data are examined based on preceding conditions and minimal statistical assumptions and, relationships are formulated within regions and between seasons.

A time series visualization of ITCZ-associated rainfall data is created by combining the regional indices for all four regions in sequential order through the period of study. By examining the change between rainfall indices of two regions, as they are traversed by the ITCZ, it is found that a forecast for the next region traversed can be made. It is further found that these predictive relationships can be improved upon towards the extremes of the data distributions and, during certain periods, associated with known and established teleconnections whose indices are accessible by worldwide users.

Session 4, Climate Variability and Forecast Strategies (Parallel with Session 5)
Tuesday, 14 May 2002, 8:00 AM-10:00 AM

Previous paper  Next paper

Browse or search entire meeting

AMS Home Page