An Evaluation of the Darwin Area Forecast Experiment Storm Occurrence Forecasts

T. Keenan Bureau of Meteorology Research Centre, Melbourne, Victoria, Australia

Search for other papers by T. Keenan in
Current site
Google Scholar
PubMed
Close
,
R. Potts Bureau of Meteorology Research Centre, Melbourne, Victoria, Australia

Search for other papers by R. Potts in
Current site
Google Scholar
PubMed
Close
, and
T. Stevenson Bureau of Meteorology Research Centre, Melbourne, Victoria, Australia

Search for other papers by T. Stevenson in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Results from real-time forecasting the occurrence of storm activity during “break” and “transition” season flow within two 10-km-radius circles for periods up to 3 h in the tropics near Darwin, Australia (12°S, 131°E), are described. The forecasts, based primarily on the detection of mesoscale convergent boundaries by Doppler radar data, are evaluated in terms of probabilistic and categorical products. Brier skill scores of the forecasts indicate significantly increased accuracy relative to those based on persistence and climatology for lead times less than 90 min, as well as significantly better accuracy than those based on persistence over the entire 3 h. The forecasts provide a more detailed forecast product than do current statistical approaches employed to forecast tropical precipitation.

Abstract

Results from real-time forecasting the occurrence of storm activity during “break” and “transition” season flow within two 10-km-radius circles for periods up to 3 h in the tropics near Darwin, Australia (12°S, 131°E), are described. The forecasts, based primarily on the detection of mesoscale convergent boundaries by Doppler radar data, are evaluated in terms of probabilistic and categorical products. Brier skill scores of the forecasts indicate significantly increased accuracy relative to those based on persistence and climatology for lead times less than 90 min, as well as significantly better accuracy than those based on persistence over the entire 3 h. The forecasts provide a more detailed forecast product than do current statistical approaches employed to forecast tropical precipitation.

Save