Potential impacts of the Saharan air layer on numerical model forecasts of north atlantic tropical cyclogenesis

Aaron S. Pratt, Jenni L. Evans

Research output: Contribution to journalArticle

11 Scopus citations

Abstract

Tropical cyclones have devastating impacts on countries across large parts of the globe, including the Atlantic basin. Thus, forecasting of the genesis of Atlantic tropical cyclones is important, but this problem remains a challenge for researchers and forecasters due to the variety of weather systems that can lead to tropical cyclogenesis (e.g., stalled frontal boundaries, African easterly waves, and extratropical cyclones), as well as the role of the surrounding environment in promoting or inhibiting the development into a tropical depression and beyond. In the North Atlantic, the effects of the Saharan air layer (SAL), a hot, dry dusty layer that moves into the eastern Atlantic basin, must be taken into account when forecasting whether genesis will occur. There are several characteristics of SAL that impact tropical cyclones (decreased mid-tropospheric moisture, increased midlevel shear, and enhanced stability). The purpose of this study is to examine the forecasting skill of the National Centers for Environmental Prediction (NCEP) Global Forecasting System (GFS) model for the 2002 and 2003 Atlantic hurricane seasons, with particular regard paid to possible SAL effects on model genesis forecast accuracy. Cyclone phase space analyses of GFS 6-hourly forecasts were divided into three possible outcomes: S (successful forecasts that verified in cyclogenesis), F1 (cyclogenesis events that were not forecast to occur), and F2 (forecasted cyclogenesis that did not occur). The spatial variabilities of these outcomes for the early, middle, and late season were analyzed for both years, as well as the background environmental conditions. The large number of F2 forecasts that were seen in both years can be partly explained by the GFS model not capturing the detrimental effects of the SAL on cyclogenesis.

Original languageEnglish (US)
Pages (from-to)420-435
Number of pages16
JournalWeather and Forecasting
Volume24
Issue number2
DOIs
StatePublished - Apr 1 2009

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Atmospheric Science

Cite this