NARA Discovery
Article Details
← Back to Search Results
Journal Article

Definition and predictability of an OLR‐based West African monsoon onset

Bernard Fontaine; Samuel Louvet; Pascal Roucou
International Journal of Climatology · Vol. 28, Issue 13 · pp. 1787-1798 · 2008

Abstract

The monsoon onset is documented in terms of latitudinal shift of deep convection areas within the ITCZ using an interpolated version of the National Oceanic and Atmospheric Administration (NOAA) Outgoing Longwave Radiation (OLR) at a 5‐day time‐step over West Africa for the period 1979–2004. Signals in moist convection derived from OLR values lower than 180 W/m2 allow better determination of onset dates (ODs) than the use of other thresholds or of the raw values of OLR. Such ODs are defined without any time filtering or spatial averaging along the meridional plane. They are also significantly correlated with ODs based on other datasets such as the CMAP and Global Precipitation Climatology Project (GPCP) rainfall estimates, and seem more realistic, especially during years when the latitudinal shift is unclear and delayed. However, the respective means [30 June for ODs based on OLR ( vs ) 25 June for ODs based on CMAP and GPCP] and standard deviations [15.6 ( vs ) 8.1 days] are slightly different. These differences illustrate the fact that the monsoon onset corresponds to a 10–15‐day transition in the West African monsoon seasonal evolution which does not affect, concomitantly, all variables describing the system. The onset is clearly linked to the time evolution of a few key‐descriptors of the monsoon system at regional scale such as the installation of northward meridional gradients of moist static energy in low levels and of the monsoon cell over the continent. It is also associated with specific variations of the relative vorticity in low troposphere and of the velocity potential in upper levels over the Sahel region. Moreover, OD time series exhibits good predictability. Multivariate linear regressions have been performed in a leave‐one‐out cross‐validation way using OLR and selected atmospheric predictors in May. The best results explain 60% of the OD variance and are obtained with 2 types of predictors: the OLR gradients between the Gulf of Guinea and the African continent, and the northward migration of the West African Monsoon cell. Copyright © 2008 Royal Meteorological Society

Bibliographic Information

JournalInternational Journal of Climatology
PublisherWiley
Publication Date2008-11-15
Publication Year2008
Volume28
Issue13
Pages1787-1798
Document TypeJournal Article
Print ISSN0899-8418
eISSN1097-0088
DOI10.1002/joc.1674
SubjectAtmospheric Sciences

Access Information

NARA Access Coverage1996-01-01~Current
Journal Homepagehttps://rmets.onlinelibrary.wiley.com/loi/10970088
Publisher PageOpen Publisher Page
Full-text access depends on NARA's subscribed coverage and institutional access.