Skip to Main content Skip to Navigation
Journal articles

Apport des données TRMM 3B42 à l'étude des précipitations au Mato Grosso.

Abstract : In tropical regions where convective rainfall is predominant while raingauge networks are deficient, satellite estimates as proposed by the Tropical Rainfall Measuring Mission (TRMM) can represent an efficient alternative way of studying rainfall spatiotemporal variability. A Principal Component Analysis based on monthly data allows mapping the various rainy regimes encountered in the south Amazonian state of Mato Grosso (Brazil). The TRMM 3B42 products are then validated based on statistical indexes. It permits determining an optimal 5-days temporal scale, which is then used to calculate rainy season parameters such as date of onset and end and duration. Those parameters are established based on the “Anomalous Accumulation” method proposed by Liebmann (2006). The date of onset occurs in October and occurs later in the south-east than in the north-west. The date of end is detected in March and progresses according to a N-S orientation. Consequently, the rainy season duration lasts from four to five months. It appears that the duration variability is especially linked to the onset variability (more than 2 months between 2 extreme years in 8 years) while the end of the rainy season is much more regular. Moreover, the onset date displays a 15 days amplitude between the main soybean productions regions. Thus, it explains the various agricultural calendars applied in Mato Grosso and help explaining the farmer's vulnerability.
Complete list of metadatas
Contributor : Vincent Dubreuil <>
Submitted on : Wednesday, January 14, 2009 - 8:01:02 AM
Last modification on : Thursday, April 2, 2020 - 1:55:54 AM


  • HAL Id : halshs-00352833, version 1


Damien Arvor, Vincent Dubreuil, Josyane Ronchail, Margareth Simões Penello Meirelles. Apport des données TRMM 3B42 à l'étude des précipitations au Mato Grosso.. Climatologie, Association internationale de climatologie, 2008, 5, pp.49-70. ⟨halshs-00352833⟩



Record views