Close Window

Long-term potential for fires in estimates of the occurrence of savannas in the tropics

Manoel Cardoso, Instituto Nacional de Pesquisas Espaciais, Centro de Previsão de Tempo e Estudos Climáticos (INPE/CPTEC) – Cachoeira Paulista 12630-000 SP Brazil., mcardoso@cptec.inpe.br (Presenting)
Carlos Nobre, Instituto Nacional de Pesquisas Espaciais, Centro de Previsão de Tempo e Estudos Climáticos (INPE/CPTEC) – Cachoeira Paulista 12630-000 SP Brazil., carlos.nobre@inpe.br
David Lapola, Instituto Nacional de Pesquisas Espaciais, Centro de Previsão de Tempo e Estudos Climáticos (INPE/CPTEC) – Cachoeira Paulista 12630-000 SP Brazil., dmlapola@yahoo.com.br
Marcos Oyama, Instituto de Aeronáutica e Espaço – Centro Técnico Aeroespacial (IAE/CTA) – São José dos Campos – 12228-040 SP Brazil, oyama@iae.cta.br
Gilvan Sampaio, Instituto Nacional de Pesquisas Espaciais, Centro de Previsão de Tempo e Estudos Climáticos (INPE/CPTEC) – Cachoeira Paulista 12630-000 SP Brazil., sampaio@cptec.inpe.br

In order to improve the formulation and results of the CPTEC Potential Vegetation Model, we developed a new parameterization for long-term occurrence of fire in regions with potential for the establishment of savannas in the tropics. Fires are important to consider because they may favor grasses and reduce tree coverage due to fast biomass consumption and mortality relatively to slow carbon uptake and growth of the vegetation. To find relations between long-term potential for fires and other environmental factors, we have assumed that at long-term, climate conditions in savannas support both the presence of fuel and flammability, and thus fire activity in these regions may be limited primarily by the presence of ignition sources. Following other studies indicating that lightning is the most important source of ignition for natural fires, we built the long-term fire parameterization for savannas based on large-scale potential for lightning activity during transitions between dry and wet seasons. The application of this new relation improved the formulation and the results of the CPTEC-PVM. In particular, important commission errors in allocating savannas instead of seasonal forests are now substantially lower than in previous studies.

Science Theme:  LC (Land Use and Land Cover Change)

Presentation Type:  Poster

Abstract ID: 49

Close Window