C3
Project C3: Climate feedbacks from changes in fire-vegetation interactions in South-America
Research Team: H. Barbosa (USP), M. Cardoso (INPE), K. Thonicke (PIK), J. Volkholz (PIK)
Outline: Fire regimes differ substantially in the South-American ecosystems forming a deep gradient from tropical rainforests, where fires naturally do not occur, to the savannah ecosystems of the Cerrado and Caatinga where fires are a frequent and natural element. Changes in land-cover and habitat fragmentation continue to influence forest and savannah ecosystems. Fires escape from pasture and burn into neighboring tropical forest contributing to degradation . Recent trends show an increase in land-use change in the savannah ecosystems with unknown effects on fire. How these trends change vegetation-fire feedbacks, thus regional climate, is a big unknown. In addition to a release of carbon to the atmosphere, fire disturbance can modify evapotranspiration and surface albedo, thus potentially influencing precipitation and air temperature. These, in turn, can affect fuel conditions and induce more fires. How these feedbacks will change with shifting fire dominance from tropical forests to pasture and savannah is highly uncertain since it may not reduce total emissions. By combining thorough data analysis, building on complex network techniques with regional vegetation-climate modelling we aim to better quantify feedbacks between land-surface processes and climate.
Research Topic: Recent development of complex network tools which identify fire clusters and burning conditions according to the land-cover type, help to analyze the land-use impact on fire regimes. We will expand this recent method to Cerrado and Caatinga ecosystems. This work will then form the basis to expand the SPITFIRE fire model with a dynamic fire-emission scheme for pasture and other land-use types. Simulated land-surface characteristics and fire-related emissions will be used to quantify the climate feedback applying the regional climate model CCLM. Simulation results will then be used to analyze changes in fire regimes and regional climate feedbacks using complex network methods for fire cluster analysis and event synchronization.