Inverse Determination of the Influence of Fire on Vegetation Carbon Turnover in the Pantropics. Issue 12 (10th December 2018)

• Record Type: Journal Article
• Authors: Exbrayat, Jean‐François; Smallman, T. Luke; Bloom, A. Anthony; Hutley, Lindsay B.; Williams, Mathew
• Is Part Of: Global biogeochemical cycles. Volume 32:Issue 12(2018:Dec.)
• Page Start: 1776
• Page End: 1789
• Publication Date: 2018-12-10
• Subjects: model‐data fusion -- fire -- terrestrial carbon cycle -- sensitivity -- inverse modeling -- tropics; Biogeochemical cycles -- Periodicals; Electronic journals; 577.1405
• DOI: 10.1029/2018GB005925
• Languages: English
• ISSNs: 0886-6236

Abstract

Abstract: Fire is a major component of the terrestrial carbon cycle that has been implemented in most current global terrestrial ecosystem models. Here we use terrestrial carbon cycle observations to characterize the importance of fire regime gradients in the spatial distribution of ecosystem functional properties such as carbon allocation, fluxes, and turnover times in the tropics. A Bayesian model‐data fusion approach is applied to an ecosystem carbon model to derive the posterior distribution of corresponding parameters for the tropics from 2000 to 2015. We perform the model‐data fusion procedure twice, that is, with and without imposing fire. Gradient of differences in model parameters and ecosystem properties in response to fire emerge between these experiments. For example, mean annual burned fraction correlates with an increase in carbon use efficiency and reductions in carbon turnover times. Further, our analyses reveal an increased allocation to more fire‐resistant tissues in the most frequently burned regions. As fire modules are increasingly implemented in global terrestrial ecosystem models, we recommend that model development includes a representation of the impact of fire on ecosystem properties as they may lead to large differences under climate change projections. Key Points: The impact of fire on pantropical ecosystems is determined using an inverse modeling approach Mean annual burned fraction correlates with increased productivity, more carbon allocated to structural pools, and shorter transit times Large‐scale patterns of shifts in allocation are in agreement with field observations which points to the need to refine parameterizations