Estimating the carbon balance of the Amazonian Rainforest: regionalization of the forest model FORMIND
The Amazonian Rainforest is a carbon pool of high global importance. However, looking at the yearly carbon balance, it is controversial whether it is a sink or source of atmospheric CO2.
This PhD project is an integrated project between the Department Computational Hydrosystems ( , ) and the Department of Ecological Modelling ( , ). The main goal of this PhD work is the regionalization of the local forest model FORMIND in order to estimate the carbon balance of the Amazonian rainforest with the help of eddy covariance (EC) data.
Compared to global vegetation models, local forest models consider disturbances and a large number of species. Studies have shown that both have strong impacts on the carbon dynamics of a forest. The understanding of carbon fluxes on a local scale, therefore, is an important first step before upscaling the model to a larger region like the Amazonian rainforest.
The project faces three main challenges:
(1) Reducing the model’s time step to link the forest model and eddy-covariance data:
While carbon flux data from the EC-method are commonly available in half-hourly time steps, forest models describe the exchange of carbon and forest dynamics with the atmosphere on much larger time scales. The forest model is adjusted to smaller time steps to assure consistency of model output and comparability of both methods.
(2) Calibration of the forest model with the help of eddy-covariance data:
simulates on the individual level of a tree, whereas the EC-method measures eco-physiological responses at the ecosystem level. EC-data is processed in order to derive information about functional relationships between individual carbon fluxes and climate variables as used in the forest model.
(3) Regionalization of local simulations:
The forest model calculates growth for each tree individually and is, therefore, computationally limited to areas of up to 50ha. The model will be regionalized from the model output of selected forest stands within the Amazonian rainforest where EC-data is available.
Contact: Edna Rödig (edna.roedig(at)ufz.de, )
Matthias Cuntz (matthias.cuntz(at)ufz.de,)
Andreas Huth (andreas.huth(at)ufz.de,