Simulation modelling
A focus of our research is the development and application of computer simulation models. In addition to theory and experiments, such models are increasingly used to answer scientific questions. Complex systems – such as forest ecosystems – are particularly well suited for model based analyses. Following an open source philosophy, all models and code are freely available.
Simulation models pack our mechanistic scientific understanding in a common frame in form of a computer program. Models are therefore prime tools for assessing ecosystem dynamics, particularly in a changing world. They help to understand how changes in climate, disturbance regimes, and management affect our systems.
iLand
iLand - the individual-based forest landscape and disturbance model - was designed to analyze interactions between climate, disturbance, and vegetation. The model is process-based and simulates individual trees at the landscape scale. iLand includes modules for forest fire, wind disturbance and spruce bark beetles as well as a general framework for simulating biotic disturbances. Furthermore, iLand features an agent based forest management engine that simulates adaptive forest management at the landscape scale.
for more information: iland-model.org
SVD
SVD - Scaling Vegetation Dynamics - is a deep learning based model of vegetation development on larger scales. It simulations vegetation dynamics as probabilistic transitions between discrete states which are learned by a deep neural network (DNN) from data.
The training data can originate from different sources including model projections from process based forest models. In the latter case, the DNN learns a representation of more detailed models, including their response to environmental drivers such as climate change. SVD can be used for large scale projections with high spatial resolution and includes dynamic disturbance modules (e.g., for fire).
Check out code and documenta/tion at svdmodel.github.io/SVD/
Our models are used all over the world
The models developed in the EDFM group at TUM are applied in many regions of the world to answer scientific questions about forest dynamics - how will forests respond in a warmer world? What are the impacts of changing disturbance regimes? What are our options for managing future forests?
Currently, our models are applied by researchers in Europe, North America, and Asia.
