Process-based modelling

Here are some of the process-based models used in the Carbon Cycle group:


For our forward modeling studies on carbon dioxide, methane and water cycles we have used the Max Planck Institute Earth System Model (MPI-ESM) model family, especially MPI-ESM land surface component JSBACH. JSBACH was developed in Max-Planck Institute for Meteorology, Hamburg, Germany.

In other words, JSBACH is a land surface model that simulates exchanges of matter and energy between the surface and the atmosphere. It uses meteorological forcing data and the domain size can be from global to a point. In our group, it is used in various contexts: to study carbon balances of urban green spaces, to simulate forest fires in Fennoscandia, afforested wasteland and various forest managements and to model drained and undrained peatland, including studies on soil carbon accumulation and hydrology. This model provides simulations of carbon and energy balances, which can be driven by different management and climate scenarios.


Landscape DeNitrification-DeComposition (LDNDC) model is a process-based model that simulates the biogeochemical cycles of carbon, nitrogen, water and microclimate. The model is based on the DNDC model, which was developed for agricultural ecosystems, but the LDNDC model has the versatility to simulate other ecosystems such as forests. In our group, it is used to simulate arable land, including agricultural peatland sites, and forestry peatland. These simulations are coupled with studies on the the effect of different management scenario until 2100 using different climate drivers.


Yasso (Yet Another Simulator of Soil Organic matter) is FMI's own soil carbon model. In our group, we use it to investigate simulating Nitrogen mineralisation, simulating Finnish croplands on national and regional scales and to compare some other model approaches.


A site-scale model for managed grasslands. Simulates vegetation and soil dynamics (C fluxes, biomass growth, etc.) on a daily time step. BASGRA-BGC is a version for peatlands and includes a more detailed, multi-layer soil model. It is in use for similar simulations to LDNDC, but BASGRA is limited to grasses, trading off for a more detailed description of plant physiology.


A crop model for agricultural simulations, can be run for perennial and annual crops, and has capacity to simulate a two species intercrop. In our group STICS is mostly run through PEcAN (Predictive Ecosystem Analyser).


CTSM (the Community Terrestrial Systems Model, a global land model) and the module FATES (Functionally Assembled Terrestrial Ecosystem Simulator) implements an ecosystem demography concept into a vegetation cohort model of competition and co-existence. In our group, it is used to develop processes and parameters for arctic PFTs, and potentially for crops.


A data assimilation method that allows calibration of parameters and state with very little extra computational work.


JULES is a land surface model and HIMMELI is a peatland methane model. They were previously used in our group to model wetland sites.