Home | deutsch  | Legals | Data Protection | KIT

Fire Model Intercomparison Project


The Fire Model Intercomparison Project (FireMIP), initiated in 2014, is a collaboration between fire modelling groups worldwide. Modelling groups participating in FireMIP will run a set of common experiments to examine fire under present-day and past climate scenarios, and will conduct systematic data-model comparisons and diagnosis of these simulations with the aim of providing an assessment of the reliability of future projections of changes in fire occurrence and characteristics. Currently, 2 experiments are in progress:


  1. 1700-2013 transient and sensitivity experiment


The baseline FireMIP simulation will use prescribed climate, CO2, lightning, population density, and land use forcings from 1700 through 2013. Examination of the simulated vegetation and fire during the 20th century will allow differences between models to be quantified, and any systematic differences between types of models or with model complexity to be identified. Separate sensitivity runs will be conducted by keeping each of the main forcing variables constant to see the sensitivity to different forcings between models. A “World Without Fire” run is also performed to determine the importance of fire disturbance within each vegetation model.

A detailed description of the protocol can be found here; and a list of variables


  1. Last Glacial Maximum out-of-sample experiment


While most models are optimized to reproduce current day fire occurrence relatively well, it is unknown how well they are able to reproduce fire occurrence under very different environmental conditions (and therefore possibly under future climate change).

The Last Glacial Maximum (LGM) experiment will try to evaluate the model performance under a set of very different environmental conditions. Model outputs will be evaluated using a set of paleo fire proxies.

Protocol: in progress


Data policy


All FireMIP Data will be made freely available to the wider scientific community via a data repository once the final version of the peer-reviewed article presenting the main study has been published.

In the meantime, the FireMIP Data must be considered provisional, and subject to change. FireMIP data will be made available in preliminary form to FireMIP investigators, and associates (on a case-by-case basis) to enable quality assurance through preliminary analysis and intercomparison with other data sets. Corrections and refinements to data products and tools are possible as the analysis proceeds. The full data policy is described in the attached document.

[Firemip data policy]


Current status of participating DGVM groups and contact details:


Responsible scientist

Contact details


Fang Li


CTEM v2.0

Joe Melton

joe meltonBne7∂canada ca


Gitta Lasslop



Stephane Mangeon



Stijn Hantson



Stijn Hantson



Matthew Forrest



Dominique Bachelet



Chao Yue



FireMIP manuscripts


Hantson, S., A. Arneth, S. P. Harrison, D. I. Kelley, I. C. Prentice, S. S. Rabin, S. Archibald, F. Mouillot, S. R. Arnold, P. Artaxo, D. Bachelet, P. Ciais, M. Forrest, P. Friedlingstein, T. Hickler, J. Kaplan, S. Kloster, W. Knorr, G. Lasslop, F. Li, S. Mangeon, J. R. Melton, A. Meyn, S. Sitch, A. Spessa, G. R. van der Werf, A. Voulgarakis, C. Yue. (2016) The status and challenge of global fire modelling. Biogeosciences, 13, 3359-3375. http://www.biogeosciences.net/13/3359/2016/


FireMIP is a non-funded community initiative and participation is open to all. For more information, contact Stijn Hantson (stijn.hantson∂kit.edu).