SPITFIRE within the MPI Earth system model: Model development and evaluation

Thumbnail Image
Date
2014
Volume
6
Issue
3
Journal
Series Titel
Book Title
Publisher
Hoboken, NJ : Blackwell Publishing Ltd
Link to publishers version
Abstract

Quantification of the role of fire within the Earth system requires an adequate representation of fire as a climate-controlled process within an Earth system model. To be able to address questions on the interaction between fire and the Earth system, we implemented the mechanistic fire model SPITFIRE, in JSBACH, the land surface model of the MPI Earth system model. Here, we document the model implementation as well as model modifications. We evaluate our model results by comparing the simulation to the GFED version 3 satellite-based data set. In addition, we assess the sensitivity of the model to the meteorological forcing and to the spatial variability of a number of fire relevant model parameters. A first comparison of model results with burned area observations showed a strong correlation of the residuals with wind speed. Further analysis revealed that the response of the fire spread to wind speed was too strong for the application on global scale. Therefore, we developed an improved parametrization to account for this effect. The evaluation of the improved model shows that the model is able to capture the global gradients and the seasonality of burned area. Some areas of model-data mismatch can be explained by differences in vegetation cover compared to observations. We achieve benchmarking scores comparable to other state-of-the-art fire models. The global total burned area is sensitive to the meteorological forcing. Adjustment of parameters leads to similar model results for both forcing data sets with respect to spatial and seasonal patterns. Key Points The SPITFIRE fire model was evaluated within the JSBACH land surface model A modified wind speed response improved the spatial pattern of burned area Regional gradients in burned area are driven by vegetation and fuel properties.

Description
Keywords
carbon cycle modeling, Earth system modeling, fire modeling, wind speed-fire interactions, Balloons, Carbon, Earth (planet), Fires, Surface measurement, Vegetation, Wind, Carbon cycle models, Comparison of models, Earth system model, Fire modeling, Land surface modeling, Meteorological forcing, Model implementation, Wind speed, Climate models, atmospheric forcing, carbon cycle, correlation, data set, fire, land surface, parameterization, quantitative analysis, satellite data, seasonality, spatial variation, wind velocity
Citation
Lasslop, G., Thonicke, K., & Kloster, S. (2014). SPITFIRE within the MPI Earth system model: Model development and evaluation. 6(3). https://doi.org//10.1002/2013MS000284
License
CC BY-NC-ND 3.0 Unported