Project Details
Description
Global fire models coupled to dynamic global vegetation models (DGVMs) have existed since the 1990s and used to simulate past and future fire occurrence. However, current fire models still struggle to reproduce recent trends in burnt area. Against this background, MERGE researcher Lars Nieradzik has developed a new-generation prognostic wildfire scheme, SIMFIRE-BLAZE, that takes advantage of the detailed horizontal and vertical description of vegetation structure in the MERGE’s flagship model LPJ-GUESS to enable more accurate simulations of fire occurrence, impact and emissions. The updated model will be used to account for fire-mediated effects in coupled biosphere-atmosphere simulations with EC-EARTH, contributing to CMIP6 and to the next IPCC assessment report.
Variations in fire regime under past climate and land use provide a unique opportunity to evaluate the skill of a regional fire model like SIMFIRE-BLAZE. Comparing model projections and proxy-based reconstructions of fire history can highlight discrepancies and guide model improvements.
Within this context, this MERGE Short Project (SP) aims to use LPJ-GUESS-SIMFIRE-BLAZE to simulate natural and anthropogenic fire dynamics over the Industrial Period (1750-2000 AD) in Fennoscandia and Canadian boreal forests. Output variables of the model (burned area and CO2 fire-related emissions) will be compared with past fire reconstructions based on sedimentary charcoal
records (point estimates of past biomass burning) from the same regions at centennial time scales.
The main aim is to answer the follow research questions:
(1) “How well can a global wildfire model reproduce reconstructed trends in past fire activity in the boreal forests?” and
(2) “What aspects of the model structure, parameterization and performance for certain conditions/regions may need to be reviewed in the light of discrepancies?”
This model-data comparison will underline the correlation between simulated and reconstructed past fire activity. The possible reasons for differences and similarities between observations and model outputs will be investigated, and priorities for improvement identified.
Variations in fire regime under past climate and land use provide a unique opportunity to evaluate the skill of a regional fire model like SIMFIRE-BLAZE. Comparing model projections and proxy-based reconstructions of fire history can highlight discrepancies and guide model improvements.
Within this context, this MERGE Short Project (SP) aims to use LPJ-GUESS-SIMFIRE-BLAZE to simulate natural and anthropogenic fire dynamics over the Industrial Period (1750-2000 AD) in Fennoscandia and Canadian boreal forests. Output variables of the model (burned area and CO2 fire-related emissions) will be compared with past fire reconstructions based on sedimentary charcoal
records (point estimates of past biomass burning) from the same regions at centennial time scales.
The main aim is to answer the follow research questions:
(1) “How well can a global wildfire model reproduce reconstructed trends in past fire activity in the boreal forests?” and
(2) “What aspects of the model structure, parameterization and performance for certain conditions/regions may need to be reviewed in the light of discrepancies?”
This model-data comparison will underline the correlation between simulated and reconstructed past fire activity. The possible reasons for differences and similarities between observations and model outputs will be investigated, and priorities for improvement identified.
| Status | Finished |
|---|---|
| Effective start/end date | 2020/01/01 → 2020/12/31 |