Quantifying the role of fire in the Earth system – Part 1: Improved global fire modeling in the Community Earth System Model (CESM1)

F Li,S Levis,D S Ward

doi:10.5194/bg-10-2293-2013

Abstract

Abstract. Modeling fire as an integral part of an Earth system model (ESM) is vital for quantifying and understanding fire–climate–vegetation interactions on a global scale and from an Earth system perspective. In this study, we introduce to the Community Earth System Model (CESM) the new global fire parameterization proposed by Li et al. (2012a, b), now with a more realistic representation of the anthropogenic impacts on fires, with a parameterization of peat fires, and with other minor modifications. The improved representation of the anthropogenic dimension includes the first attempt to parameterize agricultural fires, the economic influence on fire occurrence, and the socioeconomic influence on fire spread in a global fire model – also an alternative scheme for deforestation fires. The global fire parameterization has been tested in CESM1's land component model CLM4 in a 1850–2004 transient simulation, and evaluated against the satellite-based Global Fire Emission Database version 3 (GFED3) for 1997–2004. The simulated 1997–2004 average global totals for the burned area and fire carbon emissions in the new fire scheme are 338 Mha yr−1 and 2.1 Pg C yr−1. Its simulations on multi-year average burned area, fire seasonality, fire interannual variability, and fire carbon emissions are reasonable, and show better agreement with GFED3 than the current fire scheme in CESM1 and modified CTEM-FIRE. Moreover, the new fire scheme also estimates the contributions of global fire carbon emissions from different sources. During 1997–2004, the contributions are 8% from agricultural biomass burning, 24% from tropical deforestation and degradation fires, 6% from global peat fires (3.8% from tropical peat fires), and 62% from other fires, which are close to previous assessments based on satellite data, government statistics, or other information sources. In addition, we investigate the importance of direct anthropogenic influence (anthropogenic ignitions and fire suppression) on global fire regimes during 1850–2004, using CESM1 with the new fire scheme. Results show that the direct anthropogenic impact is the main driver for the long-term trend of global burned area, but hardly contributes to the long-term trend of the global total of fire carbon emissions.

Highlights

Fire is an important Earth system process on a global scale.ahnIteadcetCpiveirntyidesos,osannpvdheggeeetnraeetiroanteschfaereadcbtearcikssby affecting biogeochemical cycles, vegetation composition and structure, land–atmosphere water and heat exchanges, atmospheric chemistry and composition, and human health and property (Bond et al, 2004; IPCC, 2007; Cochrane and Ryan, 2009; Bond-Lamberty et al, 2009; van der Werf et al, 2010; Bowman et al, 2011)
Existing global fire parameterizations suitable for Earth system model (ESM) aim to best match the observed fire regimes for the contemporary time period, given that only the contemporary global fire product is available with a sufficient quality to be used as evaluation data (Kloster et al, 2010; Prentice et al, 2011)
Its global performance was evaluated using a modified CLM-DGVM (Levis et al, 2004; Zeng et al, 2008; Zeng, 2010), and the results showed that the simulated global total amount and spatial distribution of burned area and fire emissions were broadly consistent with the satellitebased Global Fire Emission Database version 3 (GFED3) fire product

Summary

Introduction

Fire is an important Earth system process on a global scale (tiBcos,wcmlimanateet,aal.n,d20h0u9mT).ahnIteadcetCpiveirntyidesos,osannpvdheggeeetnraeetiroanteschfaereadcbtearcikssby affecting biogeochemical cycles, vegetation composition and structure, land–atmosphere water and heat exchanges, atmospheric chemistry and composition, and human health and property (Bond et al, 2004; IPCC, 2007; Cochrane and Ryan, 2009; Bond-Lamberty et al, 2009; van der Werf et al, 2010; Bowman et al, 2011). Existing global fire parameterizations suitable for ESMs aim to best match the observed fire regimes for the contemporary time period, given that only the contemporary global fire product is available with a sufficient quality to be used as evaluation data (Kloster et al, 2010; Prentice et al, 2011). These global fire parameterizations are generally large scale (103–105 km2), and simplify the real world at various levels to improve the efficiency of computations. A modified version of SPITFIRE was used in LPX-DGVM by Prentice et al (2011)

Objectives

Results

Conclusion