Introduction

CO₂ and CH₄ are known as important greenhouse effect gases and CO₂ is considered to be the most important contributor to global warming and CH₄ is the second. Soil is one of the source of atmospheric CO₂ and sink of atmospheric methane. Therefore land-use change should affect gas fluxes of soils (IPCC, 2000). Forest fire is major massive destruction of forest ecosystems. Since forest fire changes physical and chemical characteristics of soils, and microbial biomass, activity and composition of microbe in soils (e.g. Fritze et al., 1992; Kennard and Gholtz, 2001), forest fire may affect soil–atmospheric exchange of CO₂ and CH₄ gases. But there has been a little attention especially on CH₄ flux (Chen-rui et al., 2003). So more information is needed forest fire effect on these gas fluxes of soil processes.

  This study aimed to investigate the effects of forest fire on CO₂ and CH₄ flux in a temperate deciduous forest in Japan. In this area, a forest fire happened on 5-6th April 2002.

Material and method

CO₂ and CH₄ fluxes were measured by closed chamber method at both unburned and burned sites in the forest. CO₂ flux was monthly measured from September 2003 to October 2004, and CH₄ flux was measured from July 2002 to October 2004.

Soil core incubation analysis was conducted in order to evaluate the degree of depth forest fire affected. Soil cores were obtained from 0-5 cm, 5-10 cm, and 10-15 cm layers except Ao layer. The following soil properties were analyzed: water content, WFPS, pH, CN ratio (Total C, Total N), NH₄⁺, microbial biomass C, FDA hydrolysis activity.

Result

  Through all seasons, there was a significant difference in CO₂ flux between unburned site (5.3 ± 1.1 g CO₂/m²/day; mean ± S.E.) and burned site (9.5 ± 1.9 g CO₂/m²/day) (p=0.002 by two-sided Wilcoxon's signed rank test). Also, there was a significant difference in CH₄ flux between unburned site (-3.57 ± 0.60 g CH₄/m²/day) and burned site (-0.85 ± 0.20 g CH₄/m²/day) (p=0.007 by two-sided Wilcoxon's signed rank test). The CO₂ fluxes at both sites had exponential correlation with soil temperature. The Q₁₀ values are 2.46 at unburned and 2.25 at burned site. Not significant but weak negative relationship between CO₂ flux and water content at both sites was observed (unburned site, R=-0.497, p=0.100; burned site, R=-0.441, p=0.151). On the other hands, CH₄ fluxes at unburned and burned sites had no significant correlation with soil temperature nor water content.

  In incubation experiments (Fig.1a), CO₂ evolution rate at 0–5 cm layer was higher than those of the other layers in both unburned and burned sites. Comparison between unburned and burned site's samples at same layer showed a significant difference at 0–5 cm layers (p=0.047), but not at 5-10 cm (p=0.46) and 10-15 cm layers (p=0.12). On methane uptake rate (Fig.1b), 5-10 cm layer in unburned site showed the highest among three layers. On the other hand, in burned site, 10–15 cm layer was the highest and no uptake was found at 0-5 cm layer. Comparison between unburned and burned site's samples at same layer, there was a significant difference in 0-5 cm (p=0.009) and 5-10 cm layers (p=0.016).

Discussion

  The CO₂ flux at burned site was about a half of that of unburned site. Since all trees were burned out and dead in the burned site, the CO₂ emission flux at burned site mainly was derived from heterotrophic respiration. Therefore decrease of CO₂ flux is mainly explained by reduction of autotrophic respiration. In core incubation the result suggests heterotrophic respiration partially distribute decline of CO₂ flux.

  The CH₄ uptake flux at burned site was reduced to about 25% of the value at unburned site. In core incubation experiment, we found that forest fire affected on 0–10 cm layer of soil, and not on 10–15 cm layer. However, methane uptake flux at field was largely decreased. It suggests that surface methane oxidation is very important.

Conclusion

  Forest fire reduces CO₂ flux and CH₄ uptake flux in temperate deciduous forest in Japan.