PARENT SESSION
Poster Session 11: Disturbance Ecology
Tuesday, August 9, 5:00 PM - 6:30 PM, Exhibit Hall 220 A-E, Level 2, Palais des congrès de Montréal
Annual release of carbon from forest fire in China during 1950-2000: Integrating remotely sensed data, forest fire inventory and models.
Lu, Aifeng*,1, 2, Tian, Hanqin2, Liu, Mingliang2, Liu, Jiyuan1, Melillo, Jerry3, 1 Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China2 School of Forestry and Wildlife Sciences, Auburn University, Auburn, AL, USA3 The Ecosystem Center, Marine Biological Laboratory, Woods Hole, MA, USA
ABSTRACT- We have estimated the emission of carbon and carbon-containing trace gases (CO2, CO, CH4, NMHC) from forest fires in China for the time period from 1950-2000, by using a combination of remote sensing, forest fire inventory and terrestrial ecosystem modeling. Our results suggest that mean annual carbon emission from forest fires in China is about 11.31 Tg per year, varying from a minimum level of 8.55 Tg per year to a maximum level of 13.9 Tg per year. This amount of carbon emission includes 40.66 Tg CO2 with a range from 29.21 to 47.53 Tg CO2, 2.71 Tg CO with a range from 1.48 to 4.30 Tg CO, 0.112 Tg CH4 with a range from 0.06 to 0.2 Tg CH4, and 0.113 Tg NMHC with a range from 0.05 to 0.19 Tg NMHC. Our study indicates that fire-induced carbon emissions show substantially interannual and decadal variations before 1980, but relatively stable from then to 2000 because of the application of fire suppression. Large spatial variation in fire-induced carbon emissions exists due to the spatial variability of climate, forest types and fire regimes. Uncertainty in the estimation of fire-induced carbon emission is primarily due to our poor estimation of burnt areas and biomass. To examine the uncertainty in the estimation of burnt area, we have generated the spatial distribution of burnt forest area over the entire nation in 2000 by using multi-scale remotely sensed data including AVHRR, SPOT-Vegetation, ATSR and Landsat TM/ETM for the year of 2000. We have found that our estimate on carbon emission can be significantly improved by using remote sensing-derived fire information so that accurate estimate on fire-induced carbon emission will required to have further integration of remotely sensed data into carbon cycle models.
Key words: Biomass burning, Carbon emission, Forest fire, Trace gas