Japan Geoscience Union Meeting 2025

Presentation information

[J] Poster

A (Atmospheric and Hydrospheric Sciences ) » A-CG Complex & General

[A-CG46] Biogeochemical Cycles in Land Ecosystem

Tue. May 27, 2025 5:15 PM - 7:15 PM Poster Hall (Exhibition Hall 7&8, Makuhari Messe)

convener:Munemasa Teramoto(Arid Land Research Center, Tottori University), Tomomichi Kato(Research Faculty of Agriculture, Hokkaido University), Kazuhito Ichii(Chiba University), Takeshi Ise(FSERC, Kyoto University)

5:15 PM - 7:15 PM

[ACG46-P07] Effect of El Niño–Southern Oscillation on the CO2 emission rate from wildfire events

*Yasuto Watanabe1, Makoto Deushi1, Seiji Yukimoto1, Yukimasa Adachi1, Masahiro Hosaka1, Hiroyuki Tsujino1 (1.Meteorological Research Institute)

Keywords:Wildfire, Global warming

Wildfire events occur in various regions on Earth and affect vegetation activities, land carbon budget, atmospheric chemistry, and human activities in a short period of time. The frequency and magnitude of wildfire events depend on many factors including lightning frequency, human ignition, and dryness of soil carbon. This means that the mechanisms of the occurrence of wildfire events would interact with multiple factors of the Earth system, which limits our understanding of the fundamental behavior of wildfire activities. Simulations using a process-based wildfire model have successfully simulated the wildfire distributions on the present Earth, while its response to climate changes differs between models. For a better understanding of the vegetation and wildfire activities, we have developed a dynamical vegetation–wildfire model driven by the state-of-the-art meteorological reanalysis data JRA-3Q (Kosaka et al., 2024). This model was developed based on the LPJ-LMfire model by Pfeiffer et al. (2013). We modified the treatment of the human-driven ignitions as a function of the population density to simulate the human-driven ignitions in the present. We have also considered the effect of the changes in the land use. Using this model, we estimated the long-term trend and interannual variations of the occurrence of wildfire events from 1948 to 2014 CE.

We show that the CO2 emission rate from wildfire events (Φ↑,fire) exhibits a slight increasing trend since 1948 CE. The value of Φ↑,fire was ~2.5 PgC yr–1 in the early 1950s, while it increased to ~2.7–2.8 PgC yr–1 in the early 2010s. This value is broadly consistent with the estimated values based on the satellite observations. The estimated values of Φ↑,fire also show strong interannual variations related to El-Niño Southern Oscillation (ENSO). During the interval of El Niño events, Φ↑,fire tends to be high while it tends to be low during the interval of La Niña events. Specifically, during the intense El Niño events in 1982–1983 and 1997–1998, Φ↑,fire reached ~3.4 PgC yr–1. The composite of the results of the interval of El Niño events show that Φ↑,fire is especially high in Indonesia, the northern part of South America, the Indochina Peninsula, and the savanna regions in North Africa. This would have decreased the net primary production in these regions in El Niño intervals, which would further increase the CO2 production rates from these regions during the El Niño events. These results may help constrain the response of the wildfire distributions under high temperature conditions in future climate changes.