AttentionFire_v1.0: interpretable machine learning fire model for burned area predictions over tropics. Li, F., Zhu, Q., Riley, W., Zhao, L., Xu, L., Yuan, K., Chen, M., Wu, H., Gui, Z., Gong, J., & Randerson, J. 2022.
AttentionFire_v1.0: interpretable machine learning fire model for burned area predictions over tropics [link]Paper  doi  abstract   bibtex   
African and South American (ASA) wildfires account for more than 70 % of global burned areas and have strong connection to local climate for sub-seasonal to seasonal wildfire dynamics. However, representation of the wildfire-climate relationship remains challenging, due to spatiotemporally heterogenous responses of wildfires to climate variability and human influences. Here, we developed an interpretable Machine Learning (ML) fire model (AttentionFire_v1.0) to resolve the complex spatial- heterogenous and time-lagged controls from climate on burned area and to better predict burned areas over ASA regions. Our ML fire model substantially improved predictability of burned area for both spatial and temporal dynamics compared with five commonly used machine learning models. More importantly, the model revealed strong time-lagged control from climate wetness on the burned areas. The model also predicted that under a high emission future climate scenario, the recently observed declines in burned area will reverse in South America in the near future due to climate changes. Our study provides reliable and interpretable fire model and highlights the importance of lagged wildfire-climate relationships in historical and future predictions.
@preprint{li2022attentionfirev10,
  abstract = {African and South American (ASA) wildfires account for more than 70 % of global burned areas and have strong connection to local climate for sub-seasonal to seasonal wildfire dynamics. However, representation of the wildfire-climate relationship remains challenging, due to spatiotemporally heterogenous responses of wildfires to climate variability and human influences. Here, we developed an interpretable Machine Learning (ML) fire model (AttentionFire_v1.0) to resolve the complex spatial- heterogenous and time-lagged controls from climate on burned area and to better predict burned areas over ASA regions. Our ML fire model substantially improved predictability of burned area for both spatial and temporal dynamics compared with five commonly used machine learning models. More importantly, the model revealed strong time-lagged control from climate wetness on the burned areas. The model also predicted that under a high emission future climate scenario, the recently observed declines in burned area will reverse in South America in the near future due to climate changes. Our study provides reliable and interpretable fire model and highlights the importance of lagged wildfire-climate relationships in historical and future predictions.},
  added-at = {2022-08-15T10:57:01.000+0200},
  author = {Li, Fa and Zhu, Qing and Riley, William and Zhao, Lei and Xu, Li and Yuan, Kunxiaojia and Chen, Min and Wu, Huayi and Gui, Zhipeng and Gong, Jianya and Randerson, James},
  biburl = {https://www.bibsonomy.org/bibtex/2ce1f7d94c85df527c7332dfd11ec38e8/pbett},
  description = {GMDD - AttentionFire_v1.0: interpretable machine learning fire model for burned area predictions over tropics},
  doi = {https://doi.org/10.5194/gmd-2022-195},
  interhash = {a16730b4bd6e19bfb24c6dadd869f767},
  intrahash = {ce1f7d94c85df527c7332dfd11ec38e8},
  issn = {1991-959X},
  journal = {Geoscientific Model Development Discussions},
  keywords = {MyFireWork machinelearning fire},
  pages = {1-28},
  publisher = {Copernicus GmbH},
  timestamp = {2022-08-15T10:57:01.000+0200},
  title = {AttentionFire_v1.0: interpretable machine learning fire model for burned area predictions over tropics},
  type = {misc},
  url = {https://gmd.copernicus.org/preprints/gmd-2022-195/},
  year = 2022
}

Downloads: 0