journal article·2026-02-27Open access

Forest Fire Forecasting and Early Warning Model Based on Satellite FY-4 and Other Multi-Source Data

Zijun Zheng, Fuzeng Wang, Ning Yang, Boshen Zhang, Zhaojun Deng, Peng Fang, Guangmin Cao, Hongxia Shi

Applied Sciences· Volume 16· Issue 5

Abstract

Forest fires pose a severe threat to ecosystems and human safety in the mountainous Liangshan Prefecture, China. To enhance early warning capabilities, this study constructs a fire risk prediction model by combining FY-4A satellite data, historical fire records, and terrain data. Through principal component analysis and multicollinearity diagnostics, critical factors including topography, vegetation indices (NDVI, NDII7), and a moisture index (TVDI) were selected for a logistic regression model. The model successfully identifies key risk drivers, with vegetation moisture content (NDII7) being the most influential. Validation shows the model achieves an AUC of 0.77 and a prediction accuracy of 71.5%, confirming its effectiveness. This work demonstrates the utility of China’s FY-4 satellite for operational forest fire risk forecasting, providing a methodological basis for improved disaster prevention.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Fire effects on ecosystems · Fire Detection and Safety Systems · Landslides and related hazards

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Liangshan Meteorological Bureau, Xichang 615000, China

The Key Laboratory of Rainstorm and Drought-Flood in Plateaus and Basins, Chengdu 610072, China

College of Electronic Engineering, Chengdu University of Information Technology, Chengdu 610225, China

Heilongjiang Meteorological Data Center, Harbin 150030, China

College of Atmospheric Sciences, Chengdu University of Information Technology, Chengdu 610225, China

Funding

Sichuan Provincial Science and Technology Plan Projects · Liangshan Prefecture Science and Technology Plan Projects · Science Development Fund Project of the Sichuan Key Laboratory for Rainstorm, Drought, and Flood Disasters in Plateau and Basin Area · Sichuan Provincial Natural Science Fund Projects · Ningxia Natural Science Fund Projects · Project of Introducing Talents for Scientific Research

References (38)

Show source-provided references
1. Remote sensing monitoring of forest fire in Liangshan Prefecture, Sichuan province
2. Assessment of forest fire danger rating systems
3. Modeling initial attack success of wildfire suppression in Catalonia, Spain10.1016/j.scitotenv.2019.02.323 ↗
4. Spatial pattern analysis and prediction of forest fire using new machine learning approach of Multivariate Adaptive Regression Splines and Differential Flower Pollination optimization: A case study at Lao Cai province (Viet Nam)10.1016/j.jenvman.2019.01.108 ↗
5. Gao, B., Jia, W., Wang, Q., and Yang, G. (2025). All-Weather Forest Fire Automatic Monitoring and Early Warning Application Based on Multi-Source Remote Sensing Data: Case Study of Yunnan. Fire, 8.10.3390/fire8090344 ↗
6. Zhang, Z., Wang, L., Xue, N., and Du, Z. (2021). Spatiotemporal Analysis of Active Fires in the Arctic Region during 2001–2019 and a Fire Risk Assessment Model. Fire, 4.10.3390/fire4030057 ↗
7. Tian, Y., Wu, Z., Li, M., Wang, B., and Zhang, X. (2022). Forest Fire Spread Monitoring and Vegetation Dynamics Detection Based on Multi-Source Remote Sensing Images. Remote Sens., 14.10.3390/rs14184431 ↗
8. Liang, H., Zheng, C., Liu, X., Tian, Y., Zhang, J., and Cui, W. (2023). Super-Resolution Reconstruction of Remote Sensing Data Based on Multiple Satellite Sources for Forest Fire Smoke Segmentation. Remote Sens., 15.10.3390/rs15174180 ↗
9. Rostami, A., Shah-Hosseini, R., Asgari, S., Zarei, A., Aghdami-Nia, M., and Homayouni, S. (2022). Active Fire Detection from Landsat-8 Imagery Using Deep Multiple Kernel Learning. Remote Sens., 14.10.3390/rs14040992 ↗
10. Wang, H., Zhang, G., Yang, Z., Xu, H., Liu, F., and Xie, S. (2024). Satellite Remote Sensing False Forest Fire Hotspot Excavating Based on Time-Series Features. Remote Sens., 16.10.3390/rs16132488 ↗
11. Xie, Z., Song, W., Ba, R., Li, X., and Xia, L. (2018). A Spatiotemporal Contextual Model for Forest Fire Detection Using Himawari-8 Satellite Data. Remote Sens., 10.10.3390/rs10121992 ↗
12. Shahriar, S.A., Choi, Y., and Islam, R. (2025). Advanced Deep Learning Approaches for Forecasting High-Resolution Fire Weather Index (FWI) over CONUS: Integration of GNN-LSTM, GNN-TCNN, and GNN-DeepAR. Remote Sens., 17.10.3390/rs17030515 ↗
13. Ntinopoulos, N., Sakellariou, S., Christopoulou, O., and Sfougaris, A. (2023). Fusion of Remotely-Sensed Fire-Related Indices for Wildfire Prediction through the Contribution of Artificial Intelligence. Sustainability, 15.10.3390/su151511527 ↗
14. Wildfire spreading prediction using multimodal data and deep neural network approach10.1038/s41598-024-52821-x ↗
15. Wildfire assessment using machine learning algorithms in different regions10.1186/s42408-024-00335-2 ↗
16. Multi-time scale features of fire weather in two major forests in China during 1961–2020
17. Study on forecasting danger weather grades of forest fire in Ningxia Autonomous region
18. Study on the forest fire danger class division of Yong’an city
19. Study on forest fire risk in Conghua District of Guangzhou City based on multi-source data10.1007/s11069-022-05510-9 ↗
20. The response of forest fires to climate change and fire trend prediction in Yunnan province
21. Research on prediction model and risk zoning of forest fire in Liangshan Prefecture
22. MODIS brightness temperature change-based forest fire monitoring10.1007/s12524-019-01071-w ↗
23. Analysis of forest fire risk in Sichuan Liangshan based on logistic model
24. Dong, Z., Yu, J., An, S., Zhang, J., Li, J., and Xu, D. (2022). Forest Fire Detection of FY-3D Using Genetic Algorithm and Brightness Temperature Change. Forests, 13.10.3390/f13060963 ↗
25. Dong, L., Wang, Y., Li, C., Zhu, W., Yu, H., and Tian, H. (2025). A Joint Transformer–XGBoost Model for Satellite Fire Detection in Yunnan. Fire, 8.10.3390/fire8100376 ↗
26. Forest Fire Susceptibility Modeling Using a Convolutional Neural Network for Yunnan Province of China10.1007/s13753-019-00233-1 ↗
27. Predictive model of spatial scale of forest fire driving factors: A case study of Yunnan Province, China10.1038/s41598-022-23697-6 ↗
28. Jang, E., Kang, Y., Im, J., Lee, D.-W., Yoon, J., and Kim, S.-K. (2019). Detection and Monitoring of Forest Fires Using Himawari-8 Geostationary Satellite Data in South Korea. Remote Sens., 11.10.3390/rs11030271 ↗
29. Das, J., Mahato, S., Joshi, P.K., and Liou, Y.-A. (2023). Forest Fire Susceptibility Zonation in Eastern India Using Statistical and Weighted Modelling Approaches. Remote Sens., 15.10.3390/rs15051340 ↗
30. Zhang, Z., Tian, Y., Wang, G., Zheng, C., and Zhao, F. (2024). A forest fire prediction method for lightning stroke based on remote sensing data. Forests, 15.10.3390/f15040647 ↗
31. Pang, Y., Li, Y., Feng, Z., Feng, Z., Zhao, Z., Chen, S., and Zhang, H. (2022). Forest Fire Occurrence Prediction in China Based on Machine Learning Methods. Remote Sens., 14.10.3390/rs14215546 ↗
32. Zhao, P., Zhang, F., Lin, H., and Xu, S. (2021). GIS-Based Forest Fire Risk Model: A Case Study in Laoshan National Forest Park, Nanjing. Remote Sens., 13.10.3390/rs13183704 ↗
33. Zhang, F., Zhang, B., Luo, J., Liu, H., Deng, Q., Wang, L., and Zuo, Z. (2023). Forest fire driving factors and fire risk zoning based on an optimal parameter logistic regression model: A case study of the Liangshan Yi Autonomous Prefecture, China. Fire, 6.10.3390/fire6090336 ↗
34. Landscape dynamics in crown fire ecosystems10.1007/BF00135079 ↗
35. Asam, S., Eisfelder, C., Hirner, A., Reiners, P., Holzwarth, S., and Bachmann, M. (2023). AVHRR NDVI Compositing Method Comparison and Generation of Multi-Decadal Time Series—A Timeline Thematic Processor. Remote Sens., 15.10.3390/rs15061631 ↗
36. Evaluating soil moisture status in China using the temperature–vegetation dryness index (TVDI)10.5589/m04-029 ↗
37. Detection of fire impact and vegetation recovery over tropical peat swamp forest by satellite data and ground-based NDVI instrument10.1080/01431160903302981 ↗
38. A comparative analysis of the use of NOAA-AVHRR NDVI and FWI data for forest fire risk assessment10.1080/01431160801958397 ↗