LinU-Mamba: Visual Mamba U-Net with Linear Attention to Predict Wildfire Spread
Abstract
1. Introduction
- LinU-Mamba shows superior performance compared to existing deep learning methods using the same dataset, while presenting a more efficient training time.
- We explore the impact of pre-training and feature selection when using remote sensing data and demonstrate the importance of using linear attention in vision Mamba-based U-Net architectures.
- To the best of our knowledge, LinU-Mamba is the first model based on state-space mechanisms in the field of wildfire spread prediction, making it a valuable contribution and opening up potential new directions for future research in this area.
2. Related Work
2.1. Wildfire Spread Prediction Techniques
2.2. State-Space Models, Mamba and Vision Mamba
2.3. Vision Mamba-Based U-Net Models
3. Methodology
3.1. Network Architecture
3.1.1. Overall Description
3.1.2. Encoder
3.1.3. Decoder
3.2. Building Blocks
3.2.1. Linear Attention
3.2.2. State-Space Models
3.2.3. The VSS Block and the SS2D Module
Algorithm 1: Forward pass of an S6 block. |
4. Experimental Setup
4.1. Dataset
4.2. Data Preprocessing and Data Augmentation
4.3. Loss
4.4. Performance Metrics
4.5. Implementation Details
5. Results
5.1. Ablation Studies
5.1.1. VSS Block
5.1.2. Pre-Training
5.1.3. Linear Attention
6. Discussion
6.1. Architecture Implications
6.2. Further Analysis
6.3. Study Limitations and Future Work
7. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
Abbreviations
NDWS | Next Day Wildfire Spread |
MODIS | Moderate Resolution Imaging Spectroradiometer |
VIIRS | Visible Infrared Imaging Radiometer Suite |
ABI | Advanced Baseline Imager |
MSI | Multispectral Instrument |
CNN | Convolutional neural network |
LSTM | Long short-term memory |
SHAP | SHapley Additive exPlanations |
FBP | Canadian Forest Fire Behavior Prediction System |
LLMs | Large language models |
FRP | Fire Radiative Power |
VSS | Visual State Space |
SSMs | State-space models |
Vim | Vision Mamba |
ODEs | Ordinary differential equations |
SS2D | 2D-Selective-Scan |
FFN | Feed-forward network |
ERC | Energy release component |
WBCE | Weighted Binary Cross Entropy |
References
- Potapov, P.; Tyukavina, A.; Turubanova, S.; Hansen, M.C.; Giglio, L.; Hernandez-Serna, A.; Lima, A.; Harris, N.; Stolle, F. Unprecedentedly high global forest disturbance due to fire in 2023 and 2024. Proc. Natl. Acad. Sci. USA 2025, 122, e2505418122. [Google Scholar] [CrossRef]
- Canada’s Record-Breaking Wildfires in 2023: A Fiery Wake-Up Call. Available online: https://natural-resources.canada.ca/stories/simply-science/canada-s-record-breaking-wildfires-2023-fiery-wake-call (accessed on 26 July 2025).
- Annual Area Burnt by Wildfires. Available online: https://ourworldindata.org/grapher/annual-area-burnt-by-wildfires (accessed on 4 June 2025).
- Billion-Dollar Weather and Climate Disasters. Available online: https://www.ncei.noaa.gov/access/billions/ (accessed on 26 July 2025).
- MODIS Moderate Resolution Imaging Spectroradiometer. Available online: https://modis.gsfc.nasa.gov/ (accessed on 4 June 2025).
- VIIRS Visible Infrared Imaging Radiometer Suite. Available online: https://www.earthdata.nasa.gov/data/instruments/viirs (accessed on 4 June 2025).
- ABI Advanced Baseline Imager. Available online: http://earthdata.nasa.gov/data/instruments/abi (accessed on 4 June 2025).
- Sentinel-2 MSI. Available online: https://www.earthdata.nasa.gov/data/instruments/sentinel-2-msi (accessed on 4 June 2025).
- Radke, D.; Hessler, A.; Ellsworth, D. FireCast: Leveraging Deep Learning to Predict Wildfire Spread. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), Macao, China, 10–16 August 2019; pp. 4575–4581. [Google Scholar]
- Masrur, A.; Yu, M.; Taylor, A. Capturing and Interpreting Wildfire Spread Dynamics: Attention-Based Spatiotemporal Models Using ConvLSTM Networks. Ecol. Inform. 2024, 82, 102760. [Google Scholar] [CrossRef]
- Li, B.S.; Rad, R. Wildfire Spread Prediction in North America Using Satellite Imagery and Vision Transformer. In Proceedings of the 2024 IEEE Conference on Artificial Intelligence (CAI), Singapore, 25–27 June 2024; pp. 1536–1541. [Google Scholar]
- Subramanian, S.G.; Crowley, M. Combining MCTS and A3C for Prediction of Spatially Spreading Processes in Forest Wildfire Settings. In Proceedings of the Advances in Artificial Intelligence, 31st Canadian Conference on Artificial Intelligence, Canadian AI 2018, Toronto, ON, Canada, 8–11 May 2018; pp. 285–291. [Google Scholar]
- Jiang, W.; Wang, F.; Su, G.; Li, X.; Wang, G.; Zheng, X.; Wang, T.; Meng, Q. Modeling Wildfire Spread with an Irregular Graph Network. Fire 2022, 5, 185. [Google Scholar] [CrossRef]
- Huot, F.; Hu, R.L.; Goyal, N.; Sankar, T.; Ihme, M.; Chen, Y.F. Next Day Wildfire Spread: A Machine Learning Dataset to Predict Wildfire Spreading from Remote-Sensing Data. IEEE Trans. Geosci. Remote Sens. 2022, 60, 4412513. [Google Scholar] [CrossRef]
- Hirsch, K. Canadian Forest Fire Behavior Prediction (FBP) System: User’s Guide; Canadian Forest Service, Northern Forestry Centre: Edmonton, AB, Canada, 1996.
- Finney, M.A. FARSITE, Fire Area Simulator-Model Development and Evaluation; U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: Washington, DC, USA, 1998.
- Rothermel, R. A Mathematical Model for Predicting Fire Spread in Wildland Fuels. In The Bark Beetles, Fuels, and Fire Bibliography; US Forest Service: Washington, DC, USA, 1972; Volume Res. Pap. INT-115, pp. 1–40. [Google Scholar]
- Prometheus. Available online: https://firegrowthmodel.ca/#/prometheus_overview (accessed on 4 June 2025).
- Alexandridis, A.; Vakalis, D.; Siettos, C.I.; Bafas, G.V. A cellular automata model for forest fire spread prediction: The case of the wildfire that swept through Spetses Island in 1990. Appl. Math. Comput. 2008, 204, 191–201. [Google Scholar] [CrossRef]
- Zheng, Z.; Huang, W.; Li, S.; Zeng, Y. Forest Fire Spread Simulating Model Using Cellular Automaton with Extreme Learning Machine. Ecol. Model. 2017, 348, 33–43. [Google Scholar] [CrossRef]
- Xu, Y.; Li, D.; Ma, H.; Lin, R.; Zhang, F. Modeling Forest Fire Spread Using Machine Learning-Based Cellular Automata in a GIS Environment. Forests 2022, 13, 1974. [Google Scholar] [CrossRef]
- Khanmohammadi, S.; Arashpour, M.; Golafshani, E.M.; Cruz, M.G.; Rajabifard, A.; Bai, Y. Prediction of Wildfire Rate of Spread in Grasslands Using Machine Learning Methods. Environ. Model. Softw. 2022, 156, 105507. [Google Scholar] [CrossRef]
- Wood, D.A. Prediction and Data Mining of Burned Areas of Forest Fires: Optimized Data Matching and Mining Algorithm Provides Valuable Insight. Artificial Intelligence in Agriculture. Artif. Intell. Agric. 2021, 5, 24–42. [Google Scholar] [CrossRef]
- Singh, A.; Yadav, R.; Sudhamshu, G.; Basnet, A.; Ali, R. Wildfire Spread Prediction Using Machine Learning Algorithms. In Proceedings of the 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India, 6–8 July 2023; pp. 1–5. [Google Scholar]
- Rubí, J.N.S.; de Carvalho, P.H.P.; Gondim, P.R.L. Application of Machine Learning Models in the Behavioral Study of Forest Fires in the Brazilian Federal District Region. Eng. Appl. Artif. Intell. 2023, 118, 105649. [Google Scholar] [CrossRef]
- Marjani, M.; Mesgari, M.S. The Large-Scale Wildfire Spread Prediction Using a Multi-Kernel Convolutional Neural Network. ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 2023, 14W1, 483–488. [Google Scholar] [CrossRef]
- Khennou, F.; Ghaoui, J.; Akhloufi, M.A. Forest Fire Spread Prediction Using Deep Learning. In Proceedings of the Geospatial Informatics XI, Bellingham, FL, USA, 10–11 December 2021; pp. 106–117. [Google Scholar]
- Chen, Y.; Hantson, S.; Andela, N.; Coffield, S.R.; Graff, C.A.; Morton, D.C.; Ott, L.E.; Foufoula-Georgiou, E.; Smyth, P.; Goulden, M.L.; et al. California Wildfire Spread Derived Using VIIRS Satellite Observations and an Object-Based Tracking System. Sci. Data 2022, 9, 249. [Google Scholar] [CrossRef]
- Marjani, M.; Mahdianpari, M.; Mohammadimanesh, F. CNN-BiLSTM: A Novel Deep Learning Model for Near-Real-Time Daily Wildfire Spread Prediction. Remote Sens. 2024, 16, 1467. [Google Scholar] [CrossRef]
- Shadrin, D.; Illarionova, S.; Gubanov, F.; Evteeva, K.; Mironenko, M.; Levchunets, I.; Belousov, R.; Burnaev, E. Wildfire Spreading Prediction Using Multimodal Data and Deep Neural Network Approach. Sci. Rep. 2024, 14, 2606. [Google Scholar] [CrossRef]
- Ramesh, M.; Sun, Z.; Li, Y.; Zhang, L.; Annam, S.K.; Fang, H.; Tong, D. Assessing WildfireGPT: A comparative analysis of AI models for quantitative wildfire spread prediction. Nat. Hazards 2025, 121, 13117–13130. [Google Scholar] [CrossRef]
- Gu, A.; Goel, K.; Re, C. Efficiently Modeling Long Sequences with Structured State Spaces. In Proceedings of the International Conference on Learning Representations, Virtual, 25 April 2022. [Google Scholar]
- Goel, K.; Gu, A.; Donahue, C.; Ré, C. It is Raw! Audio Generation with State-Space Models. arXiv 2022, arXiv:2202.09729. [Google Scholar] [CrossRef]
- Ma, X.; Zhou, C.; Kong, X.; He, J.; Gui, L.; Neubig, G.; May, J.; Zettlemoyer, L. Mega: Moving Average Equipped Gated Attention. In Proceedings of the Eleventh International Conference on Learning Representations, Kigali, Rwanda, 1–5 May 2023. [Google Scholar]
- Gu, A.; Dao, T. Mamba: Linear-Time Sequence Modeling with Selective State Spaces. arXiv 2024, arXiv:2312.00752. [Google Scholar] [CrossRef]
- Hwang, S.; Lahoti, A.S.; Puduppully, R.; Dao, T.; Gu, A. Hydra: Bidirectional State Space Models Through Generalized Matrix Mixers. In Proceedings of the Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024, NeurIPS 2024, Vancouver, BC, Canada, 10–15 December 2024; Curran Associates, Inc.: Red Hook, NY, USA, 2024; pp. 1–33. [Google Scholar]
- Nguyen, E.; Goel, K.; Gu, A.; Downs, G.W.; Shah, P.; Dao, T.; Baccus, S.A.; Ré, C. S4ND: Modeling Images and Videos as Multidimensional Signals with State Spaces. In Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, 28 November–9 December 2022; Koyejo, S., Mohamed, S., Agarwal, A., Belgrave, D., Cho, K., Oh, A., Eds.; Curran Associates, Inc.: Red Hook, NY, USA, 2022. [Google Scholar]
- Xiao, C.; Li, M.; ZHANG, Z.; Meng, D.; Zhang, L. Spatial-Mamba: Effective Visual State Space Models via Structure-Aware State Fusion. In Proceedings of the Thirteenth International Conference on Learning Representations, Singapore, 24–28 April 2025. [Google Scholar]
- Zhu, L.; Liao, B.; Zhang, Q.; Wang, X.; Liu, W.; Wang, X. Vision mamba: Efficient visual representation learning with bidirectional state space model. In Proceedings of the 41st International Conference on Machine Learning, ICML’24, Vienna, Austria, 21–27 July 2024. [Google Scholar]
- Huang, T.; Pei, X.; You, S.; Wang, F.; Qian, C.; Xu, C. LocalMamba: Visual State Space Model with Windowed Selective Scan. arXiv 2024, arXiv:2403.09338. [Google Scholar] [CrossRef]
- Ruan, J.; Li, J.; Xiang, S. VM-UNet: Vision Mamba UNet for Medical Image Segmentation. arXiv 2024, arXiv:2402.02491. [Google Scholar] [CrossRef]
- Ma, J.; Li, F.; Wang, B. U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation. arXiv 2024, arXiv:2401.04722. [Google Scholar] [CrossRef]
- Wu, R.; Liu, Y.; Liang, P.; Chang, Q. H-vmunet: High-order Vision Mamba UNet for medical image segmentation. Neurocomputing 2025, 624, 129447. [Google Scholar] [CrossRef]
- Liu, J.; Yang, H.; Zhou, H.Y.; Xi, Y.; Yu, L.; Li, C.; Liang, Y.; Shi, G.; Yu, Y.; Zhang, S.; et al. Swin-UMamba: Mamba-Based UNet with ImageNet-Based Pretraining. In Medical Image Computing and Computer Assisted Intervention—MICCAI 2024; Linguraru, M.G., Dou, Q., Feragen, A., Giannarou, S., Glocker, B., Lekadir, K., Schnabel, J.A., Eds.; Springer: Cham, Switzerland, 2024; pp. 615–625. [Google Scholar] [CrossRef]
- Su, C.; Luo, X.; Li, S.; Chen, L.; Wang, J. VMKLA-UNet: Vision Mamba with KAN linear attention U-Net. Sci. Rep. 2025, 15, 13258. [Google Scholar] [CrossRef] [PubMed]
- Zhong, X.; Lu, G.; Li, H. Vision Mamba and xLSTM-UNet for medical image segmentation. Sci. Rep. 2025, 15, 8163. [Google Scholar] [CrossRef]
- Xie, F.; Nie, J.; Tang, Y.; Zhang, W.; Zhao, H. Mamba-Adaptor: State Space Model Adaptor for Visual Recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 11–15 June 2025; pp. 20124–20134. [Google Scholar]
- Lu, D.; Xu, L.; Zhou, J.; Gao, K.; Gong, Z.; Zhang, D. 3D-UMamba: 3D U-Net with state space model for semantic segmentation of multi-source LiDAR point clouds. Int. J. Appl. Earth Obs. Geoinf. 2025, 136, 104401. [Google Scholar] [CrossRef]
- Zhao, K.; Zhang, Q.; Wan, C.; Pan, Q.; Qin, Y. Visual Mamba UNet fusion multi-scale attention and detail infusion for unsound corn kernels segmentation. Sci. Rep. 2025, 15, 10933. [Google Scholar] [CrossRef]
- Han, D.; Wang, Z.; Xia, Z.; Han, Y.; Pu, Y.; Ge, C.; Song, J.; Song, S.; Zheng, B.; Huang, G. Demystify Mamba in Vision: A Linear Attention Perspective. In Proceedings of the NeurIPS, Vancouver, BC, Canada, 10–15 December 2024. [Google Scholar]
- Li, Y.; Wang, Y.; Shao, X.; Zheng, A. An efficient fire detection algorithm based on Mamba space state linear attention. Sci. Rep. 2025, 15, 11289. [Google Scholar] [CrossRef] [PubMed]
- Liu, Y.; Tian, Y.; Zhao, Y.; Yu, H.; Xie, L.; Wang, Y.; Ye, Q.; Liu, Y. VMamba: Visual State Space Model. arXiv 2024, arXiv:2401.10166. [Google Scholar]
- Google Earth Engine. Available online: https://earthengine.google.com/ (accessed on 26 July 2025).
- MODIS/Terra Thermal Anomalies/Fire Daily L3 Global 1km SIN Grid V006. Available online: https://www.earthdata.nasa.gov/data/catalog/lpcloud-mod14a1-006 (accessed on 26 July 2025).
- Farr, T.G.; Rosen, P.A.; Caro, E.; Crippen, R.; Duren, R.; Hensley, S.; Kobrick, M.; Paller, M.; Rodriguez, E.; Roth, L.; et al. The Shuttle Radar Topography Mission. Rev. Geophys. 2007, 45, RG2004. [Google Scholar] [CrossRef]
- Abatzoglou, J.T. Development of gridded surface meteorological data for ecological applications and modelling. Int. J. Climatol. 2013, 33, 121–131. [Google Scholar] [CrossRef]
- Abatzoglou, J.T.; Rupp, D.E.; Mote, P.W. Seasonal Climate Variability and Change in the Pacific Northwest of the United States. J. Clim. 2014, 27, 2125–2142. [Google Scholar] [CrossRef]
- Didan, K.; Barreto, A. VIIRS/NPP Vegetation Indices 16-Day L3 Global 500m SIN Grid V001; NASA Land Processes Distributed Active Archive Center: Sioux Falls, SD, USA, 2018. [CrossRef]
- Gridded Population of the World. Available online: https://www.earthdata.nasa.gov/data/projects/gpw (accessed on 26 July 2025).
- Marjani, M.; Mahdianpari, M.; Ali Ahmadi, S.; Hemmati, E.; Mohammadimanesh, F.; Saadi Mesgari, M. Application of Explainable Artificial Intelligence in Predicting Wildfire Spread: An ASPP-Enabled CNN Approach. IEEE Geosci. Remote Sens. Lett. 2024, 21, 2504005. [Google Scholar] [CrossRef]
- Fitzgerald, J.; Seefried, E.; Yost, J.E.; Pallickara, S.; Blanchard, N. Paying Attention to Wildfire: Using U-Net with Attention Blocks on Multimodal Data for Next Day Prediction. In Proceedings of the 25th International Conference on Multimodal Interaction, New York, NY, USA, 9 October 2023; pp. 470–480. [Google Scholar]
- Lahrichi, S.; Johnson, J.; Malof, J. Predicting Next-Day Wildfire Spread with Time Series and Attention. arXiv 2025, arXiv:2502.12003. [Google Scholar] [CrossRef]
Methods | Precision (%) Mean ± SD | Recall (%) Mean ± SD | F1-Score (%) Mean ± SD |
---|---|---|---|
Multikernel CNN [26] | 24.41 ± 5.14 | 17.58 ± 2.74 | 20.06 ± 1.88 |
CNN-ASPP [60] | 26.20 ± 1.09 | 57.71 ± 1.49 | 36.01 ± 0.78 |
WPN [61] | 31.40 | 45.10 | 37.02 |
Conv autoencoder [14] | 33.6 | 43.1 | 37.76 |
ASUFM [11] | 37.28 | 43.01 | 41.09 |
LinU-Mamba | 37.25 ± 1.49 | 49.33 ± 2.43 | 42.22 ± 0.29 |
Methods | Parameters | FLOPs | Training Time (1 Epoch) | F1-Score (%) Mean ± SD |
---|---|---|---|---|
Multikernel CNN [26] | 12.3 M | 0.59 G | 0.13 mn | 20.06 ± 1.88 |
CNN-ASPP [60] | 27.5 M | 1.12 G | 0.67 mn | 36.01 ± 0.78 |
WPN [61] | 8.7 M | - | - | 37.02 |
ASUFM [11] | 35 M | - | - | 41.09 |
LinU-Mamba | 13 M | 0.2 G | 0.22 mn | 42.22 ± 0.29 |
Methods | Precision (%) Mean ± SD | Recall (%) Mean ± SD | F1-Score (%) Mean ± SD |
---|---|---|---|
No VSSBlock | 35.75 ± 0.84 | 42.69 ± 1.08 | 38.90 ± 0.30 |
Encoder | 37.14 ± 1.69 | 49.13 ± 2.46 | 42.20 ± 0.36 |
Decoder | 36.27 ± 1.14 | 49.48 ± 1.62 | 41.82 ± 0.32 |
Both | 37.25 ± 1.49 | 49.33 ± 2.43 | 42.22 ± 0.29 |
Pre-Training | Features | Precision (%) Mean ± SD | Recall (%) Mean ± SD | F1-Score (%) Mean ± SD |
---|---|---|---|---|
No | 3 | 36.65 ± 1.25 | 49.56 ± 1.75 | 42.09 ± 0.35 |
Yes | 3 | 37.25 ± 1.49 | 49.33 ± 2.43 | 42.22 ± 0.29 |
No | 12 | 35.31 ± 0.86 | 50.48 ± 1.65 | 41.52 ± 0.27 |
Methods | Precision (%) Mean ± SD | Recall (%) Mean ± SD | F1-Score (%) Mean ± SD |
---|---|---|---|
No attention | 35.69 ± 0.76 | 49.84 ± 1.40 | 41.57 ± 0.35 |
Encoder | 36.38 ± 1.14 | 50.24 ± 1.43 | 42.17 ± 0.38 |
Decoder | 36.51 ± 0.99 | 50.01 ± 1.20 | 42.18 ± 0.33 |
Both | 37.25 ± 1.49 | 49.33 ± 2.43 | 42.22 ± 0.29 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Andrianarivony, H.S.; Akhloufi, M.A. LinU-Mamba: Visual Mamba U-Net with Linear Attention to Predict Wildfire Spread. Remote Sens. 2025, 17, 2715. https://doi.org/10.3390/rs17152715
Andrianarivony HS, Akhloufi MA. LinU-Mamba: Visual Mamba U-Net with Linear Attention to Predict Wildfire Spread. Remote Sensing. 2025; 17(15):2715. https://doi.org/10.3390/rs17152715
Chicago/Turabian StyleAndrianarivony, Henintsoa S., and Moulay A. Akhloufi. 2025. "LinU-Mamba: Visual Mamba U-Net with Linear Attention to Predict Wildfire Spread" Remote Sensing 17, no. 15: 2715. https://doi.org/10.3390/rs17152715
APA StyleAndrianarivony, H. S., & Akhloufi, M. A. (2025). LinU-Mamba: Visual Mamba U-Net with Linear Attention to Predict Wildfire Spread. Remote Sensing, 17(15), 2715. https://doi.org/10.3390/rs17152715