Search

Menu
Search
in Ecology

An explainable GeoAI framework for spatial assessment of wildfire susceptibility in the Upper Ravi sub-basin, Indian Himalaya

98 Views


Abstract

Wildfires have emerged as a significant environmental concern in the Himalayan region, particularly in the Upper Ravi sub-basin of Himachal Pradesh, India. This study aims to map wildfire susceptibility by integrating Geographic Information Systems (GIS), remote sensing data, and advanced ensemble machine learning techniques. A total of sixteen biophysical and anthropogenic conditioning factors, including topography, climatic variables, vegetation indices, and human activity indicators, were used to develop wildfire susceptibility models. Five machine learning algorithms were evaluated, including Random Forest, XGBoost, LightGBM, CatBoost, and a stacking ensemble model. Among these, the stacking model demonstrated the best predictive performance with an AUC of 0.95.To enhance model interpretability and robustness, explainability, uncertainty, and sensitivity analyses were conducted on the best-performing stacking model using SHapley Additive exPlanations (SHAP), Monte Carlo uncertainty analysis, and Sobol global sensitivity analysis. SHAP results identified temperature, soil moisture, distance to villages, and relative humidity as the most influential wildfire conditioning factors. Monte Carlo simulations (1000 iterations) yielded a mean AUC of 0.847, indicating stable model performance under input perturbations. Sobol sensitivity analysis further confirmed soil moisture and temperature as the most influential variables, with total-order sensitivity indices of 0.45 and 0.21, respectively. Spatial analysis revealed that approximately 20.75% of the study area falls within high to very high wildfire susceptibility zones, primarily associated with regions characterized by steep terrain, low soil moisture conditions, and significant anthropogenic influence. Overall, the study presents an interpretable and uncertainty-aware GeoAI framework that integrates ensemble learning with explainable artificial intelligence and sensitivity analysis, providing a reproducible approach for wildfire susceptibility assessment in complex mountainous environments.

Data availability

The data that support the findings of this study are available on request from the corresponding author.

References

  1. Feng, J. G. et al. Case-based evaluation of forest ecosystem service function in China. Ying Yong Sheng Tai Xue Bao 27, 1375–1382 (2016).

    Google Scholar 

  2. Jaafari, A., Termeh, S. V. R. & Bui, D. T. Optimized neuro-fuzzy prediction of wildfire probability using genetic and firefly algorithms. J. Environ. Manag. 243, 358–369 (2019).

    Google Scholar 

  3. Zhang, G., Wang, M. & Liu, K. Forest fire susceptibility modeling using convolutional neural networks in China. Int. J. Disaster Risk Sci. 10, 386–403 (2019).

    Google Scholar 

  4. Jain, P. et al. Machine learning applications in wildfire science and management: A review. Environ. Rev. 28, 478–505 (2020).

    Google Scholar 

  5. Pechony, O. & Shindell, D. T. Driving forces of global wildfires over the past millennium and the forthcoming century. Proc. Natl. Acad. Sci. USA 107, 19167–19170 (2010).

    Google Scholar 

  6. Giglio, L. et al. Collection 6 MODIS burned area mapping algorithm and product. Remote Sens. Environ. 217, 72–85 (2018).

    Google Scholar 

  7. Collins, B. et al. The rising threats of wildland-urban interface fires in the era of climate change: The Los Angeles 2025 fires. Innovation 6, 100835 (2025).

    Google Scholar 

  8. Michailidis, K., Pseftogkas, A., Koukouli, M. E., Biskas, C. & Balis, D. Los Angeles wildfires 2025: Satellite-based emissions monitoring and air-quality impacts. Atmosphere 17, 50 (2025).

    Google Scholar 

  9. Sagar, N. et al. Forest fire dynamics in India (2005–2022): Unveiling climatic impacts, spatial patterns, and interface with anthrax incidence. Ecol. Indic. 166, 112454 (2024).

    Google Scholar 

  10. Sarkar, M. S. et al. Ensembling machine learning models to identify forest fire-susceptible zones in Northeast India. Ecol. Inf. 81, 102598 (2024).

    Google Scholar 

  11. Reddy, C. S. et al. Identification and characterization of spatio-temporal hotspots of forest fires in South Asia. Environ. Monit. Assess. 191, 1–17 (2019).

    Google Scholar 

  12. Mohanty, A. & Mithal, V. Managing forest fires in a changing climate. Council on Energy, Environment and Water Report 1–23 (2022).

  13. Murthy, K. K., Sinha, S. K., Kaul, R. & Vaidyanathan, S. A fine-scale state-space model to understand drivers of forest fires in the Himalayan foothills. Ecol. Manag. 432, 902–911. https://doi.org/10.1016/j.foreco.2018.10.009 (2019).

    Google Scholar 

  14. Bargali, H., Calderon, L. P. P., Sundriyal, R. C. & Bhatt, D. Impact of forest fire frequency on floristic diversity in the forests of Uttarakhand, western Himalaya. Trees People 9, 100300 (2022).

    Google Scholar 

  15. Kumar, M., Sheikh, M. A., Bhat, J. A. & Bussmann, R. W. Effect of fire on soil nutrients and understory vegetation in chir pine forest in Garhwal Himalaya, India. Acta Ecol. Sin. 33, 59–63 (2013).

    Google Scholar 

  16. Mishra, M. et al. Spatial analysis and machine learning prediction of forest fire susceptibility: A comprehensive approach for effective management and mitigation. Sci. Total Environ. 926, 171713 (2024).

    Google Scholar 

  17. Guria, R. et al. Predicting forest fire probability in Similipal Biosphere Reserve (India) using Sentinel-2 MSI data and machine learning. Remote Sens. Appl. Soc. Environ. 36, 101311 (2024).

    Google Scholar 

  18. Mabdeh, A. N. et al. Forest fire susceptibility assessment using support vector regression and ANFIS-based evolutionary algorithms. Sustainability 14, 9446 (2022).

    Google Scholar 

  19. Rihan, M. et al. Forest fire susceptibility mapping with sensitivity and uncertainty analysis using machine learning and deep learning algorithms. Adv. Space Res. 72, 426–443 (2023).

    Google Scholar 

  20. Guria, R., Mishra, M., Mohanta, S. & Paul, S. Forest fire probability zonation using dNBR and machine learning models: A case study at the Similipal Biosphere Reserve, Odisha, India. Environ. Sci. Pollut Res. 32, 1–22 (2025).

    Google Scholar 

  21. Iban, M. C. & Aksu, O. SHAP-driven explainable AI framework for wildfire susceptibility mapping using MODIS active fire pixels in Izmir, Türkiye. Remote Sens. 16, 2842 (2024).

    Google Scholar 

  22. Nguyen Van, L. & Lee, G. Optimizing stacked ensemble machine learning models for accurate wildfire severity mapping. Remote Sens. 17(5), 854 (2025).

    Google Scholar 

  23. Hang, H. T. et al. Exploring forest fire susceptibility and management strategies in Western Himalaya: Integrating ensemble machine learning and explainable AI for accurate prediction and comprehensive analysis. Environ. Technol. Innov. 35, 103655 (2024).

    Google Scholar 

  24. Moumane, A. et al. Advancing wildfire susceptibility mapping through machine learning and SHAP-integrated geospatial analysis in Northern Morocco’s Mediterranean region. Front. Glob Change. 8, 1705341 (2025).

    Google Scholar 

  25. Sharma, N. Physical and Social Analysis of Ravi River Basin in Himachal Pradesh (Rating Academy, 2020).

  26. Higuera, P. E. et al. Changing strength and nature of fire–climate relationships in the northern Rocky Mountains, USA (1902–2008). PLoS One. 10, e0127563 (2015).

    Google Scholar 

  27. Tehrany, M. S., Kumar, L. & &Drielsma, M. J. Review of native vegetation condition assessment concepts and methods. J. Nat. Conserv. 40, 12–23 (2017).

    Google Scholar 

  28. Durlević, U., Ilić, V. & Aleksova, B. Wildfire probability mapping in Southeastern Europe using deep learning and machine learning models based on open satellite data. AI 7, 21 (2026).

    Google Scholar 

  29. Rahmati, O., Pourghasemi, H. R. & &Zeinivand, H. Flood susceptibility mapping using frequency ratio and weights-of-evidence models in Golestan Province, Iran. Geocarto Int. 31, 42–70 (2016).

    Google Scholar 

  30. Sannigrahi, S. et al. Effects of forest fire on terrestrial carbon emission and ecosystem production in India using remote sensing. Sci. Total Environ. 725, 138331 (2020).

    Google Scholar 

  31. Tran, T. T. K. et al. Improving wildfire susceptibility prediction using explainable hybrid machine learning. J. Environ. Manag. 351, 119724 (2024).

    Google Scholar 

  32. Guo, M. et al. Importance degree of weather elements in driving wildfire occurrence in mainland China. Ecol. Indic. 148, 110152 (2023).

    Google Scholar 

  33. Bilucan, F., Teke, A. & &Kavzoglu, T. Susceptibility mapping of wildfires using XGBoost, random forest and AdaBoost: A Mediterranean case study. in Proc. Int. Conf. Mediterranean Geosciences Union 99–101 (2022).

  34. Sazib, N., Bolten, J. D. & &Mladenova, I. E. Assessing fire susceptibility using NASA SMAP over Australia and California. IEEE J. Sel. Top. Appl. Earth Obs Remote Sens. 15, 779–787 (2021).

    Google Scholar 

  35. Chuvieco, E. et al. Combining NDVI and surface temperature for estimation of live fuel moisture content. Remote Sens. Environ. 92, 322–331 (2004).

    Google Scholar 

  36. Da Silva, S. S. et al. Dynamics of forest fires in the southwestern Amazon. Ecol. Manag. 424, 312–322 (2018).

    Google Scholar 

  37. Kuhn, M. & Johnson, K. Applied Predictive Modeling (Springer, 2013).

  38. Ageenko, A. et al. Landslide susceptibility mapping using machine learning: A Danish case study. ISPRS Int. J. Geo-Inf. 11, 324 (2022).

    Google Scholar 

  39. Naikoo, M. W. et al. Peri-urban land use/land cover change and drivers using geospatial techniques and GWR. Environ. Sci. Pollut Res. 30, 116421–116439 (2023).

    Google Scholar 

  40. Gigović, L. et al. Testing a new ensemble model based on SVM and random forest in forest fire susceptibility assessment and its mapping in Serbia’s Tara National Park. Forests 10, 408 (2019).

    Google Scholar 

  41. Chen, W. et al. Spatial prediction of landslide susceptibility using data mining-based kernel logistic regression, naive Bayes and RBFNetwork models for the Long County area (China). Bull. Eng. Geol. Environ. 78, 247–266 (2019).

    Google Scholar 

  42. Breiman, L. Random forests. Mach. Learn. 45, 5–32 (2001).

    Google Scholar 

  43. Chen, T. & &Guestrin, C. XGBoost: A scalable tree boosting system. in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining 785–794 (2016).

  44. Ke, G. et al. LightGBM: A highly efficient gradient boosting decision tree. Adv Neural Inf. Process. Syst 30 (2017).

  45. Zhang, Y., Zhao, Z. & Zheng, J. CatBoost for estimating daily reference crop evapotranspiration in arid regions. J. Hydrol. 588, 125087 (2020).

    Google Scholar 

  46. Prasanna Venkatesh, N. et al. CatBoost-based improved detection of P-wave changes in sinus rhythm and tachycardia conditions: A lead selection study. Phys. Eng. Sci. Med. 46, 925–944 (2023).

    Google Scholar 

  47. Wolpert, D. H. Stacked generalization. Neural Netw. 5, 241–259 (1992).

    Google Scholar 

  48. Dou, J. et al. Improved landslide assessment using SVM with ensemble learning in Japan. Landslides 17, 641–658 (2020).

    Google Scholar 

  49. Inan, M. S. K. & Rahman, I. Explainable AI integrated feature selection for landslide susceptibility mapping using TreeSHAP. SN Comput. Sci. 4, 482 (2023).

    Google Scholar 

  50. Muhammad, S. et al. Machine learning-based forest fire vulnerability assessment in subtropical chir pine forests of Pakistan. Fire Ecol. 21, 1–17 (2025).

    Google Scholar 

  51. Bouzeraa, Y. et al. Machine learning-based wildfire susceptibility mapping: A GIS-integrated predictive framework. Appl. Sci. 15, 12188 (2025).

    Google Scholar 

  52. Symeonidis, P., Vafeiadis, T., Ioannidis, D. & Tzovaras, D. Wildfire susceptibility mapping in Greece using ensemble machine learning. Earth 6, 75 (2025).

    Google Scholar 

  53. Ghasemian, B. et al. A robust deep-learning model for landslide susceptibility mapping. Sensors 22, 1573 (2022).

    Google Scholar 

  54. Naderpour, M., Rizeei, H. M., Khakzad, N. & Pradhan, B. Forest fire-induced Natech risk assessment: A geospatial technology survey. Reliab. Eng. Syst. Saf. 191, 106558 (2019).

    Google Scholar 

  55. Youssef, A. M. & &Pourghasemi, H. R. Landslide susceptibility mapping using machine learning in Saudi Arabia. Geosci. Front. 12, 639–655 (2021).

    Google Scholar 

  56. Agrawal, N. & Dixit, J. Assessment of landslide susceptibility for Meghalaya (India) using bivariate and multi-criteria decision analysis models. All Earth. 34, 179–201 (2022).

    Google Scholar 

  57. Lundberg, S. M. & Lee, S. I. A unified approach to interpreting model predictions. Adv Neural Inf. Process. Syst. 30 (2017).

  58. Lundberg, S. M., Erion, G. G. & Lee, S. I. Consistent individualized feature attribution for tree ensembles. arXiv:1802.03888 (2018).

  59. Rodriguez, M. A. & &Dabdub, D. Monte Carlo uncertainty and sensitivity analysis of the CACM chemical mechanism. J Geophys. Res. Atmos 108 (2003).

  60. Didi, A. et al. Monte Carlo transport code for simulating the neutron yield of spallation targets for an accelerator based on high proton beam. in Proc. 4th Int. Conf. Optimization and Applications (ICOA), 1–7 (2018).

  61. Sobol’, I. M. Sensitivity estimates for nonlinear mathematical models. Math. Model. Comput. Exp. 1, 407 (1993).

    Google Scholar 

  62. Saltelli, A., Chan, K. & Scott, E. M. (eds) Sensitivity Analysis: Gauging the Worth of Scientific Models (Wiley, 2000).

  63. Tang, X. et al. Flood susceptibility assessment using a random naïve Bayes method. Catena 190, 104536 (2020).

    Google Scholar 

  64. Li, Y. et al. Forest fire risk prediction using stacking ensemble learning in Yunnan province, China. Fire 7, 13 (2023).

    Google Scholar 

  65. Shahzad, F. et al. Multi-layer stacking ensemble model for forest fire prediction. Earth Sci. Inf. 18, 270 (2025).

    Google Scholar 

  66. Lundberg, S. M. et al. From local explanations to global understanding with explainable AI for trees. Nat. Mach. Intell. 2, 56–67 (2020).

    Google Scholar 

  67. Suryabhagavan, K. V., Alemu, M. & Balakrishnan, M. GIS-based multi-criteria decision analysis for forest fire susceptibility mapping in Ethiopia. Trop. Ecol. 57, 33–43 (2016).

    Google Scholar 

  68. Tan, C. & Feng, Z. Mapping forest fire risk zones using machine learning in Hunan Province, China. Sustainability 15, 6292 (2023).

    Google Scholar 

  69. Pragya et al. Integrated spatial analysis of forest fire susceptibility using GIS-based fuzzy AHP in the western Himalayas. Remote Sens. 15, 4701 (2023).

    Google Scholar 

  70. Ribeiro, M. T., Singh, S. & &Guestrin, C. Explaining the predictions of any classifier. in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 1135–1144 (2016).

  71. Cilli, R. et al. Explainable artificial intelligence detects wildfire occurrence in Mediterranean countries of southern Europe. Sci. Rep. 12, 17560 (2022).

    Google Scholar 

  72. Sun, Y., Zhang, F., Lin, H. & Xu, S. Forest fire susceptibility modeling using the LightGBM algorithm. Remote Sens. 14, 4362 (2022).

    Google Scholar 

  73. Ambrose, G. P. Monte Carlo simulation in the evaluation of susceptibility breakpoints: Predicting the future. Pharmacotherapy 26, 129–134 (2006).

    Google Scholar 

  74. Dahri, N. & Abida, H. Monte Carlo simulation-aided AHP for flood susceptibility mapping in Gabes Basin, Tunisia. Environ. Earth Sci. 76, 302 (2017).

    Google Scholar 

  75. Kondylatos, S., Camps-Valls, G. & Papoutsis, I. Uncertainty-aware deep learning for wildfire danger forecasting. arXiv 2509–2517 (2025).

  76. Ott, C. W. et al. Predicting fire propagation across heterogeneous landscapes using WyoFire: A Monte Carlo-driven wildfire model. Fire 3, 71 (2020).

    Google Scholar 

  77. Rihan, M. et al. Improving landslide susceptibility prediction in Uttarakhand through hyper-tuned artificial intelligence and global sensitivity analysis. Earth Syst. Environ. 9, 3405–3424 (2024).

    Google Scholar 

  78. Hou, X. & Orth, R. Observational evidence of wildfire-promoting soil moisture anomalies. Sci. Rep. 10, 1–8 (2020).

    Google Scholar 

  79. Yu, G. et al. Performance of fire danger indices and their utility in predicting future wildfire danger over the conterminous United States. Earth’s Future. 11, e2023EF003823 (2023).

    Google Scholar 

  80. Trucchia, A. et al. On the merits of sparse surrogates for global sensitivity analysis of multi-scale nonlinear problems: Application to turbulence and fire-spotting model in wildland fire simulators. Commun. Nonlinear Sci. Numer. Simul. 73, 120–145 (2019).

    Google Scholar 

Download references

Acknowledgements

We highly acknowledge the European Space Agency (ESA) and the National Aeronautics and Space Administration (NASA) for the acquisition of satellite datasets that we employed for study. Authors are thankful to anonymous reviewers for their critical evaluation that helps to improve the quality of manuscript.

Author information

Authors and Affiliations

  1. Department of Geography, Faculty of Sciences, Jamia Millia Islamia, New Delhi, 110025, India

    Suheb, Md Nawazuzzoha, Md Shahid Ali, Md. Mamoon Rashid & Hasan Raja Naqvi

  2. Division Communication, School of Business, Galgotias University, Greater Noida, Uttar Pradesh, 203201, India

    Darakhsha Fatma Naqvi

  3. ACRI-ST, Route Du Pin Montard, Biot, France

    Honey Qaisar

  4. INCDS Marin Drăcea, 128 Eroilor Bvd., 077030, Voluntari, Romania

    Pierre Sicard

  5. Department of Applied Geology, College of Applied Natural Science, Adama Science and Technology University, P.O. Box 1888, Adama, Ethiopia

    Shankar Karuppannan

  6. Department of Research Analytics, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences (SIMATS), Saveetha University, Chennai, 600077, Tamil Nadu, India

    Shankar Karuppannan

Authors

  1. Suheb
    View author publications

    Search author on:PubMed Google Scholar

  2. Md Nawazuzzoha
    View author publications

    Search author on:PubMed Google Scholar

  3. Md Shahid Ali
    View author publications

    Search author on:PubMed Google Scholar

  4. Md. Mamoon Rashid
    View author publications

    Search author on:PubMed Google Scholar

  5. Darakhsha Fatma Naqvi
    View author publications

    Search author on:PubMed Google Scholar

  6. Honey Qaisar
    View author publications

    Search author on:PubMed Google Scholar

  7. Pierre Sicard
    View author publications

    Search author on:PubMed Google Scholar

  8. Shankar Karuppannan
    View author publications

    Search author on:PubMed Google Scholar

  9. Hasan Raja Naqvi
    View author publications

    Search author on:PubMed Google Scholar

Contributions

Suheb: research framework, data curation, software, original draft review and editing; Md. Nawazuzzoha: Data curation, formal analysis, visualization; Md Shahid Ali: Data curation, formal analysis, visualization; Md. Mamoon Rashid: Data curation, software, visualization; Darakhsha Fatma Naqvi: Data curation, software, visualization; Honey Qaiser: visualization, review and editing Shankar Karuppannan: Formal analysis, Reviewing and Editing; Pierre Sicard: Formal analysis, Reviewing and Editing; Hasan Raja Naqvi: Formal analysis, supervision, resources, editing original draft-Reviewing and Editing. All authors have read and agreed to the published version of the manuscript.

Corresponding authors

Correspondence to
Shankar Karuppannan or Hasan Raja Naqvi.

Ethics declarations

Competing interests

The authors declare no competing interests.

Consent for publication

We have carefully reviewed all the images in our manuscript and confirm that no human faces are present.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Suheb, Nawazuzzoha, M., Ali, M.S. et al. An explainable GeoAI framework for spatial assessment of wildfire susceptibility in the Upper Ravi sub-basin, Indian Himalaya.
Sci Rep (2026). https://doi.org/10.1038/s41598-026-46924-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41598-026-46924-w

Keywords

  • Wildfire
  • Susceptibility
  • Ensemble machine learning
  • SHAP explainability
  • Himalayan Region

Source: Ecology - nature.com

Digitally native species are a necessary shift in taxonomic practice

The impact of visual-auditory interactions on well-being: colour and soundscapes in urban spaces

This portal is not a newspaper as it is updated without periodicity. It cannot be considered an editorial product pursuant to law n. 62 of 7.03.2001. The author of the portal is not responsible for the content of comments to posts, the content of the linked sites. Some texts or images included in this portal are taken from the internet and, therefore, considered to be in the public domain; if their publication is violated, the copyright will be promptly communicated via e-mail. They will be immediately removed.

Back to Top
Close