Introduction
The growing concern about climate change has increased awareness about the essential role of forest ecosystems. About 36% of forest lands in the world have affected by forest fires. Fire affects the functioning of the ecosystem and forests and causes soil erosion and changes in the carbon cycle over the climate, vegetation and lands. One of the most valuable forest reserves in Iran is the Zagros forests, which are spread over 11 provinces of the country with an area of ​6 million hectares and make up 40% of Iran’s forests. According to the statistics of natural resources of Lorestan province in Iran, 109 fires had occurred in the forests of this province from 2008 to 2020. In these fires, 17,755 hectares of Lorestan forests burned. Considering the necessity of protecting forests and pastures and preventing the destruction of valuable natural reserves of the country, in this study, by focusing on the causes of man-made forest fires (intentional or accidental) in Zagros forests in Lorestan province, we aim to design a dynamic model of the forest fire management system to analyze the policies for managing man-made fires in Zagros forests.

Methods
This study uses a system dynamics approach. First, the dynamic model of the forest fire management system was designed in Vensim software, version DSS 6.4E with the participation of natural resource policy makers of Lorestan province and using the fire data of North Zagros forests during 2016-2020. For validation, in addition to using the Boundary Adequacy test to address the problem and match the model structure with the descriptive knowledge of the system based on the criteria of credibility, transferability, dependability and transferability, we used the structural validity test, the integration error, and the behavior reproduction test. After validation, the model was simulated over a 20-year horizon. According to the results of Monte Carlo sensitivity analysis, forest fire management policies, three strategies were identified: Prevention of fire incident in the forests and pastures of North Zagros, command and management of the fire incident in the forests and pastures of North Zagros and reconstruction of the forests and pastures of North Zagros after the fire incident. By applying the policies, each strategy was tested separately on the model and the results were compared.

Results
Based on the findings of the model simulation, the selected combined policies were identified as follows, considering the budget limit of about 40 thousand million Iranian Rials (IRR): 
1. Provision of forest protection equipment by increasing the annual budget for the protection of natural resources of Lorestan province and increasing the allocation of financial resources to protection equipment (estimated budget=800 million IRR,
 2. Matching the number of rangers with the density of the forests and pastures of North Zagros by applying an increase estimate of 1.3 fire management specialists, and a 5% increase in the annual budget for the protection of natural resources of Lorestan province (estimated budget=20,000 million IRR), 
3. Specialized education for the rescue and protection forces, the increase of 2% in the annual budget for the protection of natural resources of the province, and the increase in the allocation of financial resources for education (estimated budget=3200 million IRR, 
4. Informing people about the consequences of forest damage using local and national media (estimated budget=1000 million IRR), 
5. Supporting NGOs and training local volunteers to support forests and pastures, and providing education to 100 native volunteers during 2 years (estimated budget=1000 million IRR), 
6. Development of fire notification systems for forests and pastures, a 3% increase in the annual budget for the protection of natural resources of the province, and an increase in the design of public notification and warning systems during 2 years (estimated budget=4000 million IRR),
 7. Development of a decision support system based on the knowledge management of forest fires and pastures, and a 1.5% increase in the allocation of financial resources to crisis management research and the design of an incident command system for inter-organizational coordination during 2 years (estimated budget=5000 million IRR), 
8. Designing and developing an incident command system for inter-organizational coordination at the time of fire incident during 2 years (estimated budget=1500 million IRR), 
9. Investigating and analyzing the root causes of every fire incident in forests and pastures and doubling the allocation of financial resources to crisis management research for the analysis of the causes of fires (estimated budget=300 million IRR).

Conclusion
The proposed model is an application of the system dynamics approach to develop policies for management of man-made fires in the north Zagros forests in Lorestan province and its validity was confirmed according to the fire data of the north Zagros forests in the province. For the future studies, it is recommended that the developed model be examined according to the natural causes of forest fires, and comprehensive policies should be investigated according to the climate change process. Due to the importance of a forest fire warning system, its development is also recommended. The model is also recommended to be analyzed according to the data of other forests at risk of fire and the results should be compared. 
Considering the role of policy makers at the national level on budgeting and planning at the regional level, it is recommended that the system dynamics model with the participation of selected policy makers and decision makers should be developed in future researches based on the soft systems approach. In addition, according to the emphasis on social participation in crisis management, studies on the development of the governance system of natural resources protection and specifically the management of fire incidents in the country’s forests and pastures should be carried out, and the fields of participation for the private sector, local communities, and NGOs should be identified based on the law.

Ethical Considerations
Compliance with ethical guidelines
In this study, no experiments on animal or human samples were conducted. All publication ethics were observed.

Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Authors' contributions
Conceptualization, investigation, and data collection: Amirreza Kouranifar; Modeling, validation, analysis, and initial draft preparation: Amirreza Kouranifar and Marzieh Samadi-Foroushani; Analysis, supervision, and review: Ali Mohammad Ahmadvand and Marzieh Samadi-Foroushani.

Conflicts of interest
The authors declared no conflict of interest.

Acknowledgments
The authors would like to thank the Iranian Natural Resources and Watershed Management Organization, General Department of Natural Resources and Watershed in Lorestan Province, and Tehran Disaster Mitigation and Management Organization for their support in this research.


References
Abdel-Latif, A., Saad-Eldien, A., & Marzouk, M. (2023). System dynamics applications in crisis management: A literature review. Journal of Simulation, 17(6), 800-817. [DOI:10.1080/17477778.2022.2088306] 
Andrade, S. R., & Hulse, D. E. (2022). Evaluation and improvement of system-of-systems resilience in a simulation of wildfire emergency response. IEEE Systems Journal, 17(2), 1877 - 1888. [DOI:10.1109/JSYST.2022.3169125] 
Budiningsih, K., Nurfatriani, F., Salminah, M., Ulya, N. A., Nurlia, A., & Setiabudi, I. M., et al. (2022). Forest management units' performance in forest fire management implementation in central Kalimantan and South Sumatra. Forests, 13(6), 894. [DOI:10.3390/f13060894] 
Dhall, A., Dhasade, A., Nalwade, A., VK, M. R., & Kulkarni, V. (2020). A survey on systematic approaches in managing forest fires. Applied Geography, 121, 102266. [DOI:10.1016/j.apgeog.2020.102266] 
General Department of Natural Resources And Watershed Management of Lorestan Province. (2021). [Fire prevention plan in forests and pastures of Lorestan province (Persian)]. Lorestan:General Department of Natural Resources And Watershed Management of Lorestan Province. 
Hetu, S. N., Gupta, S., Vu, V. A., & Tan, G. (2018). A simulation framework for crisis management: Design and use. Simulation Modelling Practice and Theory, 85, 15-32. [DOI:10.1016/j.simpat.2018.03.001] 
Kalogiannidis, S., Chatzitheodoridis, F., Kalfas, D., Patitsa, C., & Papagrigoriou, A. (2023). Socio-psychological, economic and environmental effects of forest fires. Fire, 6(7), 280. [DOI:10.3390/fire6070280] 
Naderpour, M., Rizeei, H. M., Khakzad, N., & Pradhan, B. (2019). Forest fire induced Natech risk assessment: A survey of geospatial technologies. Reliability Engineering & System Safety, 191, 106558. [DOI:10.1016/j.ress.2019.106558] 
Natural Resources and Watershed Management Organization. (2022). [Statistics related to fires in pastures and forests in the country (Persian)]. Tehran: Natural Resources and Watershed Management Organization.
Oliveira, A. C., Botega, L. C., Saran, J. F., Silva, J. N., Melo, J. O., & Tavares, M. F., et al. (2019). Crowdsourcing, data and information fusion and situation awareness for emergency management of forest fires: The project DF100Fogo (FDWithoutFire). Computers, Environment and Urban Systems, 77, 101172. [DOI:10.1016/j.compenvurbsys.2017.08.006] 
Prasad, A., Pedlar, J., Peters, M., Matthews, S., Iverson, L., & McKenney, D., et al. (2024). Understanding climate change dynamics of tree species: Implications for future forests. In: S. G. McNulty (Ed.), Future forests: Mitigation and adaptation to climate change (pp. 151-175). Amsterdam: Elsevier. [DOI:10.1016/B978-0-323-90430-8.00002-2] 
Sondej, I., & Domisch, T. (2024). Impact of Large-Scale fire and habitat type on ant nest density and species abundance in Biebrza National Park, Poland. Forests, 15(1), 123. [DOI:10.3390/f15010123] 
Sterman, J. (2002). System dynamics: Systems thinking and modeling for a complex world. Massachusetts Institute of Technology Engineering Systems Division, 1-31. [Link]
Thom, D., & Seidl, R. (2016). Natural disturbance impacts on ecosystem services and biodiversity in temperate and boreal forests. Biological Reviews of the Cambridge Philosophical Society, 91(3), 760–781. [DOI:10.1111/brv.12193] [PMID] [PMCID] 
Tian, G., Fathollahi-Fard, A. M., Ren, Y., Li, Z., & Jiang, X. (2022). Multi-objective scheduling of priority-based rescue vehicles to extinguish forest fires using a multi-objective discrete gravitational search algorithm. Information Sciences, 608, 578-596. [DOI:10.1016/j.ins.2022.06.052] 
Tian, Y., Wu, Z., Li, M., Wang, B., & Zhang, X. (2022). Forest fire spread monitoring and vegetation dynamics detection based on multi-source remote sensing images. Remote Sensing, 14(18), 4431. [DOI:10.3390/rs14184431] 
Tietze, N., Gerhold, L., & Ibisch, P. L. (2023). Policy narratives in forest fire management. Environmental Research Letters, 18(11), 114023. [DOI:10.1088/1748-9326/acf6d2] 
Vigna, I., Besana, A., Comino, E., & Pezzoli, A. (2021). Application of the socio-ecological system framework to forest fire risk management: A systematic literature review. Sustainability, 13(4), 2121. [DOI:10.3390/su13042121] 
Zamarreño-Aramendia, G., Cristòfol, F. J., de-San-Eugenio-Vela, J., & Ginesta, X. (2020). Social-media analysis for disaster prevention: Forest Fire in Artenara and Valleseco, Canary Islands. Journal of Open Innovation: Technology, Market, and Complexity, 6(4), 169. [DOI:10.3390/joitmc6040169]