Using remote sensing and artificial intelligence to forecast future peatland fire risk caused by climate change and human intervention : a case study of the Kuan Kreng Landscape in Thailand | |
| Author | Kasambara, Talengi |
| Call Number | AIT Thesis no.WM-24-21 |
| Subject(s) | Peatlands--Thailand--Nakhon Si Thammarat--Case studie Wildfires--Thailand--Nakhon Si Thammarat--Case studies Remote sensing--Thailand--Nakhon Si Thammarat |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Water Engineering and Management |
| Publisher | Asian Institute of Technology |
| Abstract | Peatland fires are caused by a combination of climate change and human activities, with impacts that range from local to global scales. While studies have shown that climate change will increase the risk of peatland fires in the future, there has been limited study on the impact of human activities such as the construction of drainage canals for agriculture, which are major contributors to these fires. The exclusion of human activities in peatland fire risk assessment can be attributed to the inability of the available fire risk models to incorporate human factors. It is important to better understand how the interaction between climate change and human factors affects peatland fire risk.Machine learning models have shown to address these challenges by their ability to incorporate human factors and to handle complex and nonlinear relationships within fire influencing factors.Therefore, the objective of the study was to forecast future peatland fire risk caused by climate change and human intervention in Kuan Kreng Landscape, Thailand. The study used different approaches to understand how these factors can increase peatland fire risk. First, the historical burn areas from 2014 to 2022 were mapped using Landsat imagery and Random Forest classification model.Then a fire risk map was developed using Random Forest model, incorporating fire inventory data along with climatic, topographic, biophysical, hydrological, and anthropogenic factors. The results revealed that areas with existing drainage canals showed the highest burn and risk to fire. A polynomial regression model was used to assess future fire risk under different scenarios, considering both business-as-usual and canal blocking interventions within three climate pathways (SSP126, SSP370, and SSP585) from 2025 to 2083. The projections showed that under BAU conditions, there will be an increased risk, while the implementation of canal blocking measures would effectively reduce fire risk, highlighting that human intervention, in form of drainage canals, has a major contribution to the risk of fires in peatland. The findings of this study can contribute to the policy decision making process by government and development partners in peatland restoration projects in the face of climate change and human intervention. |
| Year | 2024 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Water Engineering and Management (WM) |
| Chairperson(s) | Nattachet Tangdamrongsub |
| Examination Committee(s) | Shanmugam, Mohana Sundaram;Shrestha, Sangam |
| Scholarship Donor(s) | Pan Merit Belt and Road Scholarships |
| Degree | Thesis (M. Sc.) - Asian Institute of Technology, 2024 |