A Meta-Modeling Approach for High-Resolution Forest Fire Risk Projection in Nepal under Past, Present, and Future Climate Scenarios

I am currently conducting an independent, nationwide research project that aims to predict forest fire risk using a robust meta-model framework built from an ensemble of ten base models:

1. Random Forest (RF)
2. Boosted Regression Tree (BRT)
3. Decision Tree (DT)
4. eXtreme Gradient Boosting (XGBoost)
5. Adaptive Boosting (AdaBoost)
6. Artificial Neural Networks (ANNs)
7. Support Vector Machines (SVMs)
8. classification and regression tree (CART)
9. Adaptive Boosting (AdaBoost)
10. multivariate adaptive regression splines (MARS)

This research evaluates the spatial and temporal dynamics of forest fire risk under three distinct climate regimes: the past (2001–2012), present (2012–2025), and future (2025–2050). To achieve this, I am integrating historical datasets with future climate projections based on Shared Socioeconomic Pathways (SSPs) and Representative Concentration Pathways (RCPs). Processing large-scale OpenStreetMap (OSM) data (>400 GB) for extracting detailed anthropogenic factors such as roads and settlements has been computationally intensive. To address these computational challenges, I am actively searching for high-performance computing resources, including supercomputers and alternative efficient processing methods. To ensure accurate and timely data extraction, I have downscaled CHIRPS precipitation data to a 1 km resolution and plan to use Landsat 8/9 imagery for temperature and vegetation indices to enhance spatial accuracy.

This meta-model was initially developed and validated using detailed data from my previous research work in the Rasuwa District, which provides a solid methodological foundation. By combining ensemble learning techniques with high-resolution climate and land-use data, this research will produce nuanced insights into forest fire risk hotspots and trends across Nepal.

I hope the findings will help policymakers, forest managers, and local communities better understand evolving fire regimes and strengthen their preparedness and adaptive management strategies in the face of climate change.