Machine learning-based assessment of deforestation, successional stages, and carbon stocks in a tropical montane forest - thesis summary

Aug 5, 2026

Upload by Rowan

Tropical montane forests (TMFs) provide essential ecosystem services, yet deforestation and degradation increasingly threaten them. In the Philippines’ TMFs, the extent of deforestation, successional stages, and carbon stocks remains insufficiently studied. This thesis examines Benguet, Philippines, using Sentinel radar and optical satellite imagery and biophysical data to: delineate deforestation via ML/DL, map successional stages through InSAR, GEDI and Sentinel products, and estimate above-ground biomass and carbon. Field assessment was conducted from Dec 2023 to Jan 2024.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages133

Reading Time5 h 35 min

AccessFree to download

Abstract

Study Objectives

Deforestation mapping using ML/DL fusion

Successional stage mapping using InSAR, GEDI and Sentinel

Biomass and carbon estimation using optical/radar and ML

Field Assessment and Systematic Review

Key Results and Highlights

Deforestation classification performance

Predictors of vegetation type distribution

Biomass/carbon estimation accuracy

Conservation and Policy Implications

MACHINE LEARNING–BASED ASSESSMENT OF DEFORESTATION, SUCCESSIONAL STAGES, AND CARBON STOCKS IN A TROPICAL MONTANE FOREST USING RADAR AND OPTICAL SATELLITE IMAGERY A Thesis submitted by Richard Dein D. Altarez B.Sc. Environmental Science (magna cum laude) M. Sc. Environmental Science For the award of DOCTOR OF PHILOSOPHY ABSTRACT Tropical montane forests (TMFs) play a crucial role in providing essential ecosystem services. However, these environments are increasingly threatened by deforestation and degradation. In the Philippines’TMFs, the extent of deforestation, successional stages, and carbon stocks remain poorly explored. Hence, this study focuses on the Province of Benguet, Philippines, with three specific objectives: 1) to demarcate deforestation using the fusion of Sentinel-1,-2, and biophysical data through a traditional classifier, machine (ML) and deep learning (DL) algorithms; 2) to map the successional stages in different vegetation types through Interferometric Synthetic Aperture Radar (InSAR), Global Ecosystem Dynamics Investigation (GEDI), Sentinel products and biophysical data with ML; and, 3) to estimate the above-ground biomass (AGB) and above-ground carbon (AGC) through optical, radar, biophysical data and ML. In addition to the field assessments conducted from December 2023 to January 2024, a systematic review of spaceborne remote sensing applications in global TMFs reinforced the significance of this study. The following results are the highlights ofthis study: 1) generally, RS investigations on TMFs are concentrated in the Americas (62%), with optical sensors (85.76%) being used more frequently than SAR (12.70%); 2) the fusion of Sentinel-1-2 and biophysical data with U-Net DL algorithm effectively demarcated the deforestation (Overall Accuracy (OA) = 86.77%, Kappa Index (KI) = 78.89); 3) elevation emerged as a significant predictor of vegetation type distribution, with Random Forest’s (RF) top 10 features yielding the best predictive performance (OA = 84.22%, KI = 81.19%); and, 4) among the various algorithms utilized for AGB assessment, RF demonstrated the highest accuracy (r = 0 .982; RMSE = 53.980 Mgha -1) . Above-ground carbon density varied from 0 to 434.94 Mgha-1. This study underscores the urgency of formulating conservation and sustainable management policies. It also emphasizes the significance of Benguet’s TMF in the context of carbon sequestration initiatives like REDD+ . CERTIFICATION OF THESIS I, Richard Dein D. Altarez, declare that the PhD Thesis entitled Machine learningbased assessment of deforestation, successional stages, and carbon stocks in a tropical montane forest using radar and optical satellite imagery is not more than 100,000 words in length, including quotes and exclusive of tables, figures, appendices, bibliography, references, and footnotes. This Thesis is the work of Richard Dein D. Altarez except where otherwise acknowledged, with the majority of the contribution to the papers presented as a Thesis by Publication undertaken by the student. The work is original and has not previously been submitted for any other award except where acknowledged. Date: 09 January 2024 Endorsed by: Prof. Armando A. Apan Principal Supervisor Prof. Tek N. Maraseni Associate Supervisor Student and supervisors’ signatures of endorsement are held at the University. STATEMENT OF CONTRIBUTION The following information outlines the mutually agreed upon distribution of contributions for the candidate and co-authors in the publications provided within this thesis: Paper 1: Altarez, R.D.D., Apan, A. & Maraseni, T. (2022) . Spaceborne satellite remote sensing of tropical montane forests: a review of applications and future trends, Geocarto International, 37:26, 11900-11928, DOI: 10. 1080/10106049 .2022.2060330 Altarez, R.D.D. contributed 70% to this paper. Collectively, Apan, A. & Maraseni, T. contributed the remainder. Paper 2: Altarez, R.D.D., Apan, A. & Mara

FAQs

What are the three main research objectives of the thesis?

The thesis aims to delineate deforestation, map successional stages across vegetation types, and estimate above-ground biomass and carbon stocks using satellite remote sensing and machine learning.

Which satellite and biophysical data sources are used for deforestation assessment?

Deforestation demarcation uses a fusion of Sentinel-1 and Sentinel-2 imagery with biophysical data, supported by machine learning and deep learning algorithms.

What is the significance of this work for carbon sequestration initiatives?

The study emphasizes the urgency of conservation and sustainable management and highlights Benguet’s TMF as important for carbon sequestration efforts such as REDD+.

Aug 5, 2026

Upload by Rowan