A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Natural Resources Management

Thesis (Ph.D.) - Asian Institute of Technology, 2023

This study examines the challenges of carbon accounting in emerging nations like Sri Lanka and proposes a novel approach to estimate carbon stocks in forests and changes as the basis for assessing the baseline carbon emissions, Forest Reference Emission Level (FREL), and revenues from emission reductions and carbon removals with or without REDD+ activities. The study utilizes combined phenology-based threshold classification (PBTC) methods in Google Earth Engine (GEE) and modelling to address the difficulty of acquiring reliable historical forest cover data, particularly at the subnational or district level. The study leverages Landsat 5 and 8 satellite imageries to assess forest cover changes and identify Enhanced Vegetation Index (EVI) threshold values for classifying open and dry monsoon forests, which are critical to accurately quantify carbon stocks. The results reveal that the open forest gained 3.62%, but the dry monsoon lost 0.3% annually. Notably, GEE overcame obstacles in collecting past data on land covers to assess forest cover and carbon stocks with high accuracy. Carbon removals and emissions during the study period were estimated to be 24,064.5 MgCO2 yr-1 and 165,306.6 MgCO2 yr-1 , respectively, due to gain and loss in open and dry monsoon forest cover. The established subnational FREL was 155,187.9 MgCO2 yr-1 for the period of the Paris Agreement (2020-2030). Based on the carbon price selected at the voluntary carbon market and the European Union emission trading system, total emission reductions and removals can be equivalent to carbon revenues ranging between USD 7.3 million and USD 87.5 million, which can be received under the Paris Agreement if REDD+ activities are commenced positively. Furthermore, qualitative comparative analysis (QCA) was conducted to determine the sufficient conditions for result-based payment. The results suggest that a combination of conditions such as pressure from the lack of forests, availability of funds for performance-based activities, and already commenced policy shifts in the presence status with the absence of transformative coalition provided sufficient conditions. Moreover, continuous monitoring and adaptation to changing conditions can facilitate successful progress toward REDD+ goals in Sri Lanka.