NCS carbon projects face challenges in measurement, additionality, baselines, permanence, and various uncertainties. Addressing these challenges can help improve the robustness of financial instruments (e.g. carbon crediting and trading, insurance and risk management) that carbon markets operate upon. There is considerable potential to use statistical and technological tools, such as data science, artificial intelligence, and remote sensing, in combination with process-based mechanistic models and tools grounded in scientific understanding, to resolve these challenges. Using such approaches, Hoong Chen’s research aims to better quantify the benefits, risks and uncertainties of nature-based climate solutions (NCS).
The first objective is to quantify permanence risks to forest carbon that arise from natural disturbance causes, using a combination of mechanistic models and artificial intelligence and machine learning data science models. This application of robust modelling techniques is a significant improvement over simple qualitative assessments of permanence risks (e.g. rating on scale of 1-5) often used in forest carbon projects, and will can help address scientific concerns over whether the provision of buffer credits, currently arbitrarily determined, is adequate.
As such, the second objective is to estimate the uncertainties involved in the range of deforestation baseline calculation methods, using statistical, artificial intelligence and machine learning models, in order to assess the reproducibility and suitability of these methods. This will help to improve deforestation baseline calculations, which form the basis for determining the amount of forest carbon credits a project can claim.