Dynamic Ex-Post Baselines for Enhancing the Credibility of Improved Forest Management Carbon Offsets
Abstract
Forest carbon offsets are pivotal to the global climate strategy, yet their claimed benefits are often undermined by flawed baselines. This study assesses five Improved Forest Management projects in China by constructing ex-post counterfactual baselines from satellite observations from 2001 to 2020. Using synthetic control, statistical matching, random forest, and XGBoost methods, the analysis finds that reported static baselines systematically underestimate natural carbon stock accumulation, leading to over-crediting and weak evidence of additionality across the studied projects.
Type
Publication
Environmental Science & Technology 2026
This paper develops dynamic ex-post baselines for improved forest management carbon offset projects by combining satellite observations with causal inference and machine learning methods.
