Carbon markets offer a promising avenue for tackling climate change, yet their advancement encounters challenges, notably in accurately measuring emissions avoidance from forest-related activities. This paper introduces a dynamic discrete choice model tailored for assessing such emissions, using a novel database of panel data on private property land use, characteristics, and carbon project participation. Our analysis reveals that approximately 23 percent of carbon stocks within forestry carbon projects on private properties in the Brazilian Amazon lack exposure to deforestation risks and should therefore not be tradable as carbon credits. Through simulated scenarios, we demonstrate that elevated carbon prices or reduced participation costs in these projects could substantially augment the supply of avoided carbon emissions. Interventions such as cost reductions, price subsidies or regulatory improvements could bolster supply and contribute to climate change mitigation efforts. Lastly, we identify suitable properties for future project participation, aiming to mitigate investment risks and optimize expected returns.