A Review of Models for Simulating Carbon Dynamics in Forests and Agroforestry System

Forests store large amounts of carbon (C) and significantly influence global C cycles. However, forest ecosystems are highly vulnerable to anthropogenic disturbances such as deforestation and agricultural expansion. Changes in forest cover affect the C stocks stored in forest biomass and soil. Modelling approaches can help to accurately estimate C emissions and give some insights to aid decision processes related to C management strategies and interventions. In this study, a scoping review was carried out to find a suitable model that can simulate C dynamics in five major pools defined by the Intergovernmental Panel on Climate Change (IPCC). Based on the findings, there are at least seven existing C models that fulfill the IPCC criteria, including CAMFor, CBM-CFS3, CENTURY, CLM, CO2Fix, EX-ACT, and FullCAM. A comparative review is conducted for each of these models based on the model structure, required parameters, produced output, performance, and applicability, including for forestry and agroforestry assessments. Furthermore, a decision tree is also given to help potential users in choosing available models for C monitoring. EX-ACT can provide C estimation based on mitigation project scenarios, which can be useful for project planning activities. CAMFor, CBM-CFS3, and FullCAM are applicable to forest C accounting, while CO2Fix can be used for C assessment in both forests and agroforestry systems. CENTURY might perform better for soil C assessment compared with other models. On the other hand, CLM is more commonly used for regional to global C assessment than for smaller scale applications. Except for EX-ACT, all models can provide C estimation in finer detail than the five C pools used by IPCC.