Introduction

Forests-based measures such as afforestation/reforestation (A/R) and reducing deforestation (RDF) are considered promising options to mitigate climate change, yet their mitigation potentials are limited by economic and biophysical factors that are largely uncertain. The range of mitigation potential estimates from integrated assessment models raises concerns about the capacity of land systems to provide realistic, cost-effective and permanent land-based mitigation.

We use the Global Change Analysis Model (GCAM) to quantify the economic mitigation potential of forests-based measures by simulating a climate policy including a tax on greenhouse gas emissions from agriculture, forestry, and other land uses. In addition, we assess how constraining unused arable land (UAL) availability, forestland expansion rates, and global bioenergy demand may influence the forests-based mitigation potential by simulating scenarios with alternative combinations of constraints. 

GCAM is used to simulate a climate policy over the period 2020–2050. In this study, the reference scenario follows the Shared Socioeconomic Pathway 2 (SSP2) 'middle-of-the-road' scenario. For the mitigation scenarios, a global GHG tax on AFOLU emissions is implemented with an annual growth rate of 5% to reach $100/tCO_2eq by 2050. The policy fosters forests-based mitigation by creating both the incentive to retain existing carbon stocks (e.g. RDF) and increase terrestrial carbon stocks (e.g. A/R). A global cap on CO₂ emissions from energy and industrial processes (EIP) is added to ensure a relevant mitigation context.

We define one scenario with constraints for each of the major uses of the AFOLU sector (agriculture, forestry and bioenergy) to represent potential limits to forests-based mitigation. The scenarios are shown in the table below.

Table 1. Scenario design.

Representation of climate policy impacts

GCAM outcomes are used to quantify changes in land allocation, CO₂ emissions from land use change, agricultural non-CO₂ emissions, net trade balances, and agricultural intensification. The following figure summarizes the causal responses triggered by the climate policy. Outcomes in terms of area changes across land uses compared to the reference by 2050 are used to calculate the mitigation potential of forests for each land use unit (LUT). GCAM estimates land allocation in 5-year time steps for 43 land uses per LUT. Here, land uses are aggregated into the following categories: forestland, cropland, unused arable land, pasture, grassland and shrubland, biomass, and others. Forestland includes managed and unmanaged (protected and unprotected) forests. 

Results on land allocation

In the reference scenario, global food demand increases by 2739 Pcal.yr⁻¹, driving an increase of 174 Mha of cropland between 2020 and 2050. Bioenergy demand leads to an increase of dedicated biomass crop area of 128 Mha by 2050. Forestland areas decrease by 11 Mha, UAL by 75 Mha, pasture by 24 Mha, and grassland and shrubland by 192 Mha between 2020 and 2050.

The climate policy affects land allocation by increasing the profitability of the land uses with relatively large carbon stocks at the expense of other uses. In the fully constrained scenario (CP_SlowFOR + LowARA + HighBIO), forestland increases globally by 86 Mha compared to reference (81 Mha and 4 Mha from net A/R and net RDF, respectively). Europe has the highest net A/R (34 Mha) while the lowest net RDF is observed in Rest_Asia (Indonesia and Southeast Asia) (−9 Mha). Globally, biomass plantations expand by 114 Mha while UAL is reduced by 11 Mha.

Results on GHG emissions potentials

Results show that the average forests-based mitigation potential in 2020–2050 increases from 738 MtCO₂.yr⁻¹ through a forestland increase of 86 Mha in the fully constrained scenario to 1394 MtCO₂.yr⁻¹ through a forestland increase of 146 Mha when all constraints are relaxed. Regional potentials in terms of A/R and RDF differ strongly between scenarios: unconstrained forest expansion rates mostly increase A/R potentials in northern regions (e.g., +120 MtCO₂.yr⁻¹ in North America); while unconstrained UAL conversion and low bioenergy demand mostly increase RDF potentials in tropical regions (e.g., +76 and +68 MtCO₂.yr⁻¹ in Southeast Asia, respectively). 

Conclusions and more details

This study shows that forests-based mitigation is limited by many factors that constrain the rates of land use change across regions. These factors, often overlooked in modelling exercises, should be carefully addressed for understanding the role of forests in global climate mitigation and defining pledges towards the Paris Agreement.

For more details, read the open access article in Environmental Research Letters. 

Rouhette, T., Escobar, N., Zhao, X., Sanz, M. J., & Ven, D.-J. van de. (2024). Limits to forests-based mitigation in integrated assessment modelling: Global potentials and impacts under constraining factors. Environmental Research Letters, 19(11), 114017. https://doi.org/10.1088/1748-9326/ad7748