When Climate Solutions Compete for Land: Food Security, Forest Restoration, and Carbon Removal in Southeast Asia

Landscapes differ in carbon stocks, emissions, biodiversity, and productive functions. Protecting, restoring, or modifying one hectare does not necessarily produce the same benefits as intervening in another.

Sasmito et al. (2025) demonstrate this distinction through Southeast Asia’s peat swamp forests and mangroves. Land-use change affecting these ecosystems generated an estimated 691.8 ± 97.2 million metric tons of carbon dioxide equivalent annually during 2001–2022, representing approximately 48 percent of regional land-use-change emissions despite occupying only 5.4 percent of regional land area.

The authors also estimate additional mitigation potential of 94.4 ± 7.4 million metric tons of carbon dioxide equivalent annually through restoration, predominantly from rewetting drained peatland plantations. Protecting these ecosystems can avoid further emissions, while restoration may reduce continuing emissions or increase carbon storage, depending on the intervention. These benefits should not be equated with reversing historical ecosystem losses.

Lim et al. (2026) identify another opportunity: approximately 42 million hectares of natural tropical moist forest within forestry and agricultural concessions across four Southeast Asian countries. Their analysis estimates that clearing these forests could threaten approximately 4.6 million hectares of Key Biodiversity Areas and release 1,216 million metric tons of carbon dioxide over 30 years.

These figures describe potential consequences under the study’s assumptions, not a forecast that all identified forests will be cleared. The estimates also concern land-use processes and areas that may overlap with those examined by Sasmito et al. (2025); they should not simply be added together.

The studies demonstrate why spatial priorities matter. Peatland conservation can prevent emissions associated with drainage and degradation, while mangrove restoration may contribute to mitigation and coastal ecosystem functions. Other landscapes may offer lower-emission agriculture or compatible tree-based production.

Van Noordwijk et al. (2014) show that relationships among tree cover, agriculture, and food security vary across Southeast Asian landscapes. Carbon potential should therefore inform land-use decisions without independently determining them.

Climate and development targets may appear compatible separately, but competition emerges when their land requirements overlap.

Land may simultaneously be identified for agriculture, bioenergy, forest protection, restoration, biodiversity conservation, or carbon projects. Compatibility depends on existing allocations, ecological conditions, economic incentives, and legitimate rights.

Lim et al. (2026) illustrate the economic dimension. Their analysis identifies natural forests within commercial concessions where conservation may compete with returns from timber extraction or agricultural development. Estimated break-even carbon prices for avoided-deforestation projects range from approximately US$33–US$1,677 per metric ton of carbon dioxide.

These are modeled thresholds, not current market prices. Their variation indicates that conservation opportunities differ financially and that carbon revenue may not be sufficient to change every commercial land-use decision. The study does not directly estimate effects on household food prices or rural livelihoods.

Competition also depends on land classification. Areas designated degraded, marginal, or unused may support farming, grazing, customary resource use, biodiversity, or cultural activities. Genuinely degraded areas may offer restoration opportunities with lower displacement risks, but classification alone cannot establish their availability.

The Food and Agriculture Organization of the United Nations (FAO 2022) emphasizes recognizing legitimate rights to land, fisheries, and forests. Such rights matter when conservation or carbon projects affect existing resource users.

Cambodia’s Oddar Meanchey Community Forestry REDD+ Project (under the framework for reducing emissions from deforestation and forest degradation and related forest-conservation measures) illustrates the distinction between forest-use and carbon rights. Yeang (2012) reports that participating communities retained forest-use and access rights, while carbon rights belonged to the government. The government agreed to allocate at least 50 percent of net carbon-credit revenue to participating communities; the 2012 study documents the agreement but does not verify subsequent disbursement.

Earlier Philippine forestry carbon projects faced challenges involving land eligibility, transaction costs, and community participation, including Indigenous Peoples (Lasco 2008). Varkkey, Miller, and Taylor (2026) likewise examine competing interests, community rights, and benefit distribution in Southeast Asian nature-based carbon governance.

Together, these studies suggest that land competition cannot be resolved through spatial mapping or carbon valuation alone. Economic feasibility, tenure, participation, and distributional effects also determine which choices can be implemented.

Food security consequences cannot be assessed solely through cultivated area. Land-use decisions may also affect productive potential, farm income, water, resource access, and food availability.

Calura et al. (2023), studying agricultural land conversion in Pura, Tarlac, Philippines, found that soils considered marginally suitable under existing conditions could become highly suitable for crop production with appropriate management. Their research concerns urban and industrial conversion rather than climate projects. Nevertheless, it illustrates why current productivity should not be equated with long-term productive potential.

Trees can also contribute to food systems without necessarily displacing crops. Van Noordwijk et al. (2014) identify pathways through which trees provide food, generate income, and support agricultural production through ecosystem services. These relationships vary with landscape and production conditions.

Local knowledge is another consideration. Galang and Vaughter (2020) surveyed 108 landscape users in Mindanao’s Libungan-Alamada Watershed and documented knowledge of agroforestry’s provisioning and regulating services. The study did not directly measure improvements in yields, food security, or carbon storage, but supports considering local ecological knowledge when assessing existing landscape functions.

Agroforestry may combine production and ecological benefits where appropriate. However, adding trees to existing farms differs from replacing natural forests or staple-producing areas with commercial tree crops. Outcomes depend on what land use changes, what remains productive, and who benefits.

Agricultural mitigation can also occur within existing farmland. In a global meta-analysis of 5,322 field experiments from 504 studies, Thai et al. (2026) found that alternate wetting and drying in rice cultivation was associated with 12 percent lower greenhouse gas emissions and 4 percent higher yields. Dry direct seeding was associated with about 13 percent lower emissions across planting-method comparisons, with yields comparable to those from transplanting.

These are global estimates, not Southeast Asia-specific outcomes. Applicability depends on irrigation, soils, climate, and management. Alternate wetting and drying may also affect methane and nitrous oxide emissions, requiring suitable water and nutrient management.

These findings show why agricultural emissions reductions need not always require additional land. Such practices complement ecosystem protection and restoration. However, higher productivity does not automatically prevent agricultural expansion into natural ecosystems; markets, commodity prices, infrastructure, and policies continue to influence land conversion.

Food security assessments should therefore examine agricultural production, income, resource access, food affordability, and vulnerable households alongside cropland area.

Restoration, forest protection, and carbon dioxide removal contribute to climate action through different mechanisms. Their ecological and livelihood outcomes should be assessed separately.

Protecting forests and peatlands can avoid emissions from conversion or degradation. Vegetation regrowth can remove atmospheric carbon dioxide, while rewetting drained peatlands primarily reduces continuing emissions from peat oxidation. Avoided emissions and atmospheric removals should not be counted as equivalent.

Sasmito et al. (2025) illustrate why protection remains important even where restoration is possible. Restoration may recover certain functions but cannot be assumed to replace all carbon stocks, biodiversity, or ecological benefits lost through conversion.

Restoration also affects livelihoods. Terzano et al. (2022) propose a community-led peatland restoration framework involving five components: rewetting, fire reduction, revegetation, livelihood revitalization, and reporting and monitoring. Their “5Rs” approach connects ecological restoration with community participation and livelihood considerations.

Rewetting can reduce peat oxidation, subsidence, and fire risks but may require adjustments to agriculture and resource use. Suitability therefore depends on ecological conditions and households’ capacity to adapt.

Agroforestry provides evidence of location-specific outcomes. Using spatial causal-inference methods, Teo et al. (2025) estimated a net reduction in deforestation associated with agroforestry of approximately 250,319 hectares annually, equivalent to 58.8 ± 15.5 million metric tons of carbon dioxide equivalent per year.

However, among 38 subnational regions with statistically significant estimated effects, 22 showed reduced deforestation while 16 showed increased deforestation relative to modeled counterfactuals. The study does not establish that agroforestry expansion reduces forest loss everywhere.

Carbon dioxide removal (CDR) introduces additional considerations. The State of Carbon Dioxide Removal: 3rd Edition distinguishes atmospheric removal and storage from emissions avoidance (Edwards et al. 2026). Afforestation and reforestation for CDR can place demands on land and water already serving other uses.

Carbon accounting cannot independently establish whether a project is ecologically appropriate or socially equitable. Varkkey, Miller, and Taylor (2026) emphasize the relevance of rights, governance, and benefit distribution.

The relevant test is whether interventions deliver intended climate outcomes while sustaining appropriate ecological, productive, and social functions—not simply whether they increase tree cover or reported carbon.

No single land-use strategy can eliminate every trade-off. Reducing unnecessary competition requires matching interventions to landscapes while making unavoidable choices transparent.

The evidence suggests four priorities.

First, assess protection before relying on restoration where conversion would cause substantial carbon and biodiversity losses, particularly in peat swamp forests, mangroves, and natural forests within commercial concessions (Sasmito et al. 2025; Lim et al. 2026).

Second, identify mitigation opportunities within existing agricultural systems. Improved rice management can reduce emissions, while agroforestry may combine production and ecological functions, although its effects vary across locations (Thai et al. 2026; Teo et al. 2025).

Third, evaluate productive potential and existing resource uses before changing land allocations. Soil suitability and landscape functions may not be reflected in present productivity or land classifications (Calura et al. 2023; van Noordwijk et al. 2014).

Fourth, integrate tenure, participation, and livelihood needs into restoration and carbon-project design (Lasco 2008; Terzano et al. 2022; Yeang 2012).

Implementing these priorities requires coordination across agriculture, forestry, energy, biodiversity, climate, and rural development institutions. Spatial assessments can identify competing uses but cannot independently determine how social, economic, and ecological interests should be balanced.

Integrated planning should compare alternatives, recognize legitimate rights, examine who benefits and bears costs, and involve affected communities before decisions are finalized. The objective is not to make every hectare serve every purpose, but to identify where functions can coexist, where particular uses warrant priority, and where trade-offs require negotiation.