Hainan Island study links food exports to growing carbon storage

Hainan Island, a tropical island province of China long celebrated for its beaches and agricultural exports, has quietly become one of the most revealing case studies in the global struggle to reconcile food production with climate goals. A new study reconstructing more than seven decades of the island’s food-system carbon budget shows how a landscape that once stored more carbon than it released transformed, within a single generation, into a persistent source of greenhouse gases—and how, under an ambitious combination of technology and structural reform, it could swing back toward becoming a net carbon sink by 2060.

The research, led by Zichen Li, Zhaohai Bai, and colleagues from Hainan University, the Chinese Academy of Sciences, and Nanjing University, and published in the journal Carbon Research, is among the most comprehensive life cycle assessments ever attempted for a tropical island food economy. The team assembled a carbon account spanning 1952 to 2020 that reaches far beyond the farm gate. Their framework incorporates coastal and offshore ecosystems, land use and land-use change, crop cultivation, livestock, aquaculture, food processing and consumption, wastewater treatment, and both on-farm and off-farm energy use. This breadth matters because food-system emissions are routinely underestimated when analyses stop at the field boundary, ignoring the fossil energy embedded in processing, cold chains, transport, and the treatment of wastes that modern diets generate.

The headline numbers are striking. Gross food-system emissions on Hainan rose from 3.29 million tonnes of carbon dioxide equivalent in 1952 to 15.13 million tonnes in 2020, roughly a fivefold increase over the study period. The authors trace this growth principally to fertilizer-related emissions, the expansion of tropical crop cultivation, booming aquaculture, and rising energy demand along the food value chain. Nowhere is the intensification more visible than in nitrogen: fertilizer application rates climbed from 223 kilograms per hectare in 1952 to 627 kilograms per hectare by 2020, a level that places Hainan among the most fertilizer-intensive agricultural regions in the world. Synthetic nitrogen not only releases nitrous oxide—a greenhouse gas nearly 300 times more potent than carbon dioxide per molecule over a century—from soils, but its production itself is energy-hungry, relying on the Haber-Bosch process that consumes natural gas at industrial scale.

Perhaps the most consequential finding concerns timing. According to the reconstruction, Hainan’s food system was a net carbon sink in the mid-twentieth century, sequestering 2.77 million tonnes of carbon dioxide equivalent in 1952 and still retaining net sequestration of 1.31 million tonnes in 1962. But sometime between those decades and the turbulent years of agricultural expansion that followed, the balance tipped. By 1970, the system was emitting 0.92 million tonnes of carbon dioxide equivalent more than it absorbed, and it has remained a net source ever since. The researchers stress the robustness of this tipping point: a sensitivity analysis of pre-1990 land-use assumptions—which are inherently uncertain for historical tropical landscapes—changed the estimated magnitude of early sequestration but did not alter the conclusion that the island crossed into net emissions by 1970. That finding carries a broader message for tropical developing regions worldwide: the transition from carbon sink to carbon source in food systems can happen remarkably fast, well before economies industrialize fully.

Trade adds a layer that is increasingly central to food-system climate accounting. Export-oriented agriculture became a dominant force in Hainan’s emissions profile over the study period. Emissions embodied in exported food rose from 0.12 million tonnes of carbon dioxide equivalent in 1952—a negligible 3.6 percent of the total—to 5.37 million tonnes in 2020, representing 35 percent of the island’s food-system emissions. In other words, more than a third of Hainan’s agricultural carbon footprint is now effectively consumed by customers elsewhere, while the atmospheric burden accumulates on the island. The authors identify high-input tropical crops and industrialized aquaculture as the principal contributors to this trade-linked carbon burden. This raises questions of climate justice and carbon leakage that resonate far beyond Hainan: when wealthy markets demand tropical produce and seafood, whose carbon ledger should carry the cost?

Looking forward, the team constructed scenarios extending to 2060, and the contrast between them is stark. Under a business-as-usual trajectory, net food-system emissions would climb to 6.78 million tonnes of carbon dioxide equivalent by 2060. Government management measures alone—regulation, incentives, and administrative controls—would moderate this to 4.11 million tonnes, a meaningful but insufficient improvement. Technological optimization, deploying practices such as precision fertilization, improved manure management, energy efficiency in processing, and enhanced sequestration in soils and biomass, is far more powerful: under that pathway, the food system could become a net sink of 2.85 million tonnes of carbon dioxide equivalent.

But the most ambitious pathway is not purely technological. The study’s central scenario combines technological measures with a structural transition of the food economy itself—one that prioritizes regional food security, moderates the expansion of resource-intensive export crops and aquaculture, and invests heavily in ecological restoration of forests, mangroves, and seagrass beds. Under this combined scenario, Hainan’s food system could achieve absolute net sequestration of 13.55 million tonnes of carbon dioxide equivalent by 2060, a swing of 20.33 million tonnes relative to the business-as-usual pathway. Because such large claims invite skepticism, the authors subjected their projections to Monte Carlo uncertainty analysis, which produced a median sequestration estimate of 13.02 million tonnes with a 95 percent uncertainty interval of 6.47 to 19.29 million tonnes. Even at the pessimistic end of that distribution, the transition scenario outperforms every alternative pathway by a wide margin.

The restoration component deserves particular attention. Mangroves and seagrass meadows are among the most carbon-dense ecosystems on Earth, burying organic carbon in waterlogged sediments where it can persist for centuries. Tropical islands, ringed by coastlines suitable for “blue carbon” restoration, hold an underappreciated mitigation asset. Hainan’s food system, uniquely in this analysis, is coupled to those coastal ecosystems, meaning that shrimp ponds and aquaculture conversions—which historically replaced mangroves across much of tropical Asia—can be reversed or redesigned to recover sequestration capacity. The study’s framework explicitly accounts for these offshore ecosystems, treating them not as scenery but as active components of the carbon budget.

The authors are careful, however, not to oversell their most ambitious scenario. They emphasize that the 13.55-million-tonne figure represents an upper-bound mitigation pathway, not a forecast of expected development. Its feasibility depends on sustained policy support, widespread adoption of low-carbon technologies, effective land-use governance, genuine ecological restoration capacity, and accommodating trade arrangements. Just as importantly, it depends on social outcomes: moderating export-oriented agriculture could affect farm income, employment, and rural livelihoods, and any transition that compromises regional food security would be politically and ethically untenable. The scenario is conditional on navigating these trade-offs—on proving that a less export-intensive, more restoration-oriented food economy can still feed people and sustain communities.

Why does a single Chinese island matter to the rest of the world? Because Hainan is, in miniature, the dilemma facing dozens of tropical and subtropical economies where agriculture is expanding, aquaculture is industrializing, export markets are lucrative, and coastal carbon sinks are disappearing. Food systems globally account for roughly a quarter to a third of all greenhouse gas emissions, yet national climate plans frequently treat them as intractable, focusing mitigation efforts on energy and transport instead. This study demonstrates that with a full life cycle accounting framework—one that spans land, sea, farm, factory, and table—it is possible to quantify precisely where the emissions reside and what a pathway to neutrality would require. The finding that a food system flipped from sink to source within roughly two decades of intensification is a warning; the demonstration that integrated intervention could restore a margin of net sequestration, even under uncertainty, is a blueprint.

The work also speaks to the growing field of consumption-based emissions accounting. With 35 percent of Hainan’s food-system emissions embedded in exports, the study underscores how production-based national inventories can obscure the true drivers of agricultural expansion. Tropical deforestation, mangrove conversion, and fertilizer overuse in exporting regions are frequently responses to demand generated far away. Frameworks like the one developed for Hainan offer a template for assigning responsibility more transparently and for designing trade policies that internalize carbon costs rather than offloading them onto developing landscapes.

What emerges from seven decades of Hainan’s carbon history is ultimately a story about choices. The island’s food system did not have to become a net emitter by 1970, and it does not have to remain one through 2060. Whether it becomes a working model of tropical carbon-neutral agriculture—or remains a cautionary example of export-driven intensification—will depend on decisions about technology, land, trade, and livelihoods that are still very much open. For the rapidly developing tropical regions watching from the Philippines to Indonesia to West Africa, Hainan’s carbon ledger offers both the mirror and the map.