For decades, our understanding of the terrestrial carbon cycle has suffered from a form of "canopy vision." Because dense, contiguous forests are easy to spot from space and measure on the ground, they have dominated global carbon accounting. However, a growing body of research suggests that by focusing primarily on these "skyscrapers" of the natural world, we have systematically ignored a massive "suburban" and "underground" carbon economy.
This research direction focuses on identifying and quantifying these "hidden" carbon pools—ranging from individual trees in agricultural fields to the vast root systems beneath our feet—to create a more accurate and actionable map of the planet's carbon sinks.
The Invisible Canopy of Scattered Trees
The most significant gap in traditional carbon inventories is the exclusion of trees outside of areas officially classified as "forests." In many regions, particularly in Africa and the Sahel, woody biomass is not concentrated in jungles but is scattered across croplands, grasslands, and urban areas.
Using sub-meter resolution satellite imagery and deep learning, Martin Brandt and Florian Reiner have demonstrated that nearly 29% of Africa’s tree cover exists in these non-forest landscapes A consistent assessment of trees inside and outside forests improves estimates of total tree cover in Africa. In fragmented landscapes like Rwanda, Maurice Mugabowindekwe and Jérôme Chave found that trees in agricultural mosaics can account for nearly half of a nation's total aboveground carbon Nation-wide mapping of tree-level aboveground carbon stocks in Rwanda.
The intuition here is simple: while a single tree in a farmer's field seems negligible, billions of such trees across a continent add up to a massive, unprotected carbon reservoir. By treating agricultural zones as "tree-free," traditional models significantly underestimate the sequestration potential of agroforestry and local land management Farmer Managed Natural Regeneration Promotes Expansion of Trees on Croplands in the Sahel.
The Underground Economy of Roots and Soil
If scattered trees are the "suburbs" of the carbon cycle, the root systems and soil organic matter are its "basements." Most ecological assessments prioritize visible above-ground metrics like timber yield or leaf area. However, Jaehong Hwang and Sang‐Tae Lee argue that excluding below-ground biomass leads to an incomplete understanding of the total ecosystem carbon budget Standing Biomass and Inorganic Nutrients Distribution for a Camellia japonica Stand at Mt. Cheon-gwan(Janghueng-gun, Jeonnam).
In dryland ecosystems, root systems can account for nearly one-fifth of total biomass. Compton J. Tucker and Ankit Kariryaa highlight that integrating allometric equations for belowground mass is vital for accurate carbon sink assessments Sub-continental-scale carbon stocks of individual trees in African drylands. Furthermore, Zhen Qian and Niklas Boers point out that these subterranean reservoirs may respond differently to climate stress than the canopy, meaning a forest that looks healthy from above might be losing carbon below Decadal sink-source shifts of forest aboveground carbon since 1988.
Beyond the Rainforest: The Role of Semiarid Biomes
There is a common misconception that the global carbon cycle is driven almost exclusively by tropical rainforests. However, Lei Fan and Jean‐Pierre Wigneron have shown that year-to-year fluctuations in atmospheric $CO_2$ are actually driven by semiarid regions Satellite-observed pantropical carbon dynamics.
Yu Feng and Stephen Sitch found that non-forest ecosystems like savannas and shrublands contribute more than half of all tropical carbon gains Global patterns and drivers of tropical aboveground carbon changes. These biomes are the "workhorses" of the carbon sink, yet they are often overlooked in favor of closed-canopy forests.
Accounting for the "Minor" Fluxes
To truly close the carbon budget, researchers are now looking at components that were previously dismissed as "noise":
- Standing Litter: Baoxian Tao and Baohua Zhang discovered that dead plants still standing upright can release up to 70% as much $CO_2$ as litter on the soil surface, a factor largely missing from wetland models CO2 emissions from reed litter in the air and on the soil surface in the Yellow River Delta, China.
- Lateral Transfers: Zhu Deng emphasizes that carbon moved through rivers or international trade of wood products creates significant discrepancies between "bottom-up" inventories and "top-down" atmospheric observations Comparing national greenhouse gas budgets reported in UNFCCC inventories against atmospheric inversions.
- Temporal Gaps: Daozhi Gong notes that restricting measurements to the active growing season ignores respiration during winter fallow periods, distorting the net annual impact of agricultural practices Warmer and Wetter Soil Stimulates Assimilation More than Respiration in Rainfed Agricultural Ecosystem on the China Loess Plateau: The Role of Partial Plastic Film Mulching Tillage.
Why This Matters
The shift toward "wall-to-wall" monitoring—mapping every tree and accounting for every pool—is not just about scientific completeness. It is about policy. When we omit trees in croplands or carbon in the soil, we fail to reward the local land management practices, such as Farmer Managed Natural Regeneration, that are actually working. By quantifying the "hidden" carbon, researchers are providing the data necessary to transform fragmented landscapes and semi-arid biomes into recognized pillars of global climate mitigation.