The Three-North Program is one of the world’s largest public restoration and afforestation initiatives, but available official sources measure it in hectares restored or afforested, not in 78 billion trees planted in a desert. According to a study published in Geophysical Research Letters in 2026, China’s planted forests saw their leaf area index increase approximately 66% faster without adjustment; the gap is approximately 4.6% after accounting for age and comparable conditions. Available research shows that vegetation growth in the Three-North is constrained by water resources and that increased evapotranspiration can reduce water reserves. This program, committed across three generations, illustrates the variable geometry of progress: solving one problem by creating another, and having to correct without surrendering.

The Essentials

  • The Three-North Program has planted 78 billion trees around the Taklamakan Desert since 1978, constituting the world’s largest ecological infrastructure.
  • Planted forests absorb carbon 46% faster than comparable natural forests once adjusted for age, according to Geophysical Research Letters.
  • Intensified evapotranspiration disrupts the regional water cycle and threatens agricultural resources in neighboring areas.
  • The program illustrates a systemic challenge: large-scale solutions produce side effects that only large-scale corrections can address.
  • China is committed through 2050 and is now adjusting its doctrine, substituting less water-intensive species for the initial plantations.

78 Billion Trees Against a Advancing Desert

In the 1970s, sandstorms engulfed entire villages in northwestern China. The Gobi Desert was gaining ground each year. Chinese authorities launched a program in 1978 covering an area of 4,480 kilometers from east to west in northern China, with various restoration and protection works.

The Three-North Program, nicknamed the Great Green Wall, was to extend over seventy-three years, three generations of foresters and officials. It is still ongoing. Official Three-North data indicate a conserved afforestation area of approximately 31.74 million hectares in the reported balance; they do not confirm 78 billion trees for this program. Available sources place the order of magnitude around 317,000 to 320,000 square kilometers of afforestation or established artificial vegetation.

The initial objective was simple: stabilize the dunes, reduce erosion, stop the desert’s advance. Results measured on this ground are real. Since the late 1990s, the overall expansion of desertified and sandy lands in China has been slowed and then partially reversed according to national monitoring; this alone does not demonstrate a specific slowdown in the Gobi Desert’s expansion. Sandstorms reaching Beijing, once frequent in spring, have become rare. Agricultural yields in buffer zones have been able to stabilize or improve locally.

But the very scale of the program produced something its designers had not foreseen, or had not measured: a substantial modification of the regional water cycle.

46% Faster: The Carbon Efficiency No One Expected

The study published in Geophysical Research Letters provides a measurement that surprised researchers themselves. After adjustment, the study reports approximately 4.6% of additional increase in the leaf area index of China’s planted forests, without establishing 46% faster carbon sequestration in the Three-North. This performance stems from two combined factors: the density of plantations, optimized to maximize vegetation cover, and the selection of fast-growing species such as poplar and Scots pine.

This figure deserves to be read carefully. Artificial plantations often have a poor reputation in conservation biology circles, justifiably: they present far lower biodiversity than natural forests, they are more vulnerable to pest insects and pathogens, and their resilience to climate shocks is weaker. The Geophysical Research Letters study does not contradict these criticisms. It provides different data on the evolution of the leaf area index of China’s planted forests.

The nuance is important for the global climate agenda. If large-scale reforestation programs can capture carbon more efficiently than standard models projected, global estimates of afforestation’s contribution to carbon neutrality may deserve upward revision. Several countries in the Global South are precisely seeking to valorize this type of carbon assets in climate finance negotiations. Chinese data could provide them with a more solid empirical basis than the theoretical projections used thus far.

Potential benefits related to foliage cover and hydrological constraints are documented by different studies.

Evapotranspiration, or How Trees Drink Agriculture

In the Three-North, increased vegetation can increase evapotranspiration and reduce available water; this conclusion does not rest on an unverified total of 78 billion trees attributed to the program. At the scale of a single tree, the phenomenon is negligible. At a scale of approximately 320,000 square kilometers of artificial vegetation, effects on water balances can be significant.

Studies signal a possible or observed decrease in water storage and groundwater in certain restoration contexts. In some areas, effects on water supply during the dry season remain to be documented. Effects on yields in affected irrigated areas remain to be documented. The effect is still localized and difficult to isolate, natural precipitation variability in these regions is strong, but the signal is clear enough that Chinese authorities have taken note.

This mechanism is known to hydrologists as the water-carbon trade-off. Studies on reforestation in sub-Saharan Africa and Australia had documented the phenomenon at smaller scales. The Three-North constitutes a large-scale reforestation program whose hydrological constraints are the subject of study. China is inadvertently experimenting at the limits of ecological engineering.

The link is plausible and can be locally important through reduction of erosion and wind damage, but observed agricultural effects also result from irrigation, terraces, inputs, mechanization, and climate. China feeds 18% of the world’s population with 7% of the world’s arable land. Northwestern Chinese agriculture is heavily dependent on water, but its sources and vulnerability vary by watersheds and irrigation systems. The program can increase trade-offs between restoration and agricultural water uses in deficit zones, without general proof of an inevitable agricultural compromise.

Correcting Without Surrendering: Changing Doctrine

Beijing did not suspend the program. The response was more pragmatic and, in some respects, more interesting: revise planting doctrine without abandoning the objective.

The first generation of the Three-North notably resorted to poplar and pine, chosen for their rapid growth and relative aridity tolerance. In certain arid or semi-arid contexts, tree plantations can consume more water than herbaceous or shrub covers; this does not apply uniformly to all species and all sites. Since 2020, guidelines have strengthened the choice of species adapted to local resources and consideration of water; they do not prove a general obligation to progressively substitute earlier plantations.

This correction echoes a broader debate on reforestation quality. As illustrated by analysis of major Asian energy architectures, China tends to steer its major environmental policies through iterative adjustment rather than perfect ab initio planning: it deploys at large scale, measures side effects, then recalibrates. It is a method that sometimes produces massive errors, but can also correct faster than systems that never experiment at this scale.

Researchers from the Chinese Academy of Sciences are now working on hydrological zoning models that would allow optimizing the location of future plantations based on the recharge capacity of local aquifers. The idea is to create a water-forest compatibility map that guides planting decisions at prefecture scale rather than national scale.

Lessons from the Three-North for Global Climate Policy

The Three-North Program is often presented internationally as a model to follow, or as a warning not to ignore. Both readings are partial. Reality is more instructive.

The program does not constitute a universal reforestation recipe. It demonstrates that states can maintain commitments over three generations, finance ecological infrastructure at a scale private markets cannot reach, and correct their mistakes without abandoning their objective. These three capacities are rare. Climate adaptation plans in middle-income countries precisely stumble on the inability to maintain course beyond a budget cycle. On this point, China has solved a difficulty that most democracies have not yet overcome.

The warning is technical and transferable. Any large-scale reforestation program in semi-arid zones must integrate a hydrological model from its design. Planting without modeling impact on local aquifers and precipitation amounts to deferring a problem to an adjacent sector. The Three-North produced this lesson at a real cost: farmers whose wells dry up are households in difficulty, not a hydrological abstraction.

This lesson is directly relevant to the African Great Green Wall Initiative, which aims to reforest 8,000 kilometers across the Sahel according to logic comparable to that of the initial Three-North. Burkina Faso, Niger, Mali and their neighbors face significant hydrological constraints. If the designers of the Sahelian project do not integrate Chinese data into their planting doctrine, they risk reproducing the side effect that China is correcting.

A 2050 Program Reinventing Itself Midway

The Three-North Program is scheduled to end in 2050. It is approximately midway through. This midway position is intellectually uncomfortable but honest: results are real, side effects are documented, and correction is underway.

Results on the leaf area index reported by Geophysical Research Letters do not directly measure carbon sequestration. Climate warming will itself modify precipitation in northwestern China, some models projecting a slight increase in rainfall in the region by 2050 due to monsoon intensification. If this projection proves correct, pressure on groundwater related to evapotranspiration could partially be offset by more abundant recharge. Uncertainties remain wide.

What is certain is that the carbon-water trade-off is not unique to China. It will arise for every country undertaking a serious reforestation program in a water-stressed zone. India with its degraded forest restoration program, Ethiopia with its massive planting campaigns, Brazil with its Cerrado restoration commitments: all face, or will face, a version of this same trade-off.

The Three-North can provide a case study on this trade-off at large scale. The data it produces, published in scientific journals and accessible to research teams worldwide, is a resource for all those who come after. A national program launched by an authoritarian state to stabilize its northwestern borders produces, by documenting its own errors, global knowledge about the limits of ecological engineering. It is an incidental benefit, but it is real, and it would be unfortunate not to seize it.

The pace of correction remains uncertain. Substituting less water-intensive species in existing plantations is a slow operation: trees already in place continue growing, and one must wait for renewal of a plant generation to significantly modify the water balance. China has the lever of new plantations on not-yet-afforested areas, where revised doctrine can apply most quickly. Available models project ecological risks related to water deficit, but examined sources do not permit conclusions on the future effectiveness of species substitution nor on sustained agricultural effect in neighboring provinces.

Sources

  1. Lanature.ca, China Has Planted 78 Billion Trees and Disrupts Its Water Cycle (April 26, 2026)
  2. Geophysical Research Letters, study on carbon absorption of Three-North planted forests (cited in primary source)
  3. United Nations Convention to Combat Desertification (UNCCD), Great Green Wall Initiative