Kan et al. (2026) Latitudinal divergence in runoff responses to global forestation due to forest-atmosphere feedbacks

Identification

• Journal: Nature Communications
• Year: 2026
• Date: 2026-02-09
• Authors: Fei Kan, Xu Lian, Hao Xu, Shuchang Tang, Jiangpeng Cui, Peter Levy, Mingze Sun, Xichen Li, Shilong Piao
• DOI: 10.1038/s41467-026-68945-9

Research Groups

• Institute of Carbon Neutrality, Sino-French Institute for Earth System Science, College of Urban and Environmental Sciences, Peking University, Beijing, China
• State Key Laboratory of Tibetan Plateau Earth System and Resources Environment (TPESRE), Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China
• UK Centre for Ecology and Hydrology, Wallingford, Oxfordshire, UK
• Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China

Short Summary

Global potential forestation leads to a latitudinal divergence in runoff responses, increasing runoff in tropical regions but decreasing it in boreal regions, a pattern primarily driven by forest-atmosphere feedbacks rather than direct effects of forest expansion.

Objective

• To comprehensively quantify the hydrological outcomes of global potential forestation, specifically investigating the latitudinal divergence in runoff responses and disentangling the contributions of direct forest expansion and forest-atmosphere feedbacks (precipitation and potential evapotranspiration changes).

Study Configuration

• Spatial Scale: Global, with analysis at regional and grid-cell scales.
• Temporal Scale: Equilibrium simulations over 100 years, with analysis focusing on the last 60 years to represent multi-decadal mean conditions.

Methodology and Data

• Models used:
  • Institute Pierre Simon Laplace climate model (IPSL-CM) (LMDZ atmospheric general circulation model, ORCHIDEE land surface model).
  • Community Earth System Model (CESM) (Community Atmosphere Model version 5, Community Land Model version 4.0) for robustness verification.
  • Budyko framework (Choudhury–Yang equation) for runoff change decomposition.
  • Moisture budget decomposition for precipitation change analysis.
  • Potential evapotranspiration (PET) calculated using Penman-OW, Penman-Monteith (PM-RC), and an energy-balance approach (PETe).
• Data sources:
  • Current Plant Functional Type (PFT) maps based on Olson categories.
  • Global potential tree cover map from Bastin et al. (machine-learning approach).
  • Multi-model averaged sea surface temperatures (SSTs) and sea ice concentrations (SICs) from CMIP6 1pctCO2 simulations.
  • Refined forestation scenarios incorporating MODIS Burned Area dataset, Hasler et al. dataset (albedo offset), ESA CCI Global Land Cover dataset, and Global Lakes and Wetlands Database (GLWD).
  • Observed data for sensitivity analysis: Multi-Source Weighted-Ensemble Precipitation (MSWEP), Global Land Evaporation Amsterdam Model (GLEAM), European Space Agency Climate Change Initiative (ESA CCI) forest cover, Climatic Research Unit Time-Series (CRU TS) air temperature, Clouds and the Earth’s Radiant Energy System (CERES) incoming shortwave radiation.
  • Global Streamflow Indices and Metadata Archive (GSIM) for model performance evaluation.

Main Results

• Global potential forestation is estimated to increase terrestrial evapotranspiration (ET) by 5.3% and global mean precipitation by 4.2%.
• Global mean soil moisture declines by 2.7%, while global annual average runoff increases by 2.8%.
• Runoff responses exhibit a strong latitudinal divergence: increases are observed in warm tropical and temperate regions (e.g., Brazil, Sahel, Congo Basin, southeastern China), while decreases occur in cold boreal areas (e.g., Europe, Russia, eastern North America).
• This latitudinal divergence is primarily driven by forest-atmosphere feedbacks (changes in precipitation and PET), rather than direct land cover changes.
• In warm regions, substantial precipitation gains (due to intensified upward moisture transport and horizontal advection) overwhelm ET enhancement, leading to runoff increases. PET decreases slightly due to reduced vapor pressure deficit (VPD).
• In cold regions, slight precipitation compensation is outweighed by enhanced evaporative losses (increased PET due to elevated surface net radiation), leading to runoff declines. PET increases due to increased surface net radiation dominating over reduced VPD.
• Direct forest expansion (land surface effect) consistently suppresses runoff globally, with peak impacts in energy-water transitional regions (where the ratio of precipitation to potential evapotranspiration, P/PET, approaches unity). This effect partially offsets tropical runoff increases and amplifies boreal runoff decreases.
• The global-scale Budyko parameter 'n' increases from 3.6 in the control scenario to 4.1 in the forestation scenario, indicating a greater partitioning of precipitation towards ET.

Contributions

• Provides a comprehensive quantification of runoff responses to global potential forestation, accounting for the complex interplay between precipitation, potential evapotranspiration, and actual evapotranspiration.
• Highlights that forest-atmosphere feedbacks, particularly precipitation modifications, are the dominant drivers of latitudinal divergence in runoff responses, challenging the conventional understanding derived from small-scale studies.
• Demonstrates that conclusions from small-scale afforestation experiments may inadequately represent, or even misrepresent, hydrological outcomes under large-scale implementation by neglecting atmospheric feedbacks.
• Emphasizes the necessity to incorporate hydrological considerations (beyond carbon sequestration or temperature modulation) into afforestation planning, advocating for latitude-specific strategies.
• Cautions against high-latitude afforestation due to potential exacerbation of water scarcity and supports the hydrological advantages of tropical afforestation.
• Reveals that large-scale forestation can reorganize atmospheric circulation patterns in tropical regions, drawing more additional moisture from nearby oceans, leading to net precipitation gains.

Funding

• National Natural Science Foundation of China (41988101)
• Second Tibetan Plateau Scientific Expedition and Research Program (2022QZKK0101)
• Peking University-BHP Carbon and Climate Wei-Ming Phd Scholars Program (Grant No. WM202411)

Citation

@article{Kan2026Latitudinal,
  author = {Kan, Fei and Lian, Xu and Xu, Hao and Tang, Shuchang and Cui, Jiangpeng and Levy, Peter and Sun, Mingze and Li, Xichen and Piao, Shilong},
  title = {Latitudinal divergence in runoff responses to global forestation due to forest-atmosphere feedbacks},
  journal = {Nature Communications},
  year = {2026},
  doi = {10.1038/s41467-026-68945-9},
  url = {https://doi.org/10.1038/s41467-026-68945-9}
}