The United Nations' Decade on Ecosystem Restoration emphasizes large-scale tree restoration for climate change mitigation and biodiversity enhancement. While tree cover expansion is expected to increase atmospheric carbon storage, it's also known to increase local evaporation, potentially reducing water availability. Recent research suggests that enhanced precipitation through atmospheric moisture recycling might offset this effect. However, the overall impact of large-scale tree restoration on the water cycle remains poorly understood, often overlooking potential downstream and downwind consequences. This study addresses this gap by quantifying the effects of a global tree restoration scenario on water availability, considering both the direct effects of increased evaporation and the indirect effects of evaporation recycling. The study aims to provide crucial insights for developing effective and sustainable tree restoration strategies that consider hydrological implications.

Traditional paired catchment studies consistently showed that tree planting increases evaporation and decreases streamflow. This is attributed to factors such as deeper roots, higher leaf area index, lower albedo, and higher aerodynamic roughness compared to other vegetation types. These studies predicted that large-scale tree restoration would decrease water availability. However, more recent large-scale research suggests a more complex impact, acknowledging the role of atmospheric feedbacks and moisture recycling. Increased evaporation from restored trees can recycle back to the land surface via atmospheric transport, potentially increasing downwind rainfall and water availability. These effects can extend beyond the river basin or even continental level. While several studies have integrated evaporation recycling in land-cover change studies, none have quantified the effects of large-scale global tree restoration on water availability by accounting for both the direct and indirect effects.

The study calculates the impact of restoring 900 million hectares of tree cover on water availability using a four-step methodology. First, it uses six data-driven Budyko models to estimate evaporation (E) and streamflow (Q) under current conditions. Second, it calculates E and Q after restoration without considering evaporation recycling, using the same Budyko models and updated tree cover data (including the tree restoration potential dataset). Third, it leverages the UTrack dataset of global evaporation recycling to determine the increase in precipitation (P) due to the increased evaporation from step 2. Finally, it recalculates E and Q after restoration, incorporating the increased P from evaporation recycling. The study uses high-resolution datasets (0.00833° spatial resolution) for precipitation (MSWEP), potential evaporation (WorldClim), and current and potential tree cover (Hansen tree cover). The Budyko models include a vegetation parameter calibrated separately for forest and non-forest conditions. Model streamflow outputs were validated against observed streamflow data from 19 large river basins. The methodology focuses on the regional distribution of evaporated water, but it does not explicitly model the effects of land-cover change on local precipitation or atmospheric circulation patterns. Yearly aggregated data from the UTrack dataset was used to maintain consistency with the Budyko model's temporal scale.

Large-scale tree restoration would lead to an average direct increase in terrestrial evaporation of 8.2 ± 5.5 mm/year (1.2%). Including the indirect effects of evaporation recycling, 68% of the extra evaporated water would rain out over land, resulting in an average increase in terrestrial precipitation of 4.8 ± 3.1 mm/year (0.7%). Without considering evaporation recycling, global mean water availability would decrease by 8.2 ± 5.3 mm/year. However, when evaporation recycling is included, the decrease is reduced to 5.3 ± 5.6 mm/year. Despite the overall net decrease, there is significant spatial variability. Some regions experience increases in water availability (up to 6%), while others experience substantial decreases (up to 38%). Analysis of 21 large river basins shows that enhanced evaporation reduces streamflow for all basins (up to 9%), but for several tropical basins with high local evaporation recycling, the increase in precipitation almost completely offsets the loss of water through evaporation. The study identifies hotspots for forest restoration that might experience a significant reduction in water availability due to insufficient compensation from evaporation recycling. These areas often coincide with regions already facing water scarcity. The findings emphasize the importance of strategic planning for tree restoration projects.

The study's findings highlight the complex and regionally varying impacts of large-scale tree restoration on water availability. While increased precipitation through moisture recycling can partially offset the increased evaporation, the net effect is a global decrease in water availability. This contradicts simplified assumptions about the benefits of afforestation. The regional variations observed underscore the importance of considering local hydrological conditions when planning restoration efforts. Future research should focus on improving understanding of land-atmosphere interactions and their impact on precipitation patterns, considering factors like changes in atmospheric circulation and cloud formation, that are not included in the model. Also, the current study uses current climate conditions; the projected effects of climate change and warming on evaporation, precipitation, and atmospheric circulation should be integrated into future models.

This study provides a comprehensive assessment of the hydrological impacts of large-scale tree restoration, showing a complex interplay between increased evaporation and enhanced precipitation. While some regions may benefit from increased water availability, many others, including water-scarce areas, could experience further reductions. This necessitates carefully planned and regionally tailored tree-restoration strategies that prioritize water security. Future research needs to focus on better quantification of land-atmosphere interactions and the impacts of climate change on these dynamics.

The study's methodology does not fully account for the complex interactions between land cover change and atmospheric circulation, potentially underestimating or overestimating the effects of land-atmosphere interactions on precipitation. The model assumes that tree restoration will simply amplify current evaporation-recycling patterns, neglecting potential changes in these patterns due to altered atmospheric dynamics. The analysis relies on yearly data, neglecting seasonal variations in water availability and the potential for increased dry-season water availability due to improved soil water storage following tree restoration. The impact of climate change on the global tree-restoration potential and hydrological patterns is not explicitly considered.