The effects of land-use change on river flows have usually been explained by changes within a river basin. However, land-atmosphere feedback such as moisture recycling can link local land-use change to modifications of remote precipitation, with further knock-on effects on distant river flows. Here, we look at river flow changes caused by both land-use change and water use within the basin, as well as modifications of imported and exported atmospheric moisture. We show that in some of the world's largest basins, precipitation was influenced more strongly by land-use change occurring outside than inside the basin. Moreover, river flows in several non-transboundary basins were considerably regulated by land-use changes in foreign countries. We conclude that regional patterns of land-use change and moisture recycling are important to consider in explaining runoff change, integrating land and water management, and informing water governance.

Urbanization is a global process that has taken billions of people from the rural countryside to concentrated urban centers, adding pressure to existing water resources. Many cities are specifically reliant on renewable freshwater regularly refilled by precipitation, rather than fossil groundwater or desalination. A precipitationshed can be considered the "watershed of the sky" and identifies the origin of precipitation falling in a given region. In this paper, we use this concept to determine the sources of precipitation that supply renewable water in the watersheds of the largest cities of the world. We quantify the sources of precipitation for 29 megacities and analyze their differences between dry and wet years. Our results reveal that 19 of 29 megacities depend for more than a third of their water supply on evaporation from land. We also show that for many of the megacities, the terrestrial dependence is higher in dry years. This high dependence on terrestrial evaporation for their precipitation exposes these cities to potential land-use change that could reduce the evaporation that generates precipitation. Combining indicators of water stress, moisture recycling exposure, economic capacity, vegetation-regulated evaporation, land-use change, and dry-season moisture recycling sensitivity reveals four highly vulnerable megacities (Karachi, Shanghai, Wuhan, and Chongqing). A further six megacities were found to have medium vulnerability with regard to their water supply. We conclude that understanding how upwind landscapes affect downwind municipal water resources could be a key component for understanding the complexity of urban water security.

The spatial and temporal dynamics of water resources are a continuous challenge for effective and sustainable national and international governance. The watershed is the most common spatial unit in water resources governance, which typically includes only surface and groundwater. However, recent advances in hydrology have revealed ‘atmospheric watersheds' – otherwise known as precipitationsheds. Water flowing within a precipitationshed may be modified by land-use change in one location, while the effect of this modification could be felt in a different province, country, or continent. Despite an upwind country's ability to change a downwind country's rainfall through changes in land-use or land management, the major legal and institutional implications of changes in atmospheric moisture flows have remained unexplored. Here we explore potential ways to approach what we denote as moisture recycling governance. We first identify a set of international study regions, and then develop a typology of moisture recycling relationships within these regions ranging from bilateral moisture exchange to more complex networks. This enables us to classify different types of possible governance principles and relate those to existing land and water governance frameworks and management practices. The complexity of moisture recycling means institutional fit will be difficult to generalize for all moisture recycling relationships, but our typology allows the identification of characteristics that make effective governance of these normally ignored water flows more tenable.

Reduced rainfall increases the risk of forest dieback, while in return forest loss might intensify regional droughts. The consequences of this vegetation-atmosphere feedback for the stability of the Amazon forest are still unclear. Here we show that the risk of self-amplified Amazon forest loss increases nonlinearly with dry-season intensification. We apply a novel complex-network approach, in which Amazon forest patches are linked by observation-based atmospheric water fluxes. Our results suggest that the risk of self-amplified forest loss is reduced with increasing heterogeneity in the response of forest patches to reduced rainfall. Under dry-season Amazonian rainfall reductions, comparable to Last Glacial Maximum conditions, additional forest loss due to self-amplified effects occurs in 10-13% of the Amazon basin. Although our findings do not indicate that the projected rainfall changes for the end of the twenty-first century will lead to complete Amazon dieback, they suggest that frequent extreme drought events have the potential to destabilize large parts of the Amazon forest.

This study presents an "Earth observation-based" method for estimating root zone storage capacity-a critical, yet uncertain parameter in hydrological and land surface modelling. By assuming that vegetation optimises its root zone storage capacity to bridge critical dry periods, we were able to use state-of-the-art satellite-based evaporation data computed with independent energy balance equations to derive gridded root zone storage capacity at global scale. This approach does not require soil or vegetation information, is model independent, and is in principle scale independent. In contrast to a traditional look-up table approach, our method captures the variability in root zone storage capacity within land cover types, including in rainforests where direct measurements of root depths otherwise are scarce. Implementing the estimated root zone storage capacity in the global hydrological model STEAM (Simple Terrestrial Evaporation to Atmosphere Model) improved evaporation simulation overall, and in particular during the least evaporating months in sub-humid to humid regions with moderate to high seasonality. Our results suggest that several forest types are able to create a large storage to buffer for severe droughts (with a very long return period), in contrast to, for example, savannahs and woody savannahs (medium length return period), as well as grasslands, shrublands, and croplands (very short return period). The presented method to estimate root zone storage capacity eliminates the need for poor resolution soil and rooting depth data that form a limitation for achieving progress in the global land surface modelling community.

An ecosystem service is a benefit derived by humanity that can be traced back to an ecological process. Although ecosystem services related to surface water have been thoroughly described, the relationship between atmospheric water and ecosystem services has been mostly neglected, and perhaps misunderstood. Recent advances in land-atmosphere modeling have revealed the importance of terrestrial ecosystems for moisture recycling. In this paper, we analyze the extent to which vegetation sustains the supply of atmospheric moisture and precipitation for downwind beneficiaries, globally.We simulate land-surface evaporation with a global hydrology model and track changes to moisture recycling using an atmospheric moisture budget model, and we define vegetation-regulated moisture recycling as the difference in moisture recycling between current vegetation and a hypothetical desert world. Our results show that nearly a fifth of annual average precipitation falling on land is from vegetation-regulated moisture recycling, but the global variability is large, with many places receiving nearly half their precipitation from this ecosystem service. The largest potential impacts for changes to this ecosystem service are land-use changes across temperate regions in North America and Russia. Likewise, in semi-arid regions reliant on rainfed agricultural production, land-use change that even modestly reduces evaporation and subsequent precipitation, could significantly affect human well-being. We also present a regional case study in the Mato Grosso region of Brazil, where we identify the specific moisture recycling ecosystem services associated with the vegetation in Mato Grosso.We find that Mato Grosso vegetation regulates some internal precipitation, with a diffuse region of benefit downwind, primarily to the south and east, including the La Plata River basin and the megacities of Sao Paulo and Rio de Janeiro. We synthesize our global and regional results into a generalized framework for describing moisture recycling as an ecosystem service. We conclude that future work ought to disentangle whether and how this vegetationregulated moisture recycling interacts with other ecosystem services, so that trade-offs can be assessed in a comprehensive and sustainable manner.