Global land water underpins livelihoods, socioeconomic development, and ecosystems. It remains unclear how water availability has changed in recent decades. Using an ensemble of observations, we quantified global land water availability over the past two decades. We show that the Southern Hemisphere has dominated the declining trend in global water availability from 2001 to 2020. The significant decrease occurs mainly in South America, southwestern Africa, and northwestern Australia. In the Northern Hemisphere, the complex regional increasing and decreasing trends cancel each other, resulting in a negligible hemispheric trend. The variability and trend in water availability in the Southern Hemisphere are largely driven by precipitation associated with climate modes, particularly the El Niño-Southern Oscillation. This study highlights their dominant role in controlling global water availability.

Evapotranspiration (ET), as a key exchanging component of the land energy, water and carbon cycles, is expected to increase in response to greening land under a warming climate. However, the relative importance of major drivers (e.g. leaf area index (LAI), climate forcing, atmospheric CO₂, etc) to long-term ET change remain largely unclear. Focusing on the Eurasia which experienced the strong vegetational greening, we aim to estimate the long-term ET trend and its drivers’ relative contributions by applying a remote sensing-based water-carbon coupling model— Penman–Monteith–Leuning version 2 (PML-V2) driven by observational climate forcing and CO₂ records, and satellite-based LAI, albedo and emissivity. The PML-V2 estimated an increasing ET trend (6.20 ± 1.13 mm year⁻¹ decade⁻¹, p < 0.01) over Eurasia during 1982–2014, which is close to the ensemble mean (6.51 ± 3.10 mm year⁻¹ decade⁻¹) from other three ET products (GLEAMv3.3a, ERA5 and CRv1.0). The PML-based ET overall agrees well with water-balance derived ET in detecting the trend directions. We find that the Eurasian ET increasing trend was mostly from vegetated regions over central and eastern Europe, Indian and southeast China where ET trends were larger than 20 mm year⁻¹ decade⁻¹. Modeling sensitivity experiments indicate that the Eurasian ET trend was mainly dominated by positive contributions from climate forcing change (40%) and increased LAI (22%), but largely offset by a negative contribution of increased CO₂ (26%). Our results highlight the importance of the suppression effect of increasing CO₂-induced stomatal closure on transpiration. This effect was rarely considered in diagnostic ET products but plays a key role to ensure that the long-term ET trend should not be overestimated by only accounting for greening-induced increases in transpiration and rainfall interception.

It is well known that global ecosystem water-use efficiency (EWUE) has noticeably increased over the last several decades. However, it remains unclear how individual environmental drivers contribute to EWUE changes, particularly from CO₂ fertilization and stomatal suppression effects. Using a satellite-driven water–carbon coupling model—Penman–Monteith–Leuning version 2 (PML-V2), we quantified individual contributions from the observational drivers (atmospheric CO₂, climate forcing, leaf area index (LAI), albedo and emissivity) across the globe over 1982–2014. The PML-V2 was well-calibrated and showed a good performance for simulating EWUE (with a determination coefficient (R ²) of 0.56) compared to observational annual EWUE over 1982–2014 derived from global 95 eddy flux sites from the FLUXNET2015 dataset. Our results showed that global EWUE increasing trend (0.04 ± 0.004 gC mm⁻¹ H₂O decade⁻¹) was largely contributed by increasing CO₂ (51%) and LAI (20%), but counteracted by climate forcing (−26%). Globally, the CO₂ fertilization effect on photosynthesis (23%) was similar to the CO₂ suppression effect on stomatal conductance (28%). Spatially, the fertilization effect dominated EWUE trend over semi-arid regions while the stomatal suppression effect controlled over tropical forests. These findings improve understanding of how environmental factors affect the long-term change of EWUE, and can help policymakers for water use planning and ecosystem management.

Soil organic carbon (SOC) loss caused by wind erosion can profoundly impact carbon (C) balance in arid and semiarid regions. Nevertheless, previous researches mainly focused on the direct effect of wind erosion through removing surface soil only but ignored its indirect effects associated with soil nitrogen (N) loss and subsequent reductions of plant productivity. To better understand the wind erosion effect on SOC storage, we conducted a large‐scale field experiment by manipulating wind erosion at 371 sites in arid and semiarid regions of northwest China from 2014 to 2016. We further integrated an observation‐based empirical equation of wind erosion process into a terrestrial biogeochemical model to evaluate the direct and indirect effects of wind erosion on SOC storage in northwest China. The observed results showed that direct SOC losses increased linearly with the square of wind speed but decreased nonlinearly with soil water content. Over the 34 years (1980–2013), simulated cumulative SOC losses associated with wind erosion in northwest China were 27.47 Tg C, among which the indirect effects contributed to 2.68 Tg C (9.76%). The indirect effect of wind erosion initially enhanced SOC storage by decreasing heterotrophic respiration from 1984 to 1988 but decreased SOC pool by reducing net primary productivity due to soil N loss under the long‐term wind erosion scenario. This work, for the first time, quantified the indirect impact of wind erosion on SOC storage via feedback of suppressed plant productivity, which is crucial for the convincing assessment on SOC storage in arid and semiarid regions.

As an essential source of freshwater river flow comprises ~80% of the water consumed in China. Per capita water resources in China are only a quarter of the global average, and its economy is demanding in water resources; this creates an urgent need to quantify the factors that contribute to changes in river flow. Here, we used an offline process-based land surface model (ORCHIDEE) at high spatial resolution (0.1° × 0.1°) to simulate the contributions of climate change, rising atmospheric CO2 concentration, and land-use change to the change in natural river flow for 10 Chinese basins from 1979 to 2015. We found that climate change, especially an increase in precipitation, was responsible for more than 90% of the changes in natural river flow, while the direct effect of rising CO2 concentration and land-use change contributes at most 6.3%. Nevertheless, rising CO2 concentration and land-use change cannot be neglected in most basins as these two factors significantly change transpiration. From 2003 to 2015, the increase in water consumption offset more than 30% of the increase in natural river flow in northern China, especially in the Yellow River basin (~140%), but it had little effect on observed river flow in southern China. Although the uncertainties of rainfall data and the statistical water consumption data could propagate the uncertainties in simulated river flow, this study could be helpful for water planning and management in China under the context of global warming.