Warming, increase in precipitation, and irrigation enhance greening in High Mountain Asia

Understanding vegetation changes is crucial for predicting and mitigating future climate change impacts. Global greening trends have been observed, primarily attributed to CO2 fertilization, but also influenced by land management and precipitation. High Mountain Asia (HMA), encompassing the Asian mountain ranges surrounding the Tibetan Plateau, is a region of significant hydrological importance, home to vast glaciers, ice, and snow, and supporting over a billion people. HMA’s water budget is controlled by cryospheric processes, monsoon and westerlies, and anthropogenic activities like irrigation. The region's topography, hydroclimate, and vegetation are highly heterogeneous, ranging from sea level to Mount Everest and including diverse land cover types. Global warming is significantly impacting HMA, with increasing temperatures and precipitation changes. While CO2 fertilization contributes to global greening, HMA's increase in vegetation greenness is predominantly moisture-induced, driven by climate and land-use changes. This study aims to holistically analyze the interplay of climatic and anthropogenic factors influencing HMA’s greening, using a large dataset of remote sensing products to study water and energy cycle changes from 2003 to 2020. The study focuses on the relationship between Leaf Area Index (LAI) changes and key land surface processes (snow dynamics, soil moisture, and terrestrial water storage), and atmospheric processes (precipitation and air temperature), to identify the principal drivers of greening.

Previous research has highlighted global greening trends, largely attributed to CO2 fertilization and nitrogen deposition. However, studies have also shown that local land management practices and precipitation patterns can significantly impact vegetation growth. In HMA, the significant increase in vegetation greenness has been linked to moisture availability, influenced by both climatic changes and human activities. Existing research indicates that the impacts of greening on hydrological connectivity, fluxes, and atmospheric dynamics are substantial, highlighting the need for a comprehensive analysis of the driving factors in HMA’s complex environment.

The study employs a multivariate analysis of remote sensing data from 2003 to 2020 to quantify vegetation changes and their links to soil moisture and snow cover. The analysis incorporates several datasets: * **MODIS LAI:** Provides leaf area index (LAI) data, a key indicator of vegetation greenness. * **Gridded surface meteorology datasets:** ERA5 (precipitation and air temperature), IMERG (precipitation), CHIRPS, APHRODITE, HAR, and PRINCETON (precipitation) provide information on atmospheric conditions. * **MODIS snow cover fraction:** Offers monthly estimates of snow cover, crucial for high-altitude regions. * **ESA CCI soil moisture:** Provides daily soil moisture data. * **GRACE TWS:** Quantifies changes in terrestrial water storage (TWS), encompassing all water stored above and below ground. The data were analyzed at a yearly temporal resolution at both basin and GRACE mascon resolution (0.5°). Yearly trends were computed using the Mann-Kendall test to identify monotonic trends. Partial Information Decomposition (PID) was used to quantify the interactions and dependencies between variables (LAI, precipitation, temperature, soil moisture, snow cover, and TWS), determining unique, redundant, and synergistic information contributions. The analysis differentiated between greening driven by precipitation, warming (decreased snow cover), and irrigation based on the relative information contributions of these factors to changes in LAI.

The study revealed a highly heterogeneous increase in vegetation greenness across HMA, with the highest rates observed in low and mid-elevation areas (<4000 m). LAI changes varied across land cover types, with evergreen and mixed forests showing an increase of 0.011 m²/m²/year, croplands 0.01 m²/m²/year, and grasslands 0.0036 m²/m²/year. Three principal drivers of greening were identified: * **Irrigation-induced greening:** Intense irrigation in the Ganges-Brahmaputra and Indus basins resulted in the highest increases in LAI (up to 0.04 m²/m²/year in the Ganges-Brahmaputra and 0.03 m²/m²/year in the Indus), despite decreasing TWS due to groundwater depletion. PID analysis showed that increases in soil moisture from irrigation were the primary driver of LAI increases in these regions, independent of precipitation and temperature. * **Warming-induced greening:** Increased air temperatures in HMA led to a decrease in snow cover fraction and a corresponding increase in soil moisture, resulting in enhanced vegetation growth. This effect was prominent in eight of the eleven HMA hydrologic basins. In the Tibetan Plateau, despite an overall decrease in snow cover, increases in precipitation counteracted some of the warming effects. Greening was linked to both increased soil moisture and longer growing seasons. * **Precipitation-driven greening:** Increased precipitation drove greening in mid-to-low elevation evergreen and mixed forests in southeastern HMA (parts of the Indus, Irrawaddy, Si, and Song Hong basins). PID analysis indicated that soil moisture changes were primarily driven by precipitation variations in these areas. While increases in precipitation generally resulted in increased TWS, some regions showed decreases likely due to groundwater abstraction and drought.

The findings address the research question by identifying the dominant drivers of vegetation greening in HMA. The study demonstrates the complex interplay between climatic and anthropogenic factors influencing vegetation dynamics. The results highlight the significant contribution of irrigation to greening in densely populated and heavily irrigated basins, underscoring the need to accurately account for human water management practices in Earth system models. The observation that warming-induced greening areas experience increased precipitation further emphasizes the feedback mechanisms between climate change and vegetation. While afforestation programs may contribute to greening, particularly in the Yangtze basin, the study indicates that climate change is the primary driver in this region, as well. The long timescale of afforestation's impact on precipitation contrasts with the quicker effect of precipitation changes on forest growth. This study reinforces the importance of considering the interactions between hydrosphere, cryosphere, and biosphere in assessing climate change impacts and predicting future water and energy cycles.

This study provides a comprehensive analysis of the drivers of vegetation greening in HMA, highlighting the significant role of irrigation, decreased snow cover, and increased precipitation. The findings emphasize the need to incorporate the effects of human activities, such as irrigation, in climate models to accurately predict future changes in water, energy, and biogeochemical cycles. Further research could explore the long-term impacts of greening on regional climate dynamics and investigate the complex feedback mechanisms between vegetation, precipitation, and temperature in greater detail.

The spatial resolution of the GRACE data (300-500 km) limits the precise identification of localized changes in terrestrial water storage. The analysis focuses on yearly trends, potentially overlooking shorter-term variations and complex interactions between variables. The study relies heavily on remote sensing data; ground-based validation data would strengthen the conclusions.