Water use efficiency (WUE, the ratio of gross primary productivity to evapotranspiration (ET)) is a robust indicator of ecosystem response to climate change, with soil moisture serving as a crucial determinant. Soil moisture varies across soil layers due to ET and soil structure, thereby differentially impacting vegetation. However, the influence of soil moisture at various soil layers on grassland WUE remains poorly understood. Using random forest and Shapley additive explanations, we examined grassland WUE sensitivity to soil moisture at different soil layers across alpine meadow and steppe ecosystems on the Tibetan Plateau. We found that grassland WUE was more sensitive to deep soil moisture (deepSM) than surface (surSM) and middle (midSM) over 2001–2020. Structural equation modeling indicated deepSM was less influenced by temperature, vapor pressure deficit (vpd) and wind speed (ws), thus providing more stable water to vegetation and exerting greater influence on WUE under the same moisture change. Additionally, alpine steppes exhibited lower sensitive to soil moisture than alpine meadows due to greater ecological adaptations and deeper root systems. Over time, WUE sensitivity to deepSM remained stable, whereas sensitivity to surSM and midSM significantly decreased in both grassland types on the Tibetan Plateau. This decrease in WUE sensitivity to surSM and midSM was associated with increased precipitation, decreased ws, and increased vpd. These findings highlight differences in WUE sensitivity to soil moisture across soil layers and ecosystems, emphasizing the importance of deepSM in sustaining grassland water‐carbon coupling. This study provides a foundation for future validations at broader spatial scales.
Climate change is projected to induce drought, resulting in the deterioration of avian food vegetation in large floodplain wetlands. The construction of water conservancy project may potentially alleviate this deterioration; however, the combined impact of climate change and water conservancy project regulation on avian food vegetation requires further investigation. In this study, a vegetation community model was developed by integrating a 2D hydrodynamic model and Cellular Automata to simulate the vegetation dynamics in floodplain wetlands. The developed model was applied in Poyang Lake, an important floodplain wetland conservation area in China under the Ramsar Convention, and the potential impact of the proposed Poyang Lake sluice on the deterioration of avian food vegetation caused by climate change was projected. We found that the avian food vegetation would continue to degrade under the influence of climate change, due to the enhanced drought conditions in the future. The extent of hygrophilous avian food vegetation experienced a significant reduction after the sluice operation, whereas the area of submerged avian food vegetation saw a notable increase. In the short term, the operation of the Poyang Lake sluice had a detrimental impact on avian food vegetation, which is primarily attributed to elevated water levels caused by sluice operation leading to submersion of hygrophilous avian food vegetation. Over time, however, the decrease in avian food vegetation caused by climate change will be mitigated by sluice operation, because the stabilized water levels and increased flooding time provide favorable conditions for the growth of submerged avian food vegetation.
Floodplain wetlands play a crucial role in the global carbon cycle, yet the spatiotemporal variation of carbon flux in floodplain wetlands remains poorly understood due to complex hydrological processes. Here, we present an integrated framework to upscale net ecosystem CO 2 exchange (NEE) in floodplain wetlands by combining eddy covariance measurements, object‐based image analysis (OBIA), hydrodynamic models, and integrated machine learning (ML) techniques. The proposed framework was instantiated in the Poyang Lake floodplain wetland, China. Results showed that OBIA could effectively capture the heterogeneous surface of wetlands through multi‐scale segmentation, thereby providing suitable space units for NEE upscaling. Hydrological regimes obtained by the hydrodynamic model, that is, maximum flood level (MFL), mean water level (WL) and water level fluctuation (WLF), were the key drivers to the NEE variability. The integrated ML model could effectively upscale NEE with higher R2 and superior robustness than commonly used individual ML model, benefiting from an optimized integration of outputs from multiple ML model through Powell optimization. The spatiotemporal distribution of the upscaled NEE results indicated that permanently inundated areas in floodplain wetlands mostly functioned as carbon sources or weak carbon sinks, while littoral zones functioned as carbon sinks. Hydrological regime changes lead to a shift between carbon sources and sinks in floodplain wetlands. The proposed framework can provide a feasible way to analyze the spatiotemporal changes of NEE and is of great benefit to achieving carbon sequestration in floodplain wetlands.
Rapid urbanization has fostered the proliferation of urban villages (UVs), high-density informal settlements that pose unique challenges for environmental management. Despite their prevalence, the dynamics of pollutant accumulation in these transitional neighborhoods remain underexplored. This study investigated nitrogen and phosphorus accumulation in road-deposited sediment (RDS) within Shenzhen, a representative megacity in southern China, utilizing field sampling and statistical analysis to identify dominant drivers. The results indicate that UVs function as significant pollution hotspots, with RDS accumulation rates approximately 3.7 times higher than in formal built-up areas. Analysis revealed that pollution intensity is primarily driven by natural factors such as slope, whereas pollution load is controlled by anthropogenic supply factors. This creates a critical input–output imbalance where high pollutant inputs exceed the natural removal capacity. Consequently, effective mitigation of urban non-point source pollution requires shifting from traditional engineering solutions to spatially sensitive planning strategies, offering practical guidance for enhancing urban sustainability in rapidly urbanizing regions.