The Turpan-Hami (Tuha) Basin of China, a critical region on the Silk Road Economic Belt and a major national energy base, occupies a significant position in energy security and in the major industrial clusters in Xinjiang Uygur Autonomous Region, China. Understanding spatial and temporal evolution of human activities in this area is essential for harmonizing ecological protection with energy development, safeguarding the ecological security of the Silk Road Economic Belt, and promoting the sustainable development of the area. However, despite rapid socioeconomic advances, the trajectories of human activity intensity and the principal driving mechanisms over the past three decades remain inadequately understood. To address these gaps, this study constructed a land use dataset for the Tuha Basin from 1990 to 2020, utilizing Google Earth Engine (GEE) and random forest classification algorithm. We assessed the intensity of human activities and their spatial autocorrelation patterns and further identified key drivers influencing spatial and temporal variations using the Geodetector model. Our findings indicated that the intensity of human activities in the Tuha Basin has exhibited a "first decline and then recovery" trend over the past 30 a, accompanied by significant spatial clustering. In recent years, the aggregation of hot spots has diminished, while clustering of cold spots has intensified, suggesting a dispersion of human activity centers. Nevertheless, urban areas in the Hami and Turpan cities, along with their surrounding areas, continued to serve as core areas of human activities. Topographic features (slope gradient and aspect) and their interactions with economic variables emerged as dominant determinants shaping the spatial patterns and temporal dynamics of human activity intensity. This result provides critical insights into fostering sustainable regional development and ecological conservation in the Tuha Basin and offers valuable methodological and empirical references for studies on land use dynamics and human activity intensity in similar arid areas.

Increasing human disturbance and climate change have threatened ecological connectivity and structural stability, especially in semi-arid mountain areas with sparse vegetation and weak hydrological regulation. Large-scale ecological restoration, such as adding ecological sources or corridors, is difficult in such environments and often faces poor operability and high implementation costs in practice. Taking the southern slope of the Qilian Mountains in China as the study area and 2020 as the baseline, this study integrated weighted complex network theory into the "ecological source-resistance surface-corridor" framework to construct a heterogeneous ecological network (EN). Circuit theory was integrated with weighted betweenness to identify critical barrier points for locally differentiated restoration, followed by assessment of the network optimization effects. The results revealed that 494 ecological sources and 1308 ecological corridors were identified in the study area. Fifty-one barrier points with restoration potential were identified along key ecological corridors and locally restored. After optimization, the network gained 11 additional ecological corridors, and the total ecological corridor length increased by approximately 1143 km. Under simulated attacks, the decline rates of maximum connected subgraph (MCS) and network efficiency (Ne) slowed compared with pre-restoration conditions, indicating improved robustness. These findings demonstrate that targeted local restoration can enhance network connectivity and stability while minimizing disturbance to the overall landscape pattern, providing a practical pathway for ecological restoration and sustainable management in semi-arid mountain areas.

Extreme temperature events have intensified across Jordan over the past 40 a, increasing risks to agriculture, water availability, urban infrastructure, and public health. The purpose of this study is to assess the long-term spatial trends and regime shifts in extreme temperature indicators across Jordan's climate zones to explore climate adaptation strategies. This study presents a high-resolution and spatially explicit assessment of thermal extremes using daily data from 1982 to 2024 across 45 grid-based study points in Jordan. Thirteen temperature indices, including percentile-based thresholds, duration metrics, and absolute extremes, were computed using RClimDex and analyzed across four Köppen climate zones: hot desert (BWh), hot semi-arid (BSh), cold desert (BWk), and Mediterranean (Csa) climates. The analysis confirmed a statistically significant warming trend: annual mean maximum temperatures increased by 2.198°C, while annual mean minimum temperatures rose by 2.035°C. Cold extremes have sharply declined, with cold days (TX10p) decreasing by 70.0%-80.0%, and the cold spell duration indicator (CSDI) dropping from 12.6 to 4.0 d/a, particularly in the BWk zone. Heat indices intensified across all zones, with warm days (TX90p) increasing by over 300.0% in BWh, warm nights (TN90p) rising by 38.1%, and the warm spell duration indicator (WSDI) extending fourfold, indicating prolonged exposure to heatwaves. Mean value of maximum temperature (TXx) reached 45.600°C in most arid areas, while minimum temperature (TNx) exceeded 31.600°C, highlighting increased nocturnal heat stress. Change-point analysis indicated that 1998 was a pivotal year, marking a structural transition in both cold and warm temperature indices. Subsequent intensifications after 2010 in TN90p, TNx, and mean of daily maximum temperature (T_maxmean) reflected an ongoing trend toward sustained thermal extremes. In addition to time-series trends, the study employed network-based correlation analysis to explore the coherence among climate indices. Strong positive correlations were observed among TXx, TX90p, and mean of daily minimum temperature (T_minmean) (r≥0.94), as well as among TN90p, T_minmean, and TNx (r≥0.87), indicating a tightly clustered heat subsystem. Duration metrics like the WSDI showed a close alignment with percentile extremes (between WSDI and TX90p; r=0.88), suggesting integrated heatwave behavior. In contrast, cold indices (TX10p, TN90p, frost days, and CSDI) exhibited weak or negative correlations and displayed peripheral positioning in the climate network, indicating their limited role under a warming regime. Absolute extremes showed weak internal linkages, suggesting episodic rather than systemic response characteristics. This structural realignment indicated a shift from a previously balanced thermal profile to a heat-dominated climate system. Regional variations revealed that BWh and BSh were experiencing the steepest warming, while Csa was transitioning more slowly but was showing signs of reduced winter cooling and increased irrigation demands. The findings establish a robust climate baseline for Jordan and offer actionable insights for climate adaptation planning. Recommended measures include precision irrigation, the development of heat-resilient crops, improvements to urban cooling infrastructure, and early warning systems for thermal extremes. By integrating spatial climate zoning, regime shift analysis, and inter-index correlation structures, this study provides a replicable framework for monitoring climatic transformations and informing resilience strategies in arid and semi-arid areas.

In dryland ecosystems, nitrogen (N) is the primary limiting factor after water availability, constraining both plant productivity and organic matter decomposition while also regulating ecosystem function and service provision. However, the distributions of different soil N fraction stocks in drylands and the factors that influence them remain poorly understood. In this study, we collected 2076 soil samples from 173 sites across the drylands of northern China during the summers of 2021 and 2022. Using the best-performing eXtreme Gradient Boosting (XGBoost) model, we mapped the spatial distributions of the soil N fraction stocks and identified the key drivers of their variability. Our findings revealed that the stocks of total nitrogen (TN), inorganic nitrogen (IN), and microbial biomass nitrogen (MBN) in the top 30 cm soil layer were 1020.4, 92.2, and 40.8 Tg, respectively, with corresponding mean densities of 164.6, 14.9, and 6.6 g/m². Climate variables—particularly mean annual temperature and aridity—along with human impacts emerged as the dominant drivers of soil N stock distribution. Notably, increased aridity and intensified human impacts exerted mutually counteracting effects on soil N fractions: aridity-driven moisture limitation generally suppressed N accumulation, whereas anthropogenic activities (e.g., fertilization and grazing) promoted N enrichment. By identifying the key environmental and anthropogenic factors shaping the soil N distribution, this study improves the accuracy of regional and global N stock estimates. These insights provide a scientific foundation for developing more effective soil N management strategies in dryland ecosystems, contributing to sustainable land use and long-term ecosystem resilience in drylands.

Oasis soils in Tunisia are characterized by low soil organic carbon (SOC) stocks, primarily due to their coarse texture and intensive irrigation practices. In the Gataaya Oasis, soils receive 3.000 to 4.000 L/m² annually through submersion irrigation, leading to a rapid decline in SOC stocks. Despite their sandy texture, which promotes good water infiltration, these soils are enriched with clay, dissolved materials, and fertilizers in deeper horizons. This study aimed to assess SOC content in the Gataaya Oasis soils, investigate the transport of labile carbon in drainage water, and clarify the destiny of this transported carbon. Soil samples were collected systematically at three depths (0-10, 10-20, and 20-30 cm), focusing on the top 30 cm depth, which is most affected by amendments. Two sampling points (P1 and P2) were selected, i.e., P1 profile near the trunk of date palms (with manure input) and P2 profile between two adjacent date palms (without manure input). Water samples were collected from drainage systems within the oasis (W1, W2, and W3) and outside the oasis (W4). A laboratory experiment simulating manure application and irrigation was conducted to complement field observations. Physical-chemical analyses revealed a significant decrease in SOC stocks with soil depths. In P1 profile, SOC stocks declined from 17.71 t/hm² at the 0-10 cm depth to 7.80 t/hm² at the 20-30 cm depth. In P2 profile, SOC stocks were lower, decreasing from 6.73 t/hm² at the 0-10 cm depth to 3.57 t/hm² at the 20-30 cm depth. Labile carbon content in drainage water increased outside the oasis, with chemical oxygen demand (COD) values rising from 73 mg/L in W1 water sample to 290 mg/L in W4 water sample, indicating cumulative leaching effects from surrounding oases. The laboratory experiment confirmed field observations, showing a decline in soil organic matter (SOM) content from 3.27% to 2.62% after 12 irrigations, highlighting the vulnerability of SOC stocks to intensive irrigation. This study underscores the low SOC stocks in the Gataaya Oasis soils and their rapid depletion under successive irrigations. The findings provide insights into the dynamics of labile carbon transport and its contribution to regional carbon cycling, offering valuable information for sustainable soil management and ecological protection in arid ecosystems.

The diversity and discontinuity of plant communities in the oasis-desert ecotone are largely shaped by variations in groundwater depth, yet the relationships between spatial distribution patterns and ecological niches at a regional scale remain insufficiently understood. This study examined the oasis-desert ecotone in Qira County located in the Tarim Basin of China to investigate the spatial distribution of plant communities and groundwater depth as well as their relationships using an integrated approach that combined remote sensing techniques, field monitoring, and numerical modeling. The results showed that vegetation distribution exhibits marked spatial heterogeneity, with coverage ranked as follows: Tamarix ramosissima>Phragmites australis>Populus euphratica>Alhagi sparsifolia. Numerical simulations indicated that groundwater depths range from 2.00 to 65.00 m below the surface, with the system currently in equilibrium, sustaining an average annual recharge of 1.06×10⁸ m³ and an average annual discharge of 1.01×10⁸ m³. Groundwater depth strongly influences vegetation composition and structure: Phragmites australis dominates at average groundwater depth of 5.83 m, followed by Populus euphratica at average groundwater depth of 7.05 m. As groundwater depth increases, the community is initially predominated by Tamarix ramosissima (average groundwater depth of 8.35 m), then becomes a mixture of Tamarix ramosissima, Populus euphratica, and Karelinia caspia (average groundwater depth of 10.50 m), and finally transitions to Alhagi sparsifolia (average groundwater depth of 14.30 m). These findings highlight groundwater-dependent ecological thresholds that govern plant community composition and provide a scientific basis for biodiversity conservation, ecosystem stability, and vegetation restoration in the arid oasis-desert ecotone.

Desert lakes are an important link in the water cycle and an important reservoir of water resources in arid and semi-arid areas, playing an important role in maintaining the stability of the regional natural environment. However, studies on the hydrochemical evolution and transformation relationships between desert lake groups and potential water sources are limited. Taking the Qixing Lake, the only lake group within the Hobq Desert in China, as the area of interest, this study collected samples of precipitation water, Yellow River water, lake water, and groundwater at different burial depths in the Qixing Lake region from July 2023 to October 2024. The hydrochemistry of different water bodies was analyzed using a combination of Piper diagrams, Gibbs diagrams, ratio of ions, and MixSIAR mixing models to reveal the transformational relationships of lake water with precipitation, groundwater, and Yellow River water. Results showed that both groundwater and surface water in the study area are weakly-to-strongly alkaline, with HCO₃^- as the dominant anion and Na⁺, Ca²⁺, and K⁺ as the main cations. The hydrochemical type of groundwater and some lakes was dominated by HCO₃^--Na⁺, whereas that of other lakes was dominated by Cl^--Na⁺ and HCO₃^--Mg²⁺. The hydrochemistry of groundwater and Yellow River water in the Qixing Lake region was controlled mainly by a combination of evaporite saline and silicate rock mineral dissolution. The local meteoric water line (LMWL) of the study area proved that regional water bodies are strongly affected by evaporative fractionation. The MixSIAR model revealed that shallow groundwater is the main recharge source of the lake group in the Qixing Lake region, accounting for 59.0%-64.2% of the total. The findings can provide references for the identification of water sources in desert lakes and the development and utilization of water resources in desert lake regions.

Spatiotemporal forecasting of surface soil moisture (SSM) is recognized as a critical scientific issue in precision agricultural irrigation, regional drought monitoring, and early warning systems for extreme precipitation. However, long-term forecasting continues to pose formidable challenges because of the complexity observed across both the spatial and temporal scales. In this study, we used a daily SSM dataset at a 0.05°×0.05° spatial resolution over the Qilian Mountains, China and proposed a hybrid Convolutional Long Short-Term Memory (ConvLSTM)-Nudging model, which combined deep neural networks with data assimilation to increase the accuracy of long-term SSM forecasting. We trained and evaluated the SSM predictive performance of four models (Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), ConvLSTM, and ConvLSTM with Squeeze-and-Excitation (SE) attention mechanism (ConvLSTM-SE)) in both short-term and long-term scenarios. The results showed that all the models perform well under short-term predictions, but the accuracy decrease substantially in long-term predictions. Therefore, we integrated Nudging technique during the long-term prediction phase to assimilate observational information and rectify model biases. Comprehensive evaluations demonstrate that Nudging significantly improves all the models, with ConvLSTM-Nudging achieving the best performance under the 200-d forecasting scenario. Relative to those of the best-performing ConvLSTM model for long-term forecasts, when observation noise δ=0.00 and observation fraction obs=50.0%, the coefficient of determination (R²) of ConvLSTM-Nudging increases by approximately 82.1%, while its mean absolute error (MAE) and root mean squared error (RMSE) decrease by approximately 84.8% and 77.3%, respectively; the average Pearson correlation coefficient (r) improves by approximately 23.6%, and Bias is reduced by 98.1%. These results demonstrated that although pure deep learning models achieve high accuracy in the short-term predictions, they are prone to error accumulation and systematic drift in long-term autoregressive predictions. Integrating data assimilation with deep learning and continuously correcting the state through observation can effectively suppress long-term biases, thereby achieving robust long-term SSM forecasting.