Search Results (4,360)

Accurate identification of eucalyptus plantation maturity, which is hierarchically defined as young forests, middle-aged forests, and mature forests, corresponding to different growth and physiological development stages of eucalyptus stands, is critical for forestry management and sustainable development. Currently, research on fine-grained semantic segmentation of different eucalyptus growth stages remains insufficient, and there is a lack of systematic evaluation of classical deep learning architectures and multimodal data for eucalyptus plantation maturity identification. This limits the application of remote sensing technology in forestry management, and constrains the understanding of growth dynamics in subtropical planted forests. Taking Gaofeng Forest Farm in Nanning, Guangxi, as the study area, this study analyzes the adaptability of four generations of deep learning segmentation architectures—U-Net (convolutional baseline), Trans-UNet (CNN-Transformer hybrid), Swin-UNet (pure Transformer), and Mamba-UNet (state-space model)—in the fine identification of eucalyptus plantation maturity based on spectral indices (SIs), C-band SAR data (S1), and multispectral data (S2). The results show the following: (1) There is no positive correlation between model complexity and recognition performance. Among all architectures, Mamba-UNet achieves the best performance, with a validation set mIoU of 79.10% and an F1-score of 88.26%. The performance ranking of the different architectures evaluated is Mamba-UNet > Swin-UNet > U-Net > Trans-UNet. (2) The S2+SI combination achieves the highest accuracy (mIoU 79.10%), outperforming S2+S1+SI (78.55%), S2+S1 (78.38%), and single S2 (77.72%), which indicates the strong correlation between spectral indices and eucalyptus physiological characteristics. The backscattering features of S1 are limited by canopy penetration in the subtropical rainforest, introducing redundancy and triggering negative fusion effects. (3) Independent verification with field survey points verifies the strong generalization ability of the proposed approach, with OA of 94.08%, mIoU of 85.62%, Precision of 93.00%, Recall of 91.37% and F1-score of 92.15%, which indicates that the methodological framework can realize the identification of eucalyptus maturity with high precision. (4) Eucalyptus accounts for 60.15% of the total area of Gaofeng Forest Farm. Within the eucalyptus stand age structure, young, middle-aged, and mature forests account for 20.67%, 13.62%, and 25.86%, respectively. The overall distribution exhibits a polarized pattern with high proportions of young and mature forests. The findings offer theoretical support and insights for dynamic monitoring of fast-growing plantations and refined management of stand development stages. Full article

The Internet of Plants (IoP) represents the convergence of Artificial Intelligence (AI), Big Data analytics, and the Internet of Things (IoT) within protected agricultural systems. This study presents an IoP platform designed to collect, process, and analyze real-time environmental data using specialized IoT sensors capable of monitoring critical variables, including carbon dioxide concentration (CO₂), pH, air temperature, and relative humidity in hydroponic production systems. The proposed framework integrates advanced machine-learning algorithms, including Random Forest Regressor and Long Short-Term Memory (LSTM) neural networks, to process large volumes of environmental data and support crop management. In addition, the platform incorporates Vapor Pressure Deficit (VPD) and Growing Degree Days (GDD) analyses to provide crop-specific recommendations and support informed decision-making. This platform establishes a benchmark for smart agriculture in Mexico’s Laja–Bajío region, facilitating informed decision-making and maximizing the sustainability of food systems. Experimental validation was conducted under both controlled and semi-controlled environments using Swiss chard (Beta vulgaris subsp. cicla L.) and lettuce (Lactuca sativa L.) cultivated in hydroponic systems. These environments represented contrasting climatic conditions, allowing evaluation of platform stability and forecasting performance under varying thermal regimes. The Random Forest Regressor model, trained using growth chamber data consisting of 19,836 valid observations, reproduced the deterministic VPD relationship with a coefficient of determination (R²) of 0.90 and a root mean square error (RMSE) of 0.08 kPa, confirming internal consistency and identifying temperature as the dominant contributing variable rather than predicting an independent outcome. The dynamic alarm system, integrated with crop phenological stages, demonstrated greater effectiveness than conventional static-threshold approaches by generating alerts according to crop developmental requirements. Furthermore, the web-based visualization platform enabled users to interpret environmental conditions through intuitive graphical representations, facilitating decision-making without requiring specialized technical expertise. The results demonstrate the feasibility of the IoP platform as a comprehensive environmental management tool for protected agricultural systems. The proposed framework provides a scalable solution for precision agriculture applications in the Laja–Bajío region of Mexico and in other regions with similar production systems. Full article

Nature reserves are among the regions with the richest forest and grass resources. They face dual pressures of protection and development. The non-timber forest products (NTFPs) industry combines ecological protection with economic development. This unique function provides a key solution to this contradiction. Drawing on the Theory of Planned Behavior (TPB), this study employs a structural equation model (SEM). We collected survey data from 361 farmers. These farmers live in communities surrounding Yunling Provincial Nature Reserve. We empirically examine how behavioral attitude, subjective norm, and perceived behavioral control influence their willingness to participate in the non-timber forest products (NTFPs) industry. The results show that subjective norm and perceived behavioral control significantly enhance participation willingness. Subjective norm emerges as the strongest predictor. In contrast, behavioral attitude has no significant effect. This suggests that external social pressure and perceived self-capability outweigh simple benefit expectations in shaping willingness. Accordingly, we recommend three measures. First, strengthen external support to translate attitudes into actual willingness. Second, leverage social networks to amplify subjective norms. Third, enhance farmers’ endogenous capacity to consolidate their participation base. These measures can foster a win–win outcome for ecological protection and community income growth. Full article

The expansion of sustainable timber harvesting on peatland forests is important for forest owners, contractors, and the forest industry. This study investigated the effects of site-specific conditions on summertime harvesting operations and modeled the influence of a light nine-ton forwarder on rut formation. Three logistic mixed-effects models predicted peat surface disturbance (ROC = 0.60, 0.63, and 0.67), while a linear mixed model predicted rut depth when rutting occurred (R² = 0.35). Rut formation was associated with the number of machine passes, cumulative load, groundwater table depth, peat layer thickness, and interactions between stand and harvesting variables. The results indicate that timber can be successfully forwarded from drained peatlands with limited bearing capacity when operations are adapted to local site conditions. The findings highlight the importance of road-network planning, load management, and operator decision-making in reducing rut formation. This modeling approach also supports operator training, thereby contributing to more sustainable timber harvesting on low-bearing-capacity sites. Full article

Dispersal mechanisms of phytopathogenic fungi and oomycetes are critical components of forest disease dynamics. While wind and water are well-studied pathways, the role of large forest herbivores as passive vectors remains significantly overlooked. This study quantifies and compares the pathogen loads carried on the hooves and hair of wild red deer (Cervus elaphus) to evaluate their epidemiological potential. Swab samples were collected from the hooves and hair of harvested deer in the Czerwony Bór Forest District, Poland. Quantitative PCR (qPCR) assays targeting the ITS1 region were deployed to detect total fungal DNA, Alternaria alternata, Fusarium avenaceum/F. tricinctum, and several Phytophthora species. A linear mixed-effects model was implemented to statistically evaluate variations in pathogen loads across anatomical sampling locations while controlling for individual animal variability. Fungal DNA was detected in 87.5% of hoof samples, showing significantly lower Ct values (13.85–18.54) compared to fur samples (17.02–29.56), which exhibited a more patchy distribution (p = 0.016). Similarly, A. alternata transfer was highly favored by hooves (p < 0.001). Conversely, F. avenaceum was more frequently detected on hair. Among oomycetes, Phytophthora pseudosyringae was detected in all sampled animals, whereas Phytophthora cactorum occurred rarely, and other tested Phytophthora species were not detected. Wild deer carry DNA of multiple fungal and oomycete pathogens and may act as potential passive carriers within forest ecosystems. Hooves constitute the primary vector for soil-borne pathogens due to sustained contact with topsoil, whereas hair facilitates the movement of specific canopy or airborne taxa. These findings suggest that wildlife movements should be considered in future forest biosecurity assessments for comprehensive forest health management and for understanding pathogen exchange between forest and agricultural ecosystems. Full article

Tomicus pilifer is an important wood-boring forest pest in China, and its host localization and intraspecific communication rely on the perception of volatile chemical cues. However, the molecular mechanisms underlying odor recognition in this species remain largely unknown. In this study, we systematically investigated the behaviorally active volatiles present in the hindgut and feces of T. pilifer and elucidated the roles of odorant-binding proteins (OBPs) in their recognition. Gas chromatography–mass spectrometry (GC–MS) identified eight volatile compounds common to both hindgut and fecal samples. Among them, five terpenoid compounds, α-pinene, 3-carene, D-limonene, camphene, and β-myrcene, elicited significant electroantennogram (EAG) responses and induced positive behavioral attraction in adults. Based on antennal transcriptome data, phylogenetic relationships with functionally characterized homologous OBPs, preliminary molecular docking analyses, and tissue-specific expression patterns, three candidate OBPs (TpilOBP5, TpilOBP16, and TpilOBP29) were selected from 51 identified TpilOBP genes and subsequently expressed as recombinant proteins. Fluorescence competitive binding assays demonstrated that all three OBPs bound to the five behaviorally active terpenoid volatiles, with TpilOBP29 exhibiting the broadest ligand-binding spectrum and the highest binding affinity. Molecular docking and interaction analyses further revealed that the binding pocket of TpilOBP29 forms a continuous hydrophobic core composed of multiple conserved hydrophobic residues, which cooperatively stabilizes ligand binding through hydrophobic interactions, π–alkyl interactions, and van der Waals forces, thereby conferring broad-spectrum and high-efficiency odorant recognition. These findings provide new insights into the molecular mechanisms underlying the recognition of key behaviorally active terpenoid volatiles in T. pilifer, identify TpilOBP29 as a key mediator of odor recognition, and provide a potential molecular target for the development of environmentally friendly semiochemical-based behavioral management strategies against bark beetle pests. Full article

High-precision estimation of forest aboveground biomass (AGB) is crucial for global carbon cycle monitoring and sustainable forest management. However, existing machine learning-based approaches often suffer from high-dimensional feature redundancy, uneven spatial distribution of training samples, and inefficient hyperparameter optimization, which collectively limit estimation accuracy and computational efficiency. To address these issues, this study proposes a synergistic feature-sample optimization framework (DFS) for high-precision forest AGB estimation. First, with the involvement of forestry experts, we constructed the Hunan and Hubei datasets covering typical subtropical forest types through multi-source remote sensing and ground plot sampling. Second, we propose the Dual-Criteria Adaptive Feature Selection (DCAFS) method, integrating ReliefF and mutual information criteria to adaptively select key features highly correlated with AGB, eliminating spectral redundancy while preserving biomass-sensitive information. Next, we introduce a Bidirectional Active Learning Sample Optimization mechanism, called BALSO, and in its forward step, plots with high uncertainty and representativeness are given priority, so samples with high AGB variability can be captured effectively; in the backward step, spatially redundant samples and feature-redundant samples are removed through density peak clustering, and by doing this, sample selection and spatial distribution are optimized at the same time, so plot balance gets improved. Finally, the framework brings in a parameter tuning structure based on Dream Optimization Algorithm, namely DOA, and through staged exploration together with local fine-tuning, DOA makes model hyperparameters and AGB data distribution characteristics align in an adaptive manner, which helps improve convergence efficiency and estimation stability. Input variables comprise Landsat 8 OLI spectral bands, GLCM texture features, vegetation indices, and Sentinel-1/2 data. On the Hunan dataset, the framework achieved an $R^2$ of 0.83 and an RMSE of 25.6 Mg·ha⁻¹; on the Hubei dataset, it achieved an $R^2$ of 0.86 and an RMSE of 26.8 Mg·ha⁻¹. The framework was further validated on an independent public dataset from Inner Mongolia. These results demonstrate that the DFS framework provides an effective and feasible approach for regional-scale forest AGB estimation and carbon monitoring. Full article

This study investigates the spatiotemporal dynamics of Land Use/Land Cover (LULC) along precipitation gradients across Türkiye by integrating high-resolution satellite-based precipitation datasets (CHIRPS, TerraClimate, and TRMM) with the European Space Agency (ESA) WorldCover (10 m) product and multi-sensor Normalized Difference Vegetation Index (NDVI) composites (Landsat, MODIS, Sentinel-2). Türkiye’s heterogeneous climate, characterized by a sharp contrast between humid coastal belts and semi-arid interiors, serves as a natural laboratory to assess ecosystem responses to moisture availability. The results reveal a systematic and non-linear transformation of LULC classes as precipitation increases. In low-rainfall zones (200–400 mm), agricultural activities and bare surfaces predominate, reflecting human-induced land management in water-constrained environments. A critical ecological threshold was identified between 400 mm and 700 mm, where grassland areas expand rapidly, becoming the dominant class. Beyond the 900 mm isohyet, forest cover exhibits a sharp increase, approaching nearly 100% dominance in regions exceeding 1200 mm, effectively displacing other LULC categories. Comparative analysis of precipitation products shows that while all datasets capture the “coastal-wet/inland-dry” pattern, TRMM tends to overestimate winter precipitation (exceeding 100 mm), whereas CHIRPS and TerraClimate provide more conservative estimates (75–80 mm). Overlay analyses between seasonal NDVI and precipitation confirm a pronounced “time-lag effect” in vegetation phenology. Despite peak precipitation occurring in winter (~75 mm), NDVI reaches its minimum (~0.03) due to thermal limitations and dormancy. Conversely, vegetation greenness peaks during the dry summer months (NDVI ~0.14 to 0.40), utilizing antecedent soil moisture stored during the spring recharge. High-resolution Sentinel-2 data proved superior in delineating micro-topographic vegetation responses compared to Landsat and MODIS. These findings provide a scientific baseline for sustainable land management and climate adaptation strategies, highlighting that precipitation thresholds are the primary determinants of Türkiye’s ecological boundaries. Full article

Excessive use of chemical fertilizers degrades soil health and threatens sustainable crop production, which can be mitigated by partially substituting chemical fertilizers with Chinese milk vetch (Astragalus sinicus L., MV, as green manure). However, the mechanisms through which MV incorporation alters soil microbial community structure and function, enhances soil quality and crop productivity, as well as its long-term effects in paddy soils, are still not fully understood. In this study, we investigated the responses of soil physical, chemical, and biological properties (to comprehensively evaluate soil quality); microbial community structure and function; rice productivity; and the sustainable yield index (SYI) to five fertilizer treatments based on a 13-year field experiment in a paddy’s soil in Henan, China. The treatments included: CK (no chemical fertilizer and no MV), F100 (100% chemical fertilizer), MVF80, MVF60 and MVF40 (80%, 60%, and 40% of the chemical fertilizer rate combined with MV, respectively). Compared with the F100 treatment, MVF60 slightly increased rice yield by 1.71% and significantly improved SYI by 5.10%. All MV treatments significantly increased soil organic carbon (SOC, by 14.4–16.3%) and microbial biomass carbon (MBC, by 16.7–20.1%). MVF60 and MVF40 significantly reduced bulk density, and increased macroaggregate content and mean weight diameter (MWD). MVF80 significantly enriched soil total phosphorus (TP), total potassium (TK), mineral nitrogen (Nmin), and urease (UE). The improvement in these soil properties resulted in a marked increase (by 11.6–20.1%) in the soil quality index (SQI) under all MV treatments. Random forest analysis identified MBC and Nmin as the most important predictors of SQI. Moreover, MV incorporation increased the relative abundance of beneficial taxa (Firmicutes, Clostridium_sensu_stricto_1, Bradyrhizobium, and Nigrospora), which were positively correlated with SQI (p < 0.05), while reducing the relative abundance of pathogenic fungal genera such as Fusarium. Furthermore, regression analysis revealed strong positive correlations between SQI and both rice yield and SYI. In summary, long-term MV incorporation with a 40% reduction in chemical fertilizer (MVF60) constitutes an effective and sustainable nutrient management approach for rice production in southern China. This practice enhances soil quality through improved physical structure, nutrient cycling, and microbial community structure and function, ultimately resulting in higher and more stable yields. Full article

Soil erosion on sloping cropland is a major threat to agricultural sustainability and ecological security in mountainous regions, yet its spatial distribution and landscape-level structural characteristics remain insufficiently quantified. Taking Zhaotong in northeastern Yunnan, China, as a typical mountainous agricultural region in the upper Yangtze River Basin, this study integrated Sentinel-2 imagery, high-resolution reference data, field survey information, the Google Earth Engine platform, a random forest classifier, the Chinese Soil Loss Equation, and landscape pattern metrics to assess soil erosion on sloping cropland. The land use classification achieved an overall accuracy of 90.70% and a Kappa coefficient of 0.88, providing a reliable basis for sloping cropland extraction. Sloping cropland covered 4218.77 km², accounting for 84.64% of total cropland area, but contributed 1.84 × 10⁷ t·yr⁻¹ of annual soil loss, equivalent to 97.51% of total cropland erosion. The mean erosion rate of sloping cropland reached 4260.50 t·km⁻²·yr⁻¹, and 95.84% of sloping cropland exceeded the soil loss tolerance threshold. County-level analysis revealed strong spatial heterogeneity, with high erosion risks concentrated in northern and eastern mountainous counties. Intensive, Severe, and Extreme erosion occupied only 26.14% of the sloping cropland area but contributed 62.21% of total soil loss. Landscape metrics further showed that Moderate erosion had the highest patch density and edge density, indicating a critical fragmentation stage in erosion development. These findings support a tiered conservation strategy in which high-intensity patches are prioritized for immediate sediment reduction, while fragmented Moderate-erosion (2500–5000 t·km⁻²·yr⁻¹) areas receive preventive management. The proposed framework provides a useful approach for identifying erosion hotspots and supporting targeted soil and water conservation in mountainous agricultural landscapes. Full article

Tropical forests sustain exceptional biodiversity and regulate global carbon and water cycles, yet degradation and incomplete recovery increasingly compromise these functions. We integrated bibliometric mapping with structured systematic synthesis to characterize research at the tropical-forest degradation–restoration interface, identify its most influential contributors and publications, and evaluate evidence on drivers, interventions, monitoring, and knowledge gaps. This study addresses three guiding scientific questions: (i) How has the field developed over time and across geographic space? (ii) Which authors, institutions, journals and publications have been most influential? (iii) What does the evidence indicate about degradation drivers, restoration strategies, monitoring approaches and knowledge gaps? Scopus and Web of Science were searched for peer-reviewed articles and reviews published from 1980 to 2025. After deduplication and PRISMA-based screening, 1075 publications were analyzed bibliometrically, and the 400 most-cited studies were coded against twenty predefined questions. Scientific output increased by 10.68% annually, with 483 publications appearing during 2020–2025. The corpus comprised 318 journals and 4470 authors. Forest Ecology and Management was the leading source (141 publications; 13.1%), while the twenty most productive journals accounted for 44.3% of the corpus. Brancalion P.H.S. was the most productive author (23 publications), followed by Chazdon R.L. and Tabarelli M. (20 each). Citation influence was concentrated: the twenty most-cited documents received 25.7% of all citations, led by Ribeiro M.C. (3339 citations), whereas Hua F. achieved the highest publication-year-normalized citation score among this group. Agricultural expansion, pasture establishment, and logging were the most frequently reported degradation pressures, but interactions among logging, fire, drought, and fragmentation were rarely quantified. Restoration evidence supported a context-dependent continuum from natural regeneration to assisted and active interventions, although planting and enrichment were more visible than direct passive–active comparisons. Carbon, biomass, plant diversity, and forest structure dominated outcome assessment, whereas fauna, ecological interactions, governance, and socioeconomic dimensions received less attention. Monitoring relied mainly on satellite imagery and field inventories, with limited evaluation of tool performance. Long-term trajectories and evidence from Africa, Southeast Asia, seasonally dry forests, and montane systems remained scarce. Overall, the field is rapidly consolidating but remains geographically and thematically uneven. Future research should quantify interacting degradation processes, evaluate multidimensional and long-term recovery, connect remote sensing with ecologically meaningful field indicators, and integrate governance and social conditions into restoration planning. The synthesis supports preventing further degradation and treating restoration as a context-specific complement, rather than a substitute, for protecting remaining native forests. Full article

Forest-based cultural heritage landscapes associated with traditional potash, tar, resin, and charcoal (PoTaRCh) production constitute valuable socio-ecological systems with considerable potential for sustainable tourism and rural development. However, little is known about how these destinations may evolve under future environmental, social, and governance uncertainties. The objective of this study was to identify the key drivers and critical uncertainties shaping the future development of PoTaRCh-based tourism systems in Europe by 2040 and to develop alternative scenarios to support strategic planning and adaptive governance. The research adopted the Intuitive Logics approach to scenario planning and the Scenario-Based Strategic Planning framework, combining a two-round Delphi study with 42 experts from 15 European countries and qualitative scenario development. Four key drivers were identified: forest ecosystem integrity, destination governance, community participation, and visitor experience and tourism demand. Their interaction resulted in five internally consistent scenarios: Forest Harmony, Guardians’ Forest, Tourist Forest, Forgotten Forest, and Regenerating Forest. The findings demonstrate that the resilience of PoTaRCh-based tourism systems depends on the dynamic interaction between ecological conditions, governance capacity, local stewardship, and tourism demand rather than on any single driver. To facilitate the practical application of the scenarios, the study also proposes a Scenario Monitoring Framework and Scenario Cockpit that translate qualitative scenario narratives into monitoring indicators for early detection of emerging development pathways. The study advances tourism futures research by providing one of the first comprehensive foresight analyses of forest-based cultural heritage tourism systems and offers practical guidance for destination managers and policy-makers seeking to strengthen the long-term sustainability and resilience of heritage landscapes. Full article

Accurate delineation of current landslide extent is important for hazard assessment, sustainable watershed management, and disaster risk reduction in tectonically active mountainous regions. This study presents a pilot machine learning framework for pixel-level landslide classification in the Laonung (Laonong) Creek Watershed, southern Taiwan, using very high-resolution digital elevation model (DEM) derivatives and SPOT-6 multispectral imagery. Thirteen geomorphometric and spectral features, including slope, curvature, and six spectral indices derived from SPOT-6 bands, were extracted from 96 landslide-containing tiles within a pilot subregion of the watershed; no landslide-free tiles were included in model training or evaluation. Landslide annotations followed a geomorphic-unit delineation protocol in which optical imagery provided the primary evidence of current activity and DEM-derived hillshade supported boundary refinement. Three classifiers were evaluated using column-quartile spatially blocked four-fold cross-validation, with each fold comprising a geographically contiguous range of columns, to reduce spatial leakage: logistic regression (LR), random forest (RF), and XGBoost. All three models substantially outperformed the no-skill baseline for the resampled evaluation dataset (average precision, AP $=0.250$), achieving mean AP values of $0.854\pm 0.040$, $0.858\pm 0.033$, and $0.846\pm 0.035$ for LR, RF, and XGBoost, respectively. The convergence of linear and nonlinear model performance suggests that the dominant discriminatory signal is largely captured by relatively simple spectral and topographic predictors within this pilot dataset, rather than reflecting a general property of landslide classification. SHapley Additive exPlanations (SHAP) analysis across all four spatial folds identified SPOT-6 Band 3 (Red) as the dominant predictor in every fold, with NDVI a robust secondary predictor, consistent with the spectral characteristics of fresh bare-soil landslide surfaces and with the optical cues used in the annotation protocol. The results are interpreted in the context of the pilot dataset’s limited spatial extent, the resampled class distribution used for model evaluation, and unquantified label uncertainty. This study provides a transferable methodological baseline for future, larger-scale landslide classification analysis in the Laonung Creek Watershed and highlights the potential contribution of spatially explicit landslide mapping to sustainability-oriented disaster management. Full article

The escalating climate crisis presents a profound challenge to global environmental equilibrium, with rapid urbanization acting as a primary catalyst for land-use transformation. This study investigates the intricate relationship between land use and land cover changes (LULC) and the intensification of the Surface Urban Heat Island (SUHI) effect within the Shillong Planning Area (SPA). By integrating remote sensing data with advanced geospatial modeling and machine learning architectures which include Random Forest (RF), Support Vector Machine (SVM), and XGBoost, the research provides a comprehensive analysis of environmental shifts from 2000 to 2024, with predictive projections extending to 2034 and 2044. The analysis reveals a significant expansion in the built environment, with the Normalized Difference Built-up Index (NDBI) rising from 0.17 to 0.26. This urban growth has come at the expense of ecological health, as evidenced by a decline in the Normalized Difference Vegetation Index (NDVI) from a peak of 0.87 down to 0.74. A strong negative correlation between vegetative density and Land Surface Temperature (LST) underscores the critical role of green infrastructure in regional climate regulation. SUHI projections using the RF model, which achieved an Area Under the Curve (AUC) of 0.868, estimate SUHI values of 6.02 °C for 2034 and 6.66 °C for 2044. Predicted LULC scenarios for 2034 and 2044 suggest continued urban expansion, likely intensifying thermal stress. The application of predictive modeling through machine learning provides a robust framework to inform climate-resilient urban planning and sustainable land management. Full article

Ecosystem water use efficiency (WUE) is a key indicator of carbon–water coupling in ecosystems under climate change. While WUE responses to meteorological droughts are well-documented, the influence of different soil layers on WUE across vegetation types in water-limited regions remains unclear. Based on ERA5-Land soil moisture and MODIS vegetation products, this study investigates the spatiotemporal variations in soil moisture indices (SSMI) across three soil layers and their synchronous, lagged, and cumulative associations with WUE and resilience in the Three-North Shelterbelt Forest Program (TNSFP) region from 2001 to 2022. Our results showed that while the shallow and middle layers exhibited general wetting trends, the deep soil layer (100–289 cm) underwent continuous depletion in 56.01% of the study area. WUE showed weak synchronous responses to soil moisture drought across all layers, with no significant threshold effect. However, the lag and cumulative associations increased with soil depth, from 3–6 months in shallow layers to 9–12 months in deep layers, suggesting that deep-layer soil moisture may reflect longer-term ecohydrological stress and delayed ecosystem responses. Ecological resilience (R_d) differed by vegetation type. Grasslands were resilient to shallow drought but vulnerable to deep moisture deficits, while forests maintained stability under deep stress. These findings emphasize the importance of aligning vegetation configuration with soil water availability in restoration efforts, providing critical insights for sustainable water resource management and forest rehabilitation. Full article