Search Results (7,321)

Pixel-level annotation of remote sensing imagery is costly, motivating weakly supervised semantic segmentation (WSSS) using only image-level labels. However, class activation maps (CAMs) often highlight only discriminative sub-regions and fail to separate adjacent land-cover regions, particularly in remote sensing scenes characterized by densely co-occurring land-cover classes and substantial variations in object scale. To address these limitations, we propose a three-stage framework that integrates complementary priors from Contrastive Language–Image Pre-training (CLIP), Self-Distillation with No Labels version 2 (DINOv2), and the Segment Anything Model (SAM). First, a lightweight CLIP adapter aligns vision–language priors with remote sensing imagery, while sigmoid-based multi-label decoupled distillation replaces class-competitive distillation with independent class-wise supervision, producing more complete CAMs. Second, DINOv2-guided feature clustering decomposes large merged regions before SAM prompt generation, while Spatial–Semantic Constraints are used to construct confidence-guided point-and-box prompts and reject excessively expanded or semantically inconsistent masks, thereby generating reliable pseudo-labels. Finally, a compact segmentation network is initialized with the weights learned in Stage 1 and retrained using the refined pseudo-labels generated in Stage 2, eliminating the need for foundation models during inference. Experiments on the Potsdam, LoveDA, and DeepGlobe datasets show that the proposed method achieves mean intersection over union (mIoU) scores of 53.16%, 52.66%, and 62.98%, respectively, outperforming state-of-the-art WSSS baselines by 6.55, 1.16, and 1.27 percentage points, respectively. These results demonstrate the effectiveness and generalizability of the proposed framework across diverse remote sensing scenarios under image-level supervision. Full article

Climate and land-use change are shifting the spatial distribution and habitat suitability of many plant species, particularly those with narrow ecological niches and limited ranges. Dracaena cochinchinensis (Lour.) S.C.Chen is a medicinally important and nationally protected plant species distributed in tropical and subtropical southern China. Predicting its habitat suitability under current and future climate scenarios is essential for understanding its responses to environmental change and guiding conservation and resource management. In this study, an optimized maximum entropy model was used to predict the current habitat suitability of D. cochinchinensis in China and to assess changes in habitat suitability under future climate scenarios in the 2090s. The model showed reliable predictive performance for estimating species distribution. Under current conditions, suitable habitats were concentrated in southern China, particularly southern Yunnan, Guangxi, Guangdong, Hainan, Taiwan, and parts of Fujian, covering 23.93 × 10⁴ km² (8.31% of the study area). Mean temperature of the driest quarter (Bio9), temperature annual range (Bio7), and precipitation seasonality (Bio15) were the dominant predictors, highlighting the importance of thermal conditions in shaping species distribution, whereas lithology (Lith) and soil type (ST) were the major non-climatic contributors influencing habitat suitability. Under future scenarios, suitable habitats showed spatial redistribution rather than uniform expansion or contraction. The spatial comparison among future scenarios identified stable habitats in the central and southwestern parts of Hainan Island, south-central Yunnan, and western Guangxi, with habitat losses mainly occurring in coastal South China and limited expansion in southwestern regions. Habitat centroid shifts were scenario dependent, with an eastward shift under SSP1-2.6 and northwestward shifts under SSP3-7.0 and SSP5-8.5, with migration distances ranging from 60.09 to 165.77 km. These results indicate that climate change may substantially reorganize the distribution pattern of D. cochinchinensis in southern China. Therefore, future conservation planning should prioritize current high-suitability areas, potential macroclimatically stable areas, and emerging suitable habitats to support long-term preservation and sustainable utilization of nationally protected medicinal species. Full article

Mesh resolution is a central source of numerical uncertainty in two-dimensional flood modelling because it controls terrain representation, wetting–drying transitions, computational cost, and the transferability of calibrated parameters. Yet controlled evidence remains limited on whether two widely used solvers exhibit the same resolution response patterns across flood magnitudes. This study compares calibrated model configurations developed in HEC-RAS 2D version 7.0 and SMS 13.0–TUFLOW Classic for a mountainous to low-gradient reach of the Little River, Tennessee, USA, using identical terrain, land cover roughness, and hydrological boundary data. The models were calibrated with the 6–11 April 2025 event and independently validated with the 11–16 March 2026 event. A factorial experiment comprising two models, five cell sizes (15–55 ft), and five flow conditions (the observed event and Q₅₀, Q₁₀₀, Q₂₀₀, and Q₅₀₀ design floods) resulted in 50 simulations. The study simultaneously evaluates resolution sensitivity within each calibrated model and inter-model convergence under a common scenario matrix. Inundation extent was most resolution-sensitive during the lower-magnitude observed event, whereas inter-model extent differences decreased as flood magnitude increased. Outlet hydrographs and peak discharges were comparatively stable, but local water-depth distributions became less consistent as the mesh was coarsened, particularly in SMS-TUFLOW. HEC-RAS retained greater depth consistency, plausibly because its sub-grid property tables preserve cell-scale volume and conveyance information. Overall, mesh adequacy was output- and solver-specific, indicating that grid selection should be based on the intended hydraulic output and supported by explicit spatial stability testing. Full article

Urban informality in amphibious territories represents a critical yet understudied dimension of contemporary urbanisation in the Global South. This article analyses the interrelations between spatial transformation, social equity, and environmental change in Las Flores, an informal settlement located between the Mallorquín Lagoon and the Magdalena River in Barranquilla, Colombia. Drawing on a mixed-methods approach combining GIS interpretation, participatory mapping, and in-depth interviews, the study examines how processes of informal territorialisation have reshaped both the physical landscape and the social fabric of this amphibious environment. Results indicate that the settlement has expanded progressively over the past two decades, occupying areas of the wetland previously covered by mangroves and natural vegetation. This expansion has been accompanied by environmental degradation, soil infilling, and declining water quality. Residents face persistent infrastructural deficits, limited access to education and healthcare, and increasing social fragmentation between the formal and informal sectors. Yet, the community also exhibits strong organisational capacity, adaptive livelihoods, and a deep sense of place that sustains local identity and resilience. These dynamics exemplify the paradox of amphibious life: coexistence with water as both a resource and a source of vulnerability. Building on these findings, the study develops an urban socio-ecological conceptualisation of Amphibious Justice, a framework for interpreting equity, recognition, and governance in hybrid territories where urbanisation and land–water dynamics intersect. The article proposes a framework of equitable amphibious urbanism that integrates environmental restoration, social inclusion, and participatory governance. The findings suggest that sustainability in such territories cannot be achieved through technocratic restoration or forced resettlement, but through co-produced strategies that recognise local knowledge, tenure security, and ecological stewardship. Ultimately, the case of Las Flores offers insights into how cities in the Global South can pursue just and adaptive coexistence with water amid growing climate and urban pressures. Full article

Xinjiang grasslands are major ecological barriers and pastoral bases exposed to climate variability, grazing, land-use change, and socioeconomic pressures. This study assessed the vulnerability of grassland social–ecological systems in 81 counties from 2001 to 2020. A driver–pressure–state–impact–response (DPSIR) framework was integrated with CRITIC weighting and the TOPSIS model. Sen’s slope estimator and the Mann–Kendall test were used to detect temporal trends, and the obstacle degree model identified key constraints. The mean vulnerability index declined from 55.82 to 50.27, accompanied by a marked increase in counties with low or relatively low vulnerability. Vulnerability was highest in southern Xinjiang, where highly vulnerable counties were concentrated in densely populated oasis agricultural areas. Significant declines occurred in 80 of the 81 counties. The response, state, and pressure dimensions showed the highest obstacle degrees. Fiscal self-sufficiency, agricultural machinery power, the grassland–cropland ratio, economic density, and the remote sensing ecological index were the main obstacles. Declining vulnerability was associated with lower grazing intensity, a smaller bare land proportion, and increases in economic density, rural disposable income, and total agricultural machinery power, whereas the comparatively high vulnerability in southern Xinjiang was associated with limited fiscal capacity, lower ecological quality, and arid conditions. These findings support region-specific grassland conservation and differentiated county-level governance. Full article

Land-use change directly and indirectly affects regional carbon storage by reshaping the composition and function of terrestrial ecosystems, and understanding this process is important for guiding land-use strategies toward carbon neutrality. This study used wilderness mapping to assess wilderness quality in Pu’er City and integrated the FLUS–InVEST model to analyze spatiotemporal changes in land use and carbon storage from 2000 to 2020 and to simulate land-use and carbon-storage outcomes for 2030. Based on the wilderness-mapping results, the study area was divided into four wilderness-quality classes, and four scenarios were established: No Protection, Limited Protection, Moderate Protection, and Strict Protection. From 2000 to 2020, cropland and forestland decreased, whereas grassland, water bodies, and built-up land increased. Carbon storage first increased slightly and then declined, with an overall decrease of approximately 1.87 × 10⁶ Mg C. Relative to the harmonized five-class 2020 FLUS baseline, the 2030 simulations showed a decline of approximately 1.46 × 10⁶ Mg C under the No Protection scenario, whereas the Moderate Protection scenario produced an increase of approximately 0.21 × 10⁶ Mg C. Under the baseline carbon-density parameters, the Moderate Protection scenario yielded the highest estimated total carbon storage among the four scenarios and resulted in a larger simulated built-up land area than the Strict Protection scenario. However, the relatively small difference in estimated total carbon storage between the Moderate and Strict Protection scenarios should be interpreted with consideration of carbon-density uncertainty. Full article

Local physicochemical variability in small post-extraction gravel pit lakes is often difficult to interpret because water chemistry may reflect both seasonal dynamics and highly local shoreline conditions. This study assessed spatial and temporal patterns in selected physicochemical parameters of near-surface water at eight fixed littoral sampling points representing contrasting riparian and land-use settings in two gravel pit lakes within the Międzyrzeckie Jeziorka lake complex near Międzyrzec Podlaski, eastern Poland. Surface-water monitoring was conducted from 9 November 2024 to 23 September 2025. The analytical dataset covered total organic carbon, inorganic carbon, total carbon, total nitrogen and laboratory-measured pH. The strongest and most consistent changes occurred between sampling dates in both lakes. Differences among fixed points were limited; total nitrogen differed between two points in Gravel pit lake no. 1, while the other parameters showed no date-adjusted point-level pattern over the monitored period. Mean total organic carbon and total carbon concentrations were similar in the two lakes, whereas Gravel pit lake no. 1 showed higher maximum values for most parameters and greater variability in selected carbon and nitrogen fractions. Laboratory-measured pH was generally weakly acidic to near-neutral, and the lowest observations occurred during individual campaigns. The findings show pronounced temporal dynamics and a restricted spatial signal in the monitored littoral zones. Repeated fixed-point sampling provided a clear basis for comparing temporal changes and local differences in these post-extraction lakes. These inferences are restricted to the monitored near-surface physicochemical variables. Full article

Land consolidation (LC) is increasingly expected to support food security, ecological restoration, rural development, and climate mitigation, yet evidence on its carbon effects remains fragmented across engineering, ecological, spatial, and governance research. This review combines bibliometric mapping with structured evidence synthesis of 355 records from WoS and CNKI to examine knowledge evolution, analytical paradigms, and their integration. The field has expanded from component-based assessments of construction emissions, soil carbon, and land-cover change toward life-cycle accounting, carbon fractions, ecosystem-service interactions, spatial optimization, and policy evaluation. WoS-indexed and Chinese-language literature show distinct but increasingly convergent orientations shaped by differences in intervention contexts, disciplinary traditions, analytical scales, and available evidence. Four complementary paradigms are identified: carbon accounting, biogeochemical processes, spatial land systems, and decision support and governance. Together, these paradigms reveal interconnected carbon pathways but remain constrained by inconsistent accounting boundaries, weak process–scale–time linkages, and limited integration with land-governance decisions. Future research should therefore advance standardized life-cycle and multi-scale accounting, mechanism-based assessment of long-term carbon and ecosystem-service dynamics, and digitally and institutionally enabled low-carbon governance. Full article

Leaf area index (LAI) is widely used to characterize foliage amount and seasonal canopy development, but it captures only selected aspects of forest structure and can be difficult to retrieve reliably in heterogeneous, multilayered stands. This study evaluates Sentinel-2-based LAI information across eight sites in the Slovenian Classical Karst encompassing post-disturbance regeneration and established forest stands in dolines and relatively level inter-doline terrain. Field effective LAI measured during six periods in 2021 was compared with six Sentinel-2 spectral variables, LAI derived using the Sentinel Application Platform (SNAP), and the Copernicus Land Monitoring Service High-Resolution LAI product. The analysis explicitly distinguished two dimensions of retrieval performance that are often conflated: seasonal fidelity within sites and preservation of structural differences among sites. Most satellite-derived variables and LAI products captured the broad phenological progression from canopy development to senescence. However, strong temporal agreement within sites did not consistently translate into preservation of the ordering or magnitude of structural differences among sites. Several methods compressed the range of high effective LAI values at dense regeneration sites with substantial lower-layer vegetation. The study therefore provides a more informative framework for evaluating LAI products by identifying which component of variation drives apparent agreement. These findings indicate that Sentinel-2 can support phenological monitoring and broad screening of post-disturbance vegetation development. However, quantitative comparisons of canopy density or structural recovery across heterogeneous stands require consideration of canopy heterogeneity, potential spectral saturation, and plot-to-pixel support. The evaluation framework and the observed retrieval limitations are relevant beyond karst forests, particularly to post-disturbance stands, fragmented forests, open woodlands, and sites with dense understory or regeneration layers. Full article

Agricultural socialized service has expanded rapidly within China’s smallholder farming system, yet evidence on their relationship with land productivity remains limited. Using three waves of data from the China Land Economic Survey between 2020 and 2022, this study examines the association between agricultural production outsourcing and land productivity. The results show that outsourcing participation is associated with 4% to 5% higher land productivity, with the positive association remaining broadly robust across alternative estimators and identification strategies. Heterogeneity analysis indicates that this association is stronger on land with better soil fertility, among older-generation farmers, and among small-scale farmers. Further analysis suggests that outsourcing participation is associated with higher fertilizer application intensity, greater mechanization, and lower household agricultural labor input, providing evidence on potential channels linking outsourcing to land productivity. These findings highlight the potential role of specialized agricultural services in relaxing production constraints faced by smallholders, while also suggesting that their productivity implications depend on household and land characteristics. Full article

The pollution generated during the synthesis and processing of synthetic fibers demands new raw materials for the textile industry. Natural fibers are a good solution due to their increased comfort while wearing, and their biodegradability. However, the use of some of these fibers, such as cotton or ramie, comes with limitations like competition with edible plants for land and water during cultivation, and the pollution generated during their processing. Thus, finding other sources of fibers that do not compete with plants for food seems to be a good solution. Pineapple fibers represent a good example of synergy, with the fruit being intended for food while the leaves, once considered waste, can be valorized as fibers. This paper describes the progress in research on obtaining pineapple fibers and their properties depending on their mode of preparation. The technical progress in preparing pineapple fibers is emphasized. Their application in textile materials, alone or as composites, is presented. According to the literature, pineapple fibers may also be applied in other fields besides the textile industry. Moreover, the fact that waste is the raw material for these fibers represents a great asset, and the development of new technologies for their production and application is recommended. Full article

Ensuring the long-term sustainable utilisation of supplemented cropland has become a critical challenge for implementing China’s requisition–compensation balance of farmland (RCBF) policy, particularly in the fragmented hilly and mountainous regions of Southwest China. Existing studies generally evaluate land suitability and utilisation performance separately, making it difficult to identify mismatches between theoretical suitability and actual utilisation and thereby limiting targeted regulation. To address this limitation, this study proposes a suitability–utilisation deviation identification (SUDI) framework, which integrates four sequential analytical components: three-dimensional suitability assessment, suitability–utilisation deviation identification, driving mechanism analysis, and sustainable regulation. Taking Beibei District of Chongqing as a case study, supplemented cropland parcels were identified using the 2020–2024 land change survey data. A three-dimensional suitability evaluation system incorporating production, ecological, and utilisation attributes was established to quantify theoretical land suitability. Actual utilisation performance was characterised using the land economic utilisation coefficient, and suitability–utilisation deviation was identified through residual analysis between theoretical suitability and utilisation intensity. A Bayesian-optimised Extreme Gradient Boosting-SHAP (XGBoost-SHAP) model was subsequently employed to reveal the nonlinear effects and interaction mechanisms of the driving factors. The results indicate the following: (1) supplemented cropland in Beibei District is predominantly characterised by medium-to-high suitability, with high-suitability patches exhibiting a mosaic spatial pattern of local aggregation and overall dispersion; (2) suitability–utilisation deviation is dominated by under-utilised plots, whereas well-matched and over-intensified plots account for substantially smaller proportions, indicating that insufficient realisation of land suitability is the prevailing utilisation pattern; and (3) the land economic utilisation coefficient is the dominant factor driving suitability–utilisation deviation, while high-standard farmland construction and plot area exhibit significant mitigating effects. Moreover, significant interaction effects between utilisation intensity and location-related variables reveal that unfavourable spatial conditions amplify deviation risk under intensive land use. The proposed SUDI framework extends conventional suitability assessment by explicitly linking suitability evaluation with utilisation performance, driving mechanism analysis, and differentiated regulation. It provides a transferable analytical framework for diagnosing suitability–utilisation mismatch and supports dynamic management and sustainable utilisation of supplemented cropland in fragmented hilly and mountainous regions. Full article

The utilisation of microwave radiances is crucial for enhancing the precision of weather forecasts. Despite existing uncertainties over land and ice-covered surfaces, recent advances have enhanced their use. This study examines the impact of assuming either Lambertian or specular surface reflection on the simulation of brightness temperatures for surface-sensitive, clear-sky AMSU-A microwave radiances across land and snow-covered areas. It represents the preliminary work before running a full assimilation and forecast impact study. Using the high-resolution HARMONIE-AROME regional modelling system, experiments were conducted to retrieve and analyse the retrieved emissivity in different conditions/seasons. The emissivity was also used as input to the radiative transfer model to simulate brightness temperatures of surface-sensitive sounding channels. The results show that the Lambertian assumption produces higher variability in dynamic surface emissivity, while the specular approach yields smaller and more consistent deviations. During winter, specular reflection shows higher first-guess departures (e.g., observations minus simulations) for surface-sensitive sounding observations, whereas in summer it performs better over land surfaces. Over snow-covered regions, the use of the Lambertian reflection to simulate the brightness temperature gives smaller mean errors for AMSU-A channels 4 (52.8 GHz) and 5 (53.59 GHz). These findings encourage further investigation into implementing a parameter that accounts for the Lambertian component of surface reflection when simulating brightness temperature in high-resolution limited-area models. Additionally, these findings provide practical guidance for configuring complex Nordic surface regional models and for future Arctic Weather Satellite microwave radiance assimilation. Full article

The East Kazakhstan Region is characterized by heterogeneous mountain, steppe, agricultural, and urban–industrial landscapes. However, spatially integrated assessments of land-cover conditions, agricultural dynamics, settlement expansion, and atmospheric trace-gas patterns remain limited. This study combines open multi-source Earth observation datasets and products to assess regional land cover, cropland dynamics from 2003 to 2019, built-up-area expansion from 1985 to 2025, and tropospheric nitrogen dioxide (NO₂) column density from 2019 to 2024. The land-cover classification was based on Sentinel-2 imagery and ancillary geospatial datasets. Its random forest component achieved an overall accuracy of 90.17%, a Cohen’s kappa coefficient of 0.892, and a macro-averaged F1 score of 0.888. Grassland was the largest mapped land-cover class, covering 39.3% of the study region, followed by dense vegetation (16.2%), rock (14.9%), and sparse vegetation (10.6%). Mapped cropland extent increased from 4771.6 km² in 2003 to 5043.9 km² in 2019, representing a net increase of 272.3 km² (5.7%), although a minor decrease occurred after 2015. The harmonized built-up area time series showed expansion within all nine major settlements, with the largest absolute increase observed in Öskemen (Ust-Kamenogorsk). Newly detected built-up pixels during 2000–2024 were mostly associated with transition from grassland and dense shrubs. Annual Sentinel-5P observations showed recurring elevated tropospheric NO₂ column densities in northwestern East Kazakhstan, particularly around Öskemen. Over the built-up footprint, the area-weighted mean increased from 24.1 µmol m⁻² in 2019 to 29.2 µmol m⁻² in 2024. The integrated framework provides a spatially consistent regional baseline while identifying descriptive patterns that require further process-based investigation in future. Full article

Uganda hosts one of the largest refugee populations in Africa, with South Sudanese refugees forming the majority. Uganda’s progressive approach to refugee protection is recognized globally as a model for economic inclusion, integration, and self-reliance. Notwithstanding global praise, empirical evidence indicates an increasingly challenging situation for refugees’ socio-economic well-being and the difficulties this poses for pursuing sustainable lives. With limited access to agricultural land, limited formal employment opportunities, and rising budget cuts, refugees in rural settlements are often compelled to diversify their livelihoods and enter the informal economy, where home-brewed alcohol production is a visible strategy. Drawing on eight months of ethnographic fieldwork in Bidibidi from January 2025 to August 2025, this paper explores the attractions and dilemmas of home-brewed alcohol production and sale as a rural livelihood from the perspectives of South Sudanese refugees and host communities in Bidibidi. This paper advances a debate on the under-researched prospects of refugee integration and inclusion through rural livelihoods. These findings suggest areas where Uganda’s current refugee settlement agenda and alcohol policies may benefit from reconsideration. In addition, this paper highlights gaps in refugee and host community involvement in policy making and proposes ways to better address their livelihood needs for socio-economic inclusion and local integration. Full article