Against the backdrop of global climate change and urbanization, sponge cities and urban rewilding, as two nature-based solutions, exhibit potential synergies in enhancing urban stormwater resilience. However, systematic and quantitative research on collaborative pathways remains lacking. This study takes Los Angeles, USA, as an example and constructs an integrated framework of “hydrological simulation–spatial analysis–intelligent optimization-future validation” to explore the integration of Eastern and Western urban practices in stormwater management. First, the SCS-CN model was used to comparatively evaluate the hydrological performance of three scenarios: the current state, sponge city-dominated, and rewilding-dominated. It was found that the rewilding scenario achieved a runoff reduction rate of 6.8%, representing a relative increase of 34.2% compared to the sponge city scenario. Second, by integrating multi-source spatial data (population density, proximity to water bodies, slope, and proportion of permeable surfaces) and employing multi-criteria decision analysis (MCDA), a Collaborative Opportunity Index (OCI) was constructed to precisely identify high-benefit priority areas. Static simulations showed that the rewilding strategy guided by OCI enhanced collaborative benefits by a factor of approximately 7.6 relative to traditional engineering measures. Subsequently, the collaborative layout was abstracted as a spatial optimization problem under budget constraints, and a multi-strategy genetic algorithm (GA) was introduced to solve for the optimal configuration across five budget levels. Results indicated that under all budget levels, the selection rate of sponge city measures was 0%, with all transformed cells assigned to rewilding strategies, highlighting their cost-benefit advantage; moreover, the cost-benefit ratio exhibited marginally diminishing returns as the budget increased. Finally, the optimized layout was validated using SCS-CN simulations across 12 future climate scenarios (72 experiments in total), from which three macro patterns were extracted: climate pathways dominate storm and flood event pressure; investment benefits show marginally diminishing returns; and the benefits of rewilding are more pronounced under extreme scenarios. The methodological framework and empirical findings of this study integrate the systematic thinking of China's sponge city initiatives with the cutting-edge concepts of Western ecological restoration, thereby providing theoretical support and a practical reference for formulating scientific climate adaptation strategies in high-density cities.
Pelagic longline fisheries interact with diverse predator assemblages, yet conventional richness-based indicators can overlook dominance stress, functional imbalance, and species-level conservation risk. We developed an integrated ecological indicator framework and applied it to Taiwanese longline catch records for 28 pelagic species across six Pacific FAO regions. Because these data reflect catch-derived assemblage structure rather than direct community sampling, indicators were interpreted as relative, effort-influenced signals. The framework combined alpha and beta diversity, community ordination, indicator-species analysis, species-ratio instability, a proposed Biodiversity–Dominance Decoupling (BDD) metric, and a composite conservation-vulnerability score. Species richness was highest in FAO 61 (Northwest Pacific) and FAO 71 (Western Central Pacific), but effective diversity and evenness were lowest in FAO 81 (Southwest Pacific) and FAO 87 (Southeast Pacific), revealing hidden imbalance where species presence did not yield balanced structure. Ordination and indicator-species analyses separated shark-associated assemblages in the western Pacific from albacore-dominated assemblages in central and southern regions. The shark-to-tuna ratio was the most spatially unstable, indicating functional decoupling between vulnerable predators and commercial tuna assemblages. BDD peaked in FAO 87 and FAO 81, while integrated ecological decoupling was highest in FAO 81, marking it the most conservation-sensitive region. Vulnerability scores were greatest for large sharks, including bigeye thresher, hammerhead sharks, and pelagic thresher, with risks concentrated in FAO 81 and FAO 87. Correlations among indicators confirmed complementary, non-redundant signals. We recommend routine decoupling monitoring, strengthened species-level bycatch reporting, and precautionary shark measures, supporting ecosystem-based fisheries management and SDG 14.
Anthropogenic nutrient enrichment has profoundly altered riverine biogeochemical cycling and ecosystem functioning worldwide. To promote a clearer definition of “oligotrophication” in ecological studies, we define the term specifically as the physicochemical decline in nutrient concentrations, distinguishing it from “biological recovery”, which encompasses algal biomass reduction, community restructuring, and functional adaptation. Here, we synthesise global evidence for long-term (≥10 years) trophic change in rivers through a systematic review of 118 studies covering 413 river basins across six continents. Trends show marked regional contrasts: oligotrophication is widespread in Europe, whereas eutrophication persists across East Asia and Africa. North American rivers frequently exhibit stalled recovery or decoupled physicochemical and biological trajectories. Across all regions, warming, hydrological alteration, and legacy nutrients repeatedly emerge as crucial recovery constraints. Crucially, where long-term biological data are available, nutrient reductions rarely guarantee a return to historical baselines; instead, ecosystems frequently shift towards novel ecological trajectories. However, because equivalent long-term data on community composition, biodiversity, and ecosystem functioning remain exceptionally scarce compared to physicochemical records, the broader global extent of these novel biological responses remains only partially resolved. Consequently, current monitoring programmes capture chemical changes far more effectively than biological recovery. Future monitoring should emphasise long-term biological responses to accurately evaluate whether nutrient reductions translate into functional ecosystem recovery.
This study aimed to quantify how spatial scale and aggregation rules alter ecosystem service value (ESV) estimates and their spatial interpretation in Northeast China from 2000 to 2024. Using China Land Cover Dataset (CLCD) data, we compared aggregation by area weighting with aggregation by dominant class at 1 km, 5 km, 10 km, county, and provincial scales, using the 1 km grid as the benchmark; evaluated total value deviation, value hotspot consistency, aggregation loss, spatial autocorrelation, and scale dependent associations; and conducted broad and land cover class specific Monte Carlo stress tests together with a targeted experiment that reduced water and wetland coefficients. Total ESV at the 1 km benchmark increased only slightly, from 2.621 × 1012 yuan in 2000 to 2.653 × 1012 yuan in 2024 (1.22%). Aggregation by dominant class increasingly underestimated total ESV at coarser scales, whereas aggregation by area weighting better preserved the regional total but changed hotspot locations. In the 10,000 draw class specific experiment, the leading positive and negative transition pathways were unchanged in every draw, but the central 95% simulation interval for the modest 2000 to 2024 net change crossed zero. We conclude that ESV evidence should be interpreted with explicit reference to spatial scale, aggregation rule, and coefficient assumptions before it is used in ecological planning or compensation.
Water use efficiency (WUE) links terrestrial carbon uptake and water loss, yet its nonlinear responses across aridity gradients remain poorly understood in China. Previous studies have largely focused on single hydroclimatic regimes, emphasizing linear relationships and static assessments of WUE drivers. Using the aridity index, China was divided into Arid, Semi-arid, Sub-humid, and Humid zones. WUE during 2001–2020 was derived from MODIS products. The mean WUE across China was 1.168 ± 0.518 gC·kg−1 H2O, exhibiting increasing patterns from west to east and from Arid to Humid zones. Trend analysis showed that 63.51% of pixels experienced increasing WUE trends, although only 16.92% were statistically significant. The Semi-arid zone exhibited the highest increasing rate (0.0025 gC·kg−1 H2O·a−1), whereas the Arid and Sub-humid zones showed the highest proportions of increasing and decreasing persistence patterns (47.50% and 59.49%, respectively), based on Hurst analysis. Zonal-specific XGBoost models combined with SHAP and GAM smoothing captured nonlinear relationships and interactions between WUE and environmental factors, with mean test R2 values ranging from 0.84 to 0.93 across the four aridity zones. Leaf area index (LAI) was the most important and consistent contributor across zones; vapor pressure deficit (VPD) dominated in Arid to Sub-humid zones and water-availability variables became more influential in Humid zones. Response curves and transition points varied along the aridity gradient, with interaction patterns shifting from heterogeneous climate–vegetation interactions in arid zones to stronger multi-factor coupling under humid conditions. These findings provide new insights into ecosystem water–carbon coupling across contrasting hydroclimatic backgrounds.
The Yellow Sea–Bohai Gulf coasts support millions of shorebirds along the East Asian–Australasian Flyway (EAAF), but natural high-tide roosts (HTRs) have degraded sharply due to reclamation, dikes and disturbance, forcing birds to use suboptimal artificial habitats. To inform evidence-based HTR restoration, we surveyed shorebird communities at two restored HTRs in southern Jiangsu, China (2021–2024), and analyzed spatiotemporal dynamics and drivers via aligned rank transform ANOVA (ART-ANOVA), generalized additive models (GAMs), and hierarchical partitioning (HP). Abundance and species richness peaked in early May and August–early September, with significantly higher values in autumn than spring. Migratory phenology contributed the highest independent explanatory power to community dynamics, followed by landscape factors. Notably, in the study region, a well-configured 48-ha restored HTR temporarily supported more than180,000 shorebirds during spring tides while sustaining high functional richness (FRic). Within the observed empirical gradient across our study sites, higher mudflat dominance (observed MLPI up to 48.6%) and complex patch shape (PAFRAC up to 1.4) both positively correlated with abundance and richness. Simultaneously, lower water patch cohesion (WCO down to 98) correlated with higher richness and FRic, a relationship we hypothesize may reflect microhabitat heterogeneity.Our findings provide a practical regional reference for HTR management in the Tiaozini wetland and a conceptual framework for the broader Yellow Sea–Bohai Gulf. Given the site-specific nature of these results, future cross-regional validations across multiple restored HTRs throughout the EAAF are required to assess transferability and inform context-tailored restoration interventions.
The Qinghai–Tibet Plateau, a globally critical climate-sensitive alpine region, is facing severe meadow degradation, habitat fragmentation and biodiversity loss driven by climate change and anthropogenic disturbances. Most existing assessments lack comprehensive ecological indicators for quantifying climate-driven habitat responses of endangered mammals, use fragmented methods that cannot capture cascading landscape processes, and over-rely on single-species models, which restricts integrated evaluation of multi-species habitat dynamics under future climate scenarios. To address these gaps, we established an integrated framework coupling land-use simulation, habitat suitability and landscape connectivity modeling. We quantified composite habitat suitability as a core indicator, together with Morphological Spatial Pattern Analysis (MSPA)-derived ecological core area and ecological connectivity cost distance. Results show that warming does not necessarily cause a generalized decline in habitat for endangered alpine mammals. Instead, habitat outcomes emerge from non-linear interactions among warming intensity, precipitation regime shifts, topographic constraints, and scenario-specific land-use disturbance patterns. SSP2–4.5 yields the smallest and most fragmented ecological sources (55,653 km2); warm-wet SSP5–8.5 creates continuous high-elevation habitats (137,261.08 km2); SSP1–2.6 boosts core habitats by 258.8% to 210,554.08 km2. We identified six habitat response patterns for nine endangered mammals, validated the effectiveness of existing protected areas and detected dynamic conservation gaps. This study clarifies climate-driven habitat indicator responses and supports climate-adaptive conservation for endangered mammals on the Qinghai–Tibet Plateau.
Agroforestry is widely promoted for soil restoration, but its effects on the coordinated responses of soil C, N and P remain uncertain. We synthesized 1902 observations from 103 studies worldwide to assess agroforestry effects on soil C, N and P contents, C:N:P stoichiometry, microbial biomass and microbial diversity relative to monoculture controls. Because bulk-density data were not consistently reported, we evaluated soil nutrient contents rather than areal nutrient stocks. Agroforestry increased SOC, TN and TP, but their responses were non-proportional: SOC increased more strongly than TN and TP, and only C:N increased significantly. SOC and TN responses were significantly coupled but sub-proportional, whereas SOC–TP and TN–TP relationships were weak or non-significant. Microbial biomass N, microbial C:N and N:P, and fungal Chao1 also responded to agroforestry. Moderator analyses identified MAP and MAT as the most supported moderators of SOC responses and initial soil N:P as the most supported moderator of TN responses; no robust moderator was identified for TP. Overall, agroforestry increased soil C, N and P contents and altered multiple microbial indicators. However, C, N and P responses were non-proportional and context dependent, and the ecological direction of these stoichiometric shifts could not be determined from the available data. Agroforestry sustainability should therefore be assessed using SOC together with soil nutrient responses, C:N:P stoichiometry and microbial indicators.
Land use associations with river water quality vary with watershed context, pollutant properties, and spatial scale. In plateau agro-urban watersheds, these relationships are complex due to intertwined agriculture, urbanization, and hydroclimatic variability. This study investigated pollutant-specific environmental associations and scale-dependent riparian responses in the Huangshui River Basin. Water quality observations were from 76 monitoring sections during six sampling campaigns, including three wet-season and three dry-season surveys between August 2020 and March 2023. Data included multi-scale riparian land use, hydroclimatic conditions, upstream concentrations, crop composition, and management data from 675 farmer surveys. Separate Random Forest (RF) models were developed for the permanganate index (CODMn), total nitrogen (TN), and total phosphorus (TP), with model interpretation via SHapley Additive exPlanations (SHAP) with spatial cross-validation. Results showed selective water quality impairment, with 23.3% of sampling records failing to meet Chinese Class III surface-water criteria (GB 3838–2002), primarily due to TN (1.97 mg L−1) and TP (0.17 mg L−1). The mean training and spatial validation coefficients of determination (R2) values were 0.873/0.495 for CODMn, 0.713/0.389 for TN, and 0.725/0.244 for TP. SHAP results indicated that CODMn was mainly associated with riparian land use (47.46%) and hydroclimatic variability (36.75%), TN with riparian land use structure (67.15%), and TP with riparian land use (48.60%), upstream water quality (29.76%), and hydroclimatic variables (8.80%). Key land use predictors appeared at different buffer widths, indicating that no single riparian scale explained all pollutants. Emission-coefficient estimates further indicated that crop-area statistics alone may underestimate agricultural TN and TP source pressures. These findings provide a pollutant-specific and scale-sensitive framework for interpreting water quality risks in plateau agro-urban basins.
Understanding how biodiversity and environmental drivers jointly regulate ecosystem multifunctionality is critical for improving forest ecosystem management and resilience under global environmental change. Despite widespread reports of positive relationships between biodiversity and ecosystem multifunctionality, their generality across environmental contexts, particularly along strong climatic gradients, remains uncertain. Here, we examined two adjacent subtropical mountain systems, Baoding Mountain and Mingyue Mountain, in southwestern China as paired elevational gradients. We assessed how taxonomic, functional, and phylogenetic diversity interact with abiotic environmental drivers to shape ecosystem multifunctionality. By integrating biodiversity metrics, climatic, soil, and topographic variables, and 16 ecosystem functions related to carbon storage, nutrient cycling, forest regeneration, and soil microbial processes, we quantified both direct environmental effects on multifunctionality and indirect pathways mediated through biodiversity. Ecosystem multifunctionality increased consistently with elevation in both mountain systems, whereas biodiversity patterns and biodiversity–multifunctionality relationships differed markedly between landscapes. In Baoding Mountain, taxonomic and phylogenetic diversity were positively associated with multifunctionality, whereas functional diversity showed a negative relationship. In contrast, multifunctionality in Mingyue Mountain was primarily associated with functional diversity, while taxonomic and phylogenetic diversity showed no significant effects. Hierarchical partitioning indicated that mean annual temperature showed the largest relative contribution among the measured abiotic predictors in both mountain systems, whereas functional diversity showed the largest relative contribution among biodiversity predictors. Structural equation modelling further revealed that abiotic factors influenced ecosystem multifunctionality both directly and indirectly by reshaping functional diversity, with the strength and direction of these pathways varying between landscapes. These findings demonstrate that biodiversity–multifunctionality relationships are strongly context-dependent, even across geographically proximate forest systems. More broadly, the results highlight the dominant role of environmental drivers in regulating ecosystem multifunctionality. They underscore the need to incorporate environmental context and multidimensional biodiversity into forest ecosystem management and conservation strategies aimed at sustaining ecosystem functioning under ongoing climate change.
Soil erosion poses a serious threat to global black soil security, yet long-term threshold feedback between erosion sensitivity and risk remains unclear. Clarifying spatiotemporal erosion patterns, sensitivity dynamics, and multi-factor drivers is critical for targeted prevention strategies. To address this gap, we quantified erosion intensity, sensitivity, and risk across the black soil region of Northeast China (BSNC) from 1980 to 2020 by integrating the Revised Universal Soil Loss Equation (RUSLE) with the Geographical Detector (GeoDetector). The results reveal that the high-risk centre of soil erosion has shifted from the Changbai Mountains to agricultural plains including the Song-Liao and flood plain. A positive feedback loop between erosion intensity and sensitivity keeps plain areas persistently locked at a mild-to-moderate intensity level. Critical soil erosion sensitivity thresholds show prominent hierarchical spatial divergence: the western steppe requires only 65% vegetation cover to curb erosion, whereas eastern agricultural plains require coverage above 82%. The steppe also exhibits an erosion-triggering rainfall erosivity threshold of only 1040 MJ·mm·hm−2·h−1·a−1, far below mountain zones. The Song-Liao plain exhibits a bimodal soil erosion susceptibility (Skd) distribution, with both Skd I and Skd V areas exceeding 30% of the total land. GeoDetector results indicate that the interaction between soil erodibility and land use reaches 0.4979, far higher than any single factor, highlighting that synergistic amplification of multiple factors drives erosion risk accumulation in the region. The nonlinear vegetation thresholds, positive feedback loops, and gentle-slope erosion mechanisms revealed here offer a transferable framework for other intensively farmed gentle-slope regions.
In the transition zone between the Loess Plateau and the Mu Us Sandy Land, apparent vegetation greening can mask ecohydrological stress in coal-mining subsidence areas. Here, we developed a multi-source remote sensing framework for 2015–2022 that integrates quadrant-based decoupling identification, moving-window interaction modeling, and multi-level residual attribution to distinguish short-term reclamation gains from hydrologically constrained vegetation responses. To address the spatial-resolution mismatch between 30 m NDVI and 1 km hydro-climatic variables, we further conducted a 1 km scale-sensitivity analysis using annual maximum NDVI and the original 1 km hydro-climatic grids. The results showed that active subsidence zones exhibited pronounced anti-phase vegetation-moisture decoupling, with a mean time-series correlation of r = −0.968 across 553,012 significantly negative pixels (p < 0.05). Groundwater-level variation also modulated the vegetation–soil moisture relationship differently between mining and non-mining areas, suggesting a shift from buffering-dominated behavior under natural background conditions to strengthened vegetation sensitivity to shallow moisture fluctuations in mining-disturbed zones. While precipitation, temperature, and shallow soil moisture together explained 59.29% of NDVI variance in the mining area and 59.36% in the non-mining area, residual analysis revealed a relative greenness surplus, defined here as a positive residual NDVI anomaly after controlling selected hydro-climatic variables (precipitation, temperature, and shallow soil moisture). Furthermore, geomorphic threshold analysis indicated that the aeolian sandy region operates within a much narrower optimal moisture window (0.1015–0.1118) than the loess hilly-gully region (0.0849–0.1163), implying greater ecohydrological vulnerability to both moisture shortage and ineffective deep percolation. Finally, a dynamic baseline risk assessment indicated a strong polarization of degradation risk, with 11.71% of the habitats classified as critical-to-moderate relative-risk areas within the 2015–2022 observation period. These findings suggest that surface greening is not equivalent to broader ecohydrological recovery, providing a resilience-oriented paradigm for ecological restoration in fragile semi-arid coalfields.
Zooplankton are pivotal components in aquatic ecosystem energy flow and material cycling, yet decoupling the complex drivers of their biodiversity remains a challenge. This study investigated zooplankton biodiversity along a river-reservoir environmental gradient in a karst reservoir system, characterized by habitat differences related to hydrodynamic conditions, water residence time, nutrient and organic matter concentrations, water temperature, and chlorophyll a concentration. To relax the linearity assumptions of conventional statistical methods and reduce the risk of interpreting model associations as causal relationships, we integrated a LightGBM model with SHAP-based interpretation and a causal inference framework. This approach was used to evaluate potential pathways linking environmental factors, functional traits, and biodiversity. Results identified water temperature, permanganate index, and chlorophyll a as core environmental drivers affecting zooplankton biodiversity and functional structure. Their associations differed between the reservoir and its inflowing river. In the river, water temperature, chlorophyll a, and permanganate index were more directly associated with biodiversity variation, indicating that zooplankton diversity was jointly regulated by thermal conditions, food-resource availability, and organic matter load under hydrodynamically variable conditions. In the reservoir, water temperature and permanganate index were closely related to biodiversity, whereas nitrogen-related variables exerted stronger constraints on functional diversity, suggesting nutrient- and organic-load-related environmental filtering under relatively stable water-retention conditions. Causal inference further suggested distinct trait-mediated pathways. Functional dispersion (FDis) showed a positive estimated effect on biodiversity, consistent with enhanced niche differentiation, whereas reproductive season showed a negative estimated effect, reflecting stronger environmental filtering. These findings provide model-supported evidence of trait-driven assembly in karst systems and offer implications for habitat-specific management, particularly nitrogen load control in reservoirs.
Ecological indicators from concentration–discharge (C–Q) relationships can characterise event-scale water quality dynamics, yet existing indices are typically applied individually and rarely capture the complexity of pollutant mobilisation in catchments with multiple competing sources. This is particularly limiting in groundwater-fed chalk streams, which are globally rare and ecologically sensitive habitats. In this paper we develop and apply a multi-index hysteresis framework, combining traditional Hysteresis Index and Flush Index with a newly introduced Crossing Index (which functions as an ecological indicator of event-scale process complexity), and apply it to a mixed land use chalk stream in England. Through a series of 18 events and high-frequency monitoring of ammonium, turbidity, conductivity and dissolved oxygen, we tested the multi-index framework and analysed the catchment-scale water quality response. Results are interpreted through a two-pathway model, whereby storm events mobilise competing water sources: a groundwater pathway that dominates during recession, and a fast urban surface runoff pathway that dominates during the rising limb. A marked persistent shift in turbidity regime was also noted, which suggests the emergence of new, faster sediment pathways during the monitoring period. In addition, an ammonium-dissolved oxygen decoupling was observed, consistent with a sequential biogeochemical cascade. Departure from the two-pathway model was identified and mechanistically interpreted through the Crossing Index, revealing mid-event shifts on pathway dominance. This also provides an ecological indication of multiple or sequential stressors from competing pollution sources. Overall, the multi-index framework resolves event-scale processes that routine sampling cannot detect, and is transferable to comparable groundwater-fed systems.
Assessing watershed health (WH) and ecological security (ES) is a fundamental pillar of environmental sustainability, yet achieving these goals is increasingly hindered by the complexity of the systems involved. Watersheds are governed by a complex web of interconnected variables, from soil and geological constraints to fluctuating climate patterns and diverse human interventions, that collectively influence WH and ES. This study adapted the WH and ES maps from a previous national PSR-based assessment. In contrast, the novelty of the present work lies in quantifying the effect sizes of the controlling factors using the Taguchi optimization method. As a result, it is critical to identify the most influential decision variables and measure their effect sizes to ensure management efforts focus on high-impact interventions. To address this complexity, this study employs the Taguchi Method, a sophisticated fractional factorial experimental design. Unlike traditional “one-factor-at-a-time” approaches, the Taguchi framework is well suited to optimizing WH and ES because it uses orthogonal arrays to examine multiple decision variables with fewer experiments. By analyzing the signal-to-noise (S/N) ratio, this method helps identify the “ideal configuration” of decision variables that remains effective even with uncontrollable environmental noise. Applying this method to 3rd-order watersheds reveals a clear hierarchy of influence, showing that human management factors, rather than natural constraints, are the main levers for WH and ES. The analysis identified the Minimum Normalized Difference Tillage Index as the most critical criterion, accounting for 18.47% and 29.32% of the variance in WH and ES, respectively. This indicates that optimizing agricultural soil management yields the highest ecological return. Additionally, the study identifies the Area of Unimplemented Biological Watershed Operations as a secondary yet important stressor, highlighting a significant gap in current conservation efforts. While geological factors such as Fault Length and climate variability set baseline boundary conditions, they are statistically subordinate to these controllable human pressures. Ultimately, this research offers an evidence-based optimization framework, demonstrating that optimized ecosystem services require improved tillage and biological restoration within integrated policies.
Rapid urbanization has intensified conflicts between urban expansion and ecological conservation in ecologically vulnerable areas worldwide, making sustainable urban development a core target of the United Nations Sustainable Development Goal 11 (SDG 11). Taking Zhangbei County on the Bashang Plateau—a typical ecologically vulnerable region and national new-type urbanization pilot in North China—as a case, this study established a 65-year (1960–2025) land-use dataset by integrating declassified KeyHole satellite imagery and multi-source remote sensing data. Additionally, using landscape pattern metrics, the Random Forest Algorithm (RFA), and the Land Transformation Model (LTM), this study analyzed the spatiotemporal evolution and hierarchical stage-dependent driving factors of urban expansion, and further simulated three alternative urban development scenarios to 2035. The results reveal three major findings: (1) Urban expansion in Zhangbei County exhibits three sequential phases, namely axial extension, single-core infilling, and multi-core leapfrogging. Coordinated changes in urban landscape metrics demonstrate that urban spatial morphology is shifting from “low-density outward sprawl” toward “high-density compact agglomeration”, accompanied by alleviated encroachment on ecological land. (2) The RF model identifies distance to the county town, nighttime light, distance to transport hub, and distance to rural settlements as dominant driving factors, whereas slope and aspect exert marginal effects, indicating that within the gentle plateau tablelands of the Bashang Plateau, water availability rather than topographic relief acts as the primary natural constraint governing urban site selection. (3) Multi-scenario simulations suggest that the New Urbanization Scenario (NUS) can effectively guide intensive urban development toward the northeastern potential zones by designating ecologically vulnerable areas (e.g., water bodies, basic farmland) as rigid exclusion layers. This scenario can achieve coordination between development and conservation while complying with territorial spatial planning, offering a practical reference for delivering SDG 11 at the county scale across ecologically vulnerable regions.
Large-scale forest restoration programs are pivotal for global environmental sustainability, yet their success is often gauged by mere area expansion rather than holistic ecological and socio-economic outcomes. Moving beyond this simplistic view to comprehensively evaluate multi-dimensional effectiveness remains a critical global challenge. Using the Qiantang River Region (QRR), China, as a strategic case study over two decades (2000−2020), this research addresses this gap by constructing an integrated assessment framework that links forest structural integrity, non-linear ecosystem service (ES) gains, and socio-economic trade-offs. We utilized multi-source datasets and employed an innovative SHAP-GAM machine learning approach to quantitatively decouple the drivers of ecological outcomes. Results indicate a substantial improvement in forest structural integrity, with the average Ecological Integrity Index (EII) increasing from 0.73 to 0.81, which fundamentally underpinned enhancements in carbon sequestration and landscape value. Crucially, a distinct non-linear threshold was identified: ecological marginal gains saturate sharply when forest restoration program intervention intensity index (FRP intensity) exceeds approximately 0.03. Socio-economic analysis revealed significant trade-offs, marked by a pivotal economic turning point around 2015 where net benefits declined despite surging investments, alongside persistent spatial inequities between ES supply and population demand. Based on these findings, a four-quadrant spatial management framework is proposed to guide precision governance. This study provides critical scientific insights for transitioning forest restoration policy from “quantity-driven” expansion to “quality-and-efficiency-oriented” sustainability.
Estuaries, where freshwater and seawater converge, are characterized by dynamic habitats and high biodiversity. However, intensive anthropogenic modifications have profoundly disrupted their ecological structure and functions. Fish communities are particularly susceptible to such alterations, requiring careful attention through effective monitoring and sustained management efforts. In this study, we used environmental DNA (eDNA) metabarcoding and conventional fishing surveys as complementary ecological indicators of fish communities in the Nakdong River Estuary, an estuarine continuum fragmented by an estuarine barrage. Two sampling campaigns were conducted at seven sampling points representing the estuarine salinity gradient. Fish were surveyed using multiple fishing gears, and DNA extracted from filtered water samples was amplified using the fish-specific MiFish-U primer. Our comparative analysis revealed distinct functional roles for each method: fishing surveys were more effective for detecting locally established freshwater fish, whereas eDNA metabarcoding served as a highly sensitive indicator for marine-origin and transient species. Along the estuarine gradient, fishing surveys reflected abrupt community changes near the barrage, while eDNA metabarcoding captured more gradual ecological transitions. Multivariate analyses further demonstrated that fishing surveys are sensitive to fine-scale spatial differences, whereas eDNA metabarcoding reflects broader spatial connectivity. These results suggest that eDNA metabarcoding detects mobile or transient species through DNA signals and effectively captures broad-scale habitat connectivity, while fishing surveys provide direct evidence of species presence and emphasize resident community structures. Integrating these dual-indicator approaches could provide a more comprehensive monitoring framework for the management and ecological restoration of human-impacted estuarine ecosystems worldwide.
Here, we propose organisms from the European Union's (EU) List of Quarantine Pests for regulation as priority pests based on their potential economic, social and environmental impacts in the EU territory. First, a shortlisting methodology was developed to recommend, from the total organisms in the Union Quarantine Pest list, a subset of pests for detailed assessment. Second, expert knowledge was elicited for 46 pests selected by EU risk managers from the shortlisted pests. Third, expert-based estimates of biological parameters were used to inform a multi-criteria analysis. Our methodology combines impacts on economic, social, and environmental domains into a composite indicator, which is used to rank the pests by their expected impacts. The robustness of our recommendation with respect to the uncertainty in biological parameters and the potential diversity of importance given by risk managers to different domains was evaluated. The contributions of the economic, social, and environmental impact-domains to the composite indicator reveal that high potential impact can be driven by diverse profiles. The regulation on priority pests mandates control and monitoring efforts in all EU Member States, which require considerable resources. Our study provides evidence-based decision support to risk managers, to inform decisions on where to allocate the limited resources for plant health surveillance and monitoring. If the list of priority pests is to remain unchanged in size, our results suggest that if based only on impacts, 9 of the 20 currently listed pests should be substituted.
Global agricultural trade has become a major driver of land-use change and ecosystem service alteration, but the absence of a composite indicator characterising the intensity of trade-offs among multiple services constrains the monitoring of its multidimensional ecological impacts. This study constructed and validated the Trade-off Intensity Index (TII), quantifying the dispersion of the standardised supply–demand ratios of four services—food production, carbon sequestration, water yield and soil conservation—on the global 1 km grid over 2000–2020, and used a two-way fixed-effects model to identify the telecoupled effects of soybean and maize trade on national ecosystem service trade-offs. The main findings are as follows. (1) The ecological consequences of trade exhibited pronounced North–South divergence: developed countries decoupled trade growth from ecological imbalance through environmental pressure displacement, whereas developing net exporters bore structural imbalance risks. (2) The effects were crop-specific: soybean exports were the principal driver of ecological imbalance in developing countries, whereas maize trade showed no significant disruptive effect. (3) Historical attribution indicated that soybean export expansion offset approximately 33.32% of the potential ecological improvement in developing net exporters. As a diagnostic tool, the TII captures multidimensional structural degradation that single indicators cannot reflect, and provides a transferable framework for monitoring telecoupled ecological change beyond agricultural trade; these findings support coordinated green supply-chain governance to counter ecologically unequal exchange.
