Haze has remained a persistent environmental issue in Nanjing, exhibiting notable temporal variation. Emerging evidence suggests that haze weather contributes to elevated cardiovascular mortality, yet the underlying mechanisms remain unclear. We analyzed annual trends in haze days in Nanjing from 2005 to 2023 and found that the number of haze days decreased mainly during the warm season, while levels remained high during the cold season. Further, we analyzed 103,833 health examination records from local hospital (2019-2023) focusing on 12 cardiovascular indicators in relation to haze exposure. We employed a linear mixed effects model combined with a distributed lag nonlinear model to evaluate the associations. Stratified analyses were conducted based on demographic characteristics including sex, age, body mass index (BMI), diabetes and hypertension. Significant associations were observed for most cardiovascular indicators at lags 0-2 days, peaked at lags 3-5 days, and weakened at lags 6-7 days. Significant cumulative percentage changes were observed in several cardiovascular indicators over lag 0-7 days, including blood pressure (0.55%-1.10%), systemic inflammation markers (1.47%-2.10%), hematocrit (0.65%), homocysteine (4.33%), and myocardial injury enzymes (4.18%-8.62%). Increasing haze intensity was associated with a higher risk of all cardiovascular indicators. Stratified analyses revealed that males, individuals aged ≥ 50, those with higher BMI, and those with hypertension were more vulnerable to haze exposure. The haze during the cold season remains a significant public health concern in Nanjing, as even short-term exposure can significantly alter both clinical and subclinical cardiovascular indicators, including elevated blood pressure, acute systemic inflammation, increased blood viscosity, and myocardial injury.

This study is aimed at examining the associations and hospitalization costs between air pollutants and hospital admissions (HAs) for type 2 diabetes mellitus (T2DM) with three typical comorbidities, namely cardiovascular disease (CVD), hypertension, and peripheral vascular disease (PVD). Conditional Poisson regression based on Poisson distribution was used to confirm the relationship between air pollution and HAs for T2DM with comorbidities. Potential effect modifiers, such as age and gender, were also involved. The attributable hospitalization number and costs due to air pollutants exposure were estimated, using WHO's air quality guidelines as reference. In total, 92,381 T2DM HAs were recorded, of which 6,855 patients with CVD, 23,403 with hypertension, and 21,207 with PVD, respectively. The significant short-term effects of NO2, PM10, PM2.5 and SO2 on HAs for T2DM with CVD, hypertension and PVD were different, mainly concentrated on lag 6 and lag 7. Stratified analysis demonstrated that mainly female and elder people (≥45 years) were more vulnerable to air pollutants. Total hospitalization costs attributable to air pollution were substantial. For T2DM patients with CVD, the hospitalization costs linked to NO2, PM10, and PM2.5 were 7.43, 3.04, and 6.79 million China Yuan (CNY), respectively. For the hypertension comorbidity group, costs reached 12.17 million CNY for NO2 and 10.89 million CNY for PM10. Additionally, NO2 exposure for 10.82 million CNY in hospitalization costs for T2DM patients with PVD. These findings underscore that air pollution not only exacerbates health risks but also imposes a significant financial strain on the healthcare system for vulnerable populations.

As urbanization accelerates, the Urban Heat Island (UHI) effect continues to intensify, posing a serious threat to the human settlement environment and urban sustainable development. However, existing methods for quantifying Urban Heat Island Intensity (UHII) in multi-city studies struggle to balance the morphological heterogeneity within cities with cross-city comparability. Furthermore, analyses of driving mechanisms often rely on statistical correlations rather than causal inferences, limiting the effective translation of research findings into the basis for planning interventions. This study examines 150 Chinese cities of varying sizes, constructing a block-Local Climate Zone (LCZ) spatial analysis framework to quantify summer surface UHII. By integrating 22 multidimensional influencing factors, the study jointly employs XGBoost-SHAP machine learning and Double Machine Learning (DML) causal inference approaches to systematically analyze the driving mechanisms and mitigation strategies of UHII at both global and local scales. The main conclusions are as follows: (1) The average UHII across the 150 cities is 1.73°C, with megacities exhibiting the highest and most stable UHII; (2) The Green spaces and Water bodies Ratio (GWR) and Normalized Difference Vegetation Index (NDVI) rank among the most important factors in both global and local models, while the local contribution of urban morphological factors increases significantly with city size; (3) The cooling effect of GWR is the most pronounced; Building Density (BD) exhibits a significant warming effect that is strongest in small cities; the causal direction of the Sky View Factor (SVF) reverses with city size—warming in megacities and cooling in small cities; (4) Machine learning and causal inference exhibit systematic discrepancies regarding variables such as Aggregation Index (AI) and Tree Height (TH), revealing the potential pitfalls of interpreting policy based solely on feature importance. The study’s findings provide empirical evidence at the causal level for formulating differentiated heat mitigation strategies tailored to cities of different sizes.

The California almond industry underwent a remarkable transformation in food safety management and culture following outbreaks of salmonellosis associated with the consumption of raw almonds in 2000-2001 and 2003-2004. However, limited studies have examined these changes from a longitudinal perspective. This study documents the transformation of food safety management in the California almond industry over an 18-year period, explores indicators of change in food safety culture, identifies the key factors driving these changes, and examines the determinants of industry-wide technology adoption. A multifaceted approach was used, consisting of document analysis and semi-structured interviews. This study provides a detailed review of the almond industry's responses to the outbreaks, highlighting the industry commodity board's proactive leadership in crisis management, collaborative research efforts, risk assessment, and the development of a mandatory Salmonella-control program to mitigate the risks associated with raw almonds. These measures significantly strengthened food safety management systems across the industry. The industry has also shown a shift in mentality toward food safety over time, evidenced by increased prioritization of food safety, stronger management commitment, and reduced resistance to change. A conceptual framework integrating institutional theory and diffusion of innovation theory is proposed to illustrate how external and internal institutional pressures, along with intervention characteristics, influenced the almond industry's adoption of Salmonella-control interventions. The study offers valuable lessons on proactive, industry-driven food safety improvements and self-regulation in enhancing food safety outcomes.

Urban net ecosystem productivity (NEPu) plays a pivotal role in enhancing urban ecological conditions and human comfort in living environments. However, the response mechanism of NEPu to urbanization remains unclear due to the challenge of accurately estimating NEPu. This study proposes a hybrid model architecture that combines Residual Networks and Carnegie-Ames-Stanford approach (CASA) model (ResNets-CASA) to estimate NEPu. The ResNets-CASA model is trained and verified at four urban eddy covariance (EC) sites in Tianjin and Shenzhen, China. The model shows an average root-mean-square-error (RMSE) of 1.09 gC/m2/day and a coefficient of determination (R2) of 0.83 for daily NEPu simulation across the four EC sites. The trained ResNets-CASA model at the site scale is further employed for regional-scale NEPu mapping and the generation of long-term historical NEPu datasets (1986-2022) for the two cities. The long-term trend analysis indicates that the NEPu of the two cities shows a significant downward trend over the past 37 years, with an average decline rate of 2.1 gC/m2/year across the two cities. The main cause for the decline trend of NEPu is the changed urban landscape pattern, including: 1) a decrease in urban vegetation coverage resulting from urbanization; 2) shifts in the composition of urban vegetation species due to the substitution of natural woodland and cultivated land with urban landscape grassland. These results emphasize the dominant role of changes in urban landscape patterns, rather than urban microclimate, in the long-term decline of NEPu trends.

Urban tree transpiration (T) can significantly lower urban temperatures and mitigate the urban heat island effect. The careful selection of suitable tree species for the urban landscape holds immense significance in enhancing the quality of the human living environment. However, there remains a deficit of effective models capable of accurately assessing urban T and its associated cooling effects. In this study, we have developed a modified Priestley-Taylor (P-Tmc) model to assess T. This P-Tmc model calculates the vegetation P-T coefficient (alpha v) through a machine learning technique that incorporates soil, meteorological, and vegetation parameters. Moreover, the P-Tmc model integrates CO2 concentration and flux data to enhance the precision of T simulations. Upon validation using data from two eddy correlation (EC) sites and stable hydrogen and oxygen isotope observation sites, it was demonstrated that the P-Tmc model exhibits enhanced simulation capabilities for T when contrasted with the original P-T model. The root-mean-square error (RMSE) for the P-Tmc model was determined to be 0.025 mm/hf in T simulation. The transpiration and cooling effectiveness of 19 representative urban tree species were further assessed using the P-Tmc model framework. Among the selected 19 urban tree species, Ficus virens demonstrated the most impressive cooling performance. For Ficus virens, the average vegetation latent heat flux (lambda T) was measured at 6.18 MJ m2/d, while the temperature reduction (Delta T) reached 4.66 degrees C m2/d. In contrast, Palmae exhibited the least effective cooling, with average lambda T and Delta T values of only 0.89 MJ m2/d and 0.67 degrees C m2/d, respectively. Correlation analysis indicated that the cooling effects of various urban tree species were primarily influenced by tree morphology. Specifically, higher values of leaf area index (LAI) and crown area (C a) were associated with better cooling performance, while increased tree height led to reduced cooling effectiveness. This study conducted a comprehensive evaluation of the cooling effects exhibited by diverse urban tree species through the utilisation of the proposed P-Tmc model. The outcomes of our investigation offer a more robust scientific foundation for urban landscape planning and the selection of tree species.

In recent years, many recalls have been linked to flour and flour-based products. However, many consumers remain unaware of these recalls and continue to perform risky flour-handling behaviors. Food recalls are an essential tool for manufacturers, distributors, and government agencies to inform consumers about foods that may cause health issues, which has the potential to change consumers' food safety behaviors. In this study, researchers constructed model-ensembles to predict and identify the top predicting factors for consumers' flour recall awareness and their safe flour-handling behaviors. Researchers also tested the impact of the volume of flour recalls within a consumer's state of residence on their recall awareness and flour-handling behaviors. Findings indicate that consumers who perceive a higher likelihood of flour recall, aged between 18 and 24, and who pay attention to the lot number, are more likely to be aware of flour recalls. Consumers who perceive the risks of eating raw dough or batter, believe raw chicken poses a microbial risk, and are younger, are more likely to have an increased flour-handling behavior score. However, the volume of recalls in a consumer's state of residence was found to have a low predictive ability for consumers' flour recall awareness and safe flour-handling behaviors. This is the first study utilizing predictive modeling to investigate the critical factors affecting consumers' flour recall awareness and handling behaviors. The findings emphasize the importance of risk perceptions in shaping consumers' behaviors and provide implications for policymakers, food safety experts, and educators in tailoring communication strategies to enhance consumers' risk perceptions and thereby reduce their likelihood of contracting foodborne illnesses due to improper flour-handling behaviors.

Nonlinear Lamb waves are highly sensitive to subtle changes in the mechanical properties of interfaces within layered structures. This study employs the finite element (FE) method to numerically simulate the propagation characteristics of nonlinear Lamb waves in layered structures with weak interfaces. The dispersion relations of Lamb waves are theoretically calculated, and the mode pair ( S 1- s 2) with strict phase-velocity matching is selected for FE simulations. The results show that minor changes in interfacial properties cause the nonlinear acoustic parameter to oscillate sinusoidally with propagation distance, exhibiting a “beat” effect. This effect becomes more pronounced with increasing interfacial degradation. The study provides insights into the second-harmonic generation of Lamb waves and offers a potential method for assessing interfacial degradation in composite plates, which enriches and advances the methodologies of nonlinear ultrasonic testing.

Urban woodland evapotranspiration (ET) is a key variable in the water cycle of the urban Earth Critical Zone (ECZ). However, the specific impacts of urbanization on ET and its individual components, include canopy interception (I), soil evaporation (E) and vegetation transpiration (T), within urban woodlands have remained largely unquantified. This study employed a hybrid modeling framework that integrates three source physical process-based modeling with machine learning to evaluate the long-term dynamics of ET components (interception, evaporation, and transpiration) and their key driving factors across urban forests in 20 global cities. The results demonstrate that the proposed model accurately simulates ET in urban forests, achieving an average root-mean-square-error (RMSE) of 34.8 ​W/m2. Over the nearly four-decade period, total ET and its components exhibited a significant decreasing trend, with average annual declines of 1.46 ​mm for ET, 1.18 ​mm for T, 0.23 ​mm for E, and 0.05 ​mm for I. Shapley Additive Explanations (SHAP)-based attribution analysis indicated that the reduction in urban forest cover driven by urban expansion was the primary factor contributing to the decline in ET and its components (interception, evaporation, and transpiration) in urban forests. While the urban heat island effect increased atmospheric evaporative demand, the biophysical limitations imposed by vegetation loss had a far greater impact. These findings underscore the necessity of expanding urban forested areas to enhance their ET cooling services. This strategy is critical for mitigating the negative effects of climate change on the urban ECZ and improving ecological resilience.

Observational epidemiological studies have demonstrated that maternal exposure to air pollution increases the risk of adverse pregnancy outcomes. However, interactions among multiple environmental exposures remain underexplored. In this study, we performed an epidemiological analysis on 147,979 pregnant women recruited from nine provinces in southeastern China between 2013 and 2023, focusing on the risk of low birth weight (LBW). We found that the critical exposure windows for PM2.5 and ozone (O3) extend from six months prior to conception through the end of second trimester, with hazard ratio of HR = 1.152 (95 % confidence interval [CI]: 1.128-1.177) per 10-μg/m3 incremental PM2.5 exposure and HR = 1.028 (95 % CI: 1.024-1.031) per 10-ppb increase in O3. Our estimates indicate that in 2021, approximately 47,500 (95 % uncertainty interval [UI]: 41,200-53,600) live-born LBW infants nationwide in China could be attributed to ambient air pollution, declining from 79,800 (95 % UI: 71,700-87,900) in 2002. We observed statistically significant synergistic risk effects, neglecting which could lead to an underestimation of 11,600 (95 % UI: 9,300-13,900) LBW cases. Although air pollution-associated LBW burden is decreasing, the rapidly rising LBW prevalence remains a significant public health concern, particularly as China is implementing the "three-child policy". Therefore, our study offers precisely quantified, evidence-based policy guidance for safeguarding reproductive health.

This study delves into the feasibility of leveraging quasi-static component (QSC) generation during primary Lamb wave propagation to discern subtle alterations in the interfacial properties of a two-layered plate. Unlike the second-harmonic generation of Lamb waves, QSC generation doesn’t necessitate precise phase-velocity matching but rather requires an approximate matching of group velocities to ensure the emergence of cumulative growth effects. This unique characteristic empowers the QSC-based nonlinear ultrasonic method to effectively surmount the limitations associated with inherent dispersion and multimode traits of Lamb wave propagation. Modeling the QSC generation reveals that the integrated amplitude of the QSC pulse, derived from the propagation of the primary Lamb wave tone burst, progressively amplifies with increasing propagation distance. Finite element simulations illustrate an overall decline in the efficiency of QSC generation amidst interfacial degradation. Experimental outcomes obtained via a nonlinear ultrasonic measurement-based setup vividly showcase the cumulative growth effect of QSC generation, evident with the propagation distance under approximate group velocity matching. To substantiate the influence of interfacial properties on QSC generation efficiency, varying thermal fatigue durations under cyclic temperature conditions are employed to simulate subtle changes in the two-layered plate’s interfacial properties. The relative nonlinear acoustic parameter exhibits a distinctly sensitive and monotonically decreasing behavior with escalating thermal fatigue durations, corroborating the impact of alterations in interfacial properties on QSC generation efficiency. The alignment between experimental findings and numerical analysis predictions suggests that the effect of QSC generation in primary Lamb wave propagation provides a promising means for sensitively assessing the early-stage degradation of interfacial properties in layered plates.

The zero-group-velocity (ZGV) mode of Lamb waves exhibits unique characteristics, where acoustic energy is trapped within localized regions of the waveguide. Previous research has established that ZGV combined harmonics — generated through the nonlinear interaction of frequency mixing response (FMR) — serve as highly sensitive tools for probing local material nonlinearity. In this study, we present a modeling and numerical analysis of ZGV combined harmonics produced by the mixing of two counter-directional Lamb waves in an adhesively bonded plate, explicitly considering the influence of interfacial properties on FMR efficiency. Based on theoretical analysis, a specific Lamb wave mode triplet is selected to ensure satisfaction of the internal resonance condition. The generation of ZGV combined harmonics at the sum frequency, arising from the interaction of counter-propagating Lamb waves within the plate, is systematically modeled. The results indicate that the efficiency of combined-harmonic generation for sensitive response correlates with the acoustic energy trapping characteristics of ZGV modes. Critically, the spatial location accuracy of the wave mixing phenomenon depends on the central position of the interaction zone rather than its length. Thus, there is no requirement to optimize the mixing zone length for both spatial resolution and signal clarity simultaneously; this inherent balance enhances the applicability of FMR-based nonlinear methods. Finite element (FE) simulations demonstrated that localized interfacial degradation can be detected and characterized by scanning the wave mixing zone of the two primary Lamb waves. The numerical analysis further validated the method’s capability to identify multiple localized degradations with varying severity and length in bonded structures. This work elucidates the physical mechanisms underlying ZGV combined-harmonic generation in adhesively bonded plates and presents a promising approach for non-destructive assessment of interfacial integrity via counter-directional Lamb wave mixing.

Rising vapor pressure deficit (VPD) due to warming has increased global land surface soil evaporation (E), whereas reduced soil moisture (SM) from global drying has suppressed E. However, the relative contributions of these two factors to global E remain poorly understood, creating significant uncertainty regarding its long-term trends. This study constructed an E model that integrated physical processes with machine learning and validated its performance using in-situ E data from 368 global flux sites. The trained hybrid model was then employed to create a dataset of global land surface E for both historical and future periods, enabling the identification of longterm trends and drivers of E across these timeframes. Our findings revealed that the negative impact of SM decline on global E outweighed the positive effects of increased VPD, resulting in a long-term downward trend in global land surface E from 1982 to 2023 (-0.28 +/- 0.07 mm/yearn). Under future climate change scenarios, global land surface E was projected to continue its decline at a faster rate than observed in historical periods (average -0.42 +/- 0.11 mm/yearn under three climate change scenarios from 2024 to 2100), with SM playing a dominant role in this trend. The long-term downward trend was further corroborated by nine additional E datasets. These results underscored the critical role of global drying in driving the persistent decline in global E and highlighted how climate change was exacerbating risks to global water resources.

Understanding plant water use patterns is crucial for comprehending the dynamics of the soil-plant-atmosphere continuum and evaluating the adaptability of plants across diverse ecosystems. However, there remains a gap in our comprehension of non-halophyte plants' water uptake patterns and driving factors in temperate coastal regions. For this reason, we used locust trees (a widely planted non-halophyte tree species in northern China) as a study subject. We collected water isotope data (delta H-2 and delta O-18) for locust trees xylem and soil over two consecutive growing seasons. The MixSIAR model was used along with five distinct sets of input data (single isotopes, uncorrected dual isotopes, and corrected dual isotopes incorporating delta H-2 data obtained by soil water line or cryogenic vacuum distillation methods) to infer water utilization patterns. The results indicated that locust trees primarily absorb shallow soil water (0-20 cm, 29.4% +/- 16.9%) and deep soil water (120-180 cm, 24.7% +/- 5.8%). Pearson's correlation analysis revealed the key driving factors behind water uptake patterns were vegetation transpiration and soil salinity. Remarkably, the build up of salts in the lower soil layer (60-120 cm) hinders the absorption of water by plants. To prevent high salt concentrations from affecting water uptake in non-halophyte plants, we recommend implementing sufficient irrigation from March to April each year to meet the water needs of plant growth and regulate the accumulation of salts in various soil layers. This study reveals the dynamic water utilization strategy of non-halophyte plants in temperate coastal regions, offering valuable information for water resources management.

PurposeIn 2020, an outbreak of Salmonella Stanley linked to imported dried wood ear mushrooms affected 55 individuals in the United States of America. These mushrooms, commonly used in Asian cuisine, require processing, like rehydration and cutting, before serving. The US Centres for Disease Control and Prevention advise food preparers to use boiling water for rehydration to inactivate vegetative bacterial pathogens. Little is known about how food handlers prepare this ethnic ingredient and which handling procedures could enable Salmonella proliferation.Design/methodology/approachThis study used content analysis to investigate handling practices for dried wood ear mushrooms as demonstrated in YouTube recipe videos and to identify food safety implications during handling of the product. A total of 125 Chinese- and English-language YouTube videos were analysed.FindingsMajor steps in handling procedures were identified, including rehydration, cutting/tearing and blanching. Around 62% of the videos failed to specify the water temperature for rehydration. Only three videos specified a water temperature of 100 °C for rehydrating the mushrooms, and 36% of the videos did not specify the soaking duration. Only one video showed handwashing, cleaning and sanitising of surfaces when handling the dried wood ear mushrooms.Practical implicationsThis study found that most YouTube videos provided vague and inconsistent descriptions of the rehydration procedure, including water temperature and soaking duration. Food preparers were advised to use boiling water for rehydration to inactivate vegetative bacterial pathogens. However, boiling water alone is insufficient to inactivate all bacterial spores. Extended periods of soaking and storage could be of concern for spore germination and bacterial growth. More validation studies need to be conducted to provide guidance on how to safely handle the mushrooms.Originality/valueThis study will make a distinctive contribution to the field of food safety by being the first to investigate the handling procedure of a unique ethnic food ingredient, dried wood ear mushrooms, which has been linked to a previous outbreak and multiple recalls in the United States of America. The valuable data collected from this study can help target food handling education as well as influence future microbial validation study design and risk assessment.