Huanjing Kexue/Environmental Science最新文献

Accurate prediction of PM₁₀ concentration is important for effectively managing PM₁₀ exposure and mitigating health and economic risks posed to humans in dry bulk ports. However, accurately capturing the time-series nonlinear variation characteristics of PM₁₀ concentration is challenging owing to the specific intensity of port operation activities and the influence of meteorological factors. To address such challenges, a novel combined deep learning model （CLAF） was proposed, merging cascaded convolutional neural networks （CNN）, long short-term memory （LSTM）, and an attention mechanism （AM）. This integrated model aimed to forecast hourly PM₁₀ concentration in dry bulk ports. First, using the random forest characteristic importance algorithm, the distinct meteorological factors were identified among the selected five meteorological factors. These selected factors were incorporated into the prediction model along with the PM₁₀ concentration. Subsequently, the CNN layer was employed to extract high-dimensional time-varying features from the input variables, while the LSTM layer captured sequential features and long-term dependencies. In the AM layer, different weights were assigned to the output components of the LSTM layer to amplify the effects of important information. Finally, three evaluation metrics were applied to compare the performance of the CLAF model with three basic models and three commonly used prediction models. Real-case data was collected and used in this study. Comparison results demonstrated that considering the meteorological factors could improve the prediction accuracy and fitting performance of PM₁₀ concentration in ports. The CLAF model reduced the mean absolute error statistic by 13.92%-56.9%, decreased the mean square error statistic by 45.99%-81.02%, and improved the goodness-of-fit statistic by 3.2%-15.5%.

This study focused on a molybdenum mining area in the Qinling Mountains （Shaanxi segment）. Crop and corresponding soil samples were collected from the vicinity of the mining area, and the concentrations of six heavy metals （Cr, Cu, Zn, As, Cd, and Pb） were determined. Soil heavy metal pollution was assessed using single-factor, comprehensive pollution, and geo-accumulation index methods. The primary sources of soil heavy metals were analyzed using the PMF model. A health risk assessment for soil and crops was conducted using the USEPA model. The results revealed severe pollution of agricultural soils by Cr, Cu, Zn, Cd, and Pb. Among these, Cr may have been primarily sourced from chrombismite nearby mining activities, contributing to 85.1% of the pollution. Cu and As were mainly sourced from agriculture, contributing 50.3% and 70.6%, respectively. Zn and Cd were primarily sourced from natural sources such as metal slag dust and rainwash from the mining area, contributing 73.5% and 48.7%, respectively. Pb was primarily sourced from transportation sources, contributing to 54.7% of the pollution. Crop metal contamination was especially severe for Cr, followed by Pb, whereas As and Cd contamination was relatively lower. Crops were significantly impacted by heavy metal pollution in agricultural soils. The health risk assessment indicated non-carcinogenic and carcinogenic risks for children due to soil heavy metals, whereas adults faced acceptable levels of risk. Both adults and children were exposed to highly significant non-carcinogenic and carcinogenic risks from heavy metals in the crops. Moreover, it is essential to implement effective measures to control heavy metal pollution from tailings to safeguard nearby residents, especially children, from adverse health risks.

Land ecosystems are the largest carbon sink in the world, and land use change is one of the main factors leading to regional carbon emissions. By studying the spatiotemporal evolution characteristics and influencing factors of land use carbon emissions in Gansu Province from 2000 to 2020, this research aimed to provide a scientific basis and reference for promoting low-carbon land use and low-carbon economic development in Gansu Province. Using land use data and the greenhouse gas emission coefficient method, the study analyzed the growth trend of land use carbon emissions at the city-regional scale in Gansu Province, and the spatiotemporal evolution characteristics at the provincial scale, and identified the controlling factors through principal component analysis. The results showed that： ① From 2000 to 2020, land use carbon emissions in Gansu Province showed an overall increasing trend, from 24.289 3 million tons to 57.739 6 million tons. The first stage from 2000 to 2014 was a significant increase period, whereas the second stage from 2014 to 2020 was a stable and slightly decreasing period. Construction land was the main carbon source, and the carbon intensity continued to increase. ② Spatially, there was an "east high, west low" pattern, with carbon emissions in the eastern part of the province significantly higher than those in the western part. ③ Based on emission characteristics, Gansu Province could be divided into five types of carbon emission zones： slow growth, relatively slow growth, moderate growth, relatively fast growth, and rapid growth. ④ The main reasons for the continuous increase in land use carbon emissions in Gansu Province were economic development level, degree of land use, and energy consumption.

To assess the impact of human activities on regional nitrogen （N） flow, based on the statistical data of 27 cities in the Yangtze River Delta Region （YRD）, N flow characteristics of the agricultural production and consumption system （APC） in the YRD from 2011 to 2020 were analyzed using substance flow analysis, and driving factors for N flow were analyzed using scenario analysis. The results showed that from 2011 to 2020, the mean N input intensity of the APC in the YRD was 194.6 kg·（hm²·a）^-1, which was more than five times the national average value； thus, the YRD was a hotspot of N input intensity in China. Chemical N fertilizer was the largest component of N input, and the YRD changed from a net export area of grain and animal products to a net import area due to the rapid growth of food consumption demand. The N output of the system was mainly N loss to the environment, accounting for 53.2% on average. The N use efficiency （NUE） of cropland and the N recycling ratio of the APC ranged from 38.7-42.2% and 15.8-21.5%, respectively, which were both at a low level. In addition, the total amount of N input and output of the APC both showed a parabolic decline trend, decreasing by 11.3% and 10.0%, respectively. Spatially, the overall N input intensity showed a pattern of "high in the north and low in the south," and the spatial heterogeneity of N input intensity among cities was significant. Cities with high input intensity were mainly located in the north and east of Jiangsu, Shanghai, and northeast of Zhejiang. A significant positive spatial autocorrelation of the distribution of mean N input intensity was observed. The uncertainty of N flows was estimated using the error propagation equation. The uncertainty interval of N input and output ranged from 4.5% to 34.6%, which was roughly equivalent to the results of related studies, indicating that the model results were reliable. Based on the scenario analysis method, the decrease of the livestock scale led to a decrease of -0.27%-7.53% in the N input, making it the main reason for the decrease of total N input in the APC. Improving the NUE of cropland and re-establishing the linkage between cropland and livestock will help reduce N loss to the environment.

Northeast China is an important ecological barrier in China, and an in-depth understanding of the spatial distribution in ecosystem services （ESs）, and the driving factors is crucial for realizing the subsequent management and protection of ESs. In the study, we quantitatively assessed the characteristics of spatial distribution in ESs in Northeastern China using the InVEST, RWEQ, and RUSLE models and identified the driving factors of ESs spatial distribution in conjunction with the geodetector based on meteorological data, remote sensing data, and socio-economic data. The results showed that the spatial distribution of ESs in Northeast China had obvious spatial heterogeneity. The high values of habitat quality （HQ）, carbon sequestration （CS） services, and soil conservation （SC） services were mainly distributed in the northern part of the four eastern leagues of the Inner Mongolia Autonomous Region, the northern part of Heilongjiang Province, and the eastern part of Northeast China, which were high in fraction vegetation cover, and low values were mainly found in southwestern and eastern Heilongjiang Province, western Jilin Province, and western Liaoning Province. The high values of the water yield （WY） service and wind prevention and sand fixation （WPSF） service were distributed in the east of the Inner Mongolia Autonomous Region and the east of Liaoning Province. The high values of WY services and WPSF services were distributed in the eastern part of Northeast China and the four eastern provinces of the Inner Mongolia Autonomous Region. According to the geodetector results, slope had the strongest explanatory power for the spatial distribution of SC services with a q-value of 0.31, land use/cover change had the strongest explanatory power for the spatial distribution of HQ and CS services with q-values of 0.64 and 0.52, respectively, and fraction vegetation coverage and annual precipitation had the strongest explanatory power for the spatial distribution of WPSF and WY services with q-values of 0.24 and 0.64, respectively, and there were interactions among all the driving factors. The spatial distribution of ESs in Northeast China was mainly influenced by natural factors. The results will provide a scientific basis for subsequent management and enhancement of ESs in Northeast China.

Aiming at the problem that the single machine learning model has low prediction accuracy of daily average ozone concentration, an ozone concentration prediction method based on the fusion class Stacking algorithm （FSOP） was proposed, which combined the statistical method ordinary least squares （OLS） with machine learning algorithms and improved the prediction accuracy of the ozone concentration prediction model by integrating the advantages of different learners. Based on the principle of the Stacking algorithm, the observation data of the daily maximum 8h ozone average concentration and meteorological reanalysis data in Hangzhou from January 2017 to December 2022 were used. Firstly, the specific ozone concentration prediction models based on the light gradient boosting machine （LightGBM） algorithm, long short-term memory model （LSTM）, and Informer model were established, respectively. Then, the prediction results of the above models were used as meta-features, and the OLS algorithm was used to obtain the prediction expression of ozone concentration to fit the observed ozone concentration. The results showed that the prediction accuracy of the model combined with the class Stacking algorithm was improved, and the fitting effect of ozone concentration was better. Among them, R², RMSE, and MAE were 0.84, 19.65 μg·m^-3, and 15.50 μg·m^-3, respectively, which improved the prediction accuracy by approximately 8% compared with that of the single machine learning model.

Based on the use of the coupling coordination model to calculate the coupling coordination degree of carbon emission and pollutant control, the national, regional, and provincial spatiotemporal characteristics of the synergistic effect of pollution control and carbon emissions reduction in China were further analyzed, facilitating the crucial to identification of key areas. The fixed effects regression models and provincial panel data from 2006 to 2020 were used to explore factors contributing to better synergizing the reduction of pollution and carbon emissions in China. On this basis, the adjustment variable of R&D investment intensity was introduced, and the regulation effect model was constructed to further explore the influence mechanism of the synergistic effect of pollution reduction and carbon reduction. The results showed that： synergy exists between carbon emission reduction and the air pollution control system, the evolution of the synergistic effect of pollution reduction and carbon reduction in China presented an inverted "U"-shaped trend from 2006 to 2020, and there was spatial aggregation and a spatial spillover effect in pollution reduction and carbon reduction. The synergistic governance of carbon emission and pollutant control was still at a relatively low level. The carbon emission and air pollutant emission systems were still in an unstable and uncoordinated state. The results showed that： The degree of coordination of eastern China, central China, and western China decreased in turn. At the national level, energy consumption structure, per capita GDP, and the proportion of green investment were the main factors affecting the synergistic effect of pollution reduction and carbon. The heterogeneity of the influencing factors existed in the central, eastern, and western regions on industrial structure, energy consumption structure, energy utilization efficiency, per capita GDP, urbanization rate, the proportion of green investment, and transportation structure. The intensity of R&D played a significant moderating effect in the whole country, eastern, and central regions. However, no significant moderating effect was identified in the western region. In the eastern region, the urbanization rate, the proportion of green investment, and the transportation structure could not have a significant effect on the synergistic effect of pollution reduction and carbon reduction alone, and it must be coordinated with the intensity of R&D.

In Shijiazhuang City, ozone （O₃） pollution occurs frequently in June every year. In June 2023, the average O₃ 8 h concentration （O₃-8h） pollution exceeded 80% of the days in the month, and O₃ was the primary pollutant, accounting for 100%. For an O₃ heavy pollution process from June 11 to 18, the air quality model WRF-CMAQ was used for simulation, and the average error data MFB and MFE were -10.47% and 17.96%, respectively, which was within the ideal error range. The CMAQ process analysis module was used to simulate the physical and chemical processes in Shijiazhuang City, and the dry deposition （DDEP） contribution concentration was -23.88 μg·m^-3, which was the main process of O₃ consumption, whereas the transport process （TRAN） was the main source of O₃, among which the contribution was more significant in vertical transport （VTRA）. At the same time, the source analysis module （ISAM） was used to analyze the O₃ contribution of local and surrounding areas in Shijiazhuang City. The results showed that the contribution rate of local industry sources in Shijiazhuang City was as follows： traffic source （12.54%） > industrial source （6.94%） > residential source （6.56%） > power source （4.75%）. The long-distance transmission source （BCON） continued to be in the first place with a high contribution rate of 63.31%. In the heavy pollution period under stable weather, the contribution concentration of BCON in the D02 layer of the nested domain to Shijiazhuang City was lower than the sum of the marked area. Among the surrounding cities, Baoding City had the highest contribution rate under stable weather, accounting for 26.21%. In the late period, the contribution concentration of Xingtai City increased rapidly under the action of high-value southwest wind. To effectively reduce O₃ pollution, it is necessary to reduce emissions in the city and to control the upwind cities in advance, and the implementation of inter-regional joint prevention and control is the key.

Exploration of the spatiotemporal changes in fractional vegetation cover （FVC） and its response characteristics to urbanization is of great significance for urban ecological protection and planning in Beijing. This study analyzed the spatiotemporal characteristics of vegetation cover changes in Beijing from 2000 to 2020 using the Theil-Sen Median and Mann-Kendall methods based on a long-term time series vegetation cover dataset. Then, this study used the urbanization index as a key indicator of spatial urbanization and utilized the transect line and global grid analysis methods to investigate the response characteristics of FVC to different urbanization gradients. The results indicated that： ① FVC changes showed spatial and temporal heterogeneity. From 2000 to 2020, Beijing was predominantly covered by high vegetation, accounting for 65.22% of the total area, which was mainly distributed in ecological conservation areas consistent with the Jundu, Xishan, and Yaji Mountain ranges. The FVC presented an overall positive development trend, with a decreasing trend of areas with low FVC. The increase in FVC was significant （by 28.68%）, mainly distributed in ecological conservation areas and within a range of 10-12 km in concentric circles centered around Tiananmen Square. The urbanization index and FVC change rate were relatively high in Haidian District, Chaoyang District, Fengtai District, Shijingshan District, and Changping District. ② The artificial land surface in 2000, 2010, and 2020 was 9.69%, 13.64%, and 21.19%, respectively, with significant spatial agglomeration and strong spatial heterogeneity. During the urbanization process in Beijing, the increase in artificial land surface reached 11.5%, with the conversion from arable land to artificial land surface accounting for 53.83% of the total land use conversion area. ③ There was a significant negative correlation between FVC and the urbanization index, indicating that urbanization had a negative impact on regional FVC. However, as the urbanization process stabilized, this negative correlation tended to gradually weaken. Although the central urban areas were mainly characterized by low FVC, there was a significant increasing trend in the FVC, indicating a positive development in the FVC and an improvement in regional ecological quality, which was closely related to the governance of the mountain-water-forest-field-lake-grass-sand system. The results of the study can provide a basis for the development of vegetation restoration programs and ecological management measures in Beijing.