工业工程最新文献

{"title":"[Effects of Chemical Fertilizer and Organic Material Application on Soil Organic Carbon Fractions in Black Soils of Northeast China: A Meta-analysis].","authors":"Ze-Mao Zhang, Lei Wu, Tian-Yu Gao, Tian-Hong Liu, Cong Wang, Ming-Gang Xu, Wen-Ju Zhang","doi":"10.13227/j.hjkx.202407132","DOIUrl":"https://doi.org/10.13227/j.hjkx.202407132","url":null,"abstract":"<p><p>As an important agricultural management practice, fertilization affects the accumulation and stabilization of SOC fractions by influencing the amounts of carbon inputs and outputs. Exploring the effects of different fertilizer types on the SOC content and its main controlling factors could provide a scientific basis for rational fertilization, efficient utilization of organic materials, and SOC content and fertility improvement in the black soil region of Northeast China. This study collected a total of 1 628 observations regarding the effects of chemical fertilizer and organic materials （including organic fertilizer and straw） application on SOC fractions in the black soil region of Northeast China from 228 papers published during 1991 to 2024. A meta-analysis was used to investigate the effects of chemical fertilizer and organic materials application on SOC, dissolved organic carbon （DOC）, particulate organic carbon （POC）, readily oxidizable organic carbon （ROC）, and microbial biomass carbon （MBC） contents. The results showed that： ① The application of chemical fertilizer and organic materials significantly increased the contents of SOC （5%-18%, referring to the range of increase, the same below）, DOC （11%-64%）, POC （30%-141%）, ROC （19%-139%）, and MBC （16%-50%）. The increases in SOC fractions were highest under the manure amendment treatment, with increased SOC content 8% higher than that under straw return and 13% higher than that under the chemical fertilizer treatment. ② The increase in SOC was significantly positively correlated with the fertilizer application duration and the amounts of applied organic fertilizer and significantly negatively correlated with the amounts of straw return, but no correlation was observed with the amounts of applied nitrogen fertilizer. ③ The magnitude of SOC response to fertilization was regulated by annual average temperature and initial soil properties （including pH and SOC）. The fertilization-induced increase in SOC was significantly positively correlated with annual average temperature and initial pH and negatively correlated with initial SOC content. The main factor affecting fertilization-induced SOC changes was initial SOC content under the chemical fertilizer and straw return treatments, while annual average temperature was the key factor under the manure fertilizer treatment. In conclusion, the type and amount of fertilizer, climate conditions, and soil properties should be comprehensively considered to optimize fertilization, so as to increase SOC component fractions as well as improve soil fertility levels in the black soil region of Northeast China.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"4947-4960"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Adsorption Characteristics of Phosphate in an Aqueous Solution by Swine Manure Biochar].","authors":"Qin Li, Cong Zhan, Shuang-Lin Gui, Tong-Hui Deng, Zhuo Jia","doi":"10.13227/j.hjkx.202406197","DOIUrl":"10.13227/j.hjkx.202406197","url":null,"abstract":"<p><p>To solve the problem of phosphate pollution in water and achieve the resource utilization of livestock and poultry manure biomass, swine manure was selected as the raw material to prepare biochar. The composition and structure of SMBC700 were characterized using elemental analysis, specific surface area analysis, and FTIR. The adsorption isotherm model and adsorption kinetics model were used to fit the phosphate adsorption characteristics of swine manure biochar, and the effects of pyrolysis temperature, biochar dosing, initial solution pH, and coexisting ions on phosphate adsorption by biochar were studied. The conditions affecting phosphate adsorption were optimized by response surface methodology （RSM）. The results showed that only SMBC700 could adsorb phosphate. The phosphate adsorption capacity of SMBC700 reached 6.127 1 mg·g^-1 at a SMBC700 dosage of 2 g·L^-1 and the initial phosphate concentration of 20 mg·L^-1. The adsorption isotherm and kinetics were better fitted by the Langmuir isotherm model and the pseudo second-order kinetics. SMBC700 exhibited an excellent performance for phosphate adsorption over a wide pH range （3-12）. The coexistence of HCO₃^- could significantly weaken the adsorption capacity of SMBC700 to phosphate. The optimum conditions for phosphate adsorption by SMBC700 were obtained by RSM analysis with a dosage of 3 g·L^-1, an initial phosphate concentration of 30 mg·L^-1, and a pH of 7.0. The possible mechanisms of phosphate adsorption by SMBC700 included electrostatic adsorption, ligand exchange, and pore filling.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"5379-5390"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Characterization and Risk Assessment of VOCs Emissions from Typical Industries in Jiangxi Province].","authors":"Zhong-Hua Zhu, Qin-Kai Li, Xiang Tu","doi":"10.13227/j.hjkx.202406185","DOIUrl":"https://doi.org/10.13227/j.hjkx.202406185","url":null,"abstract":"<p><p>The study further clarified and compared the source emission characteristics of VOCs from different industries by sampling VOCs from a total of 65 enterprises with organized and unorganized emissions from seven typical industries in Jiangxi Province offline and analyzing their components and species by gas chromatography-mass spectrometry （GC-MS）. Meanwhile, the generation potentials of organized and unorganized emissions from different industries for atmospheric ozone and SOA were estimated based on the concentrations of key species of VOCs, and the potential carcinogenic and non-carcinogenic risks of the VOCs species emitted from each industry were evaluated. The results showed significant differences in VOCs emissions from different industries, with relatively low organized and unorganized VOCs emissions from the automobile manufacturing and printing industries, whereas the emissions from the furniture manufacturing industry were significantly higher. Alkanes and OVOCs accounted for a relatively high percentage of the organized emissions in each industry, while aromatics and OVOCs accounted for a relatively high percentage of the unorganized emissions. In addition, the results of ozone formation potential （OFP） calculations showed that the OFPs of the organic chemical industry, plastic products industry, and furniture manufacturing industry were higher, mainly contributed by OVOCs and aromatic hydrocarbons. The secondary organic aerosol （SOA） generation potential of the furniture manufacturing industry was higher than that of the other industries, with benzene as the main contributing species. In the human health risk assessment, acetaldehyde was found to be a high carcinogenic risk substance in all industries, with its most prominent carcinogenic risk in the plastics industry （risk value as high as 3.05×10^-5） and a significant non-carcinogenic risk for acrolein in the pharmaceutical manufacturing industry （HI value as high as 18.96）. Jiangxi Province should focus on olefins, aromatics, and OVOCs emitted by the organic chemical, pharmaceutical manufacturing, plastic products, and furniture manufacturing industries in terms of synergistic control of PM_2.5 and O₃. Additionally, in terms of reducing the impacts on human health, the control of unorganized OVOCs emissions from the furniture manufacturing, plastic products, and organic chemical industries should be strengthened.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"5035-5044"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Analysis and Prediction of Soil Erosion Under Different Land Use Change Scenarios in Kuye River Basin Based on PLUS-CSLE Model].","authors":"Yu-Chi Chen, Yuan He, Ahmd Ehab Talat, Jian Wang, Ze-Kang Cai","doi":"10.13227/j.hjkx.202406130","DOIUrl":"https://doi.org/10.13227/j.hjkx.202406130","url":null,"abstract":"<p><p>The distribution of land use types is a comprehensive reflection of natural conditions and human activities, which affects the process of runoff and sediment transport by changing the surface morphology, and then affects the process of soil erosion. Based on the land use data in the Kuye River Basin from 2010 to 2020, this study uses the PLUS model to predict the land use distribution in 2025 under three scenarios and evaluates the soil erosion intensity from 2010 to 2025 on CSLE model, so as to explore the impact of land use change on soil erosion. The results showed that： ① From 2010 to 2020, the land use distribution in the Kuye River Basin was dominated by grassland and farmland. The area of farmland and grassland decreased, while the construction land and unused land increased. The expansion of farmland and grassland was mainly affected by elevation, rainfall, and population, and the expansion of construction land was mainly because of human activities. ② The soil erosion intensities in 2010, 2015, and 2020 were 25.91, 22.82, and 27.58 t·（hm²·a）^-1, respectively, showing a trend of first weakening and then rebounding. ③ The soil erosion intensities under the simulated natural development, ecological protection, and economic growth scenarios in 2025 were 25.62, 24.60, and 27.41 t·（hm²·a）^-1. The ecological protection scenario was better for controlling the soil erosion. ④ The areas with the most soil erosion in the basin were mainly the grassland and farmland in the loess gully in the central and southern parts, as well as the degraded land around some industrial/mining concentration parts. To effectively reduce the soil erosion, increasing forest and grassland while reducing the farmland and industrial/mining zones is suggested. Notably, focus should be on bolstering the upkeep and preserving terraced farmlands.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"4971-4981"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Decoupling Effect and Interactive Relationship Among Transportation Infrastructure, Economic Growth, and Carbon Emissions in China].","authors":"Zhi-Guo Shao, Ke-Xin Li, Meng-di Li","doi":"10.13227/j.hjkx.202408256","DOIUrl":"https://doi.org/10.13227/j.hjkx.202408256","url":null,"abstract":"<p><p>The construction of transportation infrastructure boosts economic growth while facing the challenge of carbon emissions pressure. Clarifying the relationship among transportation infrastructure, economic growth, and carbon emissions is important in order to promote the realization of the goal of \"dual-carbon.\" Based on the panel data of 30 provinces in China from 2002 to 2021, the research period was divided into four stages （2002-2006, 2007-2011, 2012-2016, and 2017-2021）. The Tapio decoupling model was used to analyze the decoupling state between carbon emissions and transportation infrastructure, as well as between carbon emissions and economic growth, and the panel vector autoregression （PVAR） model was used to study the dynamic relationship and internal influence mechanism among the three in each region. The results showed that： ① The overall decoupling relationship between carbon emissions and transportation infrastructure in China showed the changing trend of \"weak decoupling → strong negative decoupling → strong decoupling → weak decoupling.\" ② The decoupling relationship between carbon emissions and economic growth in 30 provinces only showed four states in the four stages： strong decoupling, weak decoupling, expansive coupling, and expansive negative decoupling. During the research period, the decoupling index between carbon emissions and economic growth decreased in most provinces of China, and the overall decoupling state improved, but the carbon emissions decoupling situation was unstable. ③ Transportation infrastructure had a positive impact on economic growth in each region, and both transportation infrastructure and economic growth had a positive impact on carbon emissions in each region, but the degree of impact varied by region. The results of the research can provide low-carbon development strategies for the construction of transportation infrastructure and help promote the stable, healthy, and sustainable development of China's economy.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"4813-4825"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Ecological Sensitivity Evaluation of Three-River-Source National Park Based on CRITIC Objective Weighting Method].","authors":"Ying He, Zhong-Qiu Zhao, Zhen-Ran Mei, Hang Bai","doi":"10.13227/j.hjkx.202407221","DOIUrl":"https://doi.org/10.13227/j.hjkx.202407221","url":null,"abstract":"&lt;p&gt;&lt;p&gt;Ecological sensitivity assessment is a probability assessment that intuitively reflects the potential ecological and environmental risks in a region under ecological imbalance. Utilizing ecological sensitivity assessment enables the precise identification of sites and areas that require priority environmental construction and protection, allowing for targeted conservation efforts in these regions. To effectively measure the comprehensive impact of environmental changes and human activities on the ecosystem of the Three-Rivers-Source National Park, as well as to explore the key areas where regional ecological issues occur, the following steps are taken： First, based on the location characteristics and ecological background of the Three-Rivers-Source National Park, a multi-dimensional evaluation index system was quantitatively constructed, including topography, climate, soil, vegetation, species, water resources, and human activities. Second, the CRITIC objective weighting method was used to comprehensively evaluate the ecological sensitivity of the Three-Rivers-Source. The CRITIC objective weighting method fully considers the magnitude of indicator variations and the correlations between indicators. Compared to subjective weight assignment methods, it is better able to reflect the influence of evaluation factors on comprehensive weights, thereby further enhancing the accuracy of the data, which is also an innovative aspect of the study. Finally, the GeoDa software was used to conduct spatial autocorrelation verification analysis on the evaluation factors of ecological sensitivity and to discuss the zoning of protection and management areas. The results showed： ① The buffering distance of wildlife habitats, land use types, and elevation were the three factors with the highest weights in evaluating the ecological sensitivity of the Three-Rivers-Source, with weights of 7.501%, 7.38%, and 7.189%, respectively. ② The Three-Rivers-Source National Park was categorized as having a moderately sensitive ecological status. The comprehensive sensitivity of the study area increased from the northwest to the southeast. The composite sensitivity index ranged between 1.65 and 4.00, with areas of high and very high sensitivity accounting for more than 50% of the total area. These regions were predominantly characterized by the presence of lakes, rivers, and vegetation conservation areas, as well as high-altitude areas, which were frequent sites for ecological issues. ③ The evaluation factors of ecological sensitivity in the Three-Rivers-Source National Park showed significant spatial autocorrelation and high spatial clustering, distributed in a contiguous and cohesive pattern on the spatial level. Hotspots that require priority ecological management were mainly concentrated in the Lancang River and Yellow River source areas. The purpose of the study was to explore the theory and methods of ecological environment protection in the Three-Rivers-Source National","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"5156-5168"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Prediction of China's Carbon Emission Intensity Based on a Grey Breakpoint Model with Inverse Accumulation].","authors":"Hui-Ping Wang, Zhun Zhang","doi":"10.13227/j.hjkx.202407216","DOIUrl":"https://doi.org/10.13227/j.hjkx.202407216","url":null,"abstract":"<p><p>Given the escalating challenges posed by global climate change, as the world's largest carbon emitter, China is facing a huge challenge in achieving its \"dual carbon\" goals. Therefore, reasonable prediction of China's carbon emission intensity is crucial for formulating effective emission reduction strategies. Considering the external shocks faced by the economic system, the time breakpoint is introduced into the traditional grey prediction model. The model is optimized from two aspects： accumulation method and background value, and a new grey breakpoint model with inverse accumulation is constructed. Based on the calculation of China's carbon emissions, the carbon emission intensity from 2023 to 2030 was predicted. The following conclusions were drawn： ① By adding time breakpoints, the new model achieved accurate prediction of the future trend of the system under external shocks, further reflecting the principle of information priority in the modeling process. ② Under the external impact of the COVID-19, the growth rate of China's GDP further slowed down, and the carbon emissions showed different characteristics in the four regions. The carbon emissions in the northeast began to decline gradually, while the carbon emissions in the eastern and western regions accelerated. ③ From 2023 to 2030, China's carbon emission intensity will considerably decrease. Compared with that in 2020, the carbon emission intensity is expected to decrease by 13.2% in 2025 and by 22.6% in 2030, with the highest decline in the northeast and the lowest in the east. However, under current conditions, China still finds it difficult to fully achieve its 2025 and 2030 emission reduction targets, with the eastern and western regions facing enormous pressure to reduce carbon emissions.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"4765-4777"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Source Apportionment of Heavy Metals in Soils Based on Machine Learning Algorithms and Receptor Model].","authors":"Jie Ma, Ming-Sheng Li, Xue Feng","doi":"10.13227/j.hjkx.202408149","DOIUrl":"https://doi.org/10.13227/j.hjkx.202408149","url":null,"abstract":"<p><p>To analyze the source apportionment and influence factors of heavy metals in soils surrounding a coal gangue heap in Chongqing, three machine learning algorithms （decision tree （DT）, random forest （RF）, and support vector machine （SVM）） and the absolute principal component scores-multiple linear regression （APCS-MLR） receptor model were used. The surface soil results showed that the average values of Cd, Hg, As, Pb, Cr, Cu, Ni, and Zn were 0.44, 0.18, 9.92, 32.3, 129, 100, 72.8, and 148 mg·kg^-1. Combined profile soil data showed that Cd, Hg, As, Pb, Cr, Cu, Ni, and Zn were affected by human activities to varying degrees. Using machine learning algorithms analysis, RF was better than DT and SVM, and <i>R</i>² values of Cd, Hg, As, Pb, Cr, Cu, Ni, and Zn were 0.783, 0.728, 0.528, 0.753, 0.753, 0.853, 0.822, and 0.756. \"The number of coal gangue units\" （<i>X</i>₁）, \"the vertical height difference between the sampling point and coal gangue heap\" （<i>X</i>₂）, and \"the distance between the sampling point and the coal gangue heap\" （<i>X</i>₃） were the key driving factors by human activities. Combined with APCS-MLR model analysis, the soil in the study area was affected by natural sources, mining sources, and mixed sources （including atmospheric deposition, agricultural production, life and traffic emissions, etc.）, with contribution rates of 42.5%, 37.1%, and 20.4%, respectively. The combined application of the machine learning algorithms and receptor model can make the results of source apportionment more comprehensive, accurate, and reliable.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"5229-5236"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Spatio-temporal Changes and Driving Factors of Carbon Storage in the Middle Reaches of the Yellow River Based on PLUS-InVEST-GeoDetector Model].","authors":"Fan-Fan Bi, Zhi-Tao Wu, Han-Xue Liang, Zi-Qiang Du, Tian-Jie Lei, Bin Sun","doi":"10.13227/j.hjkx.202406176","DOIUrl":"https://doi.org/10.13227/j.hjkx.202406176","url":null,"abstract":"<p><p>Studying the temporal and spatial variation characteristics and driving factors of carbon reserves in the middle reaches of the Yellow River is crucial for achieving sustainable development and regional ecological conservation against the backdrop of the \"double carbon\" plan. Based on the five-year interval, the land use data of the middle reaches of the Yellow River from 2000 to 2020 were selected, and the spatio-temporal evolution characteristics of carbon reserves were estimated and analyzed by coupling with the PLUS-InVEST-GeoDetector model, and the driving factors affecting the spatio-temporal differentiation of carbon reserves were discussed. Finally, the carbon reserves of the middle reaches of the Yellow River in 2030 were predicted under four developmental scenarios： natural development, ecological protection, economic development, and cultivated land protection. The findings indicate that： ① The middle reaches of the Yellow River's carbon storage showed a consistent growth trend between 2000 and 2020, exhibiting an increase by 5.75×10⁷ t. The evolution of the spatial distribution was reasonably stable, exhibiting the characteristics of \"southeast is higher than northwest.\" ② The middle reaches of the Yellow River's carbon storage differentiated both spatially and temporally between 2000 and 2020, with two-factor enhancement and nonlinear enhancement observed in the interaction detection of each driving element. The main driving force was the NDVI. ③ From 2020 to 2030, the carbon storage of the four scenarios in the Yellow River's middle reaches showed an increasing trend in comparison to that in 2020. Of them, the carbon storage of the ecological preservation scenario rose the highest at 3.93×10⁷ t, while the carbon storage of the economic growth scenario increased the least at 4.8×10⁶ t. The findings of the study will offer some evidence in favor of the middle reaches of the Yellow River's long-term development and ecological environment management.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"4742-4753"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"[Water Quality Analysis and Prediction for the Middle Route of South-to-North Water Diversion Project Based on EDM-LSTM].","authors":"Bing Bai, Fei Dong, Wen-Qi Peng, Xiao-Bo Liu","doi":"10.13227/j.hjkx.202407244","DOIUrl":"https://doi.org/10.13227/j.hjkx.202407244","url":null,"abstract":"<p><p>To deeply analyze the causal relationships among various water quality indicators in the Middle Route of South-to-North Water Diversion Project and achieve high-precision predictions, a method combining empirical dynamic modeling （EDM） and deep learning is proposed. Empirical dynamic modeling is utilized to conduct causal analysis among water quality indicators. Based on this, a dataset is constructed to train long short-term memory （LSTM） neural networks for water quality prediction. The prediction accuracy and computational time of different LSTM structures are compared. The results showed that： ① The water quality of the Middle Route of South-to-North Water Diversion was stable, with no significant abrupt changes along the route. ② There was a bidirectional causal relationship between total nitrogen and dissolved oxygen, as well as pH, in the Middle Route of South-to-North Water Diversion Project. ③ The neural network trained based on causal analysis results could achieve high-precision water quality predictions for the Middle Route of South-to-North Water Diversion Project, with the Nash efficiency coefficient of the predictions generally exceeding 0.85. This method can deeply analyze the causal relationships among variables and achieve high-precision predictions, providing scientific support for water quality management and subsequent analysis and prediction of water ecological factors in the Middle Route of South-to-North Water Diversion Project.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"46 8","pages":"5103-5111"},"PeriodicalIF":0.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}