On overview of PCA application strategy in processing high dimensionality forensic data

Machine learning applications in forensic DNA profiling: A critical review

Machine learning (ML) is a range of powerful computational algorithms capable of generating predictive models via intelligent autonomous analysis of relatively large and often unstructured data. ML has become an integral part of our daily lives with a plethora of applications, including web, business, automotive industry, clinical diagnostics, scientific research, and more recently, forensic science. In the field of forensic DNA, the manual analysis of complex data can be challenging, time-consuming, and error-prone. The integration of novel ML-based methods may aid in streamlining this process while maintaining the high accuracy and reproducibility required for forensic tools. Due to the relative novelty of such applications, the forensic community is largely unaware of ML capabilities and limitations. Furthermore, computer science and ML professionals are often unfamiliar with the forensic science field and its specific requirements. This manuscript offers a brief introduction to the capabilities of machine learning methods and their applications in the context of forensic DNA analysis and offers a critical review of the current literature in this rapidly developing field.

Synergizing Machine Learning Algorithm with Triboelectric Nanogenerators for Advanced Self-Powered Sensing Systems

The advancement of the Internet of Things (IoT) has increased the demand for large-scale intelligent sensing systems. The periodic replacement of power sources for ubiquitous sensing systems leads to significant resource waste and environmental pollution. Human staffing costs associated with replacement also increase the economic burden. The triboelectric nanogenerators (TENGs) provide both an energy harvesting scheme and the possibility of self-powered sensing. Based on contact electrification from different materials, TENGs provide a rich material selection to collect complex and diverse data. As the data collected by TENGs become increasingly numerous and complex, different approaches to machine learning (ML) and deep learning (DL) algorithms have been proposed to efficiently process output signals. In this paper, the latest advances in ML algorithms assisting solid-solid TENG and liquid-solid TENG sensors are reviewed based on the sample size and complexity of the data. The pros and cons of various algorithms are analyzed and application scenarios of various TENG sensing systems are presented. The prospects of synergizing hardware (TENG sensors) with software (ML algorithms) in a complex environment and their main challenges for future developments are discussed.

Advancing forensic research: An examination of compositional data analysis with an application on petrol fraud detection

In recent years, numerous studies have examined the chemical compounds of petrol and petrol data for forensic research. Standard quantitative methods often assume that the variables or compounds do not have compositional constraints or are not part of a constrained whole, operating within an Euclidean vector space. However, chemical compounds are typically part of a whole, and the appropriate vector space for their analysis is the simplex. Biased and arbitrary results result when statistical analysis are applied on such data without proper pre-processing of such data. Compositional analysis of data has not yet been considered in forensic science. Therefore, we compare classical statistical analysis as applied in forensic research and the new proposed paradigm of compositional data analysis (CoDa). It is demonstrated how such analysis improves the analysis in petrol and forensic science. Our study shows how principal component analysis (PCA) and classification results are affected by the preprocessing steps performed on the raw data. Our results indicate that results from a log ratio analysis provides a better separation between subgroups of the data and leads to an easier interpretation of the results. In addition, with a compositional analysis a higher classification accuracy is obtained. Even a non-linear classification method - in our case a random forest - was shown to perform poorly when applied without using compositional methods. Moreover, normalization of samples due to laboratory/unit-of-measurement effects is no longer necessary, since the composition of an observation is in compositional thinking equivalent to a multiple of it, because the used (log) ratios on raw and log ratio transformed data are equal. Petrol data from different petrol stations in Brazil are used for the demonstration. This data is highly susceptible to counterfeit petrol. Forensic analysis of its chemical elements requires non-biased statistical analysis designed for compositional data to detect fraud. Based on these results, we recommend the use of compositional data methods for gasoline and petrol chemical element analysis and gasoline product characterization, authentication and fraud detection in forensic sciences.

Research progress on occurrence characteristics and source analysis of microfibers in the marine environment

Synthetic microfiber pollution is a growing concern in the marine environment. However, critical issues associated with microfiber origins in marine environments have not been resolved. Herein, the potential sources of marine microfibers are systematically reviewed. The obtained results indicate that surface runoffs are primary contributors that transport land-based microfibers to oceans, and the breakdown of larger fiber plastic waste due to weathering processes is also a notable secondary source of marine microfibers. Additionally, there are three main approaches for marine microplastic source apportionment, namely, anthropogenic source classification, statistical analysis, and numerical simulations based on the Lagrangian particle tracking method. These methods establish the connections between characteristics, transport pathways and sources of microplastics, which provides new insights to further conduct microfiber source apportionment. This study helps to better understand sources analysis and transport pathways of microfibers into oceans and presents a scientific basis to further control microfiber pollution in marine environments.

Assessment of Mechanical Damage and Germinability in Flaxseeds Using Hyperspectral Imaging

The high demand for flax as a nutritious edible oil source combined with increasingly restrictive import regulations for oilseeds mandates the exploration of novel quantity and quality assessment methods. One pervasive issue that compromises the viability of flaxseeds is the mechanical damage to the seeds during harvest and post-harvest handling. Currently, mechanical damage in flax is assessed via visual inspection, a time-consuming, subjective, and insufficiently precise process. This study explores the potential of hyperspectral imaging (HSI) combined with chemometrics as a novel, rapid, and non-destructive method to characterize mechanical damage in flaxseeds and assess how mechanical stresses impact the germination of seeds. Flaxseed samples at three different moisture contents (MCs) (6%, 8%, and 11.5%) were subjected to four levels of mechanical stresses (0 mJ (i.e., control), 2 mJ, 4 mJ, and 6 mJ), followed by germination tests. Herein, we acquired hyperspectral images across visible to near-infrared (Vis-NIR) (450-1100 nm) and short-wave infrared (SWIR) (1000-2500 nm) ranges and used principal component analysis (PCA) for data exploration. Subsequently, mean spectra from the samples were used to develop partial least squares-discriminant analysis (PLS-DA) models utilizing key wavelengths to classify flaxseeds based on the extent of mechanical damage. The models developed using Vis-NIR and SWIR wavelengths demonstrated promising performance, achieving precision and recall rates >85% and overall accuracies of 90.70% and 93.18%, respectively. Partial least squares regression (PLSR) models were developed to predict germinability, resulting in R2-values of 0.78 and 0.82 for Vis-NIR and SWIR ranges, respectively. The study showed that HSI could be a potential alternative to conventional methods for fast, non-destructive, and reliable assessment of mechanical damage in flaxseeds.

Rapid and nondestructive identification of adulterate capsules by NIR spectroscopy combined with chemometrics

This study aims to develop a rapid and non-destructive method to identify counterfeit and substandard drugs, addressing the critical need for better quality control in drug production. According to the reasons for counterfeit products in actual production, the commonly used solid preparation excipients such as HPMC, MCC, Mg-St and Pregelatinized Starch, as well as three chemical drugs with similar efficacy to Guizhi-Fuling (GZFL) Capsule as adulterants, including Aspirin, Ibuprofen and Sinomenine Hydrochloride were selected and designed as adulteration samples with different levels of adulteration. NIR spectra were collected in a non-invasive mode and analyzed by one-class classification methods. The feasibility of using Near-infrared (NIR) spectroscopy as a detection method to qualitatively identify adulterated samples was explored at three packaging levels of powder, intact capsules and capsules in PVC. The differences between the samples were analyzed by NIR spectra comparison, cluster analysis and principal component analysis. The performance of SVM, OCPLS and DD-SIMCA models in dealing with the authentication of genuine and counterfeit products was established and compared. The results show that the spectra contain sample information and the adulterated samples could be discriminated correctly by established models. Moreover, applying appropriate spectral preprocessing methods can further improve the model's performance. In addition, a PLS regression model was developed to predict the adulteration levels of the three packing level samples, which yielded satisfactory results. This study highlights the potential of NIR spectroscopy combined with Chemometrics as a rapid and non-destructive testing analysis method to accurately identify counterfeit and substandard drugs, thereby ensuring drug quality.

A Critical Review of Soil Sampling and Data Analysis Strategies for Source Tracing of Soil in Forensic Investigations

Soil is one type of Earth material demonstrating a wide range of physical, chemical, and biological properties. As the compositional profile of soil is a product of interaction between numerous abiotic and biotic components, it tends to be unique by its geographic origin. Hence, soil is paramount for predicting source or origin in forensic provenance and intelligence, food provenance, biosecurity, and archaeology. In the context of forensic investigation, source tracing of soil could be executed by a comparison or provenance analysis. Soil compositional fingerprints acquired using analytical methods must be carefully interpreted via suitable mathematical and statistical tools since multiple sources can contribute to the variability of soil other than its provenance. This article reviews recent trends in soil sampling and data interpretation strategies proposed for source tracing of soil evidence. Performances of soil provenance indicators are also described. Then, perspectives on possible research directions guiding forensic soil provenance are proposed. This timely critical review reveals the essential idea and gap in forensic soil provenance for stimulating the development of more efficient and effective provenance strategies.

Comparison of decision tree and naïve Bayes algorithms in detecting trace residue of gasoline based on gas chromatography-mass spectrometry data

Fire debris analysis aims to detect and identify any ignitable liquid residues in burnt residues collected at a fire scene. Typically, the burnt residues are analysed using gas chromatography-mass spectrometry (GC-MS) and are manually interpreted. The interpretation process can be laborious due to the complexity and high dimensionality of the GC-MS data. Therefore, this study aims to compare the potential of classification and regression tree (CART) and naïve Bayes (NB) algorithms in analysing the pixel-level GC-MS data of fire debris. The data comprise 14 positive (i.e. fire debris with traces of gasoline) and 24 negative (i.e. fire debris without traces of gasoline) samples. The differences between the positive and negative samples were first inspected based on the mean chromatograms and scores plots of the principal component analysis technique. Then, CART and NB algorithms were independently applied to the GC-MS data. Stratified random resampling was applied to prepare three sets of 200 pairs of training and testing samples (i.e. split ratio of 7:3, 8:2, and 9:1) for estimating the prediction accuracies. Although both the positive and negative samples were hardly differentiated based on the mean chromatograms and scores plots of principal component analysis, the respective NB and CART predictive models produced satisfactory performances with the normalized GC-MS data, i.e. majority achieved prediction accuracy >70%. NB consistently outperformed CART based on the prediction accuracies of testing samples and the corresponding risk of overfitting except when evaluated using only 10% of samples. The accuracy of CART was found to be inversely proportional to the number of testing samples; meanwhile, NB demonstrated rather consistent performances across the three split ratios. In conclusion, NB seems to be much better than CART based on the robustness against the number of testing samples and the consistent lower risk of overfitting.

Detection of apple fruit damages through Raman spectroscopy with cascade forest

Apple fruit damages seriously cause product and economic losses, infringe consumer rights and interests, and have harmful effects on human and livestock health. In this study, Raman spectroscopy (RS) and cascade forest (CForest) were adopted to determine apple fruit damages. First, the RS spectra of healthy, bruised, Rhizopus-infected, and Botrytis-infected apples were measured. Spectral changes and band attribution were analyzed. Different modeling methods were combined with various pre-processing and dimension reduction methods to construct recognition models. Among all models, CForest constructed with full spectra processed by Savitsky-Golay smoothing obtained the best performance with accuracies of 100%, 91.96%, and 92.80% in the training, validation, and test sets (ACC_TE). And the modeling time is reduced to 1/3 of the full-spectra model with a similar ACC_TE of 91.56% after principal component analysis. Overall, RS and CForest provided a non-destructive, rapid, and accurate identification of apple fruit damages and could be used in disease recognition and safety assurance of other fruits.

Antimicrobial Agent against Methicillin-Resistant Staphylococcus aureus Biofilm Monitored Using Raman Spectroscopy

The prevalence of antimicrobial-resistant bacteria has become a major challenge worldwide. Methicillin-resistant Staphylococcus aureus (MRSA)-a leading cause of infections-forms biofilms on polymeric medical devices and implants, increasing their resistance to antibiotics. Antibiotic administration before biofilm formation is crucial. Raman spectroscopy was used to assess MRSA biofilm development on solid culture media from 0 to 48 h. Biofilm formation was monitored by measuring DNA/RNA-associated Raman peaks and protein/lipid-associated peaks. The search for an antimicrobial agent against MRSA biofilm revealed that Eugenol was a promising candidate as it showed significant potential for breaking down biofilm. Eugenol was applied at different times to test the optimal time for inhibiting MRSA biofilms, and the Raman spectrum showed that the first 5 h of biofilm formation was the most antibiotic-sensitive time. This study investigated the performance of Raman spectroscopy coupled with principal component analysis (PCA) to identify planktonic bacteria from biofilm conglomerates. Raman analysis, microscopic observation, and quantification of the biofilm growth curve indicated early adhesion from 5 to 10 h of the incubation time. Therefore, Raman spectroscopy can help in monitoring biofilm formation on a solid culture medium and performing rapid antibiofilm assessments with new antibiotics during the early stages of the procedure.

Spatial heterogeneity and source apportionment of soil metal(loid)s in an abandoned lead/zinc smelter

Metal smelting have brought severe metal(loid)s contamination to the soil. Spatial distribution and pollution source analysis for soil metal(loid)s in an abandoned lead/zinc smelter were studied. The results showed that soil was contaminated heavily with metal(loid)s. The mean of lead (Pb), arsenic (As), cadmium (Cd), mercury (Hg) and antimony (Sb) content in topsoil is 9.7, 8.2, 5.0, 2.3, and 1.2 times higher than the risk screening value for soil contamination of development land of China (GB36600-2018), respectively. Cd is mainly enriched in the 0-6 m depth of site soil while As and Pb mainly deposited in the 0-4 m layer. The spatial distribution of soil metal(loid)s is significantly correlated with the pollution source in the different functional areas of smelter. As, Hg, Sb, Pb and copper (Cu) were mainly distributed in pyrometallurgical area, while Cd, thallium (Tl) and zinc (Zn) was mainly existed in both hydrometallurgical area and raw material storage area. Soil metal(loid)s pollution sources in the abandoned smelter are mainly contributed to the anthropogenic sources, accounting for 84.5%. Specifically, Pb, Tl, As, Hg, Sb and Cu mainly from atmospheric deposition (55.9%), Cd and Zn mainly from surface runoff (28.6%), While nickel (Ni) mainly comes from parent material (15.5%). The results clarified the spatial distribution and their sources in different functional areas of the smelter, providing a new thought for the risk prevention and control of metal(loid)s in polluted site soil.

Flavoromics Approach in Critical Aroma Compounds Exploration of Peach: Correlation to Origin Based on OAV Combined with Chemometrics

It is essential to seek the critical aroma compounds to identify the origins of peach as well as provide a guidance for quality evaluation. In this study, the peach was characterized by HS-SPME/GC-MS. Subsequently, the odor activity value (OAV) was calculated to specify the primary aroma-active compounds. Afterwards, the chemometrics methods were employed to explore the potentially critical aroma on the basis of p value, fold change (FC), S-plot, jack-knifing confidence interval, variable importance for projection (VIP), and the Shared and Unique Structures (SUS) plots. As a result, five compounds (methyl acetate, (E)-hex-2-enal, benzaldehyde, [(Z)-hex-3-enyl] acetate, and 5-ethyloxolan-2-one) were considered as critical aromas. Moreover, the multi-classification model was developed with an outstanding performance (accuracy of 100%) using the five critical aroma. Moreover, the potential chemical basis of odors was sought through sensory evaluation. In addition, this study provides the theoretical and practical foundation for geographical origin traceability and quality evaluation.

Odor emission assessment of different WWTPs with Extended Aeration Activated Sludge and Rotating Biological Contactor technologies in the province of Cordoba (Spain)

In this study, five urban WWTPs (Wastewater Treatment Plant) with different biological treatment (Extended Aeration Activated Sludge - EAAS; Rotating Biological Contactor - RBC), wastewater type (Urban; Industrial) and size, were jointly evaluated. The aim was twofold: (1) to analyze and compare their odor emissions, and (2) to identify the main causes of its generation from the relationships between physico-chemical, respirometric and olfactometric variables. The results showed that facilities with EAAS technology were more efficient than RBC, with elimination yields of organic matter higher than 90%. In olfactometric terms, sludge managements facilities (SMFs) were found to be the critical odor source in all WWTPs compared to the Inlet point (I) or Post primary treatment (PP), and for seasonal periods with ambient temperature higher than 25 °C. Moreover, the global odor emissions quantified in all SMFs revealed that facilities with EAAS (C-WWTP, V-WWTP and Z-WWTP) had a lower odor contribution (19,345, 14,800 and 11,029 ou_E/s·m², respectively) than for those with RBC technology (P-WWTP and NC-WWTP) which accounted for 19,747 ou_E/s·m² and 80,061 ou_E/s·m², respectively. In addition, chemometric analysis helped to find groupings and differences between the WWTPs considering the wastewater (71.27% of total variance explained) and sludge management (64.52% of total variance explained) lines independently. Finally, odor emissions were adequately predicted from the physico-chemical and respirometric variables in the wastewater (r² = 0.8738) and sludge (r² = 0.9373) lines, being pH, volatile acidity and temperature (wastewater line), and pH, moisture, temperature, SOUR (Specific Oxygen Uptake Rate) and OD₂₀ (Cumulative Oxygen Demand at 20 h) (sludge line) the most influential variables.

Research on non-destructive testing of hotpot oil quality by fluorescence hyperspectral technology combined with machine learning

Eating repeatedly used hotpot oil will cause serious harm to human health. In order to realize rapid non-destructive testing of hotpot oil quality, a modeling experiment method of fluorescence hyperspectral technology combined with machine learning algorithm was proposed. Five preprocessing algorithms were used to preprocess the original spectral data, which realized data denoising and reduces the influence of baseline drift and tilt. The feature bands extracted from the spectral data showed that the best feature bands for the two-classification model and the six-classification model were concentrated between 469 and 962 nm and 534-809 nm, respectively. Using the PCA algorithm to visualize the spectral data, the results showed the distribution of the six types of samples intuitively, and indicated that the data could be classified. Based on the modeling analysis of the feature bands, the results showed that the best two-classification models and the best six-classification models were MF-RF-RF and MF-XGBoost-LGB models, respectively, and the classification accuracy reached 100 %. Compared with the traditional model, the error was greatly reduced, and the calculation time was also saved. This study confirmed that fluorescence hyperspectral technology combined with machine learning algorithm could effectively realize the detection of reused hotpot oil.

Coal gangue recognition based on spectral imaging combined with XGBoost

The identification of coal gangue is of great significance for its intelligent separation. To overcome the interference of visible light, we propose coal gangue recognition based on multispectral imaging and Extreme Gradient Boosting (XGBoost). The data acquisition system is built in the laboratory, and 280 groups of spectral data of coal and coal gangue are collected respectively through the imager. The spectral intensities of all channels of each group of spectral data are averaged, and then the dimensionality is reduced by principal component analysis. XGBoost is used to identify coal and coal gangue based on the reduced dimension spectral data. The results show that PCA combined with XGBoost has the relatively best classification performance, and its recognition accuracy of coal and coal gangue is 98.33%. In this paper, the ensemble-learning algorithm XGBoost is combined with spectral imaging technology to realize the rapid and accurate identification of coal and coal gangue, which is of great significance to the intelligent separation of coal gangue and the intelligent construction of coal mines.

Serum Raman spectroscopy combined with multiple classification models for rapid diagnosis of breast cancer

Breast cancer is a malignant tumor with the highest incidence rate in women. Current diagnostic methods are time-consuming, costly, and dependent on physician experience. In this study, we used serum Raman spectroscopy combined with multiple classification algorithms to implement an auxiliary diagnosis method for breast cancer, which will help in the early diagnosis of breast cancer patients. We analyzed the serum Raman spectra of 171 invasive ductal carcinoma (IDC) and 100 healthy volunteers; The analysis showed differences in nucleic acids, carotenoids, amino acids, and lipid concentrations in their blood. These differences provide a theoretical basis for this experiment. First, we used adaptive iteratively reweighted penalized least squares (airPLS) and Savitzky-Golay (SG) for baseline correction and smoothing denoising to remove the effect of noise on the experiment. Then, the Principal component analysis (PCA) algorithm was used to extract features. Finally, we built four classification models: support vector machine (SVM), decision tree (DT), linear discriminant analysis (LDA), and Neural Network Language Model (NNLM). The LDA, SVM, and NNLM achieve 100% accuracy. As supplementary, we added the classification experiment of the raw data. By comparing the experimental results of the two groups, We concluded that the NNLM was the best model. The results show the reliability of the combination of serum Raman spectroscopy and classification models under large sample conditions.

Chemometric techniques in oil spill identification: A case study in Dalian 7.16 oil spill accident of China

Chemometric methods have unique advantages regarding comprehensive consideration of multiple parameters and the classification of samples or variables. Classification of oil spill sources was carried out by using chemometric techniques, such as Repeatability Limit, hierarchical cluster analysis (HCA), Student's t-test and Principal component analysis (PCA) Biplot. In addition, this paper takes the fingerprint identification of a Dalian "7.16″ oil spill accident as an example to illustrate the effectiveness of chemometric techniques in oil identification. PCA scores plot (explaining 82.77% of variance accounted for three PCs) showed that samples belong to four clusters and result of HCA method further confirmed that. The residual oil in Jinshatan Beach and Haibei Square may be caused by the explosion of Dalian "7-16" oil pipeline accident. The use of chemometric techniques is significant in providing independent validation for classifying the types of spilled oil in the investigation of oil spill pollution. The results will be of great significance to improve the accuracy and efficiency of oil spill identification based on oil fingerprint.

Rapid nondestructive detection of peanut varieties and peanut mildew based on hyperspectral imaging and stacked machine learning models

Moldy peanut seeds are damaged by mold, which seriously affects the germination rate of peanut seeds. At the same time, the quality and variety purity of peanut seeds profoundly affect the final yield of peanuts and the economic benefits of farmers. In this study, hyperspectral imaging technology was used to achieve variety classification and mold detection of peanut seeds. In addition, this paper proposed to use median filtering (MF) to preprocess hyperspectral data, use four variable selection methods to obtain characteristic wavelengths, and ensemble learning models (SEL) as a stable classification model. This paper compared the model performance of SEL and extreme gradient boosting algorithm (XGBoost), light gradient boosting algorithm (LightGBM), and type boosting algorithm (CatBoost). The results showed that the MF-LightGBM-SEL model based on hyperspectral data achieves the best performance. Its prediction accuracy on the data training and data testing reach 98.63% and 98.03%, respectively, and the modeling time was only 0.37s, which proved that the potential of the model to be used in practice. The approach of SEL combined with hyperspectral imaging techniques facilitates the development of a real-time detection system. It could perform fast and non-destructive high-precision classification of peanut seed varieties and moldy peanuts, which was of great significance for improving crop yields.

Near-infrared spectroscopy as a feasible method for the differentiation of Neisseria gonorrhoeae from Neisseria commensals and antimicrobial resistant from susceptible gonococcal strains

Rapid and cost-effective diagnosis of Neisseria gonorrhoeae (NG) are important measures for the control and management of gonococcal infection. Current diagnostic tools such as nucleic acid amplification tests and bacterial culture are not feasible in many resource-poor settings, and so syndromic patient management is commonplace. Alternative cost-effective diagnostic tools are therefore needed. Here, we sought to explore the utility and feasibility of Near Infrared Spectroscopy (NIRS) to (1) identify and differentiate NG from Neisseria commensals and (2) to differentiate fully susceptible NG from resistant NG. NIRS correctly classified NG from Neisseria commensals (R²= 0.89; SECV 0.164) and to a lesser capacity, susceptible NG from resistant (R² = 0.60; SECV 0.32). To the best our knowledge, this is the first proof of concept study in the field. Further evaluations are now warranted to enhance capacity and accuracy of this diagnostic approach.

Spatial and temporal differentiation and influencing factors of environmental governance performance in the Yangtze River Delta, China

As the development of Yangtze River Delta urban agglomeration has been upgraded to a national level development strategy in China, relevant regions need to pay more attention to the environmental governance issues. Based on the 2013-2018 development data of 41 cities and the established evaluation indicator system with "press-state-response" (PSR) model, we mainly use a combination of global principal component analysis (GPCA) and entropy method to comprehensively measure regional environmental governance performance (EGP) in the Yangtze River Delta urban agglomeration. Next, the spatial relationship is explored and discussed with spatial autocorrelation analysis. Finally, the panel Tobit model is used to perform a regression analysis on the factors affecting the performance of environmental governance in the Yangtze River Delta region. The results are summarized as follows. (1) From 2013 to 2018, the overall environmental governance performance of the Yangtze River Delta region maintained a steady growth trend, of which the environmental governance performance of Jiangsu Province and Shanghai maintained a steady increase, and the environmental governance performance of Anhui and Zhejiang Province fluctuated greatly. (2) These cities with better EGP are mainly located in the central and southern regions of the Yangtze River Delta, and those cities with poor EGP are mainly located in the northern part of the Yangtze River Delta. The environmental governance performance gap between provinces and cities is obviously large, and there are significant spatial positive correlations, spatial spillover effects and a trend of agglomeration with increased volatility. (3) The regression analysis shows that the economic development level has a significant positive impact on the EGP of the Yangtze River Delta region. Meanwhile, the negative impact of industrial structure and foreign investment is significant, but the positive impact of R&D investment intensity is insignificant, on the performance of environmental governance in the Yangtze River Delta region.