Search

Global volcanic and plutonic olivines record the compositional characteristics and physicochemical conditions of the parental magmas. Thus, they have significant potential for use as petrogenetic discriminators of the olivine formation environment and prospecting indicators for potential host rocks of magmatic sulfide deposits. Several data visualization approaches have been proposed by researchers to determine olivine origins. However, they can only discriminate specific olivine populations and require the incorporation of trace elements for which data are lacking globally. In this study, a machine-learning method consisting of the random forest algorithm and the synthetic minority oversampling technique (SMOTE) is used to discriminate the crystallization environments of olivine and predict the sulfide potential of olivine-bearing mafic-ultramafic intrusions. We employ a global data set of 24 341 olivine samples from 12 environments to determine the contents of MgO, FeO, Ni, Ca, Mn, and Cr and the Fo number [100 × Mg/(Mg+Fe)]. The results indicate that the proposed method can classify olivine into genetically distinct populations and distinguish olivine derived from mineralized intrusions from that derived from sulfide-barren intrusions with high accuracies (higher than 99% on average). We develop a dimensionality reduction algorithm to visualize the olivine classifications using low-dimensional vectors and an olivine classifier (accessible at http://101.33.204.62:8080/olivine%5fweb/main.html China University of Geosciences, Beijing). The model is used successfully to identify the contributions of distinct sources to regional magmatism using olivines from the late Permian picrite and basalt along the western margin of the Yangtze block (SW China) and to predict the sulfide potential of the newly discovered Qixin mafic-ultramafic complex in the southern Central Asian Orogenic Belt (NW China). The findings suggest that the proposed approach enables the accurate identification of olivine origins in diferent formation environments and is a reliable indicator suitable for global Ni-Cu-platinum group element (PGE) exploration. [ABSTRACT FROM AUTHOR]

Accurate prediction of PM2.5 concentrations can provide a solid foundation for preventing and controlling air pollution. When the Long Short-Term Memory (LSTM) is applied to predict PM2.5 concentration, the influential factors strongly correlated with PM2.5 concentration are directly fed into the LSTM network. However, the influence of these factors on PM2.5 concentration is different. To address this issue, a weighted Random Forest (RF)-LSTM model was proposed to predict PM2.5 concentration for the next six hours in this study. This model first uses the RF to select the factors that are more important for predicting PM2.5 concentration and then uses a fully connected neural network to learn the weight value of each factor. Finally, the weighted data is fed into the LSTM network. The model is trained, validated, and tested using hourly air pollutant and meteorological data collected from four monitoring stations in Beijing, China, from November 1, 2019 to February 28, 2022. The prediction performance of the weighted RF-LSTM model was compared to the RF-LSTM and LSTM models. The results show that the RMSE and MAE of the weighted RF-LSTM model are the smallest, and the R2 is the largest for the next six hours' prediction of PM2.5 concentration at four stations. Compared to the LSTM model, the weighted RF-LSTM model decreases RMSE by 2.3%-5.3%, MAE by 5.6%-9.6%, and improves R2 by 2.0%-4.8%, showing that the weighted RF-LSTM model proposed in this study can achieve better prediction performance and has strong generalization ability. [ABSTRACT FROM AUTHOR]

Chestnut trees hold a prominent position in China as an economically significant forest species, offering both high economic value and ecological advantages. Identifying the distribution of chestnut forests is of paramount importance for enhancing efficient management practices. Presently, many studies are employing remote sensing imaging methods to monitor tree species. However, in comparison to the common classification of land cover types, the accuracy of tree species identification is relatively lower. This study focuses on accurately mapping the distribution of planted chestnut forests in China, particularly in the Huairou and Miyun regions, which are the main producing areas for Yanshan chestnuts in northeastern Beijing. We utilized the Google Earth Engine (GEE) cloud platform and Sentinel-2 satellite imagery to develop a method based on vegetation phenological features. This method involved identifying three distinct phenological periods of chestnut trees: flowering, fruiting, and dormancy, and extracting relevant spectral, vegetation, and terrain features. With these features, we further established and compared three machine learning algorithms for chestnut species identification: random forest (RF), decision tree (DT), and support vector machine (SVM). Our results indicated that the recognition accuracy of these algorithms ranked in descending order as RF > DT > SVM. We found that combining multiple phenological characteristics significantly improved the accuracy of chestnut forest distribution identification. Using the random forest algorithm and Sentinel-2 phenological features, we achieved an impressive overall accuracy (OA) of 98.78%, a Kappa coefficient of 0.9851, and a user's accuracy (UA) and producer's accuracy (PA) of 97.25% and 98.75%, respectively, for chestnut identification. When compared to field surveys and official area statistics, our method exhibited an accuracy rate of 89.59%. The implementation of this method not only offers crucial data support for soil erosion prevention and control studies in Beijing but also serves as a valuable reference for future research endeavors in this field. [ABSTRACT FROM AUTHOR]

Land surface temperature (LST) is one of the most important parameters of the interface between the earth surface and the atmosphere, and it plays a significant role in many research fields, such as agriculture, climate, hydrology, and the environment. However, the thermal infrared band of remote sensors is easily affected by clouds and aerosols, leading to many data gaps in LST products, which restricts the subsequent application of these products. In this paper, Beijing, China, is selected as the study area, and the LST data retrieved from Fengyun 4A (FY-4A) Advanced Geosynchronous Radiation Imager (AGRI) are reconstructed based on the two-point machine learning method. Firstly, the two-point machine learning model is built to reconstruct the theoretical clear-sky LST from simulated and actual images, and the accuracy of the reconstruction results is evaluated compared with the random forest algorithm and the inverse distance weighted method. Secondly, the actual LST under the influence of clouds is reconstructed by using the ERA5 reanalysis LST data as the auxiliary data, and the reconstruction accuracy is then evaluated by the field measurement LST data. The experimental results show that (1) the prediction accuracy of the two-point machine learning method is higher than that of the random forest method in both simulated data and actual data experiments; (2) the R2 of reconstructed LST under theoretical clear-sky conditions is 0.6860 and the root mean square error (RMSE) is 2.9 K, while the R2 of the reconstructed accuracy of actual LST under clouds is 0.7275 and the RMSE is 2.6 K, i.e., the RMSE decreases by 10.34%; (3) the two-point machine method combined with the auxiliary ERA5 LST data can well reconstruct LST under cloudy conditions and present a reasonable LST distribution. [ABSTRACT FROM AUTHOR]

Background: The gut mycobiome of patients with lung adenocarcinoma (LUAD) remains unexplored. This study aimed to characterize the gut mycobiome in patients with LUAD and evaluate the potential of gut fungi as non-invasive biomarkers for early diagnosis. Methods: In total, 299 fecal samples from Beijing, Suzhou, and Hainan were collected prospectively. Using internal transcribed spacer 2 sequencing, we profiled the gut mycobiome. Five supervised machine learning algorithms were trained on fungal signatures to build an optimized prediction model for LUAD in a discovery cohort comprising 105 patients with LUAD and 61 healthy controls (HCs) from Beijing. Validation cohorts from Beijing, Suzhou, and Hainan comprising 44, 17, and 15 patients with LUAD and 26, 19, and 12 HCs, respectively, were used to evaluate efficacy. Results: Fungal biodiversity and richness increased in patients with LUAD. At the phylum level, the abundance of Ascomycota decreased, while that of Basidiomycota increased in patients with LUAD. Candida and Saccharomyces were the dominant genera, with a reduction in Candida and an increase in Saccharomyces, Aspergillus, and Apiotrichum in patients with LUAD. Nineteen operational taxonomic unit markers were selected, and excellent performance in predicting LUAD was achieved (area under the curve (AUC) = 0.9350) using a random forest model with outcomes superior to those of four other algorithms. The AUCs of the Beijing, Suzhou, and Hainan validation cohorts were 0.9538, 0.9628, and 0.8833, respectively. Conclusions: For the first time, the gut fungal profiles of patients with LUAD were shown to represent potential non-invasive biomarkers for early-stage diagnosis. [ABSTRACT FROM AUTHOR]

The escalation of anthropogenic heat emissions poses a significant threat to the urban thermal environment as cities continue to develop. However, the impact of urban spatial form on anthropogenic heat flux (AHF) in different urban functional zones (UFZ) has received limited attention. In this study, we employed the energy inventory method and remotely sensed technology to estimate AHF in Beijing's central area and utilized the random forest algorithm for UFZ classification. Subsequently, linear fitting models were developed to analyze the relationship between AHF and urban spatial form indicators across diverse UFZ. The results show that the overall accuracy of the classification was determined to be 87.2%, with a Kappa coefficient of 0.8377, indicating a high level of agreement with the actual situation. The business/commercial zone exhibited the highest average AHF value of 33.13 W m−2 and the maximum AHF value of 338.07 W m−2 among the six land functional zones, indicating that business and commercial areas are the primary sources of anthropogenic heat emissions. The findings reveal substantial variations in the influence of urban spatial form on AHF across different UFZ. Consequently, distinct spatial form control requirements and tailored design strategies are essential for each UFZ. This research highlights the significance of considering urban spatial form in mitigating anthropogenic heat emissions and emphasizes the need for customized planning and renewal approaches in diverse UFZ. [ABSTRACT FROM AUTHOR]

Based on a study of the spatial distribution of coffee shops in the main urban area of Beijing, the main influencing factors were selected based on the multi-source space data. Subsequently, three regression models were compared, and the best site selection model was found. A comparison was performed between the prediction model functioning with a buffer and without one, and the accuracy of the location model was verified by comparing the actual change trend and the predicted trend in two years. The following conclusions were obtained: (1) coffee shops in the main urban area of Beijing are clustered in an area within 12 km of the main urban center, and also around the core commercial agglomeration area; (2) the random forest (RF) model is the best model in this study, and the accuracy values before and after buffer analysis were 0.915 and 0.929, respectively; and (3) after verifying the accuracy of the model through two years of data, we recommend the establishment of a main road buffer zone for site selection, and the success rate of site selection was found to reach 72.97%. This study provides crucial insight for coffee shop prediction model selection and potential store location selection, which is significant to improving the layout of leisure spaces and promoting economic development. [ABSTRACT FROM AUTHOR]

The damage of expansion joints is a common problem of bridges, which adversely affects not only driving comfort and safety but also the durability of the structure. Jointless bridges can fundamentally solve this fragile problem. However, how to accurately assess its environmental and economic benefits has not been studied, specifically as climate change intensifies. To address these issues, this study first examines the effects of climate change on design expansion allowance of bridge expansion joints based on climate projections. Subsequently, the life-cycle environmental and economic benefits of two jointless bridges located in Beijing–Hong Kong–Macau Expressway are quantified under climate change scenarios. Moreover, uncertainty and random forest algorithm-based importance analyses are carried out to help decision-makers identify major environmental and economic factors and develop eco-friendly plans to facilitate jointless bridge construction. Results indicate the following: (1) the design expansion allowance of bridge expansion joints tends to increase with the time-dependent temperature and humidity; (2) the avoidance of traffic detours caused by the maintenance and replacement of expansion joints is a major source of life-cycle environmental and economic benefits; (3) traffic detour distance is the most important factor influencing the life-cycle environmental and economic benefits. [ABSTRACT FROM AUTHOR]

How to recognize the land use change in urban villages during dynamic transformation in Haidian District, Beijing, has become a hot topic with the promotion of urban renewal. The GF-1 high-resolution remote sensing images of 2013, 2015, and 2020 were used in this study to reflect the land use change in urban villages before and after urban renewal by using a hierarchical machine learning recognition method based on scene-based and random forest classification. The overall scale of urban village blocks in Haidian was 10.46 km2, showing the distribution pattern along the traffic arteries in 2013. In 2015, it dropped to 10.11 km2. The scale of urban village blocks in 2020 decreased to 1.02 km2, 9.75% of that in 2013. Three kinds of urban village renewal logic are revealed by further taking Chuanying Village as an example: "urban village–blue–green space", "urban village–real estate", and "urban village–municipal facilities". [ABSTRACT FROM AUTHOR]

Air pollution is a problem that directly affects human health, the global environment and the climate. The air quality index (AQI) indicates the degree of air pollution and effect on human health; however, when assessing air pollution only based on AQI monitoring data the fact that the same degree of air pollution is more harmful in more densely populated areas is ignored. In the present study, multi-source data were combined to map the distribution of the AQI and population data, and the analyze their pollution population exposure of Beijing in 2018 was analyzed. Machine learning based on the random forest algorithm was adopted to calculate the monthly average AQI of Beijing in 2018. Using Luojia-1 nighttime light remote sensing data, population statistics data, the population of Beijing in 2018 and point of interest data, the distribution of the permanent population in Beijing was estimated with a high precision of 200 m × 200 m. Based on the spatialization results of the AQI and population of Beijing, the air pollution exposure levels in various parts of Beijing were calculated using the population-weighted pollution exposure level (PWEL) formula. The results show that the southern region of Beijing had a more serious level of air pollution, while the northern region was less polluted. At the same time, the population was found to agglomerate mainly in the central city and the peripheric areas thereof. In the present study, the exposure of different districts and towns in Beijing to pollution was analyzed, based on high resolution population spatialization data, it could take the pollution exposure issue down to each individual town. And we found that towns with higher exposure such as Yongshun Town, Shahe Town and Liyuan Town were all found to have a population of over 200 000 which was much higher than the median population of townships of 51 741 in Beijing. Additionally, the change trend of air pollution exposure levels in various regions of Beijing in 2018 was almost the same, with the peak value being in winter and the lowest value being in summer. The exposure intensity in population clusters was relatively high. To reduce the level and intensity of pollution exposure, relevant departments should strengthen the governance of areas with high AQI, and pay particular attention to population clusters. [ABSTRACT FROM AUTHOR]

Commuting flow prediction is a crucial issue for transport optimization and urban planning. However, the two existing types of solutions have inherent flaws. One is traditional models, such as the gravity model and radiation model. These models rely on fixed and simple mathematical formulas derived from physics, and ignore rich geographic semantics, which makes them difficult to model complex human mobility patterns. The other is the machine learning models, most of which simply leverage the features of Origin-Destination (OD), ignoring the topological nature of the interaction network and the spatial correlation brought by the nearby areas. In this paper, we propose a 'preprocessing-encoder-decoder' hybrid learning model, which can make full use of geographic semantic information and spatial neighborhood effects, thereby significantly improving the prediction performance. Specifically, in the preprocessing part, we divide the study area into grids, and then incorporates features such as location, population, and land use types. The second step of the encoder designs a convolutional neural network (CNN) to achieve the fusion of neighborhood features, constructs a spatial interaction network with the grids as nodes and the flows as edges, and then uses the graph convolutional network (GCN) to extract the embeddings of the nodes. In the last step of the decoder, a random forest regressor is trained to predict the commuting flow based on the learned embedding vectors. An empirical study on a commuter dataset in Beijing shows that our proposed model is approximately 20% better than XGBoost (state-of-the-art), thus proving its effectiveness. [ABSTRACT FROM AUTHOR]

Battery swapping stations effectively address the challenges of long charging times, lack of charging stations, and safety hazards for electric two-wheelers. With the rapid development of shared electric bicycles and takeaways, the scale of electric two-wheeler users is expanding while generating a huge demand for battery swapping. The research on the planning and operation of battery swapping stations (BSSs) for electric two-wheelers has yet to be widely discussed. This study developed a data-driven optimization model based on machine learning algorithms using Beijing's battery swapping stations and point of interest (POI) dataset. First, through the spatial features of BSS analyzed by ArcGIS, we found that the coverage of BSSs was mainly concentrated within the fifth ring road, and the utilization rate was unbalanced. Then, on a 3000 m grid scale, a prediction model of BSS quantity with random forest, support vector regression, and gradient-boosting decision tree algorithm was built. The final stacking model was constructed by strengthening three single models with an accuracy of 86.21%. Compared with the original BSSs layout, the machine-learning algorithm proposed in this study can cover more factors and avoid the subjectivity of site selection. Finally, the queuing model for BSSs based on the Monte Carlo simulation was proposed. Through two scenarios, we found that the key parameters m (the number of charging slots) and λ (the user arrival rate) were influential to the outputs of service capability. [ABSTRACT FROM AUTHOR]

Objective: To explore if random forest (RF) model can predict the prognosis of hospital-acquired Klebsiella pneumoniae infection as well as traditional logistic regression(LR) model. Methods: A total of 254 cases of hospital-acquired Klebsiella pneumoniae infection in a tertiary hospital in Beijing from January 2016 to December 2020 were retrospectively collected. Appropriate influencing factors were selected by referring to relevant articles from the aspects of basic clinical information and contact history before infection, and divided into a training set and a test set. Both the RF and LR models were trained by the training set, and using testing set to compare these two models. Results: The prediction accuracy of the LR model was 87.0%, the true positive rate of the LR model was 94.7%; the false negative rate of the LR model was 5.3%; the false positive rate of the LR model was 35%; the true negative rate of the LR model was 65%; the sensitivity of the LR model for the prognosis prediction of hospital-acquired Klebsiella pneumoniae infection was 94.7%; and the specificity was 65%. The prediction accuracy of the RF model was 89.6%; the true positive rate of the RF model was 92.1%; the false negative rate of the RF model was 7.9%; the false positive rate of the RF model was 21.4%; the true negative rate of the RF model was 78.6%; the sensitivity of the RF model for the prognosis prediction of hospital-acquired Klebsiella pneumoniae infection was 92.1%; and the specificity was 78.6%. ROC curve shows that the area under curve(AUC) of the LR model was 0.91, and that of the RF model was 0.95. Conclusion: The RF model has higher specificity, sensitivity, and accuracy for the prognostic prediction of hospital-acquired Klebsiella pneumoniae infection than the LR model and has greater clinical application prospects. [ABSTRACT FROM AUTHOR]

Due to concealed initial symptoms, many diabetic patients are not diagnosed in time, which delays treatment. Machine learning methods have been applied to increase the diagnosis rate, but most of them are black boxes lacking interpretability. Rule extraction is usually used to turn on the black box. As the number of diabetic patients is far less than that of healthy people, the rules obtained by the existing rule extraction methods tend to identify healthy people rather than diabetic patients. To address the problem, a method for extracting reduced rules based on biased random forest and fuzzy support vector machine is proposed. Biased random forest uses the k-nearest neighbor (k-NN) algorithm to identify critical samples and generates more trees that tend to diagnose diabetes based on critical samples to improve the tendency of the generated rules for diabetic patients. In addition, the conditions and rules are reduced based on the error rate and coverage rate to enhance interpretability. Experiments on the Diabetes Medical Examination Data collected by Beijing Hospital (DMED-BH) dataset demonstrate that the proposed approach has outstanding results (MCC = 0.8802) when the rules are similar in number. Moreover, experiments on the Pima Indian Diabetes (PID) and China Health and Nutrition Survey (CHNS) datasets prove the generalization of the proposed method. [ABSTRACT FROM AUTHOR]

Random forests are one of the most popular machine learning methods due to their accuracy and variable importance assessment. However, random forests only provide variable importance in a global sense. There is an increasing need for such assessments at a local level, motivated by applications in personalized medicine, policy-making, and bioinformatics. We propose a new nonparametric estimator that pairs the flexible random forest kernel with local sufficient dimension reduction to adapt to a regression function's local structure. This allows us to estimate a meaningful directional local variable importance measure at each prediction point. We develop a computationally efficient fitting procedure and provide sufficient conditions for the recovery of the splitting directions. We demonstrate significant accuracy gains of our proposed estimator over competing methods on simulated and real regression problems. Finally, we apply the proposed method to seasonal particulate matter concentration data collected in Beijing, China, which yields meaningful local importance measures. The methods presented here are available in the drforest Python package. for this article are available online. [ABSTRACT FROM AUTHOR]

We propose a novel machine learning approach to reconstruct meshless surface wind speed fields, i.e., to reconstruct the surface wind speed at any location, based on meteorological background fields and geographical information. The random forest method is selected to develop the machine learning data reconstruction model (MLDRM-RF) for wind speeds over Beijing from 2015–19. We use temporal, geospatial attribute and meteorological background field features as inputs. The wind speed field can be reconstructed at any station in the region not used in the training process to cross-validate model performance. The evaluation considers the spatial distribution of and seasonal variations in the root mean squared error (RMSE) of the reconstructed wind speed field across Beijing. The average RMSE is 1.09 m s−1, considerably smaller than the result (1.29 m s−1) obtained with inverse distance weighting (IDW) interpolation. Finally, we extract the important feature permutations by the method of mean decrease in impurity (MDI) and discuss the reasonableness of the model prediction results. MLDRM-RF is a reasonable approach with excellent potential for the improved reconstruction of historical surface wind speed fields with arbitrary grid resolutions. Such a model is needed in many wind applications, such as wind energy and aviation safety assessments. [ABSTRACT FROM AUTHOR]

Urban scenes consist of visual and semantic features and exhibit spatial relationships among land-use types (e.g. industrial areas are far away from the residential zones). This study applied a graph convolutional network with neighborhood information (henceforth, named the neighbour supporting graph convolutional neural network), to learn spatial relationships for urban scene classification. Furthermore, a co-occurrence analysis with visual and semantic features proceeded to improve the accuracy of urban scene classification. We tested the proposed method with the fifth ring road of Beijing with an overall classification accuracy of 0.827 and a Kappa coefficient of 0.769. In comparison with other methods, such as support vector machine, random forest, and general graph convolutional network, the case study showed that the proposed method improved about 10% in urban scene classification. [ABSTRACT FROM AUTHOR]

Objectives：Outdoor advertisement can attract the attention of target user groups and increase brand influence, and site selection is the most important influencing factor for its effectiveness. Reasonable site selection plays a positive role in improving brand awareness and expanding the market. However, due to the difficulty in obtaining commercial data, previous studies on a macro and rough scales failed to conduct detailed analysis on the actual effects of site selection. Methods：We propose a framework to solve the above problems. Firstly, we extract feature sets from geographical attributes and commercial economic attributes and preprocess them using mathematical statistics and the geographical processing basic method. Then, by coupling of the in⁃store conversion rate with road network characteristics, point of interest (POI), and other multi⁃source spatial features, a random forest model is constructed to mine the correlation between them. Finally, the importance of each feature is quantified. Results：We choose Beijing as the study area, and build a prediction model for the in⁃store conversion rate of a home decoration brand based on this framework. The model shows a good performance (Standard R²=0.758). Then we obtain the spatial distribution and factors influencing the suitability of outdoor advertising for this home decoration brand in Beijing. The in⁃store conversion rate of the brand is high in the center and is low in the periphery of Beijing, which features the phenomena of strong spatial autocorrelation and high⁃value aggregation. Meanwhile, the in⁃store conversion rate has a significant correlation with social economy, commercial politics, and crowd activities. The location of continuous advertising exposure to the same group has a great influence on the in⁃store conversion rate. Conclusions：The fine mapping results of the model constructed in this study can quantitatively evaluate the advertising effect of each location and maximize the efficiency, which can provide a reference and theoretical basis for relevant studies on the outdoor advertisement or commercial site selection. [ABSTRACT FROM AUTHOR]

Background: Chronic pruritus is a common problem among older adults, with a significant impact on their quality of life. However, it is under‐studied epidemiologically, especially among Chinese older adults. Objective: The aim of this study was to identify the prevalence and explore the risk factors of chronic pruritus among the middle‐aged and older adults in Beijing, China. Methods: A total of 929 participants aged ≥35 years from six communities in Beijing, China, were interviewed. The survey collected the demographic characteristics, the experience of chronic pruritus (>6 weeks), chronic disease history, the level of physical activities and sleep quality. A population‐based case–control study was conducted, including 178 chronic pruritus cases and 697 controls. A multivariate logistic regression model was performed to explore the risk factors of chronic pruritus. Additionally, a random forest algorithm was used to rank the importance of potential risk factors and analyse the overall interpretation of risk factors. Results: The prevalence of chronic pruritus was 19.48% (181/929) among the Beijing middle‐aged and elderly population. The findings indicated that older adults aged 65 years old or above, male, with college or higher degree, alcohol drinking, hypertension, hyperlipidaemia, chronic lung disease, cardiovascular disease, digestive system disease and osteoarthritis/rheumatism, and middle or low sleep quality were associated with the increased risk of chronic pruritus respectively. Physical activity level (≥3000 Met) was associated with a decreased risk of chronic pruritus. The rank according to the most contribution of chronic pruritus risk was sleep quality, education, physical activity level, osteoarthritis/rheumatism, age and gender. Conclusion: Prevalence of chronic pruritus was high among the Chinese middle‐aged and elderly population. Age, gender, high education, alcohol drinking, hypertension, hyperlipidaemia, chronic lung disease, cardiovascular disease, digestive system disease, osteoarthritis /rheumatism and poor sleep quality may serve as risk factors of chronic pruritus. Moderate and high physical activity levels may serve as protective factors of chronic pruritus risk. [ABSTRACT FROM AUTHOR]

Due to concealed initial symptoms, many diabetic patients are not diagnosed in time, which delays treatment. Machine learning methods have been applied to increase the diagnosis rate, but most of them are black boxes lacking interpretability. Rule extraction is usually used to turn on the black box. As the number of diabetic patients is far less than that of healthy people, the rules obtained by the existing rule extraction methods tend to identify healthy people rather than diabetic patients. To address the problem, a method for extracting reduced rules based on biased random forest and fuzzy support vector machine is proposed. Biased random forest uses the k-nearest neighbor (k-NN) algorithm to identify critical samples and generates more trees that tend to diagnose diabetes based on critical samples to improve the tendency of the generated rules for diabetic patients. In addition, the conditions and rules are reduced based on the error rate and coverage rate to enhance interpretability. Experiments on the Diabetes Medical Examination Data collected by Beijing Hospital (DMED-BH) dataset demonstrate that the proposed approach has outstanding results (MCC = 0.8802) when the rules are similar in number. Moreover, experiments on the Pima Indian Diabetes (PID) and China Health and Nutrition Survey (CHNS) datasets prove the generalization of the proposed method. [ABSTRACT FROM AUTHOR]

High-resolution land surface temperature (LST) data are essential for fine-scale urban thermal environment studies. Urban LST downscaling studies mostly remain focused on only two-dimensional (2-D) data, and neglect the impact of three-dimensional (3-D) surface structure on LST. In addition, the choice of window size is also important for LST downscaling over heterogeneous surfaces. In this study, we downscaled Landsat-LST using localized and stepwise approaches in a random forest model (RF). In addition, both 2- and 3-D building morphologies were included. Our results show that: (1) The performances of a local moving window and stepwise downscaling are dependent on the extent of surface heterogeneity. For mixed surfaces, a localized window performed better than the global window, and a stepwise approach performed better than a single-step approach. However, for monotonous surfaces (e.g., urban impervious surfaces), the global window performed better than a localized window; (2) That multi-scale geographically weighted regression (MGWR) could provide a possibility for selection of the optimal moving window. 7 × 7 windows derived from MGWR by the minimum bandwidth of predictors, performed better than other windows (3 × 3, 5 × 5, and 11 × 11) in the Beijing area; (3) That the morphology of buildings has a non-negligible impact and scaling effect on urban LST. When building morphologies were included in downscaling, the performance of the RF model improved. Furthermore, the importance of the sky view factor, building height, and building density was greater at a higher resolution than at a lower resolution. [ABSTRACT FROM AUTHOR]

The cutterhead torque and thrust, reflecting the obstruction degree of the geological environment and the behavior of excavation, are the key operating parameters for the tunneling of tunnel boring machines (TBMs). In this paper, a hybrid hidden Markov model (HMM) combined with ensemble learning is proposed to predict the value intervals of the cutterhead torque and thrust based on the historical tunneling data. First, the target variables are encoded into discrete states by means of HMM. Then, ensemble learning models including AdaBoost, random forest (RF), and extreme random tree (ERT) are employed to predict the discrete states. On this basis, the performances of those models are compared under different forms of the same input parameters. Moreover, to further validate the effectiveness and superiority of the proposed method, two excavation datasets including Beijing and Zhengzhou from the actual project under different geological conditions are utilized for comparison. The results show that the ERT outperforms the other models and the corresponding prediction accuracies are up to 0.93 and 0.99 for the cutterhead torque and thrust, respectively. Therefore, the ERT combined with HMM can be used as a valuable prediction tool for predicting the cutterhead torque and thrust, which is of positive significance to alert the operator to judge whether the excavation is normal and assist the intelligent tunneling. [ABSTRACT FROM AUTHOR]

The formation of ground-level ozone (O 3) is dependent on both atmospheric chemical processes and meteorological factors. In this study, a random forest (RF) model coupled with the reactivity of volatile organic compound (VOC) species was used to investigate the O 3 formation sensitivity in Beijing, China, from 2014 to 2016, and evaluate the relative importance (RI) of chemical and meteorological factors to O 3 formation. The results showed that the O 3 prediction performance using concentrations of measured/initial VOC species (R2=0.82/0.81) was better than that using total VOC (TVOC) concentrations (R2=0.77). Meanwhile, the RIs of initial VOC species correlated well with their O 3 formation potentials (OFPs), which indicate that the model results can be partially explained by the maximum incremental reactivity (MIR) method. O 3 formation presented a negative response to nitrogen oxides (NOx) and relative humidity (RH), and a positive response to temperature (T), solar radiation (SR), and VOCs. The O 3 isopleth calculated by the RF model was generally comparable with those calculated by the box model. O 3 formation shifted from a VOC-limited regime to a transition regime from 2014 to 2016. This study demonstrates that the RF model coupled with the initial concentrations of VOC species could provide an accurate, flexible, and computationally efficient approach for O 3 sensitivity analysis. [ABSTRACT FROM AUTHOR]

Nitrous acid (HONO) is of great interest due to its contribution to hydroxyl (OH) radicals by self-photolysis. Nowadays, machine learning (ML) algorithms are good at capturing complicated non-linear relationships between predictors and dependent variables. Here, using the whole year of 2018 of observed HONO and related pollutant data at an urban site in Beijing, an ML-RF (random forest) model is carried out to predict HONO concentrations and explore the main factors influencing HONO formation mechanisms. ML-RF models show satisfactory performance during the heating, non-heating and whole year periods with R values of 0.95, 0.96 and 0.95, respectively. Primary emissions and diffusion have an obvious influence on ambient HONO during the heating period, while chemical formation processes such as NO 2 heterogeneous reaction and photolysis of nitrate are important for HONO during the non-heating period with higher RH and stronger solar intensity. O 3 and NH 3 are the most important variables for HONO in both periods, indicating the close relationship of HONO with atmospheric oxidation and the important role of NH 3 in HONO formation processes. Although there are deviations due to some variability in HONO formation mechanisms between years, ML-RF models based on 2018 data are able to roughly predict HONO for three periods in 2017 and 2021. Overall, machine learning with limited meteorological and pollutant parameters offers great advantages in HONO prediction, and it can also provide some clues to improve the chemical mechanisms of HONO by finding related variables of ambient HONO. [Display omitted] • Machine learning predicted HONO shows satisfactory performance in 2018. • O 3 and NH 3 are the most important HONO predictors. • Main HONO sources are different between heating and non-heating periods. • Machine learning model is feasible to predict HONO across years. [ABSTRACT FROM AUTHOR]

• Experiments contrasted ten machine learning models in susceptibility prediction. • Hybrid ensemble models significantly outperformed widely-used individual models. • Study profiled the first urban waterlogging susceptibility map in Beijing, China. • Urban road network and densely populated areas face severer waterlogging risk. Urban waterlogging has emerged as a significant problem worldwide, particularly in densely populated cities. Accurate assessment of waterlogging susceptibility at the city scale is crucial for mitigating the risks associated with waterlogging and optimizing municipal design accordingly. However, existing studies on urban waterlogging susceptibility assessment have primarily relied on individual machine learning models. It is worthwhile to explore whether hybrid ensemble models have the potential to enhance the predictive performance. This research presents two hybrid ensemble machine learning models, namely Stacking and Blending, for assessing urban waterlogging susceptibility in the metropolitan area of Beijing, China. The performances of these models are compared with those of the widely used individual machine learning models. Evaluation of all the models is based on metrics such as Accuracy rate and Area Under Curve (AUC) score. The results demonstrate that the Stacking and Blending models consistently outperform the traditional machine learning models, such as Random Forest, Logistic Regression, etc. Through susceptibility analysis and model interpretation with SHAP method, this paper obtains several key findings that low lying areas may not necessarily be areas with severe waterlogging; urban roads and densely populated areas are highly susceptible to becoming high-risk areas for waterlogging in the study area. This study not only highlights the effectiveness of the Stacking and Blending models for urban waterlogging susceptibility assessment but also provides valuable insights for waterlogging mitigation strategies in urban planning. [ABSTRACT FROM AUTHOR]

The formation of ground-level ozone (O3) is dependent on both atmospheric chemical processes and meteorological factors. Traditional models have difficulty assessing O3 formation sensitivity in a timely manner due to the limitations of flexibility and computational efficiency. In this study, a random forest (RF) model coupled with the reactivity of volatile organic compound (VOC) species was used to investigate the O3 formation sensitivity in Beijing from 2014 to 2016, and evaluate the relative importance (RI) of chemical and meteorological factors to O3 formation. The results showed that the O3 prediction performance using initial concentrations of VOC species (R² = 0.87) was better than that using total VOCs (TVOCs) concentrations (R² = 0.77). Meanwhile, the RIs of VOC species correlated well with their O3 formation potentials (OFPs). O3 formation presented a negative response to NOx, PM2.5 and relative humidity, and a positive response to temperature, solar radiation and VOCs. The O3 isopleth curves calculated by the RF model were generally comparable with those calculated by the box model. O3 formation shifted from a VOC-limited regime to a transition regime from 2014 to 2016. This study demonstrates that the RF model coupled with the initial concentrations of VOC species could provide an accurate, flexible, and computationally efficient approach for O3 sensitivity analysis. [ABSTRACT FROM AUTHOR]

Hyperspectral data play an important role in monitoring and protecting wetlands because it not only offers high resolution and wide observation range but also reveal plant information. To explore the hyperspectral characteristics of wetland plants and their applicability in identification, in situ hyperspectral data for 15 species of wetland plants were collected from the Hanshiqiao Wetland Nature Reserve in Beijing, a constructed wetland. Then, first derivative, second derivative, and logarithmic transformation of the reciprocal of the hyperspectral reflectance data were performed using three data conversion methods (principal component analysis, vegetation index, and full-band spectrum) to identify characteristic hyperspectral variables. Four classifiers were compared, namely, support vector machine, decision tree, back-propagation neural network, and random forest. The results demonstrated that: (1) for a comprehensive comparison of the four spectral features data, logarithmic treatment of the reciprocal significantly improved the identification accuracy; (2) for the three data conversion methods for identifying spectral characteristics, the identification accuracy based on the full-band spectrum was better than the other two data forms, and the identification effectiveness of principal component analysis was better than that of the vegetation index; (3) because of differences in plant types and data formats, none of the four classifiers had obvious advantages or disadvantages over the others; and (4) the identification results for all plant statistics further verified the applicability of full-band spectrum data for plant identification. The results showed that the vegetation index and principal component variables reduced the calculation time but did not improve the identification accuracy. Overall, the spectral variables and classifiers examined in this study could meet the needs of wetland plant identification to a certain extent. [ABSTRACT FROM AUTHOR]

In order to improve the accuracy and calculation efficiency of aeroengine rotor vibration reliability analysis, a time-varying rotor vibration reliability analysis method under the aeroengine operating state is proposed. Aiming at the highly nonlinear and strong coupling of factors affecting the reliability of aeroengine rotor vibration, an intelligent neural network modeling framework (short form-INNMF) is proposed. The proposed method is based on DEA, with QAR information as the analysis data, and four factors including engine working state, fuel/oil working state, aircraft flight state, and external conditions are considered to analyse the rotor vibration reliability. INNMF is based on the artificial neural network (ANN) algorithm through improved particle swarm optimization (PSO) algorithm and Bayesian Regularization (BR) optimization. Through the analysis of the rotor vibration reliability of the B737-800 aircraft during a flight mission from Beijing to Urumqi, the time-varying rotor vibration reliability was obtained, which verified the effectiveness and feasibility of the method. The comparison of INNMF, random forest (RF), and ANN shows that INNMF improves analysis accuracy and calculation efficiency. The proposed method and framework can provide useful references for aeroengine rotor vibration analysis, special treatment, maintenance, and design. [ABSTRACT FROM AUTHOR]

Estimating cloud-base height (CBH) from satellite-based passive imaging radiometers is very challenging because the conventional visible and infrared radiance measured by satellite cannot be employed to profile clouds from top to base. The Advanced Himawari Imager (AHI) aboard Himawari-8 is the first of Japanese new-generation visible-infrared imaging spectroradiometer with an improved temporal and spatial resolution. A CBH estimation algorithm for Himawari-8 is imminently needed for aviation safety and climate research. This study presents a machine learning methodology for CBH estimation from Himawari-8 data using a random forest (RF) algorithm, which was conducted for the daytime focusing on single-layer clouds. About 2,876,230 matchups in the period from August 2015 to July 2016 are used in both RF training and validation processes for the purpose of learning the relationship between inputs from Himawari-8 Level-2 cloud products together with geolocation information and target CBH from CloudSat/CALIPSO combined measurements. With the trained RF model, the CBH prediction can be independently generated using Himawari-8 upstream products. Data of ground-based Ka radar in Beijing during 2017 and 2019 are used for independent validation. The results indicate that the Himawari-8 CBH estimation agrees well with those of the Ka radar. The CBH estimations for single-layer nonprecipitation clouds show the best performance with a mean difference of −0.2 ± 1.7 km. The presence of multi-layer clouds mostly leads to overestimated CBHs relative to the radar values, particularly for situations that cirrus clouds overlapping clouds at low-level. Since the ability of the Ka radar is limited to retrieve CBH in precipitation cases, significant overestimation also occurs for precipitation clouds with mean difference of 1.7 ± 2.4 km. [ABSTRACT FROM AUTHOR]

In this paper, the model output machine learning (MOML) method is proposed for simulating weather consultation, which can improve the forecast results of numerical weather prediction (NWP). During weather consultation, the forecasters obtain the final results by combining the observations with the NWP results and giving opinions based on their experience. It is obvious that using a suitable post-processing algorithm for simulating weather consultation is an interesting and important topic. MOML is a post-processing method based on machine learning, which matches NWP forecasts against observations through a regression function. By adopting different feature engineering of datasets and training periods, the observational and model data can be processed into the corresponding training set and test set. The MOML regression function uses an existing machine learning algorithm with the processed dataset to revise the output of NWP models combined with the observations, so as to improve the results of weather forecasts. To test the new approach for grid temperature forecasts, the 2-m surface air temperature in the Beijing area from the ECMWF model is used. MOML with different feature engineering is compared against the ECMWF model and modified model output statistics (MOS) method. MOML shows a better numerical performance than the ECMWF model and MOS, especially for winter. The results of MOML with a linear algorithm, running training period, and dataset using spatial interpolation ideas, are better than others when the forecast time is within a few days. The results of MOML with the Random Forest algorithm, year-round training period, and dataset containing surrounding gridpoint information, are better when the forecast time is longer. [ABSTRACT FROM AUTHOR]

• One-month dockless bikeshare trip data and metro smart card data from the same period in Beijing is used. • The random forest model and accumulated local effects plots were used to examine the nonlinear effects of factors associated with the links between dockless bikeshare and metro trips. • Findings suggest metro ridership and DBS density around service areas are the most crucial factors with certain thresholds and notable joint effects. • The results draw attention to the importance of how bikes are allocated spatially. In this study, we examine the factors associated with the integrated use of dockless bikesharing (DBS) and metros. Special attention is paid to the nonlinear effects of these factors, using machine learning; in this case, a random forest model and accumulated local effects (ALE) plots. We measure the integrated use of DBS and metros based on DBS data and metro smart card data in Beijing. Explanatory variables include metro station characteristics, road infrastructure, public transportation services, land use, and urban density, as well as meteorological variables. We find that metro ridership and DBS density around service areas are the most crucial factors for promoting cycling access/ egress near metro stations. After reaching the thresholds of 6,000 trips for metro ridership and 200 bikes/ per km2 for DBS density, notable joint effects are evident between the two variables, which are negative in egress trips but positive in access trips. Bicycle parking and bus transfer distance and their connectivity to metro stations are important for integrated use. We find that buses will be less competitive when exceeding 25 lines per station. Population density has a nonlinear effect, with a positive association below 9,000 persons/ km2 and a decreasing effect above this level. The land-use mix index was also found to be negatively associated with integrated use, and bicycle lanes are not associated with integrated use. Our results demonstrate that public transit service planning should consider the constraints of resources and public space, and we provide recommendations for governments and stakeholders to reallocate DBS and other public transit services around metro stations. [ABSTRACT FROM AUTHOR]

Combined cooling, heating and power (CCHP) systems have recently received considerable attention, which can meet multiple energy demands of the users. In this study, a novel CCHP system combining the organic flash cycle (OFC) and solar thermal utilization (ST) system is proposed. Two operating strategies based on the following electric load (FEL) are designed for the building complex in Beijing: the FEL strategy with the negative feedback (NFEL) and the FEL strategy considering flexibility (FFEL). The multi-objective optimization of the operating strategy is performed and solved by the genetic algorithm and the random forest algorithm. The results show that the time of the simplified model by the random forest is reduced by 94.5%. The cost of the three strategies decreases and then increases as the power rating of the internal combustion engine (ICE) increases. When the weight factors are inconsistent, the optimal power rating of the ICE increases and the optimal flash temperature decreases for different strategies. The contribution of the heating provision of the NFEL strategy is 8% higher than that of the FEL strategy. The levelized cost of the FFEL strategy is 6.1% and 2.9% lower than the FEL and NFEL strategies. • A novel CCHP system with organic flash cycle and ST system is proposed. • A FEL strategy considering flexibility are designed for the building complex. • A multi-objective optimization is performed using the genetic algorithm. • The random forest algorithm is used to simplify the OFC thermodynamic model. [ABSTRACT FROM AUTHOR]

Traffic noise, characterized by its highly fluctuating nature, is the second biggest environmental problem in the world. Highly dynamic noise maps are indispensable for managing traffic noise pollution, but two key difficulties exist in generating these maps: the lack of large amounts of fine-scale noise monitoring data and the ability to predict noise levels in the absence of noise monitoring data. This study proposed a new noise monitoring method, the Rotating Mobile Monitoring method, that combines the advantages of stationary and mobile monitoring methods and expands the spatial extent and temporal resolution of noise data. A monitoring campaign was conducted in the Haidian District of Beijing, covering 54.79 km of roads and a total area of 22.15 km2, and gathered 18,213 A-weighted equivalent noise (LAeq) measurements at 1-s intervals from 152 stationary sampling sites. Additionally, street view images, meteorological data and built environment data were collected from all roads and stationary sites. Using computer vision and GIS analysis tools, 49 predictor variables were measured in four categories, including microscopic traffic composition, street form, land use and meteorology. Six machine learning models and linear regression models were trained to predict LAeq, with random forest performing the best (R2 = 0.72, RMSE = 3.28 dB), followed by K-nearest neighbors regression (R2 = 0.66, RMSE = 3.43 dB). The optimal random forest model identified distance to the major road, tree view index, and the maximum field of view index of cars in the last 3 s as the top three contributors. Finally, the model was applied to generate a 9-day traffic noise map of the study area at both the point and street levels. The study is easily replicable and can be extended to a larger spatial scale to obtain highly dynamic noise maps. • A new noise monitoring method, Rotating Mobile Monitoring method, was proposed in this study. • The random forest model performed best (R2 = 0.72, RMSE = 3.28 dB) among the noise prediction models. • A highly dynamic noise map of 1s LAeq was constructed. • Traffic noise can be accurately predicted from photographs of urban roads and knowledge of where the photographs were taken. [ABSTRACT FROM AUTHOR]

Atmospheric aerosols have significant impacts on climate and environment, so the study on aerosol optical properties is of great scientific significance for climate change and atmospheric environment. The variations in column-integrated aerosol optical properties over Beijing were investigated from March 2016 to December 2019 based on the SKYNET data products. Higher values of AOD and AE found in summer indicate that it is important to pay more attention to fine particle pollution in summer. By fitting the AOD data of SKYNET with AERONET at same waveband, we found that monthly variations in aerosol optical depth estimated from SKYNET observation were mostly consistent with the AERONET data. Aerosols were divided into several modes according to optical characteristics, which deepen some insight into aerosol optical and radiative properties and revealed that the weak absorption aerosol dominated and contributed much to the radiation transport in Beijing. Data derived from the polarization lidar CALIPSO were combined with aerosol optical depth (AOD) from SKYNET and these results were consistent, which further identified the scientific of data. The random forest algorithm was used to interpolate the missing AOD 500nm observations. The original results and the interpolation results were fitted with MODIS respectively, which proved that the random forest algorithm can supplement the missing results well. This work will be of great significance in addressing the impacts of aerosols on climate and environment, and will help reduce the uncertainty of the impacts of aerosols on the Earth's radiation budget. All work will provide scientific support for the prevention and control of specific pollution. • Four years aerosol optical characteristics were analyzed based on Beijing SKYNET station. • Lidar and SKYNET data were integrated to distinguish aerosol types. • The random forest algorithm was used to interpolate missing values. [ABSTRACT FROM AUTHOR]

• We identified 13 target and 20 nontarget PFAS in the Chaobai river. • A random forest regression model was developed to quantify nontarget PFAS. • The prediction errors of random forest regression model were ≤ 2.7 times. • The developed risk-based prioritization approach flagged 4 high risk PFAS. Though thousands of per- and polyfluoroalkyl substances (PFAS) have been on the global market, most research focused on only a small fraction, potentially resulting in underestimated environmental risks. Here, we used complementary target, suspect, and nontarget screening for quantifying and identifying the target and nontarget PFAS, respectively, and developed a risk model considering their specific properties to prioritize the PFAS in surface waters. Thirty-three PFAS were identified in surface water in the Chaobai river, Beijing. The suspect and nontarget screening by Orbitrap displayed a sensitivity of > 77%, indicating its good performance in identifying the PFAS in samples. We used triple quadrupole (QqQ) under multiple-reaction monitoring for quantifying PFAS with authentic standards due to its potentially high sensitivity. To quantify the nontarget PFAS without authentic standards, we trained a random forest regression model which presented the differences up to only 2.7 times between measured and predicted response factors (RFs). The maximum/minimum RF in each PFAS class was as high as 1.2–10.0 in Orbitrap and 1.7−22.3 in QqQ. A risk-based prioritization approach was developed to rank the identified PFAS, and four PFAS (i.e., perfluorooctanoic acid, hydrogenated perfluorohexanoic acid, bistriflimide, 6:2 fluorotelomer carboxylic acid) were flagged with high priority (risk index > 0.1) for remediation and management. Our study highlighted the importance of a quantification strategy during environmental scrutiny of PFAS, especially for nontarget PFAS without standards. [Display omitted] [ABSTRACT FROM AUTHOR]

Rare earth elements (REEs) have been increasingly diffused to the environment due to their extensive use and application in industries, agriculture, and high-tech devices, which have been regarded as emerge pollutants. However, the study concerning REEs in urban soils is still limited. Therefore, the objectives of this study were to investigate the potential source and risk of REEs in urban environment. We analyzed the concentration and distribution of REEs in urban park soils, and performed a combination of micro geochemical method and random forest method to characterize the pollution sources of REEs. The results showed that the ΣREE concentrations in Beijing urban park soils ranged from 117.19 to 198.09 mg/kg. Spatial distribution indicated that the high concentrations of REEs were mainly concentrated in the west of Beijing near an industrial area. The geochemical parameters, micro spherules and random forest results confirmed the anthropogenic pollution sources from industry and traffic. Risk assessment showed that the average daily doses of total REEs for children and adults were far below the reference threshold with values of 0.08 and 0.02 µg/kg/day, respectively. Our study has exhibited that though the reconstruction of parks from abandoned industrial sites showed an accumulation of REEs, the health risk of REEs for human beings are negligible. [Display omitted] • The REEs in park soils were accumulated near the industrial area of Beijing city. • Obvious fractionation of LREEs enrichment were observed. • SEM-EDS was used in the identification of potential sources of REEs. • Estimated average daily ingestion dose of REEs exposure were considerably low. [ABSTRACT FROM AUTHOR]

The phenomenon of fly-tipping of bulky waste is becoming increasingly serious with the development of economy and the improvement of living standards. The fly-tipping of bulky waste causes a waste stream with a high proportion of good quality recyclable materials and a wide range of social and environmental problems including damaging the environment and aggravating traffic in a megacity like Beijing. In this study, we analyzed the quantitative spatial distribution of fly-tipping bulky waste in Beijing for the first time and identified its driving forces to explore the fly-tipping rule of bulky waste. We used Anselin's Local Moran I method to reveal the spatial characteristics, and Random Forest machine learning method to examine the driving factors based on multiple data sources such as geographical, population and survey data. The results showed that the spatial agglomeration of fly-tipping presented a typical core edge diffusion spatial distribution pattern. High-High clusters of cases were found in most regions of Dongcheng District and the east part of Xicheng District. In the multifactorial drivers, the three most important driving factors were found to be accommodation, floating population and income level. Surprisingly, any kind of road and educational level had small effect on the fly-tipping case. These results provide a theoretical basis for the government to advance comprehensive prevention and control strategies of bulky waste fly-tipping. And it is helpful to formulate management policies of fly-tipping bulky waste and provide targeted guidance and suggestions for senior decision makers and managers. • Random forest machine learning method was used to analyze bulky waste fly-tipping. • Fly-tipping presented a typical core edge diffusion spatial distribution pattern. • 15 driving factors were extracted from multiple data sources. • Accommodation and floating population had the highest impact on fly-tipping. [ABSTRACT FROM AUTHOR]

The Chinese Spring Festival (CSF) is the most solemn traditional festival in China, and the substantial changes in anthropogenic activities in megacities provide a unique natural experiment to assess the influence of short-term emission changes on air quality. Here we applied a machine learning based random forest algorithm to six-year aerosol composition measurements in urban Beijing during the CSFs of 2012–2020 to quantify the relative contributions of meteorology and emission changes to air quality. Our results demonstrate large variabilities of air pollutants during the CSF due to the meteorological changes and holiday effect. By removing the meteorological effect, we found that the reduced emissions during CSF caused an average decrease of 5.1% for non-refractory PM 2.5 with chloride and primary organic aerosol being the largest (8.8–18.7%) while the changes in secondary species were small. The COVID-19 lockdown during 2020 led to additional reductions of primary species by 16.3–36.8%, yet increases in nitrate and secondary organic aerosol due to enhanced secondary production. Our study has a significant implication that reducing local traffic and cooking emissions is far from enough for mitigating air pollution in winter in megacities due to the nonlinear effect of secondary production and regional transport. A synergetic control of multiple precursors, e.g., NO x and ammonia, is of great importance to reduce secondary aerosol and improve air quality. • Large variabilities of air pollutants during CSF due to the meteorological changes and holiday effect. • The meteorologically normalized changes in secondary species were small. • The COVID-19 lockdown led to additional reductions of primary species by 16.3–36.8%. [ABSTRACT FROM AUTHOR]

China has implemented two national clean air actions in 2013–2017 and 2018–2020, respectively, with the aim of reducing primary emissions and hence improving air quality at a national level. It is important to examine the effectiveness of such emission reductions and assess the resulting changes in air quality. However, such evaluation is difficult as meteorological factors can amplify, or obscure the changes of air pollutants, in addition to the emission reduction. In this study, we applied the random forest machine learning technique to decouple meteorological influences from emissions changes, and examined the deweathered trends of air pollutants in 12 Chinese mega-cities during 2013–2020. The observed concentrations of all criteria pollutants except O 3 showed significant declines from 2013 to 2020, with PM 2.5 annual decline rates of 6–9% in most cities. In contrast, O 3 concentrations increased with annual growth rates of 1–9%. Compared with the observed results, all the pollutants showed smoothed but similar variation in trend and annual rate-of-change after weather normalization. The response of O 3 to NO 2 concentrations indicated significant regional differences in photochemical regimes, and the differences between observed and deweathered results provided implications for volatile organic compound emission reductions in O 3 pollution mitigation. We further evaluated the effectiveness of first and second clean air actions by removing the meteorological influence. We found that the meteorology can make negative or positive contribution in reducing pollutant concentrations from emission reduction, depending on type of pollutants, locations, and time period. Among the 12 mega-cities, only Beijing showed a positive meteorological contribution in amplifying reductions in main pollutants except O 3 during both clean air action periods. Considering the large and variable impact of meteorological effects in changing air quality, we suggest that similar deweathered analysis is needed as a routine policy evaluation tool on a regional basis. [Display omitted] • We used a random forest method to remove meteorological effect for air quality data. • The deweathered results show smoothed but similar variation in annual trends. • The meteorology can play positive or negative role depending on specific conditions. • Beijing shows positive meteorology in reducing pollutants for two clean air actions. [ABSTRACT FROM AUTHOR]

• Develop an agent-based model of electric vehicle (EV) and vehicle-to-grid (V2G) adoption. • The EV-V2G model can explicitly simulate social influences on the EV and V2G adoption. • Only 3.3% of survey participants were familiar with V2G. • V2G selling price and advertisement were not very influential to the diffusion of V2G. The potential widespread adoption of Electric Vehicles (EVs) has received considerable attention across the globe. However, as a promising technology for both EVs and smart grid, Vehicle-to-Grid (V2G) tended to receive much less attention. This paper developed an agent-based joint EV and V2G model to simultaneously simulate how EVs and V2G might diffuse across space and over time, with empirical findings from a questionnaire survey in Beijing. In particular, random forest models were developed with the survey data to generate each agent's preferences and attitudes towards EVs and V2G. The joint model also considered three typical levels of social influence, i.e., global influence, neighbor effect, and friendship effect, in the diffusion of EVs and V2G. Finally, the joint model was tested through several "what-if" scenarios, considering different V2G prices, EV/V2G advertisement intensities, and vehicle purchase restrictions. The survey results suggested that 67.7% of the respondents were familiar with EVs, but only 3.3% of them were familiar with V2G. However, over 70% of them would/might try V2G given that they had an EV. The model results suggested that the number of CV applicants was 6.19 times that of BEV applicants in 2030 in the baseline scenario, and only 27.8% of BEV users adopted V2G. Furthermore, V2G selling price, EV/V2G advertisement, and dedicated PHEV purchase permits were not very influential to the diffusion of V2G. The outcomes would be helpful for EV- and V2G-related stakeholders in policy making and technology investment. [ABSTRACT FROM AUTHOR]

Land surface temperature (LST), retrieved from thermal infrared (TIR) bands of remote sensing satellites, is an important parameter for various climate and environmental models. TIR bands detect a range of low-energy wavelengths, resulting in a coarser spatial resolution than other multispectral bands, and limiting applicability in heterogeneous urban regions. In this study, a new nonlinear method for LST downscaling, called Three Layers Composition (TLC), was proposed. The TLC integrates large-scale temperature variations, re-constructed detailed characteristics of LSTs, and strong boundary information. The performance of TLC is compared with disaggregation of radiometric surface temperature (DisTrad), thermal imagery sharpening (TsHARP), and random forest (RF) for a complex landscape in Beijing city, which has agriculture, forest, and massive impervious surfaces. The scale effects on the downscaled LSTs (DLST) were analyzed from the aspects of spatial resolution and spatial contexts. The experimental results indicate that the nonlinear algorithms (TLC and RF) perform better than linear methods (DisTrad and TsHARP). Indicated by coefficient of determination (R 2 ), centered root-mean-square error (CRMSE), and correlation coefficient (CC), TLC (R 2 = 0.901, CRMSE = 0.319, CC = 0.951) was the most effective and workable technique for predicting LSTs, followed by RF (0.768, 0.502, 0.874), TsHARP (0.544, 0.652, 0.734), and DisTrad (0.518, 0.751, 0.719). Larger experimental regions and larger ratios between initial and target resolution weaken the accuracy of DLST. TLC indicated a stronger ability to resist the influence of such scale effects. Traditional downscaling methods (DisTrad, TsHARP, and RF) are trained with global LST-predictor relationships and predict the DLST point by point, which can result in significantly biased estimates for very high or very low temperatures. Addressing this issue, TLC advantageously preserves the texture similarity between LST and its predictors, and yields more precise DLST, which showed higher consistency with the reference LST. Considering high accuracy and low computation time, TLC may be a safe technique for LST downscaling in other regions and different remote sensing sensors. • New nonlinear downscaling LST method was proposed based on image processing. • Larger region and larger ratios to target resolution cause lower accuracy of DLST. • TLC outperforms traditional methods for DLST with improved error indicators. • TLC had a stronger ability to resist the influence of scale effect on DLST. [ABSTRACT FROM AUTHOR]

An accurate regional estimate of soil moisture content (SMC) is important for water management and drought monitoring. Traditional ground measurement methods of SMC are limited by the disadvantages of high cost and small scale. The development of remote sensing (RS) technology provides a cost-effective tool for estimating SMC at regional scale. However, the estimation of SMC by using the combination of multiple sensors has yet to be thoroughly discussed. Furthermore, the way in which vegetation types, fraction of vegetation coverage (FVC) and soil layer depth affect the SMC-estimation performance remains unclear. Therefore, the objectives in this study are to (1) evaluate the SMC-estimation performance provided by Landsat-8 data and random forest regression (RFR) algorithm; (2) discuss the accuracy of RS-based method for SMC estimation at different soil layer depths, and (3) explore how vegetation types and FVC affect the performance of SMC-estimation. The results can be summarized as: (1) the SMC estimation performance of multispectral (MS)- and thermal infrared (TIR)-based indices single used were comparable, in which TIR-based indices performed better in shallow soil layer while MS-based indices performed better in deep soil layer. Generally, MS- and TIR-based indices jointly used outperformed the index single used. (2) the accuracy of the proposed method for estimating SMC decreased with soil depth. (3) the proposed method performed greatest in grassland with relatively low height among the three vegetation types. Moreover, the SMC estimation in moderate vegetation coverage (FVC ranged from 0.3 to 0.5) was best. These results indicate that RS-based multimodal data combined with RFR could provide relatively repeatable and accurate SMC estimation. This approach can thus be used for the regional SMC monitoring and water resources management. [Display omitted] • A low-cost remote sensing method for regional-scale soil moisture monitoring was developed. • Multimodal data fusion showed effectiveness in field soil moisture estimation. • Random forest regression provided promising results in soil moisture estimation across soil depth and vegetation types. • Modest canopy coverage is more conducive to soil moisture monitoring. [ABSTRACT FROM AUTHOR]

Urbanization processes greatly change urban landscape patterns and the urban thermal environment. Significant multi-scale correlation exists between the land surface temperature (LST) and landscape pattern. Compared with traditional linear regression methods, the regression model based on random forest has the advantages of higher accuracy and better learning ability, and can remove the linear correlation between regression features. Taking Beijing's metropolitan area as an example, this paper conducted multi-scale relationship analysis between 3D landscape patterns and LST using Pearson Correlation Coefficient (PCC), Multiple Linear Regression and Random Forest Regression (RFR). The results indicated that LST was relatively high in the central area of Beijing, and decreased from the center to the surrounding areas. The interpretation effect of 3D landscape metrics on LST was more obvious than that of the 2D landscape metrics, and 3D landscape diversity and evenness played more important roles than the other metrics in the change of LST. The multi-scale relationship between LST and the landscape pattern was discovered in the fourth ring road of Beijing, the effect of the extent of change on the landscape pattern is greater than that of the grain size change, and the interpretation effect and correlation of landscape metrics on LST increase with the increase in the rectangle size. Impervious surfaces significantly increased the LST, while the impervious surfaces located at low building areas were more likely to increase LST than those located at tall building areas. It seems that increasing the distance between buildings to improve the rate of energy exchange between urban and rural areas can effectively decrease LST. Vegetation and water can effectively reduce LST, but large, clustered and irregularly shaped patches have a better effect on land surface cooling than small and discrete patches. The Coefficients of Rectangle Variation (CORV) power function fitting results of landscape metrics showed that the optimal rectangle size for studying the relationship between the 3D landscape pattern and LST is about 700 m. Our study is useful for future urban planning and provides references to mitigate the daytime urban heat island (UHI) effect. [ABSTRACT FROM AUTHOR]

Accurate forecasting of future meteorological elements is critical and has profoundly affected human life in many aspects from rainstorm warning to flight safety. The conventional numerical weather prediction (NWP) sometimes leads to unsatisfactory performance due to inappropriate initial state settings. In this paper, a short-term weather forecasting model based on wavelet packet denoising and Catboost is proposed, which takes advantage of the fusion information combining the historical observation data with the prior knowledge from NWP. The feature selection and spatiotemporal feather addition are also explored to further improve performance. The proposed method is evaluated on the datasets provided by Beijing weather stations. Experimental results demonstrate that compared with many deep-learning or machine-learning methods such as LSTM, Seq2Seq, and random forest, the proposed Catboost model incorporated with wavelet packet denoising can achieve shorter convergence time and higher prediction accuracy. [ABSTRACT FROM AUTHOR]

Harmful algal bloom is prevalent in the reclaimed-water-source (RWS) river caused by the excessive nutrient's inputs. Rainfall water may be the sole nutrient-diluted water source for the RWS river. However, the effects of storm events on the algal bloom in the RWS river are poorly understood. This study presents chlorophyll-a (Chl-a) variations before, during, and after the initial storm events (Pre-storm, In-storm, and Post-storm) at four representative sites with distinct hydraulic conditions in a dam-regulated RWS river system, Beijing. The response of Chl-a to the initial storm events mostly depends on the ecosystem status that caused by the river hydraulic properties. The upstream is more river-like and downstream is more lake-like. In the river-like system, elevated water temperature (WT, increased by 2 %) could support the dominating algae (diatom) growth (Chl-a increased by 130 %) from Pre-storm to In-storm period. In the lake-like system, the dominant algae (blue algae) declined (Chl-a sharply decreased by 96%–99 %) due to the lower WT (decreased by 3%–7%) and increased flow velocities from Pre-storm to In-storm period. During the Post-storm period, the dominant algae break out (Chl-a surged by 20%–319 %) in the lake-like system caused by the recovered WT (increased by 3%–6%) and flow velocity. The occurrence of algal bloom can be predicted by the Random Forest (RF) model based on water quality parameters such as total nitrogen (TN). The thresholds of algal bloom for the Pre-storm, In-storm, and Post-storm periods were identified as 30 μg/L, 10 μg/L, and 10 μg/L, respectively. The two driven factors were WT and nitrate (NO 3 -N) for the Pre-storm period and were WT and TN for the In- & Post-storm periods. A higher risk of algal bloom is highlighted during the initial storm events in the RWS river. We propose recommendations for improving water quality in the RWS river systems under the climatic change. [Display omitted] • Reveal algal bloom response to the initial storm event in a reclaimed-water-source river. • The chlorophyll a (Chl-a) shows a big fluctuation during the initial storm events. • Algal bloom caused by the elevated water temperature and sufficient nutrients during post-storm. • The variations of algal bloom can be highly predicated by Random Forest model. [ABSTRACT FROM AUTHOR]

In order to achieve the United Nations 2030 Sustainable Development Goals (SDGs) related to green spaces, monitoring dynamic urban green spaces (UGSs) in cities around the world is crucial. Continuous dynamic UGS mapping is challenged by large computation, time consumption, and energy consumption requirements. Therefore, a fast and automated workflow is needed to produce a high-precision UGS map. In this study, we proposed an automatic workflow to produce up-to-date UGS maps using Otsu's algorithm, a Random Forest (RF) classifier, and the migrating training samples method in the Google Earth Engine (GEE) platform. We took the central urban area of Beijing, China, as the study area to validate this method, and we rapidly obtained an annual UGS map of the central urban area of Beijing from 2016 to 2020. The accuracy assessment results showed that the average overall accuracy (OA) and kappa coefficient (KC) were 96.47% and 94.25%, respectively. Additionally, we used six indicators to measure quality and temporal changes in the UGS spatial distribution between 2016 and 2020. In particular, we evaluated the quality of UGS using the urban greenness index (UGI) and Shannon's diversity index (SHDI) at the pixel level. The experimental results indicate the following: (1) The UGSs in the center of Beijing increased by 48.62 km2 from 2016 to 2020, and the increase was mainly focused in Chaoyang, Fengtai, and Shijingshan Districts. (2) The average proportion of relatively high and above levels (UGI > 0.5) in six districts increased by 2.71% in the study area from 2016 to 2020, and this proportion peaked at 36.04% in 2018. However, our result revealed that the increase was non-linear during this assessment period. (3) Although there was no significant increase or decrease in SHDI values in the study area, the distribution of the SHDI displayed a noticeable fluctuation in the northwest, southwest, and northeast regions of the study area between 2016 and 2020. Furthermore, we discussed and analyzed the influence of population on the spatial distribution of UGSs. We found that three of the five cold spots were located in the east and southeast of Haidian District. Therefore, the proposed workflow could provide rapid mapping and dynamic evaluation of the quality of UGS. [ABSTRACT FROM AUTHOR]

As the capital city of China, Beijing has experienced unprecedented economic and population growth and dramatic impervious surface changes during the last few decades. An application of the classification method combining the spectral and textural features based on Random Forest was conducted to monitor the spatial and temporal changes of Beijing's impervious surfaces. This classification strategy achieved excellent performance in the impervious surface extraction in complex urban areas, as the Kappa coefficient reached 0.850. Based on this strategy, the impervious surfaces inside Beijing's sixth ring road in 1997, 2002, 2007, 2013, and 2017 were extracted. As the development of Beijing has a special regional feature, the changes of impervious surfaces within the sixth ring road were assessed. The findings are as follows: (1) the textural features can significantly improve the classification accuracy of land cover in urban areas, especially for the impervious surface with high albedo. (2) Impervious surfaces within the sixth ring road expanded dramatically from 1997 to 2017, had three expanding periods: 1997–2002, 2002–2007, and 2013–2017, and only shrank in 2007–2013. There are different possible major driving factors for each period. (3) The region between the fifth and sixth ring roads in Beijing underwent the most significant changes in the two decades. (4) The inner three regions are relatively highly urbanized areas compared to the outer two regions. Urbanization processes in the interior regions tend to be completed compared to the exterior regions. [ABSTRACT FROM AUTHOR]

Pollutant gases, such as CO, NO2, O3, and SO2 affect human health, and low-cost sensors are an important complement to regulatory-grade instruments in pollutant monitoring. Previous studies focused on one or several species, while comprehensive assessments of multiple sensors remain limited. We conducted a 12-month field evaluation of four Alphasense sensors in Beijing and used single linear regression (SLR), multiple linear regression (MLR), random forest regressor (RFR), and neural network (long short-term memory (LSTM)) methods to calibrate and validate the measurements with nearby reference measurements from national monitoring stations. For performances, CO > O3 > NO2 > SO2 for the coefficient of determination (R2) and root mean square error (RMSE). The MLR did not increase the R2 after considering the temperature and relative humidity influences compared with the SLR (with R2 remaining at approximately 0.6 for O3 and 0.4 for NO2). However, the RFR and LSTM models significantly increased the O3, NO2, and SO2 performances, with the R2 increasing from 0.3–0.5 to >0.7 for O3 and NO2, and the RMSE decreasing from 20.4 to 13.2 ppb for NO2. For the SLR, there were relatively larger biases, while the LSTMs maintained a close mean relative bias of approximately zero (e.g., <5% for O3 and NO2), indicating that these sensors combined with the LSTMs are suitable for hot spot detection. We highlight that the performance of LSTM is better than that of random forest and linear methods. This study assessed four electrochemical air quality sensors and different calibration models, and the methodology and results can benefit assessments of other low-cost sensors. [ABSTRACT FROM AUTHOR]

• A CNN-based approach is proposed for urban flood susceptibility assessment. • The proposed approach considers topological information for model implementation. • This approach was demonstrated in an urban catchment based on three input strategies. • Two CNN architectures were implemented and compared with two widely used models. • The proposed approach provides a more accurate result than the comparative models. In this study, a convolutional neural network (CNN)-based approach is proposed to assess flood susceptibility for urban catchment. Nine explanatory factors covering precipitation, topographical, and anthropogenic aspects were selected and two CNNs, SCNN and LeNet-5, were implemented to identify the relationship between the explanatory factors and flood inventory between 2004 and 2014 in the Dahongmen catchment in Beijing, China. The performance of the CNNs was compared with that of support vector machine (SVM) and random forest (RF) models for three model input strategies (point-based, array-based and imaged-based strategies). The results showed that: (1) The two CNNs performed better than the SVM and RF, with an accuracy of 0.90 for SCNN and 0.88 for Lenet-5 in the testing period. (2) The CNN-based approach provided more reliable flood susceptibility maps than the comparative models with the highest area under the curve (AUC) index of 0.9 validating by another flood inventory. (3) The upstream inundations induced by pluvial flood cannot be accurately identified by using point-based and array-based strategies. These errors were corrected by considering topographical information using the imaged-based CNN approach. (4) The fixed architecture LeNet-5 produced satisfactory results and avoided the time-consuming process of architecture selection of CNN. We conclude that the proposed CNN-based approach is a valid approach and provides a high-quality susceptibility map for urban mitigation and flood management. [ABSTRACT FROM AUTHOR]