Satellite-based ground PM2.5 estimation using a gradient boosting decision tree

Estimating the spatiotemporal distribution of PM2.5 concentrations in Tianjin during the Chinese Spring Festival: Impact of fireworks ban

SETTING

off fireworks during the Spring Festival (SF) is a traditional practice in China. However, because of its environmental impact, the Chinese government has banned this practice completely. Existing evaluations of the effectiveness of firework prohibition policies (FPPs) lack spatiotemporal perspectives, making it difficult to comprehensively assess their effects on air quality. Consequently, this study used remote sensing technology based on aerosol optical depth and multiple variables, compared nine statistical learning methods, and selected the optimal model, transformer, to estimate daily spatiotemporal continuous PM2.5 concentration datasets for Tianjin from 2016 to 2020. The overall model accuracy reached a root mean square error of 15.30 μg/m³, a mean absolute error of 9.55 μg/m³, a mean absolute percentage error of 21.07%, and an R2 of 0.88. Subsequently, we analysed the variations in PM2.5 concentrations from three time dimensions-the entire year, winter, and SF periods-to exclude the impact of interannual variations on the experimental results. Additionally, we quantitatively estimated firework-specific PM2.5 concentrations based on time-series forecasting. The results showed that during the three years following the implementation of the FPPs, firework-specific PM2.5 concentrations decreased by 52.70%, 49.76%, and 86.90%, respectively, compared to the year before the implementation of the FPPs. Spatially, the central urban area and industrial zones are more affected by FPPs than the suburbs. However, daily variations of PM2.5 concentrations during the SF showed that high concentrations of PM2.5 produced in a short period will return to normal rapidly and will not cause lasting effects. Therefore, the management of fireworks needs to consider both environmental protection and the public's emotional attachment to traditional customs, rather than simply imposing a blanket ban on fireworks. We advocate improving firework policies in four aspects-production, sales, supervision, and control-to promote sustainable development of the ecological environment and human society.

A review of geospatial exposure models and approaches for health data integration

BACKGROUND

Geospatial methods are common in environmental exposure assessments and increasingly integrated with health data to generate comprehensive models of environmental impacts on public health.

OBJECTIVE

Our objective is to review geospatial exposure models and approaches for health data integration in environmental health applications.

METHODS

We conduct a literature review and synthesis.

RESULTS

First, we discuss key concepts and terminology for geospatial exposure data and models. Second, we provide an overview of workflows in geospatial exposure model development and health data integration. Third, we review modeling approaches, including proximity-based, statistical, and mechanistic approaches, across diverse exposure types, such as air quality, water quality, climate, and socioeconomic factors. For each model type, we provide descriptions, general equations, and example applications for environmental exposure assessment. Fourth, we discuss the approaches used to integrate geospatial exposure data and health data, such as methods to link data sources with disparate spatial and temporal scales. Fifth, we describe the landscape of open-source tools supporting these workflows.

Is replacing missing values of PM2.5 constituents with estimates using machine learning better for source apportionment than exclusion or median replacement?

East Asian countries have been conducting source apportionment of fine particulate matter (PM2.5) by applying positive matrix factorization (PMF) to hourly constituent concentrations. However, some of the constituent data from the supersites in South Korea was missing due to instrument maintenance and calibration. Conventional preprocessing of missing values, such as exclusion or median replacement, causes biases in the estimated source contributions by changing the PMF input. Machine learning (ML) can estimate the missing values by training on constituent data, meteorological data, and gaseous pollutants. Complete data from the Seoul Supersite in 2018 was taken, and a random 20% was set as missing. PMF was performed by replacing missing values with estimates. Percent errors of the source contributions were calculated compared to those estimated from complete data. Missing values were estimated using a random forest analysis. Estimation accuracy (r2) was as high as 0.874 for missing carbon species and low at 0.631 when ionic species and trace elements were missing. For the seven highest contributing sources, replacing the missing values of carbon species with estimates minimized the percent errors to 2.0% on average. However, replacing the missing values of the other chemical species with estimates increased the percent errors to more than 9.7% on average. Percent errors were maximal at 37% on average when missing values of ionic species and trace elements were replaced with estimates. Missing values, except for carbon species, need to be excluded. This approach reduced the percent errors to 7.4% on average, which was lower than those due to median replacement. Our results show that reducing the biases in source apportionment is possible by replacing the missing values of carbon species with estimates. To improve the biases due to missing values of the other chemical species, the estimation accuracy of the ML needs to be improved.

Integrated machine learning methods with oversampling technique for regional suitability prediction of waste-to-energy incineration projects

China's tiered strategy to enhance county-level waste incineration for energy aligns with the sustainable development goals (SDGs), emphasizing the need for comprehensive assessments of waste-to-energy (WtE) plant suitability. Traditional assessment methodologies face challenges, particularly in suggesting innovative site alternatives, adapting to new data sets, and their dependence on strict assumptions. This study introduced enhancements in three pivotal dimensions. Methodologically, it leverages data-driven machine learning (ML) approaches to capture the complex relationships essential for site selection, reducing dependency on strict assumptions. In terms of predictive performance, the integration of oversampling with stacked ensemble models enhances the diversity and generalizability of ML models. The area under curve (AUC) scores from four ML models, enhanced by the oversampled dataset, demonstrated significant improvements compared to the original dataset. The stacking model excelled, achieving a score of 92%. It also led in overall Precision and Recall, reaching 85.2% and 85.08% respectively. Nevertheless, a noticeable discrepancy existed in Precision and Recall for positive classes. The stacking model topped Precision scores at 83.1%, followed by eXtreme Gradient Boosting (XGBoost) (82.61%). In terms of Recall, XGBoost recorded the lowest at 85.07%, while the other three classifiers all marked 88.06%. From an industry applicability standpoint, the stacking model provides innovative location alternatives and demonstrates adaptability in Hunan province, offering a reusable tool for WtE location. In conclusion, this study not only enhances the methodological aspects of WtE site selection but also provides practical and adaptable solutions, contributing positively to sustainable waste management practices.

Validation and calibration of aerosol optical depth and classification of aerosol types based on multi-source data over China

A refined classification of aerosol types is essential to identify and control air pollution sources. This study focused on improving the resolution and accuracy of aerosol optical depth (AOD) and further refining the classification of aerosol types in China. We validated the accuracy of the AOD acquired using the Modern-Era Retrospective Analysis for Research and Applications version 2 (MERRA2) and Copernicus Atmosphere Monitoring Service (CAMS) by comparing it with that acquired using from the Aeronet Robotic Network (AERONET). We simulated the AOD with high spatial resolution and accuracy based on the extremely randomized trees (ERT), adaptive boosting (AdaBoost), and gradient boosting decision trees (GBDT) models and identified aerosol types based on the Angstrom Exponent (AE) from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the calibrated AOD. The results showed that CAMS overestimates AOD (21.4 %) and MERRA2 underestimates AOD (-17.3 %). Among the three machine learning models, the ERT model performed best, with a determination coefficient (R2) of 0.825 and the root-mean-square error (RMSE) of 0.174. Biomass burning/urban-industrial aerosols dominated China, with the largest contributions to southern, eastern, and central China in spring and summer. Clean continental aerosols contributed the most to southwestern China in fall and winter, whereas desert dust aerosols contributed the most to northwestern and eastern China in spring.

Reconstructing annual XCO2 at a 1 km×1 km spatial resolution across China from 2012 to 2019 based on a spatial CatBoost method

Long-time-series, high-resolution datasets of the column-averaged dry-air mole fraction of carbon dioxide (XCO2) have great practical importance for mitigating the greenhouse effect, assessing carbon emissions and implementing a low-carbon cycle. However, the mainstream XCO2 datasets obtained from satellite observations have coarse spatial resolutions and are inadequate for supporting research applications with different precision requirements. Here, we developed a new spatial machine learning model by fusing spatial information with CatBoost, called SCatBoost, to fill the above gap based on existing global land-mapped 1° XCO2 data (GLM-XCO2). The 1-km-spatial-resolution dataset containing XCO2 values in China from 2012 to 2019 reconstructed by SCatBoost has stronger and more stable predictive power (confirmed with a cross-validation (R2 = 0.88 and RSME = 0.20 ppm)) than other traditional models. According to the estimated dataset, the overall national XCO2 showed an increasing trend, with the annual mean concentration rising from 392.65 ppm to 410.36 ppm. In addition, the spatial distribution of XCO2 concentrations in China reflects significantly higher concentrations in the eastern coastal areas than in the western inland areas. The contributions of this study can be summarized as follows: (1) It proposes SCatBoost, integrating the advantages of machine learning methods and spatial characteristics with a high prediction accuracy; (2) It presents a dataset of fine-scale and high resolution XCO2 over China from 2012 to 2019 by the model of SCatBoost; (3) Based on the generated data, we identify the spatiotemporal trends of XCO2 in the scale of nation and city agglomeration. These long-term and high resolution XCO2 data help understand the spatiotemporal variations in XCO2, thereby improving policy decisions and planning about carbon reduction.

Ecological risk assessment based on soil adsorption capacity for heavy metals in Taihu basin, China

Due to the toxicity, bioaccumulation, non-biodegradability and perseverance of heavy metals, their risk assessment is essential for soil quality management. The Hakanson potential ecological risk index (RI), which considers the effects of heavy metal concentration and toxicity, has been widely used in soil ecological risk assessment. However, RI overlooks the influence of soil properties on the mobility and availability of heavy metals in risk assessment. To fill this gap, this study sought to develop an improved ecological risk index (IRI), which incorporates soil adsorption into RI, and applied it to evaluate the ecological risk of heavy metals in the soil of the Taihu basin, China. The soil adsorption models based on the Gradient Boosting Decision Tree (GBDT) was used to predict the soil adsorption capacity of five heavy metals (i.e. cadmium, chromium, copper, lead, zinc). The soil adsorption capacity in 1446 sites in the Taihu basin was predicted by the GBDT models and was assigned as the weight of IRI. The risk assessment results of the five metals in the Taihu basin showed that 40% of the sites were at a moderate risk level and 60% of the sites were at a slight risk level based on the RI. The value of IRI in the basin ranged from 11.1 to 75.5, with a mean value of 28.1. IRI differed from RI in spatial distribution due to the influence of soil adsorption. The comparative analysis between the metal contents in sediments and surrounding soils confirmed the tremendous influence of soil adsorption on ecological risks, indicating that soil adsorption should be taken into consideration in soil risk assessment.

PM2.5 concentration assessment based on geographical and temporal weighted regression model and MCD19A2 from 2015 to 2020 in Xinjiang, China

PM2.5 is closely linked to both air quality and public health. Many studies have used models combined with remote sensing and auxiliary data to inverse a large range of PM2.5 concentrations. However, the data's spatial resolution is limited. and better results might have been obtained if higher resolution data had been used. Therefore, this paper establishes a geographical and temporal weighted regression model (GTWR) and estimates the PM2.5 concentration in Xinjiang from 2015 to 2020 using 1 km resolution MCD19A2 (MODIS/Terra+Aqua Land Aerosol Optical Thickness Daily L2G Global 1km SIN Grid V006) data and 9 auxiliary variables. The findings indicate that the GTWR model performs better than the simple linear regression (SLR) and geographically weighted regression (GWR) models in terms of accuracy and feasibility in retrieving PM2.5 concentrations in Xinjiang. Simultaneously, by combining the GTWR model with MCD19A2 data, a spatial distribution map of PM2.5 with better spatial resolution can be obtained. Next, the regional distribution of annual PM2.5 concentrations in Xinjiang is consistent with the terrain from 2015 to 2020. The low value area is primarily found in the mountainous area with higher terrain, while the high value area is primarily in the basin with lower terrain. Overall, the southwest is high and the northeast is low. In terms of time change, the six-year PM2.5 shows a single peak distribution with 2016 as the inflection point. Lastly, from 2015 to 2020, the seasonal average PM2.5 concentration in Xinjiang has a significant difference, thereby showing winter (66.15μg/m3)>spring (52.28μg/m3)>autumn (40.51μg/m3)>summer (38.63μg/m3). The research shows that the combination of MCD19A2 data and GTWR model has good applicability in retrieving PM2.5 concentration.

Identifying sensitivity of factor cluster based gully erosion susceptibility models

The present study has attempted to address the issue of sensitivity of different clusters of factors towards gully erosion in the Mayurakshi river basin. Firstly, the gully erosion susceptibility of the basin area has been mapped by integrating using 18 parameters divided into four factor-cluster, viz. erodibility, erosivity, resistance, and topographical cluster, with the help of four machine learning (ML) models such as random forest (RF), gradient boost (GBM), extreme gradient boost (XGB), and support vector machine (SVM). Results show that almost 20% and 25% of the upper catchment of the basin belongs to extreme and high gully erosion susceptibility. Among the applied algorithms, RF is appeared as the best performing model. The spatial association of factor cluster-based models with the final susceptibility model is found the highest for the erosivity cluster, followed by the erodibility cluster. From the sensitivity analysis, it becomes clear that geology and soil texture are dominant contributing factors to gully erosion susceptibility. The geological formation of unclassified granite gneiss and geomorphological formation of denudational origin pediment-pediplain complex is dominant over the entire upper catchment of the basin, and therefore, can be considered regional factors of importance. Since the study has figured out the different grades of susceptible areas with dominant factors and factor cluster, it would be useful for devising planning for gully erosion check measures. From economic particularly food security purpose, it is very essential since it is concerned with precious soil loss and negative effects on agriculture.

Applying machine learning to construct braking emission model for real-world road driving

Brake emissions from vehicles are increasing as the number of vehicles increases. However, current research on brake emissions, particularly the intensity and characteristics of emissions under real road conditions, is significantly inadequate compared to exhaust emissions. To this end, a dataset of 600 (200 unique real-world braking events simulated using three types of brake pads) real-world braking events (called brake pad segments) was constructed and a mapping function between the average brake emission intensity of PM2.5 from the segments and the segment features was established by five algorithms (multiple linear regression (MLR) and four machine learning algorithms). Based on the five algorithms, the importance of the different features of the fragments was discussed and brake energy intensity (BEI) and metal content (MC) of the brake pad emissions were identified as the most significant factors affecting brake emissions and used as the final modeling features. Among the five algorithms, categorical boosting (CatBoost) had the best prediction performance, with a mean R2 and RMSE of 0.83 and 0.039 respectively for the tenfold cross-validation. In addition, the CatBoost-based model was further compared with the MOVES model to demonstrate its applicability. The CatBoost-based model has better prediction performance than the MOVES model. The MOVES model overpredicts brake fragment emissions for urban roads and underpredicts brake fragment emissions for motorways. Furthermore, the CatBoost-based model was interpreted and visualized by an individual conditional expectation (ICE) plot to break the machine learning "black box", with BEI and MC showing nonlinear monotonic increasing relationships with braking emissions. ICE plot also provides viable technical solutions for controlling brake emissions in the future. Both avoiding aggressive braking driving behavior (e.g., the application of smart transportation technologies) and using brake pads with less metal content (e.g., using ceramic brake pads) can effectively reduce brake emissions. The construction of a machine learning-based brake emission model and the white-boxing of its model provide excellent insights for the future detailed assessment and control of brake emissions.

Full-Coverage PM2.5 Mapping and Variation Assessment during the Three-Year Blue-Sky Action Plan Based on a Daily Adaptive Modeling Approach

Owing to a series of air pollution prevention and control policies, China’s PM2.5 pollution has greatly improved; however, the long-term spatial contiguous products that facilitate the analysis of the distribution and variation of PM2.5 pollution are insufficient. Due to the limitations of missing values in aerosol optical depth (AOD) products, the reconstruction of full-coverage PM2.5 concentration remains challenging. In this study, we present a two-stage daily adaptive modeling framework, based on machine learning, to solve this problem. We built the annual models in the first stage, then daily models were constructed in the second stage based on the output of the annual models, which incorporated the parameter and feature adaptive tuning strategy. Within this study, PM2.5 concentrations were adaptively modeled and reconstructed daily based on the multi-angle implementation of atmospheric correction (MAIAC) AOD products and other ancillary data, such as meteorological factors, population, and elevation. Our model validation showed excellent performance with an overall R2 = 0.91 and RMSE = 9.91 μg/m3 for the daily models, along with the site-based cross-validation R2s and RMSEs of 0.86–0.87 and 12–12.33 μg/m3; these results indicated the reliability and feasibility of the proposed approach. The daily full-coverage PM2.5 concentrations at 1 km resolution across China during the Three-Year Blue-Sky Action Plan were reconstructed in this study. We analyzed the distribution and variations of reconstructed PM2.5 at three different time scales. Overall, national PM2.5 pollution has significantly improved with the annual average concentration dropping from 33.67–28.03 μg/m3, which demonstrated that air pollution control policies are effective and beneficial. However, some areas still have severe PM2.5 pollution problems that cannot be ignored. In conclusion, the approach proposed in this study can accurately present daily full-coverage PM2.5 concentrations and the research outcomes could provide a reference for subsequent air pollution prevention and control decision-making.

Energy Consumption Estimation for Electric Buses Based on a Physical and Data-Driven Fusion Model

The energy consumption of electric vehicles is closely related to the problems of charging station planning and vehicle route optimization. However, due to various factors, such as vehicle performance, driving habits and environmental conditions, it is difficult to estimate vehicle energy consumption accurately. In this work, a physical and data-driven fusion model was designed for electric bus energy consumption estimation. The basic energy consumption of the electric bus was modeled by a simplified physical model. The effects of rolling drag, brake consumption and air-conditioning consumption are considered in the model. Taking into account the fluctuation in energy consumption caused by multiple factors, a CatBoost decision tree model was constructed. Finally, a fusion model was built. Based on the analysis of electric bus data on the big data platform, the performance of the energy consumption model was verified. The results show that the model has high accuracy with an average relative error of 6.1%. The fusion model provides a powerful tool for the optimization of the energy consumption of electric buses, vehicle scheduling and the rational layout of charging facilities.

Spatiotemporal PM2.5 estimations in China from 2015 to 2020 using an improved gradient boosting decision tree

Fine particulate matter (PM_2.5) with spatiotemporal continuity can provide important basis for the assessment of adverse effects on human health. In recent years, researchers have done a lot of work on the surface PM_2.5 simulation. However, due to the limitations of data and models, it is difficult to accurately evaluate the spatial and temporal PM_2.5 variations on a fine scale. In this study, we adopted the multi-angle implementation of atmospheric correction (MAIAC) aerosol products, and proposed a spatiotemporal model based on the gradient boosting decision tree (GBDT) algorithm to retrieve PM_2.5 concentration across China from 2015 to 2020 at 1-km resolution. Our model achieved excellent performance, with overall CV-R² of 0.92, and annual CV-R² of 0.90-0.93. In addition, the model can also be used for evaluation on different time scales. Compared with previous studies, the model developed in our study performed better and more stable, which showed the highest accuracies in PM_2.5 estimation works at 1-km resolution. During the study period, the overall national PM_2.5 pollution showed a downward trend, with the annual mean concentration dropping from 42.42 μg/m³ to 27.91 μg/m³. The largest decrease occurred in Beijing-Tianjin-Hebei (BTH), with a trend of -5.17 μg/m³/yr, while it remains the most polluted region. The area meeting the secondary national air quality standard (<35 μg/m³) increased from ∼34% to ∼79%. These results indicate that the atmospheric environment has improved significantly. Moreover, different regions have different time nodes for the start of the continuous standard-met day during the year, and the duration is different as well. Overall, this study can provide reliable large-scale PM_2.5 estimations.

Mapping China’s Forest Fire Risks with Machine Learning

Forest fires are disasters that are common around the world. They pose an ongoing challenge in scientific and forest management. Predicting forest fires improves the levels of forest-fire prevention and risk avoidance. This study aimed to construct a forest risk map for China. We base our map on Visible Infrared Imaging Radiometer Suite data from 17,330 active fires for the period 2012–2019, and combined terrain, meteorology, social economy, vegetation, and other factors closely related to the generation of forest-fire disasters for modeling and predicting forest fires. Four machine learning models for predicting forest fires were compared (i.e., random forest (RF), support vector machine (SVM), multi-layer perceptron (MLP), and gradient-boosting decision tree (GBDT) algorithm), and the RF model was chosen (its accuracy, precision, recall, F1, AUC values were 87.99%, 85.94%, 91.51%, 88.64% and 95.11% respectively). The Chinese seasonal fire zoning map was drawn with the municipal administrative unit as the spatial scale for the first time. The results show evident seasonal and regional differences in the Chinese forest-fire risks; forest-fire risks are relativity high in the spring and winter, but low in fall and summer, and the areas with high regional fire risk are mainly in the provinces of Yunnan (including the cities of Qujing, Lijiang, and Yuxi), Guangdong (including the cities of Shaoguan, Huizhou, and Qingyuan), and Fujian (including the cities of Nanping and Sanming). The major contributions of this study are to (i) provide a framework for large-scale forest-fire risk prediction having a low cost, high precision, and ease of operation, and (ii) improve the understanding of forest-fire risks in China.

Prediction Performance Comparison of Risk Management and Control Mode in Regional Sites Based on Decision Tree and Neural Network

The traditional risk management and control mode (RMCM) in regional sites has the defects of low efficiency, high cost, and lack of systematism. Trying to resolve these defects and explore the application possibility of machine learning, a characteristic dataset for RMCM in regional sites was established. Three decision tree (DT) algorithms (CHAID, EXHAUSTIVE CHAID, and CART) and two artificial neural network (ANN) algorithms [back propagation (BP) and radial basis function (RBF)] were implemented to predict RMCM in regional sites. The results showed that in the aspects of accuracy (ACC), precision (PRE), recall ratio (REC), and F₁ value, CART–DT was superior to CHAID–DT and EXHAUSTIVE CHAID–DT (E-CHAID–DT); and BP–ANN was superior to RBF–ANN. However, CART–DT was inferior to BP–ANN in ACC, PRE, REC, and F₁ value. BP–ANN model is good at non-linear mapping, and it has a flexible network structure and a low risk of over-fitting. The case study of a typical county demonstration area confirmed the extensibility of the method, and the method has great potential in RMCM prediction in regional sites in the future.

Progress in Dust Modelling, Global Dust Budgets, and Soil Organic Carbon Dynamics

Dust emission is an important corollary of the soil degradation process in arid and semi-arid areas worldwide. Soil organic carbon (SOC) is the main terrestrial pool in the carbon cycle, and dust emission redistributes SOC within terrestrial ecosystems and to the atmosphere and oceans. This redistribution plays an important role in the global carbon cycle. Herein, we present a systematic review of dust modelling, global dust budgets, and the effects of dust emission on SOC dynamics. Focusing on selected dust models developed in the past five decades at different spatio-temporal scales, we discuss the global dust sources, sinks, and budgets identified by these models and the effect of dust emissions on SOC dynamics. We obtain the following conclusions: (1) dust models have made considerable progress, but there are still some uncertainties; (2) a set of parameters should be developed for the use of dust models in different regions, and direct anthropogenic dust should be considered in dust emission estimations; and (3) the involvement of dust emission in the carbon cycle models is crucial for improving the accuracy of carbon assessment.

Interpolation-Based Fusion of Sentinel-5P, SRTM, and Regulatory-Grade Ground Stations Data for Producing Spatially Continuous Maps of PM2.5 Concentrations Nationwide over Thailand

Atmospheric pollution has recently drawn significant attention due to its proven adverse effects on public health and the environment. This concern has been aggravated specifically in Southeast Asia due to increasing vehicular use, industrial activity, and agricultural burning practices. Consequently, elevated PM2.5 concentrations have become a matter of intervention for national authorities who have addressed the needs of monitoring air pollution by operating ground stations. However, their spatial coverage is limited and the installation and maintenance are costly. Therefore, alternative approaches are necessary at national and regional scales. In the current paper, we investigated interpolation models to fuse PM2.5 measurements from ground stations and satellite data in an attempt to produce spatially continuous maps of PM2.5 nationwide over Thailand. Four approaches are compared, namely the inverse distance weighted (IDW), ordinary kriging (OK), random forest (RF), and random forest combined with OK (RFK) leveraging on the NO2, SO2, CO, HCHO, AI, and O3 products from the Sentinel-5P satellite, regulatory-grade ground PM2.5 measurements, and topographic parameters. The results suggest that RFK is the most robust, especially when the pollution levels are moderate or extreme, achieving an RMSE value of 7.11 μg/m3 and an R2 value of 0.77 during a 10-day long period in February, and an RMSE of 10.77 μg/m3 and R2 and 0.91 during the entire month of March. The proposed approach can be adopted operationally and expanded by leveraging regulatory-grade stations, low-cost sensors, as well as upcoming satellite missions such as the GEMS and the Sentinel-5.

Modeling household online shopping demand in the U.S.: a machine learning approach and comparative investigation between 2009 and 2017

Despite the rapid growth of online shopping and research interest in the relationship between online and in-store shopping, national-level modeling and investigation of the demand for online shopping with a prediction focus remain limited in the literature. This paper differs from prior work and leverages two recent releases of the U.S. National Household Travel Survey (NHTS) data for 2009 and 2017 to develop machine learning (ML) models, specifically gradient boosting machine (GBM), for predicting household-level online shopping purchases. The NHTS data allow for not only conducting nationwide investigation but also at the level of households, which is more appropriate than at the individual level given the connected consumption and shopping needs of members in a household. We follow a systematic procedure for model development including employing Recursive Feature Elimination algorithm to select input variables (features) in order to reduce the risk of model overfitting and increase model explainability. Among several ML models, GBM is found to yield the best prediction accuracy. Extensive post-modeling investigation is conducted in a comparative manner between 2009 and 2017, including quantifying the importance of each input variable in predicting online shopping demand, and characterizing value-dependent relationships between demand and the input variables. In doing so, two latest advances in machine learning techniques, namely Shapley value-based feature importance and Accumulated Local Effects plots, are adopted to overcome inherent drawbacks of the popular techniques in current ML modeling. The modeling and investigation are performed at the national level, with a number of findings obtained. The models developed and insights gained can be used for online shopping-related freight demand generation and may also be considered for evaluating the potential impact of relevant policies on online shopping demand.

Exploring solutions to alleviate the regional water stress from virtual water flows in China

China has long faced an uneven distribution of physical water resources, which has been further exacerbated by the virtual water transfers embodied in the interregional trade. To alleviate such unfavorable influences of interregional virtual water flows on regional water scarcity, this paper first combined a multi-regional input-output model and a structural decomposition analysis to identify the major driving forces behind the changes in interregional virtual water flows from 2002 to 2012, and then conducted a scenario analysis to explore solutions for sustainable water resource management in China. Results indicated that the virtual water outflows from water-deficient developing regions (Northwest and Northeast) to water-abundant developed regions, such as East Coast and South Coast, have been increasingly intensified from 2002 to 2012. During the period, the final demand predominated the increase of virtual water transfers, while the improvement of water use efficiency dominated the decline in virtual water flows from 2002 to 2012. Results from the designed scenarios indicated that the negative impacts of interregional virtual water flows on the water stress can be effectively relieved, indicating the high priority of regional water use efficiency improvement, especially in water-starved regions.

Spatio-temporal variations of the major meteorological disasters and its response to climate change in Henan Province during the past two millennia

In China, historical documents have recorded large quantities of information related to natural disasters, and these disasters have had long-lasting effects on economic and social activities. Understanding the occurrence of the natural disasters and their spatio-temporal variation characters is crucial for sustainable of our society. Therefore, based on the collection and collation of historical documents, and adopting mathematical statistics, Kriging interpolation, correlation analysis and other methods, we systematically explored the meteorological disasters in Henan Province during the past two millennia in analyzing their spatio-temporal distribution characters and driving forces. The results demonstrate that there were five major types of meteorological disasters in Henan Province, including drought, flood, hails, low temperature and frost and insect pests, which presented obvious spatio-temporal variations and have occurred frequently during the past two millennia. According to the historical documents, the major meteorological disasters occurred 1,929 times in Henan from 221 BCE to 2000 CE. On the whole, the disaster frequency show that the occurrence cycle of the meteorological disasters has obvious changes, which mainly occurred in the middle and late stages during the past two millennia, especially after 1300 CE. Furthermore, we also find that the variation of meteorological disaster events is consistent with the variation of temperature in eastern China and the frequency of meteorological disaster increases in the cold period, but decreases in the warm period. In addition, there are obvious differences in the spatial distribution of the major meteorological disaster, which were mainly distributed in the northwest and southern part region of the Henan Province before 1911 CE. While after 1911 CE, the northern and southeastern parts were the meteorological disaster-prone areas in this region during this period. Spatial correlation analysis of each meteorological disaster before and after 1911 CE points out the droughts disaster frequency-occurring district has transferred in different periods, while the hail and low temperature and frost disasters just have a smaller transferred during these two periods. Conversely, the frequency-occurring districts of floods and insect pest disasters have no obviously transferred in different periods. These results can provide an important scientific basis for governmental decision makers and local people to prevent and mitigate meteorological disaster in the future.

Estimation of air temperature and the mountain-mass effect in the Yellow River Basin using multi-source data

Quantitative studies of the multiple factors influencing the mountain-mass effect, which causes higher temperatures in mountainous than non-mountainous regions, remain insufficient. This study estimated the air temperature in the Yellow River Basin, which spans three different elevation ranges, using multi-source data to address the uneven distribution of regional meteorological stations. The differences in mountain-mass effect for different geomorphic regions at the same altitude were then compared. The Manner-Kendall nonparametric test was used to analyse time series changes in temperature. Moreover, we employed the geographically weighted regression (GWR) model, with MODIS land-surface and air-temperature data, station-based meteorological data, vertical temperature gradients corresponding to the 2000-2015 period, and elevation data, to estimate the correlation between monthly mean surface temperature and air temperature in the Yellow River Basin. The following major results were obtained. (1) The GWR method and ground station-based observations enhanced the accuracy of air-temperature estimates with an error of only ± 0.74°C. (2) The estimated annual variations in the spatial distributions of 12-month average temperatures showed that the upper Tibetan Plateau is characterised by low annual air temperatures with a narrow spatial distribution, whereas north-eastern areas upstream of the Inner Mongolia Plateau are characterised by higher air temperatures. Changes in the average monthly air temperature were also high in the middle and lower reaches, with a narrow spatial distribution. (3) Considering the seasonal variation in the temperature lapse rate, the mountain-mass effect in the Yellow River Basin was very high. In the middle of each season, the variation of air temperature at a given altitude over the Tibetan Plateau was higher than that over the Loess Plateau and Jinji Mountain. The results of this study reveal the unique temperature characteristics of the Yellow River Basin according to its geomorphology. Furthermore, this research contributes to quantifying mountain-mass effects.

A National-Scale 1-km Resolution PM2.5 Estimation Model over Japan Using MAIAC AOD and a Two-Stage Random Forest Model

Satellite-based models for estimating concentrations of particulate matter with an aerodynamic diameter less than 2.5 μm (PM2.5) have seldom been developed in islands with complex topography over the monsoon area, where the transport of PM2.5 is influenced by both the synoptic-scale winds and local-scale circulations compared with the continental regions. We validated Multi-Angle Implementation of Atmospheric Correction (MAIAC) aerosol optical depth (AOD) with ground observations in Japan and developed a 1-km-resolution national-scale model between 2011 and 2016 to estimate daily PM2.5 concentrations. A two-stage random forest model integrating MAIAC AOD with meteorological variables and land use data was applied to develop the model. The first-stage random forest model was used to impute the missing AOD values. The second-stage random forest model was then utilised to estimate ground PM2.5 concentrations. Ten-fold cross-validation was performed to evaluate the model performance. There was good consistency between MAIAC AOD and ground truth in Japan (correlation coefficient = 0.82 and 74.62% of data falling within the expected error). For model training, the model showed a training coefficient of determination (R2) of 0.98 and a root mean square error (RMSE) of 1.22 μg/m3. For the 10-fold cross-validation, the cross-validation R2 and RMSE of the model were 0.86 and 3.02 μg/m3, respectively. A subsite validation was used to validate the model at the grids overlapping with the AERONET sites, and the model performance was excellent at these sites with a validation R2 (RMSE) of 0.94 (1.78 μg/m3). Additionally, the model performance increased as increased AOD coverage. The top-ten important predictors for estimating ground PM2.5 concentrations were day of the year, temperature, AOD, relative humidity, 10-m-height zonal wind, 10-m-height meridional wind, boundary layer height, precipitation, surface pressure, and population density. MAIAC AOD showed high retrieval accuracy in Japan. The performance of the satellite-based model was excellent, which showed that PM2.5 estimates derived from the model were reliable and accurate. These estimates can be used to assess both the short-term and long-term effects of PM2.5 on health outcomes in epidemiological studies.

FPGA Accelerator for Gradient Boosting Decision Trees

A decision tree is a well-known machine learning technique. Recently their popularity has increased due to the powerful Gradient Boosting ensemble method that allows to gradually increasing accuracy at the cost of executing a large number of decision trees. In this paper we present an accelerator designed to optimize the execution of these trees while reducing the energy consumption. We have implemented it in an FPGA for embedded systems, and we have tested it with a relevant case-study: pixel classification of hyperspectral images. In our experiments with different images our accelerator can process the hyperspectral images at the same speed at which they are generated by the hyperspectral sensors. Compared to a high-performance processor running optimized software, on average our design is twice as fast and consumes 72 times less energy. Compared to an embedded processor, it is 30 times faster and consumes 23 times less energy.

A Satellite-Based Land Use Regression Model of Ambient NO2 with High Spatial Resolution in a Chinese City

Previous studies have reported that intra-urban variability of NO2 concentrations is even higher than inter-urban variability. In recent years, an increasing number of studies have developed satellite-derived land use regression (LUR) models to predict ground-level NO2 concentrations, though only a few have been conducted at a city scale. In this study, we developed a satellite-derived LUR model to predict seasonal NO2 concentrations at a city scale by including satellite-retrieved NO2 tropospheric column density, population density, traffic indicators, and NOx emission data. The R2 of model fitting and 10-fold cross validation were 0.70 and 0.61 for the satellite-derived seasonal LUR model, respectively. The satellite-based LUR model captured seasonal patterns and fine gradients of NO2 variations at a 100 m × 100 m resolution and demonstrated that NO2 pollution in winter is 1.46 times higher than that in summer. NO2 concentrations declined significantly with increasing distance from roads and with increasing distance from the city center. In Suzhou, 84% of the total population lived in areas with NO2 concentrations exceeding the annual-mean standard at 40 μg/m3 in 2014. This study demonstrated that satellite-retrieved data could help increase the accuracy and temporal resolution of the traditional LUR models at a city scale. This application could support exposure assessment at a high resolution for future epidemiological studies and policy development pertaining to air quality control.