Search

Along with the rapid development of industries and the acceleration of urbanisation, the problem of air pollution is becoming more serious. Exploring the relevant factors affecting air quality and accurately predicting the air quality index are significant in improving the overall environmental quality and realising green economic development. Machine learning algorithms and statistical models have been widely used in air quality prediction and ranking assessment. In this paper, based on daily air quality data for the city of Xi'an, China, from 1 October 2022 to 30 September 2023, we construct support vector regression (SVR), gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), random forests (RF), neural network (NN) and long short-term memory (LSTM) models to analyse the influence of the air quality index for Xi'an and to conduct comparative tests. The predicted values and 95% prediction intervals of the AQI for the next 15 days for Xi'an, China, are given based on the Bootstrap-XGBoost algorithm. Further, the ordinal logit regression and ordinal probit regression models are constructed to evaluate and accurately predict the AQI ranks of the data from 1 October 2023 to 15 October 2023 for Xi'an. Finally, this paper proposes some suggestions and policy measures based on the findings of this paper. [ABSTRACT FROM AUTHOR]

This paper proposes to analyze the performance increase in the forecasting of solar irradiance and wind speed by implementing a dynamic ensemble architecture for intra-hour horizon ranging from 10 to 60 min for a 10 min time step data. Global horizontal irradiance (GHI) and wind speed were computed using four standalone forecasting models (random forest, k-nearest neighbors, support vector regression, and elastic net) to compare their performance against two dynamic ensemble methods, windowing and arbitrating. The standalone models and the dynamic ensemble methods were evaluated using the error metrics RMSE, MAE, R2, and MAPE. This work's findings showcased that the windowing dynamic ensemble method was the best-performing architecture when compared to the other evaluated models. For both cases of wind speed and solar irradiance forecasting, the ensemble windowing model reached the best error values in terms of RMSE for all the assessed forecasting horizons. Using this approach, the wind speed forecasting gain was 0.56% when compared with the second-best forecasting model, whereas the gain for GHI prediction was 1.96%, considering the RMSE metric. The development of an ensemble model able to provide accurate and precise estimations can be implemented in real-time forecasting applications, helping the evaluation of wind and solar farm operation. [ABSTRACT FROM AUTHOR]

Among the critical global crises curbing world development and sustainability, air quality degradation has been a long-lasting and increasingly urgent one and it has been sufficiently proven to pose severe threats to human health and social welfare. A higher level of model prediction accuracy can play a fundamental role in air quality assessment and enhancing human well-being. In this paper, four types of machine learning models—random forest model, ridge regression model, support vector machine model, extremely randomized trees model—were adopted to predict PM2.5 concentration in ten cities in the Jing-Jin-Ji region of north China based on multi-sources spatiotemporal data including air quality and meteorological data in time series. Data were fed into the model by using the rolling prediction method which is proven to improve prediction accuracy in our experiments. Lastly, the comparative experiments show that at the city level, RF and ExtraTrees models have better predictive results with lower mean absolute error (MAE), root mean square error (RMSE), and higher index of agreement (IA) compared to other selected models. For seasonality, level four models all have the best prediction performances in winter time and the worst in summer time, and RF models have the best prediction performance with the IA ranging from 0.93 to 0.98 with an MAE of 5.91 to 11.68 μg/m3. Consequently, the demonstration of how each model performs differently in each city and each season is expected to shed light on environmental policy implications. [ABSTRACT FROM AUTHOR]

The factors influencing precipitation in western China are quite complex, which increases the difficulty in determining accurate predictors. Hence, this paper models the monthly measured precipitation data from 240 meteorological stations in mainland China and the precipitation data from the European Centre for Medium-Range Weather Forecasts and the National Climate Centre and employs 88 atmospheric circulation indices to develop a precipitation prediction scheme. Specifically, a high-quality grid-point field is created by fusing and revising the precipitation data from multiple sources. This field is combined with the Empirical Orthogonal Function decomposition and the causal information flow. Next, the best predictors are screened through Empirical Orthogonal Function decomposition and causal information flow, and a data-driven precipitation prediction model is established using a Back Propagation Neural Network and a Random Forest algorithm to conduct the 1-month, 3-month, and 6-month precipitation predictions. The results show that: The machine learning-based precipitation prediction model has high accuracy and is generally able to predict the precipitation trend in the western region better. The Random Forest algorithm significantly outperforms the Back Propagation Neural Network algorithm in the prediction of the three starting times, and the prediction ability of both models gradually decreases as the starting time increases. Compared with the 2022 flood season prediction scores of the Institute of Atmospheric Sciences of the Chinese Academy of Sciences, the model improves the prediction of 1-month and 3-month precipitation in the western region and provides a new idea for the short-term climate prediction of precipitation in western China. [ABSTRACT FROM AUTHOR]

Particulate matter (PM) in the air can cause various health problems and diseases in humans. In particular, the smaller size of PM 2.5 enable them to penetrate deep into the lungs, causing severe health impacts. Exposure to PM 2.5 can result in respiratory, cardiovascular, and allergic diseases, and prolonged exposure has also been linked to an increased risk of cancer, including lung cancer. Therefore, forecasting the PM 2.5 concentration in the surrounding is crucial for preventing these adverse health effects. This paper proposes a method for forecasting the PM 2.5 concentration after 1 h using bidirectional long short-term memory (Bi-LSTM). The proposed method involves selecting input variables based on the feature importance calculated by random forest, classifying the data to assign weight variables to reduce bias, and forecasting the PM 2.5 concentration using Bi-LSTM. To compare the performance of the proposed method, two case studies were conducted. First, a comparison of forecasting performance according to preprocessing. Second, forecasting performance between deep learning (long short-term memory, gated recurrent unit, and Bi-LSTM) and conventional machine learning models (multi-layer perceptron, support vector machine, decision tree, and random forest). In case study 1, The proposed method shows that the performance indices (RMSE: 3.98%p, MAE: 5.87%p, RRMSE: 3.96%p, and R 2 :0.72%p) are improved because weights are given according to the input variables before the forecasting is performed. In case study 2, we show that Bi-LSTM, which considers both directions (forward and backward), can effectively forecast when compared to conventional models (RMSE: 2.70, MAE: 0.84, RRMSE: 1.97, R 2 : 0.16). Therefore, it is shown that the proposed method can effectively forecast PM 2.5 even if the data in the high-concentration section is insufficient. [ABSTRACT FROM AUTHOR]

Most of the current, popular approaches to monitoring real driving NOx emissions are based on direct measurement. However, due to the uncertainty of sensor-based measurements, such methods cannot always be used to accurately screen out the malfunctions of an emission control system. In this paper, a random forest (RF) model which extracts information from on-board diagnostics (OBD) data streams transmitted by a remote emission management vehicle terminal (REMVT) is proposed to provide a specific emission method for the online screening of high NOx emissions. First, two particular forms of modeling, random forest and logistic regression (LR), are laid out as representatives of nonparametric models and specified linear models. These two models were trained, validated and compared using OBD data collected from three China-VI heavy-duty diesel vehicles (HDDVs). The results show that as a data-driven, highly adaptive and robust learning method, the RF model can more accurately identify an abnormal emission state. Finally, a further validation was conducted, in which another China-VI HDDV was tested in two typical states, including a fault state and a normal state. The results indicated that the RF model could clearly distinguish the out-of-control emission condition from the normal operation state. The outcome of this research verifies the feasibility of using a machine learning model to process remote OBD data on HD vehicles and to identify high emissions in the case of an in-use fleet. On this basis, more sophisticated combined models and multi-stage models could be developed. [ABSTRACT FROM AUTHOR]

FAO Blaney-Criddle has been generally an accepted method for estimating reference crop evapotranspiration. In this regard, it is inevitable to estimate the b-factor provided by the Food and Agriculture Organization (FAO) of the United Nations Irrigation and Drainage Paper number 24. In this study, five soft computing methods, namely random forest (RF), M5 model tree (M5), support vector regression with the polynomial function (SVR-poly), support vector regression with radial basis function kernel (SVR-rbf), and random tree (RT), were adapted to estimate the b-factor. And Their performances were also compared. The suitable hyper-parameters for each soft computing method were investigated. Five statistical indices were deployed to evaluate their performance, i.e., the coefficient of determination (r2), the mean absolute relative error (MARE), the maximum absolute relative error (MXARE), the standard deviation of the absolute relative error (DEV), and the number of samples with an error greater than 2% (NE > 2%). Findings reveal that SVR-rbf gave the highest performance among five soft computing models, followed by the M5, RF, SVR-poly, and RT. The M5 also derived a new explicit equation for b estimation. SVR-rbf provided a bit lower efficacy than the radial basis function network but outperformed the regression equations. Models' Applicability for estimating monthly reference evapotranspiration (ETo) was demonstrated. [ABSTRACT FROM AUTHOR]

Atmospheric forcings are significant physical factors that influence the variation of sea surface temperature (SST) and are often used as essential input variables for ocean numerical models. However, their contribution to the prediction of SST based on machine-learning methods still needs to be tested. This study presents a prediction model for SST in the East China Sea (ECS) using two machine-learning methods: Random Forest and SA-ConvLSTM algorithms. According to the Random Forest feature importance scores and correlation coefficients R, 2 m air temperature and longwave radiation were selected as the two most important key atmospheric factors that can affect the SST prediction performance of machine-learning methods. Four datasets were constructed as input to SA-ConvLSTM: SST-only, SST-T2m, SST-LWR, and SST-T2m-LWR. Using the SST-T2m and SST-LWR, the prediction skill of the model can be improved by about 9.9% and 9.43% for the RMSE and by about 8.97% and 8.21% for the MAE, respectively. Using the SST-T2m-LWR dataset, the model's prediction skill can be improved by 10.75% for RMSE and 9.06% for MAE. The SA-ConvLSTM can represent the SST in ECS well, but with the highest RMSE and AE in summer. The findings of the presented study requires much more exploration in future studies. [ABSTRACT FROM AUTHOR]

Storm surge disasters result in severe casualties and economic losses. Accurate prediction of storm surge water level is crucial for disaster assessment, early warning, and effective disaster management. Machine learning methods are relatively more efficient and straightforward compared to numerical simulation approaches. However, most of the current research on storm surge water level prediction based on machine learning methods is primarily focused on point predictions. In this study, we explore the feasibility of spatial water level prediction using the ConvLSTM model. We focus on the coastal area of Guangdong Province and employ MIKE21(2019) software to simulate historical typhoons that have made landfall in the region from 1991 to 2018. We construct two datasets: one for direct water level prediction and the other for indirect water level prediction based on water level changes. Utilizing the ConvLSTM network, we employ it to forecast storm surges on both datasets, effectively capturing both temporal and spatial characteristics and thus ensuring the production of dependable results. When directly predicting water levels, we achieve an MAE (mean absolute error) of 0.026 m and an MSE (mean squared error) of 0.0038 m2. In contrast, the indirect prediction approach yields even more promising results, with an MAE of 0.014 m and an MSE of 0.0007 m2. Compared to traditional numerical simulation methods, the ConvLSTM-based approach is simpler, faster, and able to predict water levels accurately without boundary conditions or topographies. Furthermore, we consider worst-case scenarios by predicting the maximum water increase value using the random forest model. Our results indicate that the random forest model can serve as a valuable reference for forecasting the maximum water increase value of typhoon storm surges, supporting effective emergency responses to disasters. [ABSTRACT FROM AUTHOR]

The Kerkennah archipelago in Tunisia is one of the most vulnerable areas where the influence of climate change is undeniable. Soil salinization has emerged as a major consequence of climate variation on this island. In this study, remote sensing techniques were implemented to develop a model for predicting soil salinity from satellite images. Machine learning algorithms, Sentinel-1 and Sentinel-2 data, and ground truth measurements were used to estimate soil salinity. Several algorithms were considered to achieve accurate findings. These algorithms are categorized as polynomial regression, random forest regression, exponential regression, and linear regression. The results demonstrate that exponential regression is the pre-eminent algorithm for estimating soil salinity with high predictive accuracy of R2 = 0.75 and RMSE = 0.47 ds/m. However, spatiotemporal soil salinity maps reveal distinct and clear distribution patterns, highlighting salty areas (i.e., sebkhas) and agricultural parcels. Thus, through the model, we explore areas of moderately high salinity within agricultural lands that could be affected by irrigation practices. The present work demonstrates a reliable model for soil salinity monitoring in the Kerkennah archipelago and inspires more successful technologies such as remote sensing and machine learning to improve the estimation of soil salinity in climate-affected vulnerable areas. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Over the years, rampant wildfires have plagued the state of California, creating economic and environmental loss. In 2018, wildfires cost nearly 800 million dollars in economic loss and claimed more than 100 lives in California. Over 1.6 million acres of land has burned and caused large sums of environmental damage. Although, recently, researchers have introduced machine learning models and algorithms in predicting the wildfire risks, these results focused on special perspectives and were restricted to a limited number of data parameters. In this paper, we have proposed two data-driven machine learning approaches based on random forest models to predict the wildfire risk at areas near Monticello and Winters, California. This study demonstrated how the models were developed and applied with comprehensive data parameters such as powerlines, terrain, and vegetation in different perspectives that improved the spatial and temporal accuracy in predicting the risk of wildfire including fire ignition. The combined model uses the spatial and the temporal parameters as a single combined dataset to train and predict the fire risk, whereas the ensemble model was fed separate parameters that were later stacked to work as a single model. Our experiment shows that the combined model produced better results compared to the ensemble of random forest models on separate spatial data in terms of accuracy. The models were validated with Receiver Operating Characteristic (ROC) curves, learning curves, and evaluation metrics such as: accuracy, confusion matrices, and classification report. The study results showed and achieved cutting-edge accuracy of 92% in predicting the wildfire risks, including ignition by utilizing the regional spatial and temporal data along with standard data parameters in Northern California. [ABSTRACT FROM AUTHOR]

Surface downward longwave radiation (SDLR) is significant with regard to surface energy budgets and climate research. The uncertainty of cloud base height (CBH) retrieval by remote sensing induces the vast majority of SDLR estimation errors under cloudy conditions; reliable CBH observation and estimation are crucial for determining the cloud radiative effect. This study presents a CBH retrieval methodology built from 10 thermal spectral data from Himawari-8 (H-8) observations, utilizing the random forest (RF) algorithm to fully account for each band's contribution to CBH. The algorithm utilizes only infrared band data, making it possible to obtain CBH 24 h a day. Considering some factors that can significantly affect the CBH estimation, RF models are trained for different clouds using inputs from multiple H-8 channels together with geolocation information to target CBH derived from CloudSat/CALIPSO combined measurements. The validation results reveal that the new methodology performs well, with a root-mean-square error (RMSE) of only 1.17 km for all clouds. To evaluate the effect of CBH on SDLR estimation, an all-sky SDLR estimation algorithm based on previous CBH predictions is proposed. The new SDLR product not only has a resolution that is noticeably higher than that of benchmark products of the SDLR, such as the Clouds and the Earth's Radiant Energy System (CERES) and the next-generation reanalysis (ERA5) of the European Centre for Medium-Range Weather Forecasts (ECMWF), but it also has greater accuracy, with an RMSE of 21.8 W m−2 for hourly surface downward longwave irradiance (SDLI). [ABSTRACT FROM AUTHOR]

Pilots typically implement the go-around protocol to avoid landings that are hazardous due to wind shear, runway excursions, or unstable approaches. Despite its rarity, it is essential for safety. First, in this study, we present three Dynamic Ensemble Selection (DES) frameworks: Meta-Learning for Dynamic Ensemble Selection (META-DES), Dynamic Ensemble Selection Performance (DES-P), and K-Nearest Oracle Elimination (KNORAE), with homogeneous and heterogeneous pools of machine learning classifiers as base estimators for the prediction of aircraft go-around in wind shear (WS) events. When generating a prediction, the DES approach automatically selects the subset of machine learning classifiers which is most probable to perform well for each new test instance to be classified, thereby making it more effective and adaptable. In terms of Precision (86%), Recall (83%), and F1-Score (84%), the META-DES model employing a pool of Random Forest (RF) classifiers outperforms other models. Environmental and situational factors are subsequently assessed using SHapley Additive exPlanations (SHAP). The wind shear magnitude, corridor, time of day, and WS altitude had the greatest effect on SHAP estimation. When a strong tailwind was present at low altitude, runways 07R and 07C were highly susceptible to go-arounds. The proposed META-DES with a pool of RF classifiers and SHAP for predicting aircraft go-around in WS events may be of interest to researchers in the field of air traffic safety. [ABSTRACT FROM AUTHOR]

This work focuses on the prediction of an air pollutant called particulate matter (PM 2.5 ) across the Paso Del Norte region. Outdoor air pollution causes millions of premature deaths every year, mostly due to anthropogenic fine PM 2.5 . In addition, the prediction of ground-level PM 2.5 is challenging, as it behaves randomly over time and does not follow the interannual variability. To maintain a healthy environment, it is essential to predict the PM 2.5 value with great accuracy. We used different supervised machine learning algorithms based on regression and classification to accurately predict the daily PM 2.5 values. In this study, several meteorological and atmospheric variables were retrieved from the Texas Commission of Environmental Quality's monitoring stations corresponding to 2014–2019. These variables were analyzed by six different machine learning algorithms with various evaluation metrics. The results demonstrate that ML models effectively detect the effect of other variables on PM 2.5 and can predict the data accurately, identifying potentially risky territory. With an accuracy of 92%, random forest performs the best out of all machine learning models. [ABSTRACT FROM AUTHOR]

This study proposes a random forest–random pixel ID (RF–RID) method, which could reduce local anomalies in the simulation of NO2 spatial distribution and significantly improve prediction accuracy in rural areas. First, the 470 nm MAIAC AOD and OMI NO2 total and tropospheric vertical column were packed using the two-step method (TWS). Second, using RID, the filled data and auxiliary variables were combined with random forest (RF) to build an RF–RID model to predict the 1 km/d NO2 spatial distribution in southwestern Fujian (SWFJ) in 2018. The results show that the RF–RID achieves enhanced performance in the CV of the observed sample (R = 0.9117, RMSE = 3.895). Meanwhile, RF–RID has a higher correlation with the road length (RL) in remote areas, and the proposed method solves the issue related to strips or patches of NO2 spatial distribution. This model offers insights into the related research on air pollutants in large areas. [ABSTRACT FROM AUTHOR]

Vehicle air pollution is a significant problem for health and climate change that can be solved by several approaches. The route is one of the many components to be considered. In this work, we propose a statistical analysis of a large FCD database in November 2017 in Lyon (France) in order to find alternative sustainable trips and evaluate potential emission reductions (CO2, NOx, PM10). To this end, an innovative framework was built. First, we assessed vehicle speeds for each network section and the fifteen-minute period, when this information was reachable. Then, we used a regression random forest (RF) algorithm to fill in the missing data. This dynamical speed map allowed us to search for fewer pollutant trips, for the first ten days of November. By using COPERT emission factors (EFs) and the time-dependent Dijkstra algorithm, we successfully identified between 51% and 72% of alternative sustainable paths, depending on the engine technology and the pollutant. We investigated the influence of vehicle technology. In all cases, the number of alternative trips found tends to be the same as soon as the emission savings exceed 5%. Moreover, about 400 trips out of 11,000 have the potential to mitigate about 20% of emissions. [ABSTRACT FROM AUTHOR]

Potential evapotranspiration (PET) is an important input variable of many ecohydrological models, but commonly used empirical models usually input numerous meteorological factors. In consideration of machine learning for complex nonlinear learning, we evaluated the applicability of three machine learning algorithms in PET estimation in the Yellow River basin (YRB), in addition to determining significant factors affecting the accuracy of machine learning. Furthermore, the importance of meteorological factors at varying altitudes and drought index grades for PET simulation were evaluated. The results show that the accuracy of PET simulation in the YRB depends on the input of various meteorological factors; however, machine learning models including average temperature (Tmean) and sunshine hours (n) as input achieved satisfactory accuracy in the absence of complete meteorological data. Random forest generally performed best among all investigated models, followed by extreme learning machine, whereas empirical models overestimated or underestimated PET. The importance index shows that Tmean is the most influential factor with respect to PET, followed by n, and the influence of Tmean on PET gradually decreased with increased altitude and drier climate, whereas the influence of n shows the opposite trend. [ABSTRACT FROM AUTHOR]

The development of machine learning (ML) provides new means and methods for accurate climate analysis and prediction. This study focuses on summer precipitation prediction using ML algorithms. Based on BCC CSM1.1, ECMWF SEAS5, NCEP CFSv2, and JMA CPS2 model data, we conducted a multi-model ensemble (MME) prediction experiment using three tree-based ML algorithms: the decision tree (DT), random forest (RF), and adaptive boosting (AB) algorithms. On this basis, we explored the applicability of ML algorithms for ensemble prediction of seasonal precipitation in China, as well as the impact of different hyperparameters on prediction accuracy. Then, MME predictions based on optimal hyperparameters were constructed for different regions of China. The results showed that all three ML algorithms had an optimal maximum depth less than 2, which means that, based on the current amount of data, the three algorithms could only predict positive or negative precipitation anomalies, and extreme precipitation was hard to predict. The importance of each model in the ML-based MME was quantitatively evaluated. The results showed that NCEP CFSv2 and JMA CPS2 had a higher importance in MME for the eastern part of China. Finally, summer precipitation in China was predicted and tested from 2019 to 2021. According to the results, the method provided a more accurate prediction of the main rainband of summer precipitation in China. ML-based MME had a mean ACC of 0.3, an improvement of 0.09 over the weighted average MME of 0.21 for 2019–2021, exhibiting a significant improvement over the other methods. This shows that ML methods have great potential for improving short-term climate prediction. [ABSTRACT FROM AUTHOR]

Despite the levels of air pollution in Macao continuing to improve over recent years, there are still days with high-pollution episodes that cause great health concerns to the local community. Therefore, it is very important to accurately forecast air quality in Macao. Machine learning methods such as random forest (RF), gradient boosting (GB), support vector regression (SVR), and multiple linear regression (MLR) were applied to predict the levels of particulate matter (PM10 and PM2.5) concentrations in Macao. The forecast models were built and trained using the meteorological and air quality data from 2013 to 2018, and the air quality data from 2019 to 2021 were used for validation. Our results show that there is no significant difference between the performance of the four methods in predicting the air quality data for 2019 (before the COVID-19 pandemic) and 2021 (the new normal period). However, RF performed significantly better than the other methods for 2020 (amid the pandemic) with a higher coefficient of determination (R2) and lower RMSE, MAE, and BIAS. The reduced performance of the statistical MLR and other ML models was presumably due to the unprecedented low levels of PM10 and PM2.5 concentrations in 2020. Therefore, this study suggests that RF is the most reliable prediction method for pollutant concentrations, especially in the event of drastic air quality changes due to unexpected circumstances, such as a lockdown caused by a widespread infectious disease. [ABSTRACT FROM AUTHOR]

Accurate, timely air quality index (AQI) forecasting helps industries in selecting the most suitable air pollution control measures and the public in reducing harmful exposure to pollution. This article proposes a comprehensive method to forecast AQIs. Initially, the work focused on predicting hourly ambient concentrations of PM2.5 and PM10 using artificial neural networks. Once the method was developed, the work was extended to the prediction of other criteria pollutants, i.e., O3, SO2, NO2, and CO, which fed into the process of estimating AQI. The prediction of the AQI not only requires the selection of a robust forecasting model, it also heavily relies on a sequence of pre-processing steps to select predictors and handle different issues in data, including gaps. The presented method dealt with this by imputing missing entries using missForest, a machine learning-based imputation technique which employed the random forest (RF) algorithm. Unlike the usual practice of using RF at the final forecasting stage, we utilized RF at the data pre-processing stage, i.e., missing data imputation and feature selection, and we obtained promising results. The effectiveness of this imputation method was examined against a linear imputation method for the six criteria pollutants and the AQI. The proposed approach was validated against ambient air quality observations for Al-Jahra, a major city in Kuwait. Results obtained showed that models trained using missForest-imputed data could generalize AQI forecasting and with a prediction accuracy of 92.41% when tested on new unseen data, which is better than earlier findings. [ABSTRACT FROM AUTHOR]

Dust weather is common and disastrous at the Tibetan Plateau. This study selected a typical case of dust weather and analyzed its main development mechanism in the northeast of the Tibetan Plateau, then applied six machine learning methods and a time series regression model to predict PM10 concentration in this area. The results showed that: (1) The 24-h pressure change was positive when the front intruded on the surface; convergence of vector winds with a sudden drop in temperature and humidity led by a trough on 700 hPa; a "two troughs and one ridge" weather situation appeared on 500 hPa while the cold advection behind the trough was strong and a cyclone vorticity was formed in the east of Inner Mongolia. (2) The trajectory of air mass from the Hexi Corridor was the main air mass path influencing Xining City, in this case, since a significant lag in the peak of PM10 concentration appeared in Xining City when compared with Zhangye City. (3) The Multiple Linear Regression was not only timely and effective in predicting the PM10 concentration but had great abilities for anticipating the transition period of particle concentration and the appearance date of maximum values in such dust weather. (4) The MA and MP in the clean period were much lower than that in the dust period; the PM10 of Zhangye City as an eigenvalue played an important role in predicting the PM10 of Xining City even in clean periods. Different from dust periods, the prediction effect of Random Forest Optimized by Bayesian hyperparameter was superior to Multiple Linear Regression in clean periods. [ABSTRACT FROM AUTHOR]

The spatiotemporal variation of PM2.5 should be accurately estimated for epidemiological studies. However, the accuracy of prediction models may change over geographical space, which is not conducive for proper exposure assessment. In this study, we developed a prediction model to estimate daily PM2.5 concentrations from 2010 to 2017 in the Kansai region of Japan with co-existing pollutant concentrations as predictors. The overall objective was to obtain daily estimates over the study domain with spatially homogeneous accuracy. We used random forest algorithm to model the relationship between the daily PM2.5 concentrations and various predictors. The model performance was evaluated via spatial and temporal cross-validation and the daily PM2.5 surface was estimated from 2010 to 2017 at a 1 km × 1 km resolution. We achieved R2 values of 0.91 and 0.92 for spatial and temporal cross-validation, respectively. The prediction accuracy for each monitoring site was found to be consistently high, regardless of the distance to the nearest monitoring location, up to 10 km. Even for distances greater than 10 km, the mean R2 value was 0.88. Our approach yielded spatially homogeneous prediction accuracy, which is beneficial for epidemiological studies. The daily PM2.5 estimates will be used in a related birth cohort study to evaluate the potential impact on human health. [ABSTRACT FROM AUTHOR]

Three Microwave Radiation Imagers (MWRI) were carried onboard the FengYun-3B/C/D satellites and have collected more than 10 years of data since 2010. To create a robust climate quality of data, MWRI level one data were reprocessed with new calibration. This study evaluates the performance of retrieving global soil moisture from recalibrated MWRI data (RCD) and quantifies the difference of retrieved soil moisture between operational calibration data (OCD) and RCD. Soil Moisture Operational Products System (SMOPS) products from NOAA on four days of different seasons were collocated with MWRI brightness temperatures, and then the collocated data were used for training an algorithm through machine learning. The retrieved soil moisture products using OCD and RCD were evaluated against the independent SMOPS products, in situ networks and SMAP soil moisture product. It is shown that the algorithm from the random forest is suitable for FY-3D recalibrated MWRI data, with a coefficient of determination (R2) of 0.7223, a mean bias of −0.0062 and an unbiased root mean square difference (ubRMSD) of 0.0476 m3 m−3 compared with SMOPS products over the period from 12 July 2018 to 31 December 2019. The difference of retrieved soil moisture using OCD and RCD is spatially heterogeneous. Both temporal and spatial coverage and accuracy of the existing FY-3D operational soil moisture products are significantly improved. [ABSTRACT FROM AUTHOR]

Poor air quality is a major environmental concern worldwide, but people living in low- and middle-income countries are disproportionately affected. Measurement of PM2.5 is essential for establishing regulatory standards and developing policy frameworks. Low-cost sensors (LCS) can construct a high spatiotemporal resolution PM2.5 network, but the calibration dependencies and subject to biases of LCS due to variable meteorological parameters limit their deployment for air-quality measurements. This study used data collected from June 2019 to April 2021 from a PurpleAir Monitor and Met One Instruments' Model BAM 1020 as a reference instrument at Alberta, Canada. The objective of this study is to identify the relevant meteorological parameters for each season that significantly affect the performance of LCS. The meteorological features considered are relative humidity (RH), temperature (T), wind speed (WS) and wind direction (WD). This study applied Multiple Linear Regression (MLR), k-Nearest Neighbor (kNN), Random Forest (RF) and Gradient Boosting (GB) models with varying features in a stepwise manner across all the seasons, and only the best results are presented in this study. Improvement in the performance of calibration models is observed by incorporating different features for different seasons. The best performance is achieved when RF is applied but with different features for different seasons. The significant meteorological features are PM2.5_LCS in Summer, PM2.5_LCS, RH and T in Autumn, PM2.5_LCS, T and WS in Winter and PM2.5_LCS, RH, T and WS in Spring. The improvement in R2 for each season (values in parentheses) is Summer (0.66–0.94), Autumn (0.73–0.96), Winter (0.70–0.95) and Spring (0.70–0.94). This study signifies selecting the right combination of models and features to attain the best results for LCS calibration. [ABSTRACT FROM AUTHOR]

Early warning and forecasting of tornadoes began to combine artificial intelligence (AI) and machine learning (ML) algorithms to improve identification efficiency in the past few years. Applying machine learning algorithms to detect tornadoes usually encounters class imbalance problems because tornadoes are rare events in weather processes. The ADASYN-LOF algorithm (ALA) was proposed to solve the imbalance problem of tornado sample sets based on radar data. The adaptive synthetic (ADASYN) sampling algorithm is used to solve the imbalance problem by increasing the number of minority class samples, combined with the local outlier factor (LOF) algorithm to denoise the synthetic samples. The performance of the ALA algorithm is tested by using the supporting vector machine (SVM), artificial neural network (ANN), and random forest (RF) models. The results show that the ALA algorithm can improve the performance and noise immunity of the models, significantly increase the tornado recognition rate, and have the potential to increase the early tornado warning time. ALA is more effective in preprocessing imbalanced data of SVM and ANN, compared with ADASYN, Synthetic Minority Oversampling Technique (SMOTE), SMOTE-LOF algorithms. [ABSTRACT FROM AUTHOR]

The PM10 prediction has received considerable attention due to its harmful effects on human health. Machine learning approaches have the potential to predict and classify future PM10 concentrations accurately. Therefore, in this study, three machine learning algorithms—namely, decision tree (DT), boosted regression tree (BRT), and random forest (RF)—were applied for the prediction of PM10 in Kota Bharu, Kelantan. The results from these three methods were compared to find the best method to predict PM10 concentration for the next day by using the maximum daily data from January 2002 to December 2017. To this end, 80% of the data were used for training and 20% for validation of the models. The performance measure of the PM10 concentration was based on accuracy, sensitivity, specificity, and precision for RF, BRT, and DT, respectively, which indicates that these three models were developed effectively, and they are applicable in the prediction of other atmospheric environmental data. The best model to use in predicting the next day's PM10 concentration classification was the random forest classifier, with an accuracy of 98.37, sensitivity of 97.19, specificity of 99.55, and precision of 99.54, but the result of the boosted regression tree was substantially different from the RF model, with an accuracy of 98.12, sensitivity of 97.51, specificity of 98.72, and precision of 98.71. The best model can assist local governments in providing early warnings to people who are at risk of acute and chronic health consequences from air pollution. [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]

Although combustion is considered a common source of ammonia (NH3) in the atmosphere, field measurements quantifying such emissions of NH3 are still lacking. In this study, online measurements of NH3 were performed by a cavity ring-down spectrometer, in the cold season at a rural site in Xianghe on the North China Plain. We found that the NH3 concentrations were mostly below 65 ppb during the study period. However, from 18 to 21 November 2017, a close burn event (~100 m) increased the NH3 concentrations to 145.6 ± 139.9 ppb. Using a machine-learning technique, we quantified that this burn event caused a significant increase in NH3 concentrations by 411%, compared with the scenario without the burn event. In addition, the ratio of ∆NH3/∆CO during the burn period was 0.016, which fell in the range of biomass burning. Future investigations are needed to evaluate the impacts of the NH3 combustion sources on air quality, ecosystems, and climate in the context of increasing burn events worldwide. [ABSTRACT FROM AUTHOR]

Extreme weather phenomena such as wind gusts, heavy precipitation, hail, thunderstorms, tornadoes, and many others usually occur when there is a change in air mass and the passing of a weather front over a certain region. The climatology of weather fronts is difficult, since they are usually drawn onto maps manually by forecasters; therefore, the data concerning them are limited and the process itself is very subjective in nature. In this article, we propose an objective method for determining the position of weather fronts based on the random forest machine learning technique, digitized fronts from the DWD database, and ERA5 meteorological reanalysis. Several aspects leading to the improvement of scores are presented, such as adding new fields or dates to the training database or using the gradients of fields. [ABSTRACT FROM AUTHOR]

This study presents a comprehensive investigation of multiple Artificial Intelligence (AI) techniques—decision tree, random forest, gradient boosting, and neural network—to generate improved precipitation estimates over the Upper Blue Nile Basin. All the AI methods merged multiple satellite and atmospheric reanalysis precipitation datasets to generate error-corrected precipitation estimates. The accuracy of the model predictions was evaluated using 13 years (2000–2012) of ground-based precipitation data derived from local rain gauge networks in the Upper Blue Nile Basin region. The results indicate that merging multiple sources of precipitation substantially reduced the systematic and random error statistics in the Upper Blue Nile Basin. The proposed methods have great potential in predicting precipitation over the complex terrain region. [ABSTRACT FROM AUTHOR]

Tropical overshooting convection has a strong impact on both heat budget and moisture distribution in the upper troposphere and lower stratosphere, and it can pose a great risk to aviation safety. Cloud-top height is one of the essential concerns of overshooting convection for both the climate system and the aviation weather forecast. The main purpose of our work is to verify the application of the machine learning method, taking the random forest (RF) model as an instance, in overshooting cloud-top height retrieval from Himawari-8 data. By using collocated CloudSat observations as a reference, we utilize several infrared indicators of Himawari-8 that are commonly recognized to relate to cloud-top height, along with some temporal and geographical parameters (latitude, month, satellite zenith angle, etc.), as predictors to construct and validate the model. Analysis of variable importance shows that the brightness temperature of 6.2 um acts as the dominant predictor, followed by satellite zenith angle, brightness temperature of 13.3 um, latitude, and month. In the comparison between the RF model and the traditional single-channel interpolation method, retrievals from the RF model agree well with observation with a high correlation coefficient (0.92), small RMSE (222 m), and small MAE (164 m), while these metrics from traditional single-channel interpolation method shows lower skills (0.70, 1305 m, and 1179 m). This work presents a new sight of overshooting cloud-top height retrieval based on the machine learning method. [ABSTRACT FROM AUTHOR]

The downward longwave radiation (Ld, 4–100 μm) is a major component of research for the surface radiation energy budget and balance. In this study, we applied five machine learning methods, namely artificial neural network (ANN), support vector regression (SVR), gradient boosting regression tree (GBRT), random forest (RF), and multivariate adaptive regression spline (MARS), to estimate Ld using ground measurements collected from 27 Baseline Surface Radiation Network (BSRN) stations. Ld measurements in situ were used to validate the accuracy of Ld estimation models on daily and monthly time scales. A comparison of the results demonstrated that the estimates on the basis of the GBRT method had the highest accuracy, with an overall root-mean-square error (RMSE) of 17.50 W m−2 and an R value of 0.96 for the test dataset on a daily time scale. These values were 11.19 W m−2 and 0.98, respectively, on a monthly time scale. The effects of land cover and elevation were further studied to comprehensively evaluate the performance of each machine learning method. All machine learning methods achieved better results over the grass land cover type but relatively worse results over the tundra. GBRT, RF, and MARS methods were found to show good performance at both the high- and low-altitude sites. [ABSTRACT FROM AUTHOR]

Meteorological disasters are the result of the interaction of multiple factors and multiple systems. In order to improve the accuracy of prediction, it is necessary not only to consider the characteristics and cycles of each subsystem, but also to study the interaction of all systems. Based on the summer precipitation data and 130 circulation indexes of 34 national meteorological observation stations in Chongqing from 1961 to 2010, the prediction model of Chongqing summer precipitation was established based on the decision tree and the stochastic forest algorithm based on machine learning, and the prediction test of 2011–2018 was carried out independently by the model. Compared with the results of the single-factor prediction model, the trend consistency rate increased by 37.5% and 12.5% respectively. In addition, when using the random forest model to predict summer precipitation in Chongqing from 2014 to 2018, the 5-year average Ps, Cc and PC scores were 84.6, 0.27 and 67.1, respectively, which were significantly improved compared with 72.4, −0.12 and 52.9 of the current climate forecasting methods, and the forecast quality of the random forest was relatively stable. The multi-system collaborative impact model based on decision tree and random forest algorithm can achieve high accuracy and stability. Thus, this method can not only be an effective means for the diagnosis and prediction of climate causes, but also has a good theoretical and practical value for the prediction of extreme disasters. [ABSTRACT FROM AUTHOR]