Comparison of Multiple Linear Regression, Cubist Regression, and Random Forest Algorithms to Estimate Daily Air Surface Temperature from Dynamic Combinations of MODIS LST Data

Reconstruction of all-sky daily air temperature datasets with high accuracy in China from 2003 to 2022

A high-accuracy, continuous air temperature (Ta) dataset with high spatiotemporal resolution is essential for human health, disease prediction, and energy management. Existing datasets consider factors such as elevation, latitude, and surface temperature but insufficiently address meteorological and spatiotemporal factors, affecting accuracy. Additionally, no high-resolution dataset currently includes daily maximum (Tmax), minimum (Tmin), and mean (Tmean) temperatures generated using a unified methodology. Here, we introduce the four-dimensional spatiotemporal deep forest (4D-STDF) model, integrating 12 multisource factors, encompassing static and dynamic parameters, and six refined spatiotemporal factors to produce Ta datasets. This approach generates three high-accuracy Ta datasets at 1 km spatial resolution covering mainland China from 2003 to 2022. These datasets, in GeoTIFF format with WGS84 projection, comprise daily Tmax, Tmin, and Tmean. The overall RMSE are 1.49 °C, 1.53 °C, and 1.18 °C for the estimates. The 4D-STDF model can also be applied to other regions with sparse meteorological stations.

Estimation of gross calorific value of coal based on the cubist regression model

The gross calorific value (GCV) of coal is an important parameter for evaluating coal quality, and regression analysis methods can be used to predict GCV. In this study, we proposed a GCV prediction model based on cubist regression. To develop a good regression model, feature selection of input variables was performed using a correlation analysis and a recursive feature elimination algorithm. Thus, in this study, we determined three sets of variables as the optimal combination for regression models: proximate analysis variables (Set 1: moisture, standard ash, and volatile matter), element analysis variables (Set 2: carbon, sulfur, and oxygen), and comprehensive index variables (Set 3: carbon, volatile matter, standard ash, sulfur, moisture, and hydrogen). Results for comparison with multiple linear regression, random forest regression, and numerous previous prediction models, such as gradient boosting regression tree, support vector regression (SVR), backpropagation neural networks, and particle swarm optimization-artificial neural network (PSO-ANN), indicate that these seven regression models have the best fitting effect on the comprehensive index variables among the three sets of input variables. The cubist model showed higher prediction accuracy and lower error than most other models (R2, mean absolute error, root mean square error, and average absolute relative deviation percentage values are 0.990, 0.476, 0.668, and 0.086% for the proximate analysis variables; 0.992, 0.381, 0.596, and 0.140% for element analysis variables; and 0.999, 0.161, 0.219, and 0.087% for comprehensive index variables, respectively). The cubist model combines the advantages of decision tree and linear regression, which not only enables it to perform well in terms of accuracy but also makes the model highly interpretable because it is based on multiple sublinear equations. In addition, the cubist model shows obvious advantages in terms of running speed, especially compared with SVR and PSO-ANN, which require complex parameter optimization. In summary, the cubist model considers the prediction accuracy, model interpretability, and computational efficiency as well as provides a new and effective method for GCV prediction.

Accurate prediction of spatial distribution of soil potentially toxic elements using machine learning and associated key influencing factors identification: A case study in mining and smelting area in southwestern China

Accurate prediction of spatial distribution of potentially toxic elements (PTEs) is crucial for soil pollution prevention and risk control. Achieving accurate prediction of spatial distribution of soil PTEs at a large scale using conventional methods presents significant challenges. In this study, machine learning (ML) models, specially artificial neural network (ANN), random forest (RF), and extreme gradient boosting (XGB), were used to predict spatial distribution of soil PTEs and identify associated key factors in mining and smelting area located in Yunnan Province, China, under the three scenarios: (1) natural + socioeconomic + spatial datasets (NS), (2) NS + irrigation pollution index (IPI) datasets, (3) NS + IPI + deposition (DEPO) datasets. The results highlighted the combination of NS+IPI+DEPO yielded the highest predictive accuracy across ML models. Particularly, XGB exhibited the highest performance for As (R2 =0.7939), Cd (R2 =0.6679), Cu (R2 =0.8519), Pb (R2 =0.8317), and Zn (R2 =0.7669), whereas RF performed the best for Ni (R2 =0.7146). The feature importance and Shapley additive explanation (SHAP) analysis revealed that DEPO and IPI were the pivotal factors influencing the distribution of soil PTEs. Our findings highlighted the important role of DEPO in spatial distribution prediction of soil PTEs, which has often been ignored in previous studies.

Evaluation of fused multisource data of air temperature based on dropsonde and satellite observation

Continuous vertical air temperature (AT) from in-situ observation is of crucial importance for understanding the atmospheric environment, but the satellite data that have complete spatial coverage lacked vertical in-situ observation data, and the vertical dropsonde data from in-situ observations only were single-point observations. Therefore, this article introduced machine learning algorithms for fusing in-situ observation and multi-satellite data to achieve spatial continuity of vertical data on a large scale. Specially, random forest (RF), support vector regression (SVR), artificial neural network (ANN) and recurrent neural network (RNN) were employed to capture the non-linear relationships between the variables and estimated AT. The pre-training process and fine-tuning process ensured the prediction of AT spatiotemporal distribution. The four models were implemented for three-dimensional AT estimating across China. Additionally, we used the radiosonde observation data to evaluate the accuracy of estimated AT data under conventional weather and typhoon conditions. Our results revealed that the RF model performed the best with the R of 0.9992, the MAE of 0.70 °C, and the RMSE of 1.04 °C at the national scale, followed by the SVR and ANN models. The RNN model exhibited promising results under typhoon conditions, which will be valuable insights for further research on the applicability of machine learning models under different weather conditions. Besides, having a larger sample size does not necessarily result in reduced errors. For instance, the MAE value for SVR in the pressure height range of 100-200 hPa was larger than that in the pressure height range of 300-400 hPa, but the former sample size was 16,324, which was 7433 higher than the latter.

Development of a Soft Sensor Using Machine Learning Algorithms for Predicting the Water Quality of an Onsite Wastewater Treatment System

Developing advanced onsite wastewater treatment systems (OWTS) requires accurate and consistent water quality monitoring to evaluate treatment efficiency and ensure regulatory compliance. However, off-line parameters such as chemical oxygen demand (COD), total suspended solids (TSS), and Escherichia coli (E. coli) require sample collection and time-consuming laboratory analyses that do not provide real-time information of system performance or component failure. While real-time COD analyzers have emerged in recent years, they are not economically viable for onsite systems due to cost and chemical consumables. This study aimed to design and implement a real-time remote monitoring system for OWTS by developing several multi-input and single-output soft sensors. The soft sensor integrates data that can be obtained from well-established in-line sensors to accurately predict key water quality parameters, including COD, TSS, and E. coli concentrations. The temporal and spatial water quality data of an existing field-tested OWTS operated for almost two years (n = 56 data points) were used to evaluate the prediction performance of four machine learning algorithms. These algorithms, namely, partial least square regression (PLS), support vector regression (SVR), cubist regression (CUB), and quantile regression neural network (QRNN), were chosen as candidate algorithms for their prior application and effectiveness in wastewater treatment predictions. Water quality parameters that can be measured in-line, including turbidity, color, pH, NH4+, NO3-, and electrical conductivity, were selected as model inputs for predicting COD, TSS, and E. coli. The results revealed that the trained SVR model provided a statistically significant prediction for COD with a mean absolute percentage error (MAPE) of 14.5% and R2 of 0.96. The CUB model provided the optimal predictive performance for TSS, with a MAPE of 24.8% and R2 of 0.99. None of the models were able to achieve optimal prediction results for E. coli; however, the CUB model performed the best with a MAPE of 71.4% and R2 of 0.22. Given the large fluctuation in the concentrations of COD, TSS, and E. coli within the OWTS wastewater dataset, the proposed soft sensor models adequately predicted COD and TSS, while E. coli prediction was comparatively less accurate and requires further improvement. These results indicate that although water quality datasets for the OWTS are relatively small, machine learning-based soft sensors can provide useful predictive estimates of off-line parameters and provide real-time monitoring capabilities that can be used to make adjustments to OWTS operations.

Assessing surface water pollution in Hanoi, Vietnam, using remote sensing and machine learning algorithms

Rapid urbanization led to significant land-use changes and posed threats to surface water bodies worldwide, especially in the Global South. Hanoi, the capital city of Vietnam, has been facing chronic surface water pollution for more than a decade. Developing a methodology to better track and analyze pollutants using available technologies to manage the problem has been imperative. Advancement of machine learning and earth observation systems offers opportunities for tracking water quality indicators, especially the increasing pollutants in the surface water bodies. This study introduces machine learning with the cubist model (ML-CB), which combines optical and RADAR data, and a machine learning algorithm to estimate surface water pollutants including total suspended sediments (TSS), chemical oxygen demand (COD), and biological oxygen demand (BOD). The model was trained using optical (Sentinel-2A and Sentinel-1A) and RADAR satellite images. Results were compared with field survey data using regression models. Results show that the predictive estimates of pollutants based on ML-CB provide significant results. The study offers an alternative water quality monitoring method for managers and urban planners, which could be instrumental in protecting and sustaining the use of surface water resources in Hanoi and other cities of the Global South.

Reference evapotranspiration estimate with missing climatic data and multiple linear regression models

The reference evapotranspiration (ETo) is considered one of the primary variables for water resource management, irrigation practices, agricultural and hydro-meteorological studies, and modeling different hydrological processes. Therefore, an accurate prediction of ETo is essential. A large number of empirical methods have been developed by numerous scientists and specialists worldwide to estimate ETo from different climatic variables. The FAO56 Penman-Monteith (PM) is the most accepted and accurate model to estimate ETo in various environments and climatic conditions. However, the FAO56-PM method requires radiation, air temperature, air humidity, and wind speed data. In this study in Adana Plain, which has a Mediterranean climate for the summer growing season, using 22-year daily climatic data, the performance of the FAO56-PM method was evaluated with different combinations of climatic variables when climatic data were missing. Additionally, the performances of Hargreaves-Samani (HS) and HS (A&G) equations were assessed, and multiple linear regression models (MLR) were developed using different combinations of climatic variables. The FAO56-PM method could accurately estimate daily ETo when wind speed (U) and relative humidity (RH) data were unavailable, using the procedures suggested by FAO56 Paper (RMSEs were smaller than 0.4 mm d^-1, and percent relative errors (REs) were smaller than 9%). Hargreaves-Samani (A&G) and HS equations could not estimate daily ETo accurately according to the statistical indices (RMSEs = 0.772-0.957 mm d^-1; REs (%) = 18.2-22.6; R² = 0.604-0.686, respectively). On the other hand, MLR models' performance varied according to a combination of different climatic variables. According to t-stat and p values of independent variables for MLR models, solar radiation (Rs) and sunshine hours (n) variables had more effect on estimating ETo than other variables. Therefore, the models that used Rs and n data estimated daily ETo more accurately than the others. RMSE values of the models that used Rs were between 0.288 to 0.529 mm d^-1; RE(%) values were between 6.2%-11.5% in the validation process. RMSE values of the models that used n were between 0.457 to 0.750 mm d^-1; RE(%) values were between 9.9%-16.3% in the validation process. The models based only on air temperature had the worst performance (RMSE = 1.117 mm d^-1; RE(%) = 24.2; R² = 0.423).

Sulfur Dioxide (SO2) anthropogenic emissions distributions over Iraq (2000-2009) using MERRA-2 data

The Sulfur dioxide (SO2) is a colorless air pollutant cannot been seen with unaided eye. The fossil fuels burning, including coal, oil and gas, are the largest source of SO2. Often the SO2 Pollution reaches hazardous levels near the coal-fired plants, oil refineries, and in industrialized areas. This study analyzed the trend, spatial and temporal distributions of anthropogenic SO2 emissions in Iraq from January 2000 to December 2009, and series and trend analyses over six stations (Baghdad, Mosul, Basra, Muthanna, Babylon , and Kirkuk) using MERRA-2 data. The monthly SO2 are analyzed for the study period. The SO2 fluctuations were checked, depending on the background of each SO2 sources. The results shows clear reductions of SO2 values from 2002 till 2006, and the SO2 values increases during 2006 to 2009 over all stations. The annual trend analyses shows positive results over Baghdad, Al-Muthanna, and Babylon, and negative results over Basra, Mosul and Kirkuk. A large differences of SO2 values were over Basra, Kirkuk and Babylon, and slight difference over Baghdad, Mosul and Al-muthana. The monthly SO2 anthropogenic emissions values shows relatively stable over most stations, and the only fluctuation over Babylon and Kirkuk during study period. Observed higher SO2 values in the winter and spring than its values in the summer. This research pretends the satellites observation efficiently shows the spatial and temporal variations of SO2 for the considered study area

A method for improving the estimation of extreme air temperature by satellite

Air temperature (Ta) data obtained from meteorological stations were spatially discontinuous. Some satellite data have complete spatial coverage and strong relationships with Ta (e.g., elevation and land surface temperature). Therefore, Ta can be mapped using in situ Ta and satellite data. However, this method may have a large bias when estimating the extreme Ta. In this study, the error prediction and correction (EPC) method, incorporating Cubist machine learning algorithm, was proposed to improve the estimation of extreme Ta. The accuracy of the EPC method was compared with that of the widely used method in previous studies in east China from 2003 to 2012. The mean absolute errors (MAEs) of the estimated daily Ta using the EPC method ranged from 0.75-1.01 °C, which were 0.57-0.96 °C lower than that of the method in the literature. The biases of the estimated Ta obtained using the two methods were close to zero. However, the biases can be as high as 7.10 °C when Ta is extremely low and as low as -3.09 °C when Ta is extremely high. Compared with the method in the literature, the EPC method can reduce the MAE by 1.41 °C, root mean square error by 1.49 °C, and bias by 1.61 °C of the estimated extreme Ta. Additionally, the EPC method produced satisfactory accuracy (MAEs <0.9 °C) of the estimated heat and cold wave magnitudes. Finally, a 1 km resolution daily Ta map in east China from 2003 to 2012 was developed, which will be useful data in multiple research fields.

Multiple Linear Regression Models with Limited Data for the Prediction of Reference Evapotranspiration of the Peloponnese, Greece

The aim of this study was to investigate the utility of multiple linear regression (MLR) for the estimation of reference evapotranspiration (ETo) of the Peloponnese, Greece, for two representative months of winter and summer during 2016–2019. Another objective was to test the number of inputs needed for satisfactorily accurate estimates via MLR. Datasets from sixty-two meteorological stations were exploited. The available independent variables were sunshine hours (N), mean temperature (Tmean), solar radiation (Rs), net radiation (Rn), wind speed (u2), vapour pressure deficit (es − ea), and altitude (Z). Sixteen MLR models were tested and compared to the corresponding ETo estimates computed by FAO-56 Penman–Monteith (FAO PM) in a previous study, via statistical indices of error and agreement. The MLR5 model with five input variables outperformed the other models (RMSE = 0.28 mm d−1, adj. R2 = 98.1%). Half of the tested models (two to six inputs) exhibited very satisfactory predictions. Models of one input (e.g., N, Rn) were also promising. However, the MLR with u2 as the sole input variable presented the worst performance, probably because its relationship with ETo cannot be linearly described. The results indicate that MLR has the potential to produce very good predictive models of ETo for the Peloponnese, based on the literature standards.

Estimation of Maize Foliar Temperature and Stomatal Conductance as Indicators of Water Stress Based on Optical and Thermal Imagery Acquired Using an Unmanned Aerial Vehicle (UAV) Platform

Climatic variability and extreme weather events impact agricultural production, especially in sub-Saharan smallholder cropping systems, which are commonly rainfed. Hence, the development of early warning systems regarding moisture availability can facilitate planning, mitigate losses and optimise yields through moisture augmentation. Precision agricultural practices, facilitated by unmanned aerial vehicles (UAVs) with very high-resolution cameras, are useful for monitoring farm-scale dynamics at near-real-time and have become an important agricultural management tool. Considering these developments, we evaluated the utility of optical and thermal infrared UAV imagery, in combination with a random forest machine-learning algorithm, to estimate the maize foliar temperature and stomatal conductance as indicators of potential crop water stress and moisture content over the entire phenological cycle. The results illustrated that the thermal infrared waveband was the most influential variable during vegetative growth stages, whereas the red-edge and near-infrared derived vegetation indices were fundamental during the reproductive growth stages for both temperature and stomatal conductance. The results also suggested mild water stress during vegetative growth stages and after a hailstorm during the mid-reproductive stage. Furthermore, the random forest model optimally estimated the maize crop temperature and stomatal conductance over the various phenological stages. Specifically, maize foliar temperature was best predicted during the mid-vegetative growth stage and stomatal conductance was best predicted during the early reproductive growth stage. Resultant maps of the modelled maize growth stages captured the spatial heterogeneity of maize foliar temperature and stomatal conductance within the maize field. Overall, the findings of the study demonstrated that the use of UAV optical and thermal imagery, in concert with prediction-based machine learning, is a useful tool, available to smallholder farmers to help them make informed management decisions that include the optimal implementation of irrigation schedules.

Estimation of Daily and Instantaneous Near-Surface Air Temperature from MODIS Data Using Machine Learning Methods in the Jingjinji Area of China

Meteorologically observed air temperature (Ta) is limited due to low density and uneven distribution that leads to uncertain accuracy. Therefore, remote sensing data have been widely used to estimate near-surface Ta on various temporal scales due to their spatially continuous characteristics. However, few studies have focused on instantaneous Ta when satellites overpass. This study aims to produce both daily and instantaneous Ta datasets at 1 km resolution for the Jingjinji area, China during 2018–2019, using machine learning methods based on remote sensing data, dense meteorological observation station data, and auxiliary data (such as elevation and normalized difference vegetation index). Newly released Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 6 surface Downward Shortwave Radiation (DSR) was introduced to improve the accuracy of Ta estimation. Five machine learning algorithms were implemented and compared so that the optimal one could be selected. The random forest (RF) algorithm outperformed the others (such as decision tree, feedforward neural network, generalized linear model) and RF obtained the highest accuracy in model validation with a daily root mean square error (RMSE) of 1.29 °C, mean absolute error (MAE) of 0.94 °C, daytime instantaneous RMSE of 1.88 °C, MAE of 1.35 °C, nighttime instantaneous RMSE of 2.47 °C, and MAE of 1.83 °C. The corresponding R2 was 0.99 for daily average, 0.98 for daytime instantaneous, and 0.95 for nighttime instantaneous. Analysis showed that land surface temperature (LST) was the most important factor contributing to model accuracy, followed by solar declination and DSR, which implied that DSR should be prioritized when estimating Ta. Particularly, these results outperformed most models presented in previous studies. These findings suggested that RF could be used to estimate daily instantaneous Ta at unprecedented accuracy and temporal scale with proper training and very dense station data. The estimated dataset could be very useful for local climate and ecology studies, as well as for nature resources exploration.

Improved anthropogenic heat flux model for fine spatiotemporal information in Southeast China

Anthropogenic heat emission (AHE) is an important driver of urban heat islands (UHIs). Further, both urban thermal environment research and sustainable development planning require an efficient estimation of anthropogenic heat flux (AHF). Therefore, this study proposed an improved multi-source AHF model, which was constructed using diverse data sources and small-scale samples, to better represent the spatiotemporal distribution of AHF. The performances of three machine learning algorithms (Cubist, gradient boosting decision tree, and simple linear regression) were quantitatively evaluated, and the impact of spatiotemporal heterogeneity on AHF estimation was considered for the first time. The results showed that multi-source datasets and sophisticated algorithms could more effectively reduce the estimation error and improve the accuracy of the spatiotemporal distribution of AHF than simple linear regression. In practical applications, the Cubist model performed better, with prediction errors being less than 0.9 W⋅m^-2. Further, the characteristics of different heat sources from the model outputs varied widely, and the building metabolic heat exhibited significant seasonal spatiotemporal variations, which were largely determined by the regional climate. In contrast, industrial and transportation heat showed marginal monthly fluctuations. Similarly, spatiotemporal heterogeneity significantly affected the estimation of building metabolic heat (0.62 W⋅m^-2), but it did not affect other heat sources. The proposed improved AHF model was verified to effectively capture the spatiotemporal variations of building heat and solve the issue of overestimation of industrial heat in urban regions. This study provides new methods and ideas for the accurate spatiotemporal quantification of AHF that can supplement future studies on climate warming, UHI, and air pollution.

Merging framework for estimating daily surface air temperature by integrating observations from multiple polar-orbiting satellites

Reconstructing spatially continuous surface air temperature (SAT) is of great significance to climate and environmental studies. Substantial efforts have been made to estimate daily SAT based on land surface temperature (LST) derived from polar-orbiting satellites. However, previous studies are nearly all limited to estimating daily SAT based on MODIS LST from NASA's Terra or Aqua by applying different statistical learning methods. Various satellites from earth observation missions, particularly the missions for meteorological satellites, are capable of acquiring thermal infrared observations, but its implications for SAT estimation are significantly ignored. In this study, for the first time, we proposed a merging framework for estimating daily mean SAT by integrating LST datasets from multiple polar-orbiting satellites, including Metop-B from EUMETSAT's Polar System (EPS), SNPP and JPSS-1 from NOAA's Joint Polar Satellites System (JPSS), and Terra and Aqua from NASA's EOS. This study is also the first to explore the estimating of daily SAT based on LST derived from the meteorological satellites in EPS and JPSS. The framework integrates 10 estimation models based on different LST from the five satellites and generates daily merged SAT by averaging the daily SAT estimates from the models. Here we show that the framework significantly improves the spatial coverage of daily SAT estimates for cloud-free areas by an overall increase of 39% with respect to the mean coverage of the LST datasets from the five satellites. Daily coverage of the merged SAT from the framework is nearly all above 75% with an average of 91%. Compared to the SAT estimated from MODIS LST, overall increases in the coverage of daily SAT are 37%-51%. Estimation models in the framework all achieved comparable and satisfactory predicative performances with an average RMSE of 1.7-1.9 K for sample-based cross-validation, and 1.9-2.2 K for site-based cross-validation.

Predicting spatiotemporally-resolved mean air temperature over Sweden from satellite data using an ensemble model

Mapping of air temperature (Ta) at high spatiotemporal resolution is critical to reducing exposure assessment errors in epidemiological studies on the health effects of air temperature. In this study, we applied a three-stage ensemble model to estimate daily mean Ta from satellite-based land surface temperature (Ts) over Sweden during 2001-2019 at a high spatial resolution of 1 × 1 km². The ensemble model incorporated four base models, including a generalized additive model (GAM), a generalized additive mixed model (GAMM), and two machine learning models (random forest [RF] and extreme gradient boosting [XGBoost]), and allowed the weights for each model to vary over space, with the best-performing model for each grid cell assigned the highest weight. Various spatial predictors were included as adjustment variables in all the base models, including land cover type, normalized difference vegetation index (NDVI), and elevation. The ensemble model showed high performance with an overall R² of 0.98 and a root mean square error of 1.38 °C in the ten-fold cross-validation, and outperformed each of the four base models. Although each base model performed well, the two machine learning models (RF [R² = 0.97], XGBoost [R² = 0.98]) had better performance than the two regression models (GAM [R² = 0.95], GAMM [R² = 0.96]). In the machine learning models, Ts was the dominant predictor of Ta, followed by day of year, NDVI, latitude, elevation, and longitude. The highly spatiotemporally-resolved Ta can improve temperature exposure assessment in future epidemiological studies.

A statistical model to predict and analyze air surface temperature based on remotely sensed observations

Air surface temperature (AST) is a crucial importance element for many applications such as hydrology, agriculture, and climate change studies. The aim of this study is to develop regression equation for calculating AST and to analyze and investigate the effects of atmospheric parameters (O₃, CH₄, CO, H₂Ovapor, and outgoing longwave radiation (OLR)) on the AST value in Iraq. Dataset retrieved from the Atmospheric Infrared Sounder (AIRS) at EOS Aqua Satellite, spanning the years of 2003 to 2016, and multiple linear regression were used to achieve the objectives of the study. For the study period, the five atmospheric parameters were highly correlated (R, 0.855-0.958) with predicted AST. Statistical analyses in terms of β showed that OLR (0.310 to 1.053) contributes significantly in enhancing AST values. Comparisons among selected five stations (Mosul, Kanaqin, Rutba, Baghdad, and Basra) for the year 2010 showed a close agreement between the predicted and observed AST from AIRS, with values ranging from 0.9 to 1.5 K and for ground stations data, within 0.9 to 2.6 K. To make more complete analysis, also, comparison between predicted and observed AST from AIRS for four selected month in 2016 (January, April, July, and October) has been carried out. The result showed a high correlation coefficient (R, 0.87 and 0.95) with less variability (RMSE ≤ 1.9) for all months studied, indicating model's capability and accuracy. In general, the results indicate the advantage of using the AIRS data and the regression analysis to investigate the impact of the atmospheric parameters on AST over the study area.

Verification of De-Identification Techniques for Personal Information Using Tree-Based Methods with Shapley Values

With the development of big data and cloud computing technologies, the importance of pseudonym information has grown. However, the tools for verifying whether the de-identification methodology is correctly applied to ensure data confidentiality and usability are insufficient. This paper proposes a verification of de-identification techniques for personal healthcare information by considering data confidentiality and usability. Data are generated and preprocessed by considering the actual statistical data, personal information datasets, and de-identification datasets based on medical data to represent the de-identification technique as a numeric dataset. Five tree-based regression models (i.e., decision tree, random forest, gradient boosting machine, extreme gradient boosting, and light gradient boosting machine) are constructed using the de-identification dataset to effectively discover nonlinear relationships between dependent and independent variables in numerical datasets. Then, the most effective model is selected from personal information data in which pseudonym processing is essential for data utilization. The Shapley additive explanation, an explainable artificial intelligence technique, is applied to the most effective model to establish pseudonym processing policies and machine learning to present a machine-learning process that selects an appropriate de-identification methodology.

COVID-19 Vaccination Effect on Stock Market and Death Rate in India

The COVID-19 epidemic has brought attention to the vulnerability of new illnesses, and immunization remains a viable option for resuming normal life. This paper examines the influence of COVID-19 vaccination on the death rate and the performance of stock market in India. For this study, COVID-19 vaccination and death rate data is gathered from the Ministry of Health and Family Welfare (MoHFW) portal, and the data for the stock index is taken from the Bombay Stock Exchange (BSE), India. In order to achieve a precise representation of feature significance and distribution, EDA (Exploratory Data Analysis) is utilized in this study. The impact of COVID-19 immunization on the mortality rate and stock market index is investigated using both statistical analysis and Machine Learning Regression-based models. The models are remarkably accurate in reproducing actual result. The empirical study suggests that vaccination has a strong positive impact on the stock market and reducing the death rate. Furthermore, the policies recommended by government and monetary authorities coupled with COVID-19 vaccine supported the stock market recovery in pandemic.

Downscaling Building Energy Consumption Carbon Emissions by Machine Learning

The rapid rate of urbanization is causing increasing annual urban energy usage, drastic energy shortages, and pollution. Building operational energy consumption carbon emissions (BECCE) account for a substantial proportion of greenhouse gas emissions, crucially influencing global warming and the sustainability of urban socioeconomic development. As a foundation of building energy conservation, determination of refined statistics of BECCE is attracting increasing attention. However, reliable and accurate representation of BECCE remains lacking. This study proposed an innovative downscaling method to generate a gridded BECCE intensity benchmark dataset with 1 km2 spatial resolution. First, we calculated BECCE at the provincial level by energy balance table application. Second, on the basis of building climate demarcation, partial least squares regression models were used to establish the BECCE behavior equations for three climate regions. Third, Cubist regression models were built, retrieving down scale at the prefecture level to 1 km2 BECCE, which well-captured the complex relationships between BECCE and multisource covariates (i.e., gross domestic product, population, ground surface temperature, heating degree days, and cooling degree days). The downscaled product was verified using anthropogenic heat flux mapping at the same resolution. In comparison with other published pixel-based datasets of building energy usage, the gridded BECCE intensity map produced in this study showed good agreement and high spatial heterogeneity. This new BECCE intensity dataset could serve as a fundamental database for studies on building energy conservation and forecast carbon emissions, and could support decision makers in developing strategies for realizing the CO2 emission peak and carbon neutralization.

Estimation of Apple Tree Leaf Chlorophyll Content Based on Machine Learning Methods

Leaf chlorophyll content (LCC) is one of the most important factors affecting photosynthetic capacity and nitrogen status, both of which influence crop harvest. However, the development of rapid and nondestructive methods for leaf chlorophyll estimation is a topic of much interest. Hence, this study explored the use of the machine learning approach to enhance the estimation of leaf chlorophyll from spectral reflectance data. The objective of this study was to evaluate four different approaches for estimating the LCC of apple tree leaves at five growth stages (the 1st, 2nd, 3rd, 4th and 5th growth stages): (1) univariate linear regression (ULR); (2) multivariate linear regression (MLR); (3) support vector regression (SVR); and (4) random forest (RF) regression. Samples were collected from the leaves on the eastern, western, southern and northern sides of apple trees five times (1st, 2nd, 3rd, 4th and 5th growth stages) over three consecutive years (2016–2018), and experiments were conducted in 10–20-year-old apple tree orchards. Correlation analysis results showed that LCC and ST, LCC and vegetation indices (VIs), and LCC and three edge parameters (TEP) had high correlations with the first-order differential spectrum (FODS) (0.86), leaf chlorophyll index (LCI) (0.87), and (SDr − SDb)/ (SDr + SDb) (0.88) at the 3rd, 3rd, and 4th growth stages, respectively. The prediction models of different growth stages were relatively good. The MLR and SVR models in the LCC assessment of different growth stages only reached the highest R2 values of 0.79 and 0.82, and the lowest RMSEs were 2.27 and 2.02, respectively. However, the RF model evaluation was significantly better than above models. The R2 value was greater than 0.94 and RMSE was less than 1.37 at different growth stages. The prediction accuracy of the 1st growth stage (R2 = 0.96, RMSE = 0.95) was best with the RF model. This result could provide a theoretical basis for orchard management. In the future, more models based on machine learning techniques should be developed using the growth information and physiological parameters of orchards that provide technical support for intelligent orchard management.

Predicting the performance of anaerobic digestion using machine learning algorithms and genomic data

Modeling of anaerobic digestion (AD) is crucial to better understand the process dynamics and to improve the digester performance. This is an essential yet difficult task due to the complex and unknown interactions within the system. The application of well-developed data mining technologies, such as machine learning (ML) and microbial gene sequencing techniques are promising in overcoming these challenges. In this study, we investigated the feasibility of 6 ML algorithms using genomic data and their corresponding operational parameters from 8 research groups to predict methane yield. For classification models, random forest (RF) achieved accuracies of 0.77 using operational parameters alone and 0.78 using genomic data at the bacterial phylum level alone. The combination of operational parameters and genomic data improved the prediction accuracy to 0.82 (p<0.05). For regression models, a low root mean square error of 0.04 (relative root mean square error =8.6%) was acquired by neural network using genomic data at the bacterial phylum level alone. Feature importance analysis by RF suggested that Chloroflexi, Actinobacteria, Proteobacteria, Fibrobacteres, and Spirochaeta were the top 5 most important phyla although their relative abundances were ranging only from 0.1% to 3.1%. The important features identified could provide guidance for early warning and proactive management of microbial communities. This study demonstrated the promising application of ML techniques for predicting and controlling AD performance.

8-Day and Daily Maximum and Minimum Air Temperature Estimation via Machine Learning Method on a Climate Zone to Global Scale

Air temperature (Ta) is a required input in a wide range of applications, e.g., agriculture. Land Surface Temperature (LST) products from Moderate Resolution Imaging Spectroradiometer (MODIS) are widely used to estimate Ta. Previous studies of these products in Ta estimation, however, were generally applied in small areas and with a small number of meteorological stations. This study designed both temporal and spatial experiments to estimate 8-day and daily maximum and minimum Ta (Tmax and Tmin) on three spatial scales: climate zone, continental and global scales from 2009 to 2018, using the Random Forest (RF) method based on MODIS LST products and other auxiliary data. Factors contributing to the relation between LST and Ta were determined based on physical models and equations. Temporal and spatial experiments were defined by the rules of dividing the training and validation datasets for the RF method, in which the stations selected in the training dataset were all included or not in the validation dataset. The RF model was first trained and validated on each spatial scale, respectively. On a global scale, model accuracy with a determination coefficient (R2) > 0.96 and root mean square error (RMSE) < 1.96 °C and R2 > 0.95 and RMSE < 2.55 °C was achieved for 8-day and daily Ta estimations, respectively, in both temporal and spatial experiments. Then the model was trained and cross-validated on each spatial scale. The results showed that the data size and station distribution of the study area were the main factors influencing the model performance at different spatial scales. Finally, the spatial patterns of the model performance and variable importance were analyzed. Both daytime and nighttime LST had a significant contribution in the 8-day Tmax estimation on all the three spatial scales; while their contribution in daily Tmax estimation varied over different continents or climate zones. This study was expected to improve our understanding of Ta estimation in terms of accuracy variations and influencing variables on different spatial and temporal scales. The future work mainly includes identifying underlying mechanisms of estimation errors and the uncertainty sources of Ta estimation from a local to a global scale.

Long-term trends of surface and canopy layer urban heat island intensity in 272 cities in the mainland of China

The canopy layer urban heat island (CLUHI) and surface urban heat island (SUHI) refer to higher canopy layer and land surface temperatures in urban areas than in rural areas, respectively. The long-term trends of CLUHIs are poorly understood at the regional scale. In this study, 1 km resolution air temperature (Ta) data for the 2001-2018 period in the mainland of China were mapped using satellite data and station-based Ta data. Subsequently, the temporal trends of the CLUHI and SUHI intensities (CLUHII and SUHII, respectively) were investigated in 272 cities in the mainland of China. The Ta was estimated with high accuracy, with a root mean square error ranging from 0.370 °C to 0.592 °C. The CLUHII and SUHII increased significantly in over half of the cities in spring and summer, over one-third of the cities in autumn, and over one-fifth of the cities in winter. The trends of the nighttime SUHII were strongly related to the CLUHII calculated using mean and minimum Ta (correlation coefficients ranging from 0.613 to 0.770), whereas the relationships between the trends of the daytime SUHII and CLUHII were relatively weak. Human activities were the major driving forces for the increase in the CLUHII and SUHII. The difference in impervious surfaces between urban and rural areas was significantly correlated with the CLUHII and SUHII in approximately half of the cities. Meteorological factors were significantly correlated with the CLUHII and SUHII in few cities. This study highlights the trends of the significant increase in the CLUHII and SUHII in the mainland of China, which may have negative effects on humans and the environment.

Integrating random forests and propagation models for high-resolution noise mapping

The adverse effects of long-term exposure to environmental noise on human health are of increasing concern. Noise mapping methods such as spatial interpolation and land use regression cannot capture complex relationships between environmental conditions and noise propagation or attenuation in a three-dimension (3D) built environment. In this study, we developed a hybrid approach by combining a traffic propagation model and random forests (RF) machine learning algorithm to map the total environment noise levels for daily average, daytime, nighttime, and day-evening-nighttime at 30 m × 30 m resolution for the island of Montreal, Canada. The propagation model was used to predict traffic noise surfaces using road traffic flow, 3D building information, and a digital elevation model. The traffic noise estimates were compared with ground-based sound-level measurements at 87 points to extract residuals between total environmental noise and traffic noise. Residuals at these points were fit to RF models with multiple environmental and geographic predictor variables (e.g., vegetation index, population density, brightness of nighttime lights, land use types, and distances to noise contour around the airport, bus stops, and road intersections). Using the sound-level measurements as baseline data, the prediction errors, i.e., mean error, mean absolute error, and root mean squared error of daily average noise levels estimated by our hybrid approach was -0.03 dB(A), 2.67 dB(A), and 3.36 dB(A). Combining deterministic and stochastic models can provide accurate total environmental noise estimates for large geographic areas where sound-level measurements are available.

Reference evapotranspiration of Brazil modeled with machine learning techniques and remote sensing

Reference evapotranspiration (ETo) is a fundamental parameter for hydrological studies and irrigation management. The Penman-Monteith method is the standard to estimate ETo and requires several meteorological elements. In developing countries, the number of weather stations is insufficient. Thus, free products of remote sensing with evapotranspiration information must be used for this purpose. In this context, the objective of this study was to estimate monthly ETo from potential evapotranspiration (PET) made available by MOD16 product. In this study, the monthly ETo estimated by Penman-Monteith method was considered as the standard. For this, data from 265 weather station of the National Institute of Meteorology (INMET), spread all over the Brazilian territory, were acquired for the period from 2000 to 2014 (15 years). For these months, monthly PET values from MOD16 product for all Brazil were also downloaded. By using machine learning algorithms and information from WorldClim as covariates, ETo was estimated through images from the MOD16 product. To perform the modeling of ETo, eight regression algorithms were tested: multiple linear regression; random forest; cubist; partial least squares; principal components regression; adaptive forward-backward greedy; generalized boosted regression and generalized linear model by likelihood-based boosting. Data from 2000 to 2012 (13 years) were used for training and data of 2013 and 2014 (2 years) were used to test the models. The PET made available by the MOD16 product showed higher values than those of ETo for different periods and climatic regions of Brazil. However, the MOD16 product showed good correlation with ETo, indicating that it can be used in ETo estimation. All models of machine learning were effective in improving the performance of the metrics evaluated. Cubist was the model that presented the best metrics for r² (0.91), NSE (0.90) and nRMSE (8.54%) and should be preferred for ETo prediction. MOD16 product is recommended to be used to predict monthly ETo, which opens possibilities for its use in several other studies.

Spatial Interconnections of Land Surface Temperatures with Land Cover/Use: A Case Study of Tokyo

As one of the most populated metropolitan areas in the world, the Tokyo Metropolitan Area (TMA) has experienced severe climatic modifications and pressure due to densified human activities and urban expansion. The surface urban heat island (SUHI) phenomenon particularly constitutes a significant threat to human comfort and geo-environmental health in TMA. This study aimed to profile the spatial interconnections between land surface temperature (LST) and land cover/use in TMA from 2001 to 2015 using multi-source spatial data. To this end, the thermal gradients between the urban and non-urban fabric areas in TMA were examined by joint analysis of land cover/use and LST. The spatiotemporal aggregation patterns, variations, and movement trajectories of SUHI intensity in TMA were identified and delineated. The spatial relationship between SUHI and the potential driving forces in TMA was clarified using geographically weighted regression (GWR) analysis. The results show that the thermal environment of TMA exhibited a polynucleated spatial structure with multiple thermal island cores. Overall, the magnitude and extent of SUHI in TMA increased and expanded from 2001 to 2015. During that time, SUHIs clustered in the compact residential quarters and redevelopment/renovation areas rather than downtown. The GWR models showed better performance than ordinary least squares (OLS) models, with Adj R2 > 0.9, indicating that the magnitude of SUHI significantly depended on its neighboring geographical setting, including land cover composition and configuration, population size, and terrain. We suggest that UHI mitigation in Tokyo should be focused on alleviating the magnitude of persistent thermal cores and controlling unstable SUHI occurrence based on partitioned or location-specific landscape design. This study’s findings have immense implications for SUHI mitigation in metropolitan areas situated in bay regions.

Comparative Analysis of Artificial Intelligence Models for Accurate Estimation of Groundwater Nitrate Concentration

Prediction of the groundwater nitrate concentration is of utmost importance for pollution control and water resource management. This research aims to model the spatial groundwater nitrate concentration in the Marvdasht watershed, Iran, based on several artificial intelligence methods of support vector machine (SVM), Cubist, random forest (RF), and Bayesian artificial neural network (Baysia-ANN) machine learning models. For this purpose, 11 independent variables affecting groundwater nitrate changes include elevation, slope, plan curvature, profile curvature, rainfall, piezometric depth, distance from the river, distance from residential, Sodium (Na), Potassium (K), and topographic wetness index (TWI) in the study area were prepared. Nitrate levels were also measured in 67 wells and used as a dependent variable for modeling. Data were divided into two categories of training (70%) and testing (30%) for modeling. The evaluation criteria coefficient of determination (R²), mean absolute error (MAE), root mean square error (RMSE), and Nash–Sutcliffe efficiency (NSE) were used to evaluate the performance of the models used. The results of modeling the susceptibility of groundwater nitrate concentration showed that the RF (R² = 0.89, RMSE = 4.24, NSE = 0.87) model is better than the other Cubist (R² = 0.87, RMSE = 5.18, NSE = 0.81), SVM (R² = 0.74, RMSE = 6.07, NSE = 0.74), Bayesian-ANN (R² = 0.79, RMSE = 5.91, NSE = 0.75) models. The results of groundwater nitrate concentration zoning in the study area showed that the northern parts of the case study have the highest amount of nitrate, which is higher in these agricultural areas than in other areas. The most important cause of nitrate pollution in these areas is agriculture activities and the use of groundwater to irrigate these crops and the wells close to agricultural areas, which has led to the indiscriminate use of chemical fertilizers by irrigation or rainwater of these fertilizers is washed and penetrates groundwater and pollutes the aquifer.

Correlation Analysis between Air Temperature and MODIS Land Surface Temperature and Prediction of Air Temperature Using TensorFlow Long Short-Term Memory for the Period of Occurrence of Cold and Heat Waves

The purpose of this study is to analyze the correlation between surface air temperature (SAT) and land surface temperature (LST) based on land use when heat and cold waves occur and to predict the distribution of SAT using the long short-term memory (LSTM) of TensorFlow. For the correlation analysis, the Terra and Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) daytime and nighttime LST and maximum, minimum, and mean SAT were measured at 79 weather stations of the Korea Meteorological Administration (KMA) from 2008 to 2018. As a result of the correlation analysis between SAT and LST, the maximum SAT (TMX) had a good correlation with the daytime LST of Terra MODIS, with a Pearson’s correlation coefficient (R) of 0.92 and root mean square error (RMSE) of 4.8 °C, and the minimum SAT (TMN) showed a good correlation with the nighttime LST of Terra MODIS, with an R of 0.93 and RMSE of 4.2 °C. When analyzing temperature characteristics by land use (urban, paddy, upland crop, forest, grass, wetland, bare field, and water), it was confirmed that the climate mitigation effect of the wetland and vegetation area appeared in the LSTs and the observed SAT. In the cold wave period, the average temperatures for urban and wetland areas was the highest, and the average temperature for wetland and forest was not higher than that of other land use classes. As the SAT results predicted through the LSTM model, the accuracy of the TMN during the cold wave period was 0.59 for the coefficient of determination (R2), 3.1 °C for RMSE, and 0.76 for the index of agreement (IoA), while the accuracy of the TMX for the heat wave period was 0.24 for R2, 2.23 °C for RMSE, and 0.63 for IoA.

A Framework to Predict High-Resolution Spatiotemporal PM2.5 Distributions Using a Deep-Learning Model: A Case Study of Shijiazhuang, China

Air-borne particulate matter, PM2.5 (PM having a diameter of less than 2.5 micrometers), has aroused widespread concern and is a core indicator of severe air pollution in many cities globally. In our study, we present a validated framework to predict the daily PM2.5 distributions, exemplified by a use case of Shijiazhuang City, China, based on daily aerosol optical depth (AOD) datasets. The framework involves obtaining the high-resolution spatiotemporal AOD distributions, estimation of the spatial distributions of PM2.5 and the prediction of these based on a convolutional long short-term memory (ConvLSTM) model. In the estimation part, the eXtreme gradient boosting (XGBoost) model has been determined as the estimation model with the lowest root mean square error (RMSE) of 32.86 µg/m3 and the highest coefficient of determination regression score function (R2) of 0.71, compared to other common models used as a baseline for comparison (linear, ridge, least absolute shrinkage and selection operator (LASSO) and cubist). For the prediction part, after validation and comparison with a seasonal autoregressive integrated moving average (SARIMA), which is a traditional time-series prediction model, in both time and space, the ConvLSTM gives a more accurate performance for the prediction, with a total average prediction RMSE of 14.94 µg/m3 compared to SARIMA’s 17.41 µg/m3. Furthermore, ConvLSTM is more stable and with less fluctuations for the prediction of PM2.5 in time, and it can also eliminate better the spatial predicted errors compared to SARIMA.

Multi-Decadal Changes in Mangrove Extent, Age and Species in the Red River Estuaries of Viet Nam

This research investigated the performance of four different machine learning supervised image classifiers: artificial neural network (ANN), decision tree (DT), random forest (RF), and support vector machine (SVM) using SPOT-7 and Sentinel-1 images to classify mangrove age and species in 2019 in a Red River estuary, typical of others found in northern Viet Nam. The four classifiers were chosen because they are considered to have high accuracy, however, their use in mangrove age and species classifications has thus far been limited. A time-series of Landsat images from 1975 to 2019 was used to map mangrove extent changes using the unsupervised classification method of iterative self-organizing data analysis technique (ISODATA) and a comparison with accuracy of K-means classification, which found that mangrove extent has increased, despite a fall in the 1980s, indicating the success of mangrove plantation and forest protection efforts by local people in the study area. To evaluate the supervised image classifiers, 183 in situ training plots were assessed, 70% of them were used to train the supervised algorithms, with 30% of them employed to validate the results. In order to improve mangrove species separations, Gram–Schmidt and principal component analysis image fusion techniques were applied to generate better quality images. All supervised and unsupervised (2019) results of mangrove age, species, and extent were mapped and accuracy was evaluated. Confusion matrices were calculated showing that the classified layers agreed with the ground-truth data where most producer and user accuracies were greater than 80%. The overall accuracy and Kappa coefficients (around 0.9) indicated that the image classifications were very good. The test showed that SVM was the most accurate, followed by DT, ANN, and RF in this case study. The changes in mangrove extent identified in this study and the methods tested for using remotely sensed data will be valuable to monitoring and evaluation assessments of mangrove plantation projects.

Creating New Near-Surface Air Temperature Datasets to Understand Elevation-Dependent Warming in the Tibetan Plateau

The Tibetan Plateau has been undergoing accelerated warming over recent decades, and is considered an indicator for broader global warming phenomena. However, our understanding of warming rates with elevation in complex mountain regions is incomplete. The most serious concern is the lack of high-quality near-surface air temperature (Tair) datasets in these areas. To address this knowledge gap, we developed an automated mapping framework for the estimation of seamless daily minimum and maximum Land Surface Temperatures (LSTs) for the Tibetan Plateau from the existing MODIS LST products for a long period of time (i.e., 2002–present). Specific machine learning methods were developed and linked with target-oriented validation and then applied to convert LST to Tair. Spatial variables in retrieving Tair, such as solar radiation and vegetation indices, were used in estimation of Tair, whereas MODIS LST products were mainly focused on temporal variation in surface air temperature. We validated our process using independent Tair products, revealing more reliable estimates on Tair; the R2 and RMSE at monthly scales generally fell in the range of 0.9–0.95 and 1–2 °C. Using these continuous and consistent Tair datasets, we found temperature increases in the elevation range between 2000–3000 m and 4000–5000 m, whereas the elevation interval at 6000–7000 m exhibits a cooling trend. The developed datasets, findings and methodology contribute to global studies on accelerated warming.

Evaluation and Comparison of Light Use Efficiency and Gross Primary Productivity Using Three Different Approaches

Light use efficiency (LUE), which characterizes the efficiency with which vegetation converts captured/absorbed radiation into organic dry matter through photosynthesis, is a key parameter for estimating vegetation gross primary productivity (GPP). Studies suggest that diffuse radiation induces a higher LUE than direct radiation in short-term and site-scale experiments. The clearness index (CI), described as the fraction of solar incident radiation on the surface of the earth to the extraterrestrial radiation at the top of the atmosphere, is added to the parameterization approach to explain the conditions of diffuse and direct radiation in this study. Machine learning methods—such as the Cubist regression tree approach—are also popular approaches for studying vegetation carbon uptake. This paper aims to compare and analyze the performances of three different approaches for estimating global LUE and GPP. The methods for collecting LUE were based on the following: (1) parameterization approach without CI; (2) parameterization approach with CI; and (3) Cubist regression tree approach. We collected GPP and meteorological data from 180 FLUXNET sites as calibration and validation data and the Global Land Surface Satellite (GLASS) products and ERA-interim data as input data to estimate the global LUE and GPP in 2014. Site-scale validation with FLUXNET measurements indicated that the Cubist regression approach performed better than the parameterization approaches. However, when applying the approaches to global LUE and GPP, the parameterization approach with the CI became the most reliable approach, then closely followed by the parameterization approach without the CI. Spatial analysis showed that the addition of the CI improved the LUE and GPP, especially in high-value zones. The results of the Cubist regression tree approach illustrate more fluctuations than the parameterization approaches. Although the distributions of LUE presented variations over different seasons, vegetation had the highest LUE, at approximately 1.5 gC/MJ, during the whole year in equatorial regions (e.g., South America, middle Africa and Southeast Asia). The three approaches produced roughly consistent global annual GPPs ranging from 109.23 to 120.65 Pg/yr. Our results suggest the parameterization approaches are robust when extrapolating to the global scale, of which the parameterization approach with CI performs slightly better than that without CI. By contrast, the Cubist regression tree produced LUE and GPP with lower accuracy even though it performed the best for model validation at the site scale.

Machine learning-based detection of soil salinity in an arid desert region, Northwest China: A comparison between Landsat-8 OLI and Sentinel-2 MSI

Accurate assessment of soil salinization is considered as one of the most important steps in combating global climate change, especially in arid and semi-arid regions. Multi-spectral remote sensing (RS) data including Landsat series provides the potential for frequent surveys for soil salinization at various scales and resolutions. Additionally, the recently launched Sentinel-2 satellite constellation has temporal revisiting frequency of 5 days, which has been proven to be an ideal approach to assess soil salinity. Yet, studies on detailed comparison in soil salinity tracking between Landsat-8 OLI and Sentinel-2 MSI remain limited. For this purpose, we collected a total of 64 topsoil samples in an arid desert region, the Ebinur Lake Wetland National Nature Reserve (ELWNNR) to compare the monitoring accuracy between Landsat-8 OLI and Sentinel-2 MSI. In this study, the Cubist model was trained using RS-derived covariates (spectral bands, Tasseled Cap transformation-derived wetness (TCW), and satellite salinity indices) and laboratory measured electrical conductivity of 1:5 soil:water extract (EC). The results showed that the measured soil salinity had a significant correlation with surface soil moisture (Pearson's r = 0.75). The introduction of TCW generated satisfactory estimating performance. Compared with OLI dataset, the combination of MSI dataset and Cubist model yielded overall better model performance and accuracy measures (R² = 0.912, RMSE = 6.462 dS m^-1, NRMSE = 9.226%, RPD = 3.400 and RPIQ = 6.824, respectively). The differences between Landsat-8 OLI and Sentinel-2 MSI were distinguishable. In conclusion, MSI image with finer spatial resolution performed better than OLI. Combining RS data sets and their derived TCW within a Cubist framework yielded accurate regional salinity map. The increased temporal revisiting frequency and spectral resolution of MSI data are expected to be positive enhancements to the acquisition of high-quality soil salinity information of desert soils.

Developing a temporally accurate air temperature dataset for Mainland China

Spatially continuous satellite data have been widely used to estimate monthly air temperature (Ta). However, it is still not clear whether the estimated monthly Ta is temporally consistent with observed Ta or not. In this study, the accuracies of interannual variations and temporal trends in estimated monthly Ta were systematically analyzed for Mainland China during 2001-2018. The differences in accuracy among five ways to input data into the model were investigated. The Cubist algorithm and ten variables were used to estimate monthly Ta. It was found that inputting data for the same month into the model can generate more accurate results than inputting all data into the model. Using temporal variables (i.e., month and year) can significantly increase the accuracy of estimated Ta. These results can be explained by different relationships between Ta and auxiliary variables that appear at different times. Thus, using temporal variables can help distinguish between different relationships and improve accuracy levels of the estimated Ta. When applying the best method (inputting data for the same month into the model and using the year as a temporal variable), the coefficient of determination (R²) of estimated monthly mean Ta, interannual variations in monthly mean Ta and temporal trends in monthly mean Ta were recorded as 0.997, 0.731 and 0.848, respectively. The root mean squared errors (RMSEs) of estimated monthly mean Ta, interannual variations in monthly mean Ta and temporal trends in monthly mean Ta were recorded as 0.629 °C, 0.593 °C and 0.201 °C/decade, respectively. An accurate, national coverage, 1 km spatial resolution and long time series (2001-2018) monthly mean, maximum and minimum Ta dataset was finally developed. The dataset can be of great use to many fields such as climatology, hydrology and ecology.

Delineation of high resolution climate regions over the Korean Peninsula using machine learning approaches

In this research, climate classification maps over the Korean Peninsula at 1 km resolution were generated using the satellite-based climatic variables of monthly temperature and precipitation based on machine learning approaches. Random forest (RF), artificial neural networks (ANN), k-nearest neighbor (KNN), logistic regression (LR), and support vector machines (SVM) were used to develop models. Training and validation of these models were conducted using in-situ observations from the Korea Meteorological Administration (KMA) from 2001 to 2016. The rule of the traditional Köppen-Geiger (K-G) climate classification was used to classify climate regions. The input variables were land surface temperature (LST) of the Moderate Resolution Imaging Spectroradiometer (MODIS), monthly precipitation data from the Tropical Rainfall Measuring Mission (TRMM) 3B43 product, and the Digital Elevation Map (DEM) from the Shuttle Radar Topography Mission (SRTM). The overall accuracy (OA) based on validation data from 2001 to 2016 for all models was high over 95%. DEM and minimum winter temperature were two distinct variables over the study area with particularly high relative importance. ANN produced more realistic spatial distribution of the classified climates despite having a slightly lower OA than the others. The accuracy of the models using high altitudinal in-situ data of the Mountain Meteorology Observation System (MMOS) was also assessed. Although the data length of the MMOS data was relatively short (2013 to 2017), it proved that the snowy, dry and cold winter and cool summer class (Dwc) is widely located in the eastern coastal region of South Korea. Temporal shifting of climate was examined through a comparison of climate maps produced by period: from 1950 to 2000, from 1983 to 2000, and from 2001 to 2013. A shrinking trend of snow classes (D) over the Korean Peninsula was clearly observed from the ANN-based climate classification results. Shifting trends of climate with the decrease/increase of snow (D)/temperate (C) classes were clearly shown in the maps produced using the proposed approaches, consistent with the results from the reanalysis data of the Climatic Research Unit (CRU) and Global Precipitation Climatology Centre (GPCC).

A PCA–OLS Model for Assessing the Impact of Surface Biophysical Parameters on Land Surface Temperature Variations

Analysis of land surface temperature (LST) spatiotemporal variations and characterization of the factors affecting these variations are of great importance in various environmental studies and applications. The aim of this study is to propose an integrated model for characterizing LST spatiotemporal variations and for assessing the impact of surface biophysical parameters on the LST variations. For this purpose, a case study was conducted in Babol City, Iran, during the period of 1985 to 2018. We used 122 images of Landsat 5, 7, and 8, and products of water vapor (MOD07) and daily LST (MOD11A1) from the MODIS sensor of the Terra satellite, as well as soil and air temperature and relative humidity data measured at the local meteorological station over 112 dates for the study. First, a single-channel algorithm was applied to estimate LST, while various spectral indices were computed to represent surface biophysical parameters, which included the normalized difference vegetation index (NDVI), soil-adjusted vegetation index (SAVI), normalized difference water index (NDWI), normalized difference built-up index (NDBI), albedo, brightness, greenness, and wetness from tasseled cap transformation. Next, a principal component analysis (PCA) was conducted to determine the degree of LST variation and the surface biophysical parameters in the temporal dimension at the pixel scale based on Landsat imagery. Finally, the relationship between the first component of the PCA of LST and each surface biophysical parameter was investigated by using the ordinary least squares (OLS) regression with both regional and local optimizations. The results indicated that among the surface biophysical parameters, variations of NDBI, wetness, and greenness had the highest impact on the LST variations with a correlation coefficient of 0.75, −0.70, and −0.44, and RMSE of 0.71, 1.03, and 1.06, respectively. The impact of NDBI, wetness, and greenness varied geographically, but their variations accounted for 43%, 38%, and 19% of the LST variation, respectively. Furthermore, the correlation coefficient and RMSE between the observed LST variation and modeled LST variation, based on the most influential biophysical factors (NDBI, wetness, and greenness) yielded 0.85 and 1.06 for the regional approach and 0.93 and 0.26 for the local approach, respectively. The results of this study indicated the use of an integrated PCA–OLS model was effective for modeling of various environmental parameters and their relationship with LST. In addition, the PCA–OLS with the local optimization was found to be more efficient than the one with the regional optimization.

Random Forests and Cubist Algorithms for Predicting Shear Strengths of Rockfill Materials

The shear strength of rockfill materials (RFM) is an important engineering parameter in the design and audit of geotechnical structures. In this paper, the predictive reliability and feasibility of random forests and Cubist models were analyzed by estimating the shear strength from the relative density, particle size, distribution (gradation), material hardness, gradation and fineness modulus, and confining (normal) stress. For this purpose, case studies of 165 rockfill samples have been applied to generate training and testing datasets to construct and validate the models. Thirteen key material properties for rockfill characterization were selected to develop the proposed models. Validation and comparison of the models have been performed using the root mean square error (RMSE), coefficient of determination (R2), and mean estimation error (MAE) between the measured and estimated values. A sensitivity analysis was also conducted to ascertain the importance of various inputs in the prediction of the output. The results demonstrated that the Cubist model has the highest prediction performance with (RMSE = 0.0959, R2 = 0.9697 and MAE = 0.0671), followed by the random forests model with (RMSE = 0.1133, R2 = 0.9548 and MAE= 0.0665), the artificial neural network (ANN) model with (RMSE = 0.1320, R2 = 0.9386 and MAE = 0.0841), and the conventional multiple linear regression technique with (RMSE = 0.1361, R2 = 0.9345 and MAE = 0.0888). The results indicated that the Cubist and random forests models are able to generate better predictive results of the shear strength of RFM than ANN and conventional regression models. The Cubist model was considered to be more promising for interpreting the complex relationships between the influential properties of RFM and the shear strengths of RFM to some extent, which can be extremely helpful in estimating the shear strength of rockfill materials.

Use of earth observations for temperature exposure assessment in epidemiological studies

PURPOSE OF REVIEW

Low, high, extreme, and variable temperatures have been linked to multiple adverse health outcomes, particularly among the elderly and children. Recent models incorporating satellite remote sensing data have mitigated several limitations of previous studies, improving exposure assessment. This review focuses on these new temperature exposure models and their application in epidemiological studies.

RECENT FINDINGS

Satellite observations of land surface temperature have been used to model air temperature across large spatial areas at high spatiotemporal resolutions. These models enable exposure assessment of entire populations and have been shown to reduce error in exposure estimates, thus mitigating downward bias in health effect estimates.

SUMMARY

Satellite-based models improve our understanding of spatiotemporal variation in temperature and the associated health effects. Further research should focus on improving the resolution of these models, especially in urban areas, and increasing their use in epidemiological studies of direct temperature exposure and vector-borne diseases.

Improving Aboveground Biomass Estimation of Pinus densata Forests in Yunnan Using Landsat 8 Imagery by Incorporating Age Dummy Variable and Method Comparison

Optical remote sensing data have been widely used for estimating forest aboveground biomass (AGB). However, the use of optical images is often restricted by the saturation of spectral reflectance for forests that have multilayered and complex canopy structures and high AGB values and by the effect of spectral reflectance from underlayer shrub, grass, and bare soil for young stands. This usually leads to overestimations and underestimations for smaller and larger values, respectively, and makes it very challenging to improve the estimation accuracy of forest AGB. In this study, a novel methodology was proposed by incorporating stand age as a dummy variable into four models to improve the estimation accuracy of the Pinus densata forest AGB in Yunnan of Southwestern China. A total of eight models, including two parametric models (LM: linear regression model and LMC: LM with combined variables), two nonparametric models (RF: random forest and ANN: artificial neural network) without the age dummy variable, and four corresponding models with the age dummy variable (DLM, DLMC, DRF, and DANN), were compared to estimate AGB. Landsat 8 Operational Land Imager (OLI) images and 147 sample plots were acquired and utilized. The results showed that (1) compared with the two parametric models, the two nonparametric algorithms resulted in significantly greater estimation accuracies of Pinus densata forest AGB, and the increases of accuracy varied from 8% to 32% for 100 modeling plots and from 12% to 35% for 47 test plots based on root mean square error (RMSE); (2) compared with the models without the age dummy variable, the models with the age dummy variable greatly reduced the overestimations for the plots with AGB values smaller than 70 Mg/ha and the underestimations for the plots with AGB values larger than 180 Mg/ha and, thus, significantly improved the overall estimation accuracy by 14% to 42% for the modeling plots and by 32% to 44% for the test plots based on RMSE; and (3) the texture measures derived from the Landsat 8 OLI images contributed more to improving the estimation accuracy than the original spectral bands and other transformations. This implied that two nonparametric models, coupled with the use of the age dummy variable and texture measures, offered a great potential for improving the estimation accuracy of Pinus densata forest AGB.

Dual Activation Function-Based Extreme Learning Machine (ELM) for Estimating Grapevine Berry Yield and Quality

Reliable assessment of grapevine productivity is a destructive and time-consuming process. In addition, the mixed effects of grapevine water status and scion-rootstock interactions on grapevine productivity are not always linear. Despite the potential opportunity of applying remote sensing and machine learning techniques to predict plant traits, there are still limitations to previously studied techniques for vine productivity due to the complexity of the system not being adequately modeled. During the 2014 and 2015 growing seasons, hyperspectral reflectance spectra were collected using a handheld spectroradiometer in a vineyard designed to investigate the effects of irrigation level (0%, 50%, and 100%) and rootstocks (1103 Paulsen, 3309 Couderc, SO4 and Chambourcin) on vine productivity. To assess vine productivity, it is necessary to measure factors related to fruit ripeness and not just yield, as an over cropped vine may produce high-yield but poor-quality fruit. Therefore, yield, Total Soluble Solids (TSS), Titratable Acidity (TA) and the ratio TSS/TA (maturation index, IMAD) were measured. A total of 20 vegetation indices were calculated from hyperspectral data and used as input for predictive model calibration. Prediction performance of linear/nonlinear multiple regression methods and Weighted Regularized Extreme Learning Machine (WRELM) were compared with our newly developed WRELM-TanhRe. The developed method is based on two activation functions: hyperbolic tangent (Tanh) and rectified linear unit (ReLU). The results revealed that WRELM and WRELM-TanhRe outperformed the widely used multiple regression methods when model performance was tested with an independent validation dataset. WRELM-TanhRe produced the highest prediction accuracy for all the berry yield and quality parameters (R2 of 0.522–0.682 and RMSE of 2–15%), except for TA, which was predicted best with WRELM (R2 of 0.545 and RMSE of 6%). The results demonstrate the value of combining hyperspectral remote sensing and machine learning methods for improving of berry yield and quality prediction.

Improvement of Clay and Sand Quantification Based on a Novel Approach with a Focus on Multispectral Satellite Images

Soil mapping demands large-scale surveys that are costly and time consuming. It is necessary to identify strategies with reduced costs to obtain detailed information for soil mapping. We aimed to compare multispectral satellite image and relief parameters for the quantification and mapping of clay and sand contents. The Temporal Synthetic Spectral (TESS) reflectance and Synthetic Soil Image (SYSI) approaches were used to identify and characterize texture spectral signatures at the image level. Soil samples were collected (0–20 cm depth, 919 points) from an area of 14,614 km2 in Brazil for reference and model calibration. We compared different prediction approaches: (a) TESS and SYSI; (b) Relief-Derived Covariates (RDC); and (c) SYSI plus RDC. The TESS method produced highly similar behavior to the laboratory convolved data. The sandy textural class showed a greater increase in average spectral reflectance from Band 1 to 7 compared with the clayey class. The prediction using SYSI produced a better result for clay (R2 = 0.83; RMSE = 65.0 g kg−1) and sand (R2 = 0.86; RMSE = 79.9 g kg−1). Multispectral satellite images were more stable for the identification of soil properties than relief parameters.