Search

Canopy chlorophyll content (CCC) is closely related to crop nitrogen status, crop growth and productivity, detection of diseases and pests, and final yield. Thus, accurate monitoring of chlorophyll content in crops is of great significance for decision support in precision agriculture. In this study, winter wheat in the Guanzhong Plain area of the Shaanxi Province, China, was selected as the research subject to explore the feasibility of canopy spectral transformation (CST) combined with a machine learning method to estimate CCC. A hyperspectral canopy ground dataset in situ was measured to construct CCC prediction models for winter wheat over three growth seasons from 2014 to 2017. Sensitive-band reflectance (SR) and narrow-band spectral index (NSI) were established based on the original spectrum (OS) and CSTs, including the first derivative spectrum (FDS) and continuum removal spectrum (CRS). Winter wheat CCC estimation models were constructed using univariate regression, partial least squares (PLS) regression, and random forest (RF) regression based on SR and NSI. The results demonstrated the reliability of CST combined with the machine learning method to estimate winter wheat CCC. First, compared with OS-SR (683 nm), FDS-SR (630 nm) and CRS-SR (699 nm) had a larger correlation coefficient between canopy reflectance and CCC; secondly, among the parametric regression methods, the univariate regression method with CRS-NDSI as the independent variable achieved satisfactory results in estimating the CCC of winter wheat; thirdly, as a machine learning regression method, RF regression combined with multiple independent variables had the best winter wheat CCC estimation accuracy (the determination coefficient of the validation set (Rv2) was 0.88, the RMSE of the validation set (RMSEv) was 3.35 and relative prediction deviation (RPD) was 2.88). Thus, this modeling method could be used as a basic method to predict the CCC of winter wheat in the Guanzhong Plain area. [ABSTRACT FROM AUTHOR]

Wheat can be cultivated by hole sowing, but its border effect has not yet been studied. Therefore, we carried out a field experiment from 2021 to 2022 at the Doukou Crop Experimental Demonstration Station (108°52′ E, 34°37′ N) of Northwest A&F University in Jingyang County, Xianyang City, Shaanxi Province, China. The response of dry matter, photosynthetic characteristics, and yield components of wheat to the border effects under the hole sowing method was studied. The results showed specific border effects on each index of five wheat varieties (XN136, XN175, XN527, XN536, and XN765), among which the border effects of XN175 and XN765 were the most significant, with the highest yield. Subsequent correlation analysis revealed that only grain per spike and intercellular carbon dioxide concentration responded negatively to the border effects, and the rest were positively correlated. Finally, we conducted a random forest model analysis of different indicators of wheat varieties with significant border effects. We found that net photosynthetic rate and aboveground dry matter per plant had the most significant impact and contribution to the border effects. In contrast, grain per spike had the most negligible impact on the border effects. Our results fill a gap in the study of the border effects of wheat under hole sowing cultivation for future researchers. [ABSTRACT FROM AUTHOR]

Soil cation exchange property (SCEP) is important in soil development and environmental buffering. However, the variations in SCEP and its correlation with soil-forming factors in subsurface soil are not fully understood. In this study, we quantified the surface and subsurface SCEP variations as a function of parent material, vegetation, hillslope position, and soil type. Fifty upland soil profiles from mid-subtropical China were selected. The cation exchange capacity (CEC) and effective CEC (ECEC) of subsurface soil were significantly higher in soils derived from slate, Quaternary red clay (QRC), and limestone than in soils derived from granite and sandstone. The subsurface soils derived from limestone had the highest base saturation (BS) and the sum of base cations, and the lowest aluminum (Al) saturation. The SCEP in surface soil significantly varied with vegetation and hillslope position. The surface soil CEC was the highest in mixed-forest vegetation, whereas the ECEC and exchangeable acidity (EA) were the highest in arable vegetation. Exchangeable potassium (K+) was lowest and the EA was highest in soil orders at the strong development phase. Exchangeable calcium (Ca2+), magnesium (Mg2+), CEC, and BS were the highest in soil orders at the intermediate development phase. The prediction accuracy of soil CEC using the random forest model was higher than that obtained using multiple stepwise regression, with the best results (R2 = 0.92) obtained in the surface soil. Our study indicated that the SCEP in surface and subsurface soils was controlled by different soil-forming factors and could be effectively predicted by soil properties in subtropical China. [ABSTRACT FROM AUTHOR]

Gridded gross domestic product (GDP) data are a crucial land surface parameter for many geoscience applications. Recently, machine learning approaches have become powerful tools in generating gridded GDP data. However, most machine learning approaches for gridded GDP estimation seldom consider the geographical properties of input variables. Therefore, in this study, a geographically weighted stacking ensemble learning approach was developed to generate gridded GDP data. Three algorithms—random forest, XGBoost, and LightGBM—were used as base models, and the linear regression in stacking ensemble learning was replaced by geographically weighted regression to locally fuse the three predictions. A case study was conducted in China to demonstrate the effectiveness of the proposed approach. The results showed that the proposed GDP downscaling approach outperformed the three base models and traditional stacking ensemble learning. Meanwhile, it had good predictive power on county-level GDP test data with R2 of 0.894, 0.976, and 0.976 for the primary, secondary, and tertiary sectors, respectively. Moreover, the predicted 1 km gridded GDP data had a high accuracy (R2 = 0.787) when evaluated by town-level GDP data. Hence, the proposed GDP downscaling approach provides a valuable option for generating gridded GDP data. The generated 1 km gridded GDP data of China from 2020 are of great significance for other applications. [ABSTRACT FROM AUTHOR]

Selecting the best planting area for blueberries is an essential issue in agriculture. To better improve the effectiveness of blueberry cultivation, a machine learning-based classification model for blueberry ecological suitability was proposed for the first time and its validation was conducted by using multi-source environmental features data in this paper. The sparrow search algorithm (SSA) was adopted to optimize the CatBoost model and classify the ecological suitability of blueberries based on the selection of data features. Firstly, the Borderline-SMOTE algorithm was used to balance the number of positive and negative samples. The Variance Inflation Factor and information gain methods were applied to filter out the factors affecting the growth of blueberries. Subsequently, the processed data were fed into the CatBoost for training, and the parameters of the CatBoost were optimized to obtain the optimal model using SSA. Finally, the SSA-CatBoost model was adopted to classify the ecological suitability of blueberries and output the suitability types. Taking a study on a blueberry plantation in Majiang County, Guizhou Province, China as an example, the findings demonstrate that the AUC value of the SSA-CatBoost-based blueberry ecological suitability model is 0.921, which is 2.68% higher than that of the CatBoost (AUC = 0.897) and is significantly higher than Logistic Regression (AUC = 0.855), Support Vector Machine (AUC = 0.864), and Random Forest (AUC = 0.875). Furthermore, the ecological suitability of blueberries in Majiang County is mapped according to the classification results of different models. When comparing the actual blueberry cultivation situation in Majiang County, the classification results of the SSA-CatBoost model proposed in this paper matches best with the real blueberry cultivation situation in Majiang County, which is of a high reference value for the selection of blueberry cultivation sites. [ABSTRACT FROM AUTHOR]

The spatial and temporal variation analysis of land cover classification is important for studying the distribution and transformation of regional land cover changes. The Qilian Mountain National Park (QMNP), an important ecological barrier in northwestern China, has lacked land cover products for long time series. The Landsat images available on the Google Earth Engine (GEE) make it possible to analyze the land cover changes over the past three decades. The purpose of this study was to generate a long time series of datasets of land cover classification based on the method of sample migration in the QMNP in Northwest China. The Landsat 5, 7, and 8 images and field sample data were combined with multiple image features and the random forest algorithm to complete the land cover classification of the QMNP from 1990 to 2020. The results indicate that (1) the method of Jeffries–Matusita (J-M) distance can reduce image feature redundancy and show that elevation and phenological features have good differentiability among land cover types that were easy to mix with feature classes; (2) the spatial distribution of land cover every 10 years between 1990 and 2020 was consistent in the QMNP, and there were obvious differences in land cover from the east to the west part of the QMNP, with a large area of vegetation distribution in Sunan county in the central part and Tianzhu county in the east part of the QMNP; (3) over the past 30 years, forests and grasslands decreased by 62.2 km2 and 794.7 km2, respectively, while shrubs increased by 442.9 km2 in the QMNP. The conversion of bare land to grassland and the interconversion between different vegetation types were the main patterns of land cover changes, and the land cover changes were mainly concentrated in pastoral areas, meaning that human activity was the main factor of land cover changes; and (4) when the samples of 2020 were migrated to 2010, 2000, and 1990, the overall classification accuracies were 89.7%, 88.0%, 86.0%, and 83.9%, respectively. The results show that the vegetation conservation process in the QMNP was closely related to human activities. [ABSTRACT FROM AUTHOR]

The alpine grassland ecosystem accounts for 53 % of the Qinghai–Tibet Plateau (QTP) area and is an important ecological protection barrier, but it is fragile and vulnerable to climate change. Therefore, continuous monitoring of grassland aboveground biomass (AGB) is necessary. Although many studies have mapped the spatial distribution of AGB for the QTP, the results vary widely due to the limited ground samples and mismatches with satellite pixel scales. This paper proposed a new algorithm using unmanned aerial vehicles (UAVs) as a bridge to estimate the grassland AGB on the QTP from 2000 to 2019. The innovations were as follows: (1) in terms of ground data acquisition, spatial-scale matching among the traditional ground samples, UAV photos, and MODIS pixels was considered. A total of 906 pairs between field-harvested AGB and UAV sub-photos and 2602 sets of MODIS pixel-scale UAV data (over 37 000 UAV photos) were collected during 2015–2019. Therefore, the ground validation samples were sufficient and scale-matched. (2) In terms of model construction, the traditional quadrat scale (0.25 m 2) was successfully upscaled to the MODIS pixel scale (62 500 m 2) based on the random forest and stepwise upscaling methods. Compared with previous studies, the scale matching of independent and dependent variables was achieved, effectively reducing the impact of spatial-scale mismatch. The results showed that the correlation between the AGB values estimated by UAV and MODIS vegetation indices was higher than that between field-measured AGB and MODIS vegetation indices at the MODIS pixel scale. The multi-year validation results showed that the constructed MODIS pixel-scale AGB estimation model had good robustness, with an average R2 of 0.83 and RMSE of 34.13 g m -2. Our dataset provides an important input parameter for a comprehensive understanding of the role of the QTP under global climate change. The dataset is available from the National Tibetan Plateau/Third Pole Environment Data Center (10.11888/Terre.tpdc.272587; H. Zhang et al., 2022). [ABSTRACT FROM AUTHOR]

PM2.5 is the key reason for the frequent occurrence of smog; therefore, identifying its key driving factors has far-reaching significance for the prevention and control of air pollution. Based on long-term remote sensing inversion of PM2.5 data, 21 driving factors in the fields of nature and humanities were selected, and the random forest model was applied to study the influencing factors of PM2.5 concentration in the Beijing–Tianjin–Hebei urban agglomeration (BTH) from 2000 to 2016. The results indicate: (1) The main factors affecting PM2.5 concentration not only include natural factors such as sunshine hours (SSH), relative humidity (RHU), elevation (ELE), normalized difference vegetation index (NDVI), wind speed (WIN), average temperature (TEM), daily temperature range (TEMR), and precipitation (PRE), but also human factors such as urbanization rate (URB), total investment in fixed assets (INV), and the number of employees in the secondary industry (INDU); (2) The concentration of PM2.5 changed into an inverted S-shape with the increase in SSH and WIN, and into an S-shape with the increase in RHU, NDVI, TEM, PRS, URB and INV. As for ELE and TEMR, it fluctuated and decreased with the increase in ELE, while it increased and then decreased with the increase in TEMR. However, its change was less pronounced with the increase in PRE and INDU; (3) The influence of natural factors is higher than that of human factors, but the role of human factors has been continuously strengthened in recent years. The adjustment and control of PM2.5 pollution sources from the perspective of human factors will become an effective way to reduce PM2.5 concentrations in the BTH. [ABSTRACT FROM AUTHOR]

Landslides are one of the most prevalent environmental disasters in the Hengduan Mountain Region. Landslides lead to severe economic damage and property loss, as well as fatalities. Furthermore, they tend to increase in the context of climate change. The purpose of this study is to comprehensively assess landslide susceptibility across the Hengduan Mountain Region in southwest China. Specifically, the analysis is focused on the eastern boundary of the Tibetan Plateau within the context of future climate change scenarios, which are based on the latest Coupled Model Intercomparison Project Phase 6 (CMIP6) global climate model ensemble. The Generalized Additive Model (GAM), Random Forest (RF), and Light Gradient Boosting Machine (LightGBM) were selected in order to map landslide susceptibility within the context of 1.5–4.0 °C warming scenarios. This was achieved by considering the changes in extreme rainfall that exceeded the landslide triggering thresholds. The results show that the frequency over extreme rainfall thresholds (FOERT) tend to increase in conjunction with warming targets, thereby ranging from 2.3/a (at a 1.5 °C warming) to 9.0/a (at a 4.0 °C warming) on average. Such elevated extreme precipitation events contribute to an increase in projected future zones of high landslide susceptibility when compared to the historical baseline period ranging from −1.2% (at a 1.5 °C warming) to 4.0% (at a 4.0 °C warming) using different machine learning models. Moreover, the extent of high susceptibility zones increases more significantly in the context of 4.0 °C warming when compared to the historical baseline results. These results indicate the importance of limiting the global temperature rise to 1.5 as well as 2 °C. The high landslide susceptibility zones estimated by the CMIP6 multi-models ensemble are mainly located in the central and southeastern regions of the Hengduan Mountain Region. The possible changes in terms of introducing extreme precipitation in order to assess landslide susceptibility in the context of climate change that is proposed in this study may be further applied to additional study areas. These projections under different targets can provide scientific guidelines for the purposes of the development of climate change adaptation strategies. [ABSTRACT FROM AUTHOR]

Accurately identifying and quantifying the factors influencing PM2.5 pollution is of great significance for the prevention and control of pollution. However, the redundancy among potential factors of PM2.5 may be overlooked. Meanwhile, the inconsistent spatial distribution of the natural and socioeconomic conditions brings unique implications for the cities within a region, which may lead to an uncertain understanding of the relationship between pollution and environmental factors. This study focused on the Beijing-Tianjin-Hebei (BTH) Region, China, which presents complex and varied background conditions. Potential impact factors on PM2.5 were firstly screened by combining systematic cluster analysis with a random forest recursive feature elimination algorithm. Then, the representative multi-factor responsible for PM2.5 pollution in the region during the key period of 2014–2018 (when the strict national air pollution control policy was implemented). The results showed that the key driving factors of PM2.5 pollution in the BTH cities are different, indicating that the uniqueness of a city will have an impact on the leading causes of pollution. Further discussion shows that air control policy provides an effective way to improve air quality. This study aims to deepen the understanding of the risk drivers of air pollution within the BTH Region. In the future, it is recommended that more attention should be paid to the specific differences between the cities when formulating PM2.5 concentration control measures. [ABSTRACT FROM AUTHOR]

Cropland abandonment is one of the most widespread types of land-use change in Southern China. Quickly and accurately monitoring spatial-temporal patterns of cropland abandonment is crucial for food security and a good ecological balance. There are still enormous challenges in the long-term monitoring of abandoned cropland in cloud and rain-prone and cropland-fragmented regions. In this study, we developed an approach to automatically obtain Landsat imagery for two key phenological periods, rather than as a time series, and mapped annual land cover from 1989 to 2021 based on the random forest classifier. We also proposed an algorithm for pixel-based, long-term annual land cover correction based on prior knowledge and natural laws, and generated cropland abandonment maps for Guangdong Province over the past 30 years. This work was implemented in Google Earth Engine. Accuracy assessment of the annual cropland abandonment maps for every five years during study period revealed an overall accuracy of 92–95%, producer (user) accuracy of 90–96% (73–87%), and Kappa coefficients of 0.81–0.88. In recent decades, the cropland abandonment area was relatively stable, at around 50 × 104 ha, while the abandonment rate gradually increased with a decrease in the cultivated area after 2000. The Landsat-based cropland abandonment monitoring method can be implemented in regions such as southern China, and will support food security and strategies for maintaining ecological balance. [ABSTRACT FROM AUTHOR]

The uncertainty from the under-estimation and over-estimation of forest aboveground biomass (AGB) is an urgent problem in optical remote sensing estimation. In order to more accurately estimate the AGB of Pinus densata forests in Shangri-La City, we mainly discuss three non-parametric models—the artificial neural network (ANN), random forests (RFs), and the quantile regression neural network (QRNN) based on 146 sample plots and Sentinel-2 images in Shangri-La City, China. Moreover, we selected the corresponding optical quartile models with the lowest mean error at each AGB segment to combine as the best QRNN (QRNNb). The results showed that: (1) for the whole biomass segment, the QRNNb has the best fitting performance compared with the ANN and RFs, the ANN has the lowest R2 (0.602) and the highest RMSE (48.180 Mg/ha), and the difference between the QRNNb and RFs is not apparent. (2) For the different biomass segments, the QRNNb has a better performance. Especially when AGB is lower than 40 Mg/ha, the QRNNb has the highest R2 of 0.961 and the lowest RMSE of 1.733 (Mg/ha). Meanwhile, when AGB is larger than 160 Mg/ha, the QRNNb has the highest R2 of 0.867 and the lowest RMSE of 18.203 Mg/ha. This indicates that the QRNNb is more robust and can improve the over-estimation and under-estimation in AGB estimation. This means that the QRNNb combined with the optimal quantile model of each biomass segment provides a method with more potential for reducing the uncertainties in AGB estimation using optical remote sensing images. [ABSTRACT FROM AUTHOR]

Remote sensing and land resource surveys have been used in recent decades for land use/land cover (LULC) mapping; however, keeping the developed LULC up-to-date and consistent with land survey statistics remains challenging. This study developed a practical and effective framework to automatically update existing LULC products and bridge the gap between remote sensing classification results and land survey data. This study employed Landsat imagery time series, change detection algorithms, sample migration, and random forests to develop a framework for updating existing LULC products in China from 1980–2015 to 1980–2022. The updated LULC maps reflect the post-2015 LULC changes well and maintain continuity with the pre-2015 products. Additionally, a statistical space allocation method based on the minimum cross-entropy strategy was proposed to optimize the LULC maps, increasing the correlation coefficient (r) with China's second and third national land survey statistics from 0.41–0.89 to 0.86–0.99. Thus, the framework and products developed in this study provide valuable tools for sustainable land use and policy planning. [ABSTRACT FROM AUTHOR]

Accurate fractional crop-planting area (FCPA) mapping is a challenging task as it requires leveraging the advantages of geographic data in detailed spatial expression and agricultural statistics in the description of crop types and quantitative characteristics. We present a robust method to disaggregate the agricultural statistics within each county unit to 1-km scale grid by combining particle swarm optimization (PSO)-based feature selection with Random Forest (RF) regression, and an iterative area-gapallocation(IAGA) method is implemented to reconcile the discrepancies between the disaggregating results and statistics. The agriculture in Gansu Province, China, is characterized by complex heterogeneous smallholder farming landscapes. We tested this methodology in Gansu and explored the synergistic estimation of FCPA for six types of crops (i.e. wheat, maize, oil-bearing, vegetable,orchards, and other crops) in 2010. The results showed that the derived FCPA maps matched well with the statistics in terms of quantity, while also providing spatial details. The quantitative evaluation results indicated that the derived FCPA had good accuracy,with a higher R2 above 0.97, a lower RMSE below 1%, and an absolute error between 1.53-5.24%. The proposed methodology provides valuable insights for practical large-scale FCPA mapping at a high spatial resolution in a cost-effective manner. [ABSTRACT FROM AUTHOR]

At present, the problem of an aging population in China is severe. The integration of existing healthcare services with elderly care services is inefficient and cannot meet the needs of the elderly. As such, China urgently needs the concerted efforts of various social forces to cope with the increasingly serious problem of aging. In accordance with Andersen's behavioral model, a survey was conducted in Tangshan City among seniors 60 years of age and older. Using logistic regression models, decision tree models, and random forest models, we examined the factors impacting senior people's desire to choose the integrated medical care and nursing care model. The results of the three models displayed that the elderly's propensity to choose the combined medical care and nursing care model is significantly influenced by the amount of insurance, life care needs, and healthcare needs. Moreover, the study found that the willingness of the elderly in Tangshan to improve the combined medical and nursing care service system is low. The government should appeal to the community to participate in multiple developments to improve the integrated medical and nursing service system. [ABSTRACT FROM AUTHOR]

There are a variety of land cover products generated from remote-sensing images. However, misclassification errors in individual products and inconsistency among them undermine their utilities for research and other applications. While it is worth developing advanced pattern classifiers and utilizing the images of finer spatial, temporal, and/or spectral resolution for increased classification accuracy, it is also sensible to increase map classification accuracy through effective map fusion by exploiting complementarity among multi-source products over a study area. This paper presents a novel fusion method that works by weighting multiple source products based on their map-reference cover type transition probabilities, which are predicted using random forest for individual map pixels. The proposed method was tested and compared with three alternatives: consensus-based weighting, random forest, and locally modified Dempster–Shafer evidential reasoning, in a case study, over Shaanxi province, China. For this case study, three types of land cover products (GlobeLand30, FROM-GLC, and GLC_FCS30) of two nominal years (2010 and 2020) were used as the base maps for fusion. Reference sample data for model training and testing were collected following a robust stratified random sampling design that allows for augmenting reference data flexibly. Accuracy assessments show that overall accuracies (OAs) of fused land cover maps have been improved (1~9% in OAs), with the proposed method outperforming other methods by 2~8% in OAs. The proposed method does not need to have the base products' classification systems harmonized beforehand, thus being robust and highly recommendable for fusing land cover products. [ABSTRACT FROM AUTHOR]

Pine wilt disease (PWD) has caused huge economic and environmental losses since it invaded China. Although early monitoring is an effective way to control this hazard, the monitoring window for the early stage is hard to identify, and varies in different hosts and environments. We used UAV-based multispectral images of Pinus thunbergii forest in East China to identify the change in the number of infected trees in each month of the growing season. We built classification models to detect different PWD infection stages by testing three machine learning algorithms—random forest, support vector machine, and linear discriminant analysis—and identified the best monitoring period for each infection stage (namely, green attack, early, middle, and late). From the obtained results, the early monitoring window period was determined to be in late July, whereas the monitoring window for middle and late PWD stages ranged from mid-August to early September. We also identified four important vegetation indices to monitor each infection stage. In conclusion, this study demonstrated the effectiveness of using machine learning algorithms to analyze multitemporal multispectral data to establish a window for early monitoring of pine wilt disease infestation. The results could provide a reference for future research and guidance for the control of pine wilt disease. [ABSTRACT FROM AUTHOR]

The invasive pest Dendroctonus valens has spread to northeast China, causing serious economic and ecological losses. Early detection and disposal of infested trees is critical to prevent its outbreaks. This study aimed to evaluate the potential of an unmanned aerial vehicle (UAV)-based hyperspectral image for early detection of D. valens infestation at the individual tree level. We compared the spectral characteristics of Pinus tabuliformis in three states (healthy, infested and dead), and established classification models using three groups of features (reflectance, derivatives and spectral vegetation indices) and two algorithms (random forest and convolutional neural network). The spectral features of dead trees were clearly distinct from those of the other two classes, and all models identified them accurately. The spectral changes of infested trees occurred mainly in the visible region, but it was difficult to distinguish infested from healthy trees using random forest classification models based on reflectance and derivatives. The random forest model using spectral vegetation indices and the convolutional neural network model performed better, with an overall accuracy greater than 80% and a recall rate of infested trees reaching 70%. Our results demonstrated the great potential of hyperspectral imaging and deep learning for the early detection of D. valens infestation. The convolutional neural network proposed in this study can provide a reference for the automatic detection of early D. valens infestation using UAV-based multispectral or hyperspectral images in the future. [ABSTRACT FROM AUTHOR]

Under the strong influence of climate change and human activities, the frequency and intensity of disturbance events in the forest ecosystem both show significant increasing trends. Pine wood nematode (Bursapherenchus xylophilus, PWN) is one of the major alien invasive species in China, which has rapidly infected the forest and spread. In recent years, its tendency has been to spread from south to north, causing serious losses to Pinus and non-Pinus coniferous forests. It is urgent to carry out remote sensing monitoring and prediction of pine wilt disease (PWD). Taking Anhui Province as the study area, we applied ground survey, satellite-borne optical remote sensing imagery and environmental factor statistics, relying on the Google Earth Engine (GEE) platform to build a new vegetation index NDFI based on time-series Landsat images to extract coniferous forest information and used a random forest classification algorithm to build a monitoring model of the PWD infection stage. The results show that the proposed NDFI differentiation threshold classification method can accurately extract the coniferous forest range, with the overall accuracy of 87.75%. The overall accuracy of the PWD monitoring model based on random forest classification reaches 81.67%, and the kappa coefficient is 0.622. High temperature and low humidity are conducive to the survival of PWN, which aggravates the occurrence of PWD. Under the background of global warming, the degree of PWD in Anhui Province has gradually increased, and has transferred from the southwest and south to the middle and northeast. Our results show that PWD monitoring and prediction at a regional scale can be realized by using long time-series multi-source remote sensing data, NDFI index can accurately extract coniferous forest information and grasp disease information in a timely manner, which is crucial for effective monitoring and control of PWD. [ABSTRACT FROM AUTHOR]

Deep learning has drawn great attention in the financial field due to its powerful ability in nonlinear fitting, especially in the studies of asset pricing. In this paper, we proposed a long short-term memory option pricing model with realized skewness by fully considering the asymmetry of asset return in emerging markets. It was applied to price the ETF50 options of China. In order to emphasize the improvement of this model, a comparison with a parametric method, such as Black-Scholes (BS), and machine learning methods, such as support vector machine (SVM), random forests and recurrent neural network (RNN), was conducted. Moreover, we also took the characteristic of heavy tail into consideration and studied the effect of realized kurtosis on pricing to prove the robustness of the skewness. The empirical results indicate that realized skewness significantly improves the pricing performance of LSTM among moneyness states except for in-the-money call options. Specifically, the LSTM model with realized skewness outperforms the classical method and other machine learning methods in all metrics. [ABSTRACT FROM AUTHOR]

Under the background of the deepening reform of China's electric power industry, how to make use of existing resources, keep existing employees, and identify and distinguish the influencing factors of turnover have become major problems faced by companies. At the same time, the identification of influencing factors also provides a reference for relevant enterprises to do a good job in talent reserve and recruitment. The influencing factors of employee turnover are numerous and interrelated. Identifying and distinguishing the influencing factors are the focus of academic research. Meanwhile, it provides a reference for relevant enterprises to do an excellent job in personnel reserve and recruitment. Aiming at the limitation of constructing a direct matrix based on experts' scoring method in the traditional DEMATEL model, a model combining random forest and DEMATEL is proposed, applied to identify the influencing factors of employee turnover in electric power enterprises in Qinghai Province. Based on the employee turnover data set of Qinghai Electric Power Company from 2012 to 2020, the empirical analysis is conducted to analyze the influencing factors of employee turnover from two perspectives of the degree of centrality and the degree of causal. The research results show that marital status, household registration, political status, type of job, and the location of the work unit are the key influencing factors, providing a theoretical basis for the early warning of employee turnover and confirming the feasibility of the random forest-DEMATEL method proposed by us. [ABSTRACT FROM AUTHOR]

The functional group (FG) system sensu Colin S. Reynolds of freshwater phytoplankton has been widely applied in many aspects of freshwater ecological assessments. However, there is no current consensus as to how and to why some functional coda shows multiple habitat templates in different water bodies. To test whether there are multiple habitat templates for some functional coda in the FG system, we used two mesotrophic deep reservoirs in China as examples, analysing phytoplankton taxonomic and environmental data collected in 4 seasons over two years. Results showed that a large proportion of the indicators had similar habitat templates within the FG system, whilst some species had multiple or different habitat templates according to their niche differentiation. The random forest (RF) model filtered a minimum set of environmental conditions and provided two potential habitat templates for Oocystis lacustris. One explanatory habitat template of clear water provided by the FG system was verified in the Nanwan Reservoir, and the other habitat template of high dissolved inorganic nitrogen (DIN) was found in the Nianyushan Reservoir. Further studies of the potential multiple habitat templates for some functional coda are recommended for its proper use. [ABSTRACT FROM AUTHOR]

Wildfire is essential in altering land ecosystems' structures, processes, and functions. As a critical disturbance in the China–Mongolia–Russia cross-border area, it is vital to understand the potential drivers of wildfires and predict where wildfires are more likely to occur. This study assessed factors affecting wildfire using the Random Forest (RF) model. No single factor played a decisive role in the incidence of wildfires. However, the climatic variables were most critical, dominating the occurrence of wildfires. The probability of wildfire occurrence was simulated and predicted using the Adaptive Network-based Fuzzy Inference System (ANFIS). The particle swarm optimization (PSO) model and genetic algorithm (GA) were used to optimize the ANFIS model. The hybrid ANFIS models performed better than single ANFIS for the training and validation datasets. The hybrid ANFIS models, such as PSO-ANFIS and GA-ANFIS, overcome the over-fitting problem of the single ANFIS model at the learning stage of the wildfire pattern. The high classification accuracy and good model performance suggest that PSO-ANFIS can be used to predict the probability of wildfire occurrence. The probability map illustrates that high-risk areas are mainly distributed in the northeast part of the study area, especially the grassland and forest area of Dornod Province of Mongolia, Buryatia, and Chita state of Russia, and the northeast part of Inner Mongolia, China. The findings can be used as reliable estimates of the relative likelihood of wildfire hazards for wildfire management in the region covered or vicinity. [ABSTRACT FROM AUTHOR]

Carbon emissions from land use (LUCE) is an increasingly serious problem in China, and the direction of carbon reduction in urban agglomerations is still unclear. This study aimed to identify the key factors that influence county-level LUCE and their response to LUCE in Beijing-Tianjin-Hebei (BTH) region. Based on multi-source data, our study calculated county-level LUCE and its spatiotemporal variations in BTH from 1992 to 2018. Random forest (RF) and geographically weighted regression (GWR) models were applied to discuss influential factors' impact and spatial response on LUCE by quantifying relative importance and spatial heterogeneity. The results show that countylevel LUCE in BTH increased by 1300.04×104t, with an average annual growth rate of 46.11% during 1992-2018, and exhibited prominent spatial autocorrelation, with higher-value counties mostly in the core areas and lower-value counties in the suburbs within BTH. The RF estimates were better than the multiple linear regression (MLR) estimates with an improvement in R2 from 0.56 to 0.98. Population, urbanization, energy intensity, industry proportion, road network density and technical progress had effective explanatory power for county-level LUCE. In addition, GWR estimation shows that the main influencing factors demonstrated significant spatial heterogeneity, and the impacts of population, urbanization, economic, energy structure and road factors in the central core counties were significantly higher than in the northwestern and southern regions. Industrial structure and technology have the greatest impact in the southern and northeast counties. This study explores different causes of LUCE in different counties of urban agglomerations and the continuation of low-carbon control measures in BTH is required. [ABSTRACT FROM AUTHOR]

Species richness is regarded as the core index of biogeography. Estimating the correlation between species richness and modern environmental factors will be of great significance for species conservation. The arid and semiarid areas of China present serious desertification, but there are rich biodiversity resources of high value. In this study, we evaluated species diversity, species richness, and the correlation between species richness and modern environmental factors using the species of Tenebrionidae in arid and semiarid areas of China, which will provide basic data for species conservation. The species richness was measured using 1° × 1° grid cells, and its determinants were explored based on generalized linear models (GLMs) and random forest models. A total of 696 species, belonging to 125 genera of 38 tribes and 7 subfamilies, were recorded in the study area. The non-uniform species richness pattern was presented, with more species in Altai, Tianshan, Nyenchen Thanglha and Helan Mountains. The species richness was affected by a variety of environmental factors. The variables representing energy availability and climate stability had stronger explanatory power, especially the annual mean temperature (BIO1) and the mean temperature of warmest quarter (BIO10). In contrast, water availability and habitat heterogeneity have relatively little correlation with species richness. [ABSTRACT FROM AUTHOR]

Urban landscape forms can be effective in reducing increasing PM2.5 concentrations due to urbanization in China, making it crucially important to accurately quantify the spatiotemporal impact of urban landscape forms on PM2.5 variations. Three landscape indices and six control variables were selected to assess these impacts in 362 Chinese cities during different time scales from 2001 to 2020, using a spatiotemporal geographically weighted regression model, random forest models and partial dependence plots. The results show that there are spatiotemporal differences in the impacts of landscape indices on PM2.5. the proportion of urban green infrastructure (PLAND-UGI) and the fractal dimension of urban green infrastructure (FRACT-UGI) exacerbate PM2.5 concentrations in the northwest, the proportion of impervious surfaces (PLAND-Impervious) mitigates air pollution in northwest and southwest China, and shannon's diversity index (SHDI) has seasonal differences in the northwest. PLAND-UGI is the landscape index with the largest contribution (30%) and interpretable range. The relationship between FRACT and PM2.5 was more complex than for other landscape indices. The results of this study contribute to a deeper understanding of the spatial and temporal differences in the impact of urban landscape patterns on PM2.5, contributing to clean urban development and sustainable development. [ABSTRACT FROM AUTHOR]

The selection of discretization criteria and interval numbers of landslide-related environmental factors generally fails to quantitatively determine or filter, resulting in uncertainties and limitations in the performance of machine learning (ML) methods for landslide susceptibility mapping (LSM). The aim of this study is to propose a robust discretization criterion (RDC) to quantify and explore the uncertainty and subjectivity of different discretization methods. The RDC consists of two steps: raw classification dataset generation and optimal dataset extraction. To evaluate the robustness of the proposed RDC method, Lushan County of Sichuan Province in China was chosen as the study area to generate the LSM based on three datasets (optimal dataset, original dataset with continuous values, and statistical dataset) using three popular ML methods, namely, convolution neural network, random forest, and logistic regression. The results show that the areas under the receiver operating characteristic curve (AUCs) of the optimal dataset for the abovementioned ML models are 0.963, 0.961, and 0.930 which are higher than those of the original dataset (0.938, 0.947, and 0.900) and statistical dataset (0.948, 0.954, and 0.897). In conclusion, the RDC method can extract the more representative features from environmental factors and outperform the other conventional discretization methods. [ABSTRACT FROM AUTHOR]

The purpose of this study was to examine the potential of Machine Vision (MV) approaches for the classification and identification of 12 varieties of guava. There are leaf images of the 12 local varieties of guava (Psidium guajava) that include Bangkok Red, China Surahi, Moti Surahi, Choti Surahi, Golden Gola, China Gola, Multani Sada Gola, Sadda Bahar Gola, Larkana Surahi, Black Guava, Hyderabadi Safeeda, Strawberry Pink Gola. A digital camera captured these images of guava varieties in a natural environment. Multi-features were extracted from these images. It was a composite of histograms, binary features, textures, rotational, spectral, and translational features (RST). Total 47 multi-features were collected for each non-overlapping guava leaf image, i.e., 256 × 256 and 512 × 512 more, the supervised correlation-based feature selection (CFS) method with the best search algorithm was used to optimize 18 features instead of 47 multi-features. Several ML classifiers, including Instant base Identifier (IBI), Random Forest (RF), and Meta Bagging (MB), using 10-fold cross-validation, were applied to the optimized multi-features. IBI results performed better than other classifiers with an average overall accuracy of 93.01% on AOIs, 512 × 512 . In addition, IBI detected 90.5%, 89.5%, 94%, 97%, 95.5%, 97%, 99%, 96.5%, 99%, 80.5%, 88%, and 81.5% accuracy values for the 12 varieties of guava leaves, namely Bangkok Red, China Surahi, Moti Surahi, Choti Surahi, Golden Gola, China Gola, Multani Sada Gola, Sadda Bahar Gola, Larkana Surahi, Black Guava, Hyderabadi Safeeda, Strawberry Pink Gola. The proposed study could play a significant role for the early and accurate identification of Guava varieties, and it would also be helpful for export quality measures for the national economy of the country. [ABSTRACT FROM AUTHOR]

Fire was a serious natural disturbance to forest ecosystems. In this study, a national-scale forest fire risk model (RF_NTL) was proposed in mainland China by incorporating a new human activity factor (NTL) from nighttime light data. For objective verification of model's performance, the logistic regression model (LR) and random forest model (RF) were also built in this paper. Results showed that the RF_NTL model fitted with an AUC value up to 0.95, which was superior to the conventional LR model and RF model. The RF_NTL model also had good generalization ability and stability on annual and monthly scales, helping to accurately forecast forest fire risk. Moreover, the spatial distribution of forest fire risk based on RF_NTL model was detailed and reasonable on a national scale. Forest fire high-risk areas in China were mainly concentrated in the northeast and southwest regions. This study could provide references for future relevant research. [ABSTRACT FROM AUTHOR]

Our aim was to explore the possibility of using machine learning (ML) in predicting the onset and trajectories of depressive symptom in home-based older adults over a 7-year period. Depressive symptom data (collected in the year 2011, 2013, 2015 and 2018) of home-based older Chinese (n = 2650) recruited in the China Health and Retirement Longitudinal Study (CHARLS) were included in the current analysis. The latent class growth modeling (LCGM) and growth mixture modeling (GMM) were used to classify different trajectory classes. Based on the identified trajectory patterns, three ML classification algorithms (i.e. gradient boosting decision tree, support vector machine and random forest) were evaluated with a 10-fold cross-validation procedure and a metric of the area under the receiver operating characteristic curve (AUC). Four trajectories were identified for the depressive symptoms: no symptoms (63.9%), depressive symptoms onset {incident increasing symptoms [new-onset increasing (16.8%)], chronic symptoms [slowly decreasing (12.5%), persistent high (6.8%)]}. Among the analyzed baseline variables, the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10) score, cognition, sleep time, self-reported memory were the top five important predictors across all trajectories. The mean AUCs of the three predictive models had a range from 0.661 to 0.892. ML techniques can be robust in predicting depressive symptom onset and trajectories over a 7-year period with easily accessible sociodemographic and health information. Supplemental data for this article is available online at http://dx.doi.org/10.1080/13607863.2022.2031868 [ABSTRACT FROM AUTHOR]

The occurrence of rockburst can seriously impact the construction and production of deep underground engineering. To prevent rockburst, machine learning (ML) models have been widely employed to predict rockburst based on some related variables. However, due to the costs and complicated geological conditions, complete datasets to evaluate rockburst cannot always be obtained in rock engineering. To fill this limitation, this study proposed an ensemble tree model suitable for incomplete datasets, i.e., the histogram gradient boosting tree (HGBT), to build intelligent models for rockburst prediction. Three hundred fourteen rockburst cases were employed to develop the HGBT model. The hunger game search (HGS) algorithm was implemented to optimize the HGBT model. The established HGBT model had an excellent testing performance (accuracy of 88.9%). An incomplete database with missing values was applied to compare the performances of HGBT and other ML models (random forest, artificial neural network, and so on). HGBT received an accuracy of 78.8% in the incomplete database, and its capacity was better than that of other ML models. Additionally, the importance of input variables in the HGBT model was analyzed. Finally, the feasibility of the HGBT model was validated by rockburst cases from Sanshandao Gold Mine, China. [ABSTRACT FROM AUTHOR]

Click fraud is the sort of deception in which traffic figures for online ads are intentionally inflated. For businesses that advertise online, click fraud may occur often, resulting in erroneous click statistics and lost funds. That is why many businesses are hesitant to advertise their products on websites and mobile apps. To market their products safely, businesses need a reliable technique for detecting click fraud. In this paper we present a stacking algorithm as a solution to this problem. The proposed method's premise is to combine multiple learners to achieve an optimal result. The Synthetic Minority Oversampling Technique (SMOTE) with a combination of undersampling are chosen to handle the unbalanced dataset. In the first-level learners, there are four supervised Machine Learning algorithms, which are AdaBoost, Random Forest, Decision Tree and Logistic Regression. Moreover, Logistic Regression is used again as a the second-level learner. To verify the efficacy of the suggested approach, comparative tests are carried out on the public dataset available on Kaggle from China's largest independent big data service platform TalkingData. Multiple indicators, such as Accuracy, F1 Score, ROC curve, Loss Log and AUC Score, are utilized to analyze the prediction outcomes. The findings reveal that the stacking method improves forecast accuracy while also maintaining a high level of stability. [ABSTRACT FROM AUTHOR]

Earth history is punctuated by voluminous magmatism and the formation of large igneous provinces (LIPs). Although anomalous mantle temperatures are known to be involved in many LIPs formation, the potentially critical role of fluids remains elusive. Here we apply machine learning methods (e.g., random forest, deep neural network, and support vector machines) to train models based on global datasets of basalts associated with different settings. The trained models predict that the basalts of Tarim LIP in northwestern China show a spatial decrease in their island arc affinity from northeast to southwest, which can be correlated to fluids released from earlier southward oceanic subduction. Temporally, the fluid activity declined from 290 Ma basalts to 270 Ma mafic dykes, suggesting that the fusible components in the mantle source were waning over time and ultimately a strengthened lithosphere was generated. Our study provides new insights into the crucial role of fluids in the generation of LIPs, particularly those related to ancient subducted slabs. Plain Language Summary: Large igneous provinces (LIPs) represent enormous volume of magmas (>0.1 MKm3) produced within a short duration (∼1–5 Ma). The involvement of fluids in generating LIPs remains equivocal. Some studies suggest that water plays an important role, whereas the others argue against this model, based on the typical anhydrous mineralogy and oceanic island basalt‐like geochemical signatures displayed by the basalts in LIPs. In this study, we apply machine learning methods to investigate how the LIP basalts were influenced by fluids. Training data were extracted from database to establish the predictive models. The trained models are in turn applied to examine the fluid influence on the Tarim LIP basalts in NW China. The results show that the Tarim LIP basalts have variable IAB‐affinity up to 37%, suggesting that the source domains of Tarim LIP were affected by fluids. Spatially, the fluid activity decreases from northeast to southwest, which we correlate to southward subduction events. Temporally, the fluid activity declined over time from 290 to 270 Ma, suggesting that the mantle was dehydrated during the plume‐lithosphere interaction, which may induce a strengthened lithosphere. These observations suggest that the fluids played a crucial role in the generation of Tarim LIP. Key Points: Machine learning methods were employed to evaluate the fluid activity in Tarim large igneous provinceA spatial decrease in the fluid activity from northeast to southwest is observedSubducted slab‐derived fluids play a crucial role in the generation of large igneous provinces [ABSTRACT FROM AUTHOR]

Correct purchase decisions are required for the successful implementation of government procurement of public services (GPPS). Based on the limitations of decision variables in a single dimension, we consider multiple environments and propose purchasing decision variables from three dimensions: service characteristics, government capability, and environmental factors. By collecting data on decision variables from cases of government purchasing and cases of government providing services, a purchase decision model is constructed based on the random forest algorithm to help governments decide between purchasing public services and providing services. The results indicate that the decision model that considers the three dimensions has better prediction accuracy than the one that only considers service characteristics. Moreover, the results of the importance of the decision variables indicate that the financial capacity and administrative level of the government, and economic environment have a significant impact on the purchase decision of the government. [ABSTRACT FROM AUTHOR]

Flash floods are one of the most serious mountain disasters in Linzhi, China, pose a grave threat to the lives and property of the residents. However, the current regional vulnerability research cannot meet the needs of refined disaster prevention and mitigation in mountainous areas. Therefore, to address this knowledge gap, this study presented a vulnerability assessment framework base on spatialization technique along with the triangular fuzzy number-based analytic hierarchy process (TFN-AHP) and random forest (RF) model. To achieve this goal, an assessment system containing 12 indicators was constructed from the three perspectives of exposure, susceptibility and coping capacity. Firstly, factor spatialization is introduced in data preprocessing. Then TFN-AHP and the RF model were used to calculate the exposure, susceptibility, and coping capacity. Finally, the spatial distribution of the vulnerability throughout Linzhi was obtained. According to the results, the exposure, susceptibility and coping capacity in most areas are extremely low and low, accounting for 74.1%, 99% and 99%, respectively. The vulnerability is similar to the exposure that most of the regions have extremely low vulnerability and low vulnerability, accounting for 83.5%. The moderate vulnerability regions account for 8.5%, and the high vulnerability regions account for only 1%. The assessment results show that highexposure areas mainly exhibit a linear distribution, and mainly include river valleys, cultivated land, and residential areas. Obviously, these areas are mainly distributed in the population and economic agglomeration areas and on low-altitude terrain. The distribution of high and comparatively high vulnerability zones, where economic and human activities are considerably high, are closely to the topography of the catchment. Thus, the results of this study provide scientific and technological evidence for resettlement, population distribution adjustment, and prevention and mitigation of flash floods in Linzhi. [ABSTRACT FROM AUTHOR]

The past landslides from landslide inventory are essential for machine learning (ML)-based landslide susceptibility assessment (LSA). They determine not only the positive samples (characterized by the past landslides) but also the negative samples (randomly generated based on past landslides) for training and validating the ML models. However, the number of past landslides is often limited because of the constraints of time, budget, and resources available, etc. In other words, the available data for establishing landslide susceptibility ML models are limited, which indicates that the accuracy and reliability of the corresponding models are insufficient. This article, therefore, proposes using a new landslide susceptibility index-based sampling strategy to enhance the positive and negative samples for model training and validation to reach an improved ML-based LSA. To realize this idea, landslide susceptibility analysis based on initial datasets compiled from landslide inventory by using three ML models, i.e., random forest (RF), gradient boosting decision tree (GBDT) and support vector machine (SVM), are first conducted to obtain the initial landslide susceptibility indices at different space locations. Then, the landslide susceptibility indices are analyzed with the proposed sampling strategy which considers directly the grid units with very high and very low landslide susceptibility indices as potential positive and negative samples, respectively, to enrich the initial dataset; and the number of these positive/negative samples is determined by a Bayesian optimization algorithm. Thereafter, the ML models are updated with the enriched datasets. Finally, to verify the effectiveness of the proposed strategy, the improved models are applied to assess the landslide susceptibility of Taojiang County, China, and the results are compared with those from initial corresponding models without updating. The results show that compared with the initial RF, GBDT and SVM models, the corresponding improved models have a better performance in accuracy, precision, recall rate, and specificity. In particular, the AUC values of the three models are increased by 11.94%, 10.72% and 0.57%, respectively. [ABSTRACT FROM AUTHOR]

With the explosive growth of atmospheric data, machine learning models have achieved great success in air pollution forecasting because of their higher computational efficiency than the traditional chemical transport models. However, in previous studies, new prediction algorithms have only been tested at stations or in a small region; a large-scale air quality forecasting model remains lacking to date. Huge dimensionality also means that redundant input data may lead to increased complexity and therefore the over-fitting of machine learning models. Feature selection is a key topic in machine learning development, but it has not yet been explored in atmosphere-related applications. In this work, a regional feature selection-based machine learning (RFSML) system was developed, which is capable of predicting air quality in the short term with high accuracy at the national scale. Ensemble-Shapley additive global importance analysis is combined with the RFSML system to extract significant regional features and eliminate redundant variables at an affordable computational expense. The significance of the regional features is also explained physically. Compared with a standard machine learning system fed with relative features, the RFSML system driven by the selected key features results in superior interpretability, less training time, and more accurate predictions. This study also provides insights into the difference in interpretability among machine learning models (i.e., random forest, gradient boosting, and multi-layer perceptron models). [ABSTRACT FROM AUTHOR]

This study compares the capability of Landsat-8, Sentinel-2, and ASTER multispectral data in mapping lithological units in the Bukadaban Peak, China, by evaluating the performance of the Random Forest (RF) classifier. Moreover, the study assesses the importance of remote sensing original bands and the bands derived from enhancement techniques such as Principal Component Analysis (PCA) and Minimum Noise Fraction (MNF) to the classification process. Results revealed that RF selected PC8, PC9, PC12, and MNF1 as the most important features in the Sentinel-2 dataset. Several MNF bands of ASTER were more important than the original and PC bands. The original bands were the strongest predictors in Landsat-8 dataset, while PC2, PC5, and MNF5 were relatively significant. Sentinal-2 and ASTER datasets achieved very similar classification accuracy but outperformed Landsat-8 dataset. ASTER dataset yielded the highest overall accuracy 81.8%, which is 0.18% higher than Sentinel-2 and 3.86% higher than Landsat-8. [ABSTRACT FROM AUTHOR]

This study investigated the accuracy of the Random Forest (RF) model in gap-filling the sensible (H) and latent heat (LE) fluxes, by using the observation data collected at a site over rice-wheat-rotation croplands in Shouxian County of eastern China from 15 July 2015 to 24 April 2019. Firstly, the variable significances of the machine learning (ML) model's five input variables, including the net radiation (Rn), winds speed (WS), temperature (T), relative humidity (RH), and air pressure (P), were examined, and it was found that Rn accounted for 78 % and 76 % of the total variable significance in H and LE calculating, respectively, showing that it was the most important input variable. Secondly, the RF model's accuracy with the five-variable (Rn, WS, T, RH, P) input combination was evaluated, and the results showed that the RF model could reliably gap-fill the H and LE with mean absolute errors (MAEs) of 5.88 Wm-2 and 20.97 Wm-2, and root mean square errors (RMSEs) of 10.67 Wm-2 and 29.46 Wm-2, respectively. Thirdly, 4-variable input combinations were tested, and it was found that the best input combination was (Rn, WS, T, P) with the MAE of H and LE reduced by 12.65 % and 7.12 %, respectively, after removing RH from the input list. At last, through the Taylor diagram, H and LE gap-filling accuracy of the RF model, the support vector machine (SVM) model, the k-nearest neighbor (KNN) model, and the gradient boosting decision tree (GBDT) model was inter-compared, and the statistical metrics showed that RF was the most accurate for both H and LE gap-filling, while the LR and KNN model performed the worst for H and LE gap-filling, respectively. [ABSTRACT FROM AUTHOR]

High-quality gridded soil moisture products are essential for many Earth system science applications, while the recent reanalysis and remote sensing soil moisture data are often available at coarse resolution and remote sensing data are only for the surface soil. Here, we present a 1 km resolution long-term dataset of soil moisture derived through machine learning trained by the in situ measurements of 1789 stations over China, named SMCI1.0 (Soil Moisture of China by in situ data, version 1.0). Random forest is used as a robust machine learning approach to predict soil moisture using ERA5-Land time series, leaf area index, land cover type, topography and soil properties as predictors. SMCI1.0 provides 10-layer soil moisture with 10 cm intervals up to 100 cm deep at daily resolution over the period 2000–2020. Using in situ soil moisture as the benchmark, two independent experiments were conducted to evaluate the estimation accuracy of SMCI1.0: year-to-year (ubRMSE ranges from 0.041 to 0.052 and R ranges from 0.883 to 0.919) and station-to-station experiments (ubRMSE ranges from 0.045 to 0.051 and R ranges from 0.866 to 0.893). SMCI1.0 generally has advantages over other gridded soil moisture products, including ERA5-Land, SMAP-L4, and SoMo.ml. However, the high errors of soil moisture are often located in the North China Monsoon Region. Overall, the highly accurate estimations of both the year-to-year and station-to-station experiments ensure the applicability of SMCI1.0 to study the spatial–temporal patterns. As SMCI1.0 is based on in situ data, it can be a useful complement to existing model-based and satellite-based soil moisture datasets for various hydrological, meteorological, and ecological analyses and models. The DOI link for the dataset is http://dx.doi.org/10.11888/Terre.tpdc.272415 (Shangguan et al., 2022). [ABSTRACT FROM AUTHOR]

Modeling the heat and carbon dioxide (CO2) exchanges in agroecosystems is critical for better understanding water and carbon cycling, improving crop production, and even mitigating climate change, in agricultural regions. While previous studies mainly focused on simulations of the energy and CO2 fluxes in agroecosystems on the North China Plain, their corrections, simulations and driving forces in East China are less understood. In this study, the dynamic variations of heat and CO2 fluxes were simulated by a standalone version of the Simple Biosphere 2 (SiB2) model and subsequently corrected using a Random Forest (RF) machine learning model, based on measurements from 1 January to 31 May 2015–2017 in eastern China. Through validation with direct measurements, it was found that the SiB2 model overestimated the sensible heat flux (H) and latent heat flux (LE), but underestimated soil heat flux (G0) and CO2 flux (Fc). Thus, the RF model was used to correct the results modeled by SiB2. The RF model showed that disturbances in temperature, net radiation, the G0 output of SiB2, and the Fc output of SiB2 were the key driving factors modulating the H, LE, G0, and Fc. The RF model performed well and significantly reduced the biases for H, LE, G0, and Fc simulated by SiB2, with higher R2 values of 0.99, 0.87, 0.75, and 0.71, respectively. The SiB2 and RF models combine physical mechanisms and mathematical correction to enable simulations with both physical meaning and accuracy. [ABSTRACT FROM AUTHOR]

Aim: Protected areas (PAs) in China have typically been designed by considering one or several focal taxa (e.g., relict plant and vertebrate species), but the effectiveness of these "protective umbrellas" in safeguarding insects remains unclear. In this study, we aim to investigate the distribution and diversity of endemic true bugs for four family groups (i.e., Miridae, Lygaeoidea, Pentatomidae, and Reduviidae) in China, the environmental factors shaping these diversity patterns, and the effectiveness of China's PAs in safeguarding their diversity. Location: China. Methods: We mapped the spatial distributions of 1028 endemic true bugs in China using stacked species distribution models and pixel grids of observed points. The environmental factors that influence diversity patterns across China were explored by generalized linear models and random forest models. The effectiveness of PAs in safeguarding species distributions was evaluated by the Zonation platform. Main results: We found that most (84.9%) diversity hotspots for total species were located in mountainous areas of southern China. Additionally, the diversity patterns are shaped by the combined effects of different environmental factors, with the minimum temperature of the coldest month, annual precipitation, and elevation showing the strongest effects. On average, Chinese PAs covered 10.2% of the distribution areas of these endemic true bugs. Main conclusions In general, our research identifies the important roles of mountainous areas in southern China in maintaining species diversity. Moreover, our research also suggests that PAs in China could not provide sufficient protection for this diversity. Protecting these endemic true bugs diversity should be valued more in mountainous areas of southern China. [ABSTRACT FROM AUTHOR]

Aim: China has dozens of well‐recognized biodiversity hotspots, but many more potential areas have not been estimated thoroughly, which is unfavourable for biodiversity conservation. Neotenic net‐winged beetles with limited dispersal ability generally occur in restricted ranges but rarely occur in China, which makes them ideal models for biogeographical studies to define biodiversity hotspots. This study will explore the potential distribution patterns of neotenic Lycidae in China to provide basic data for the implementation of biological conservation. Location: China. Methods: We conducted maximum entropy (MaxEnt) and random forest (RF) modelling to simulate the habitat suitability for neotenic Lycidae occurring in China under different climate scenarios by using all available distribution information in Southeast Asia (a total of 305 occurrence records) and several environmental variables. The dynamic changes in the potentially suitable habitats and centroids of neotenic Lycidae were simulated under all climate scenarios. Results: The results indicated that potentially suitable habitats for neotenic Lycidae were mostly located in the montane areas and mountainous islands in southern China, including the eastern Himalayas, Gaoligong Mountains, Ailao Mountains, Hengduan Mountains, Wumeng Mountains, Miaoling Mountains, Daba Mountains, Wu Mountains, Yunkai Mountains, Wuzhi Mountains and Central Mountains. In addition, the dynamic analyses showed that their changes also occurred in montane areas, whose affinity and significance were confirmed in the biological conservation. Main conclusions: Taking the unique neotenic Lycidae as the subject, we verify that several montane areas are biodiversity hotspots that have already been well‐recognized, while it is determined that some additional mountains could be potential hotspots. Therefore, more attention and biological conservation efforts should be used in these areas. In addition, using the ensemble modelling approach to identify potential distributions is a helpful tool to develop strategies for biological conservation, even if it has several limitations. [ABSTRACT FROM AUTHOR]

Accurately and precisely grasping the spatial distribution and changing trends of China's regional population is of great significance in new urbanization, economic development, public health, disaster assessment, and ecological environmental protection. To monitor and evaluate the long-term spatiotemporal characteristics of the population distribution in China, a population monitoring estimation model was proposed. Based on remote sensing data such as nighttime light (NTL) images, land use data, and data from the fifth, sixth, and seventh censuses of China, the population spatiotemporal distribution in China from 2000 to 2020 was analyzed with a random forest algorithm. This study obtained spatial distribution maps of population density at a 1 km x 1 km resolution in 2000, 2010, and 2020. The results revealed the trend of the spatiotemporal pattern of population change from 2000 to 2020. It shows that: the accuracy assessment using the 2020 census population of townships/streets as a reference shows an R2 of 0.67 and a mean relative error (MRE) of 0.44. The spatial pattern of the population in 2000 and 2010 is generally unchanged. In 2020, population agglomeration is evident in the east, with a slight increase in the proportion of the population in the west. The patterns of population agglomeration and urbanization also change over time. The population spatiotemporal distribution obtained in this study can provide a scientific reference for urban sustainable development and promote the rational allocation of urban resources. [ABSTRACT FROM AUTHOR]

In China, as elsewhere, amphibians are highly endangered. Anthropogenic environmental change has affected the distribution and population dynamics of species, and species distributions at a broad scale are strongly driven by climate and species' ability to disperse. Yet, current knowledge remains limited on how widespread human activity affects the distribution patterns of amphibians in China and whether this effect extends beyond climate. We compiled a relatively comprehensive database on the distribution of 196 amphibian species in China from the literature, public databases, and field data. We obtained 25,826 records on almost 50% of known species in China. To test how environmental factors and human activities influence the current distribution of amphibians (1960–1990), we used range filling, which is species realized ranges relative to their potential climate distribution. We used all species occurrence records to represent realized range and niche models to predict potential distribution range. To reduce uncertainty, we used 3 regression methods (beta regression, generalized boosted regression models, and random forest) to test the associations of species range filling with human activity, climate, topography, and range size. The results of the 3 approaches were consistent. At the species level, mean annual precipitation (climate) had the most effect on spatial distribution pattern of amphibians in China, followed by range size. Human activity ranked last. At the spatial level, mean annual precipitation remained the most important factor. Regions in southeastern of China that are currently moist supported the highest amphibian diversity, but were predicted to experience a decline in precipitation under climate change scenarios. Consequently, the distributions of amphibians will likely shift to the northwest in the future, which could affect future conservation efforts. [ABSTRACT FROM AUTHOR]

Objective. The effect of vaginal microbiota on spontaneous preterm birth (sPTB) has not been fully addressed, and few studies have explored the associations between vaginal taxa and sPTB in the gestational diabetes mellitus (GDM) and non-GDM groups, respectively. Study Design. To minimize external interference, a total of 41 pregnant women with sPTB and 308 controls (pregnant women without sPTB) from same regain were enrolled in this case-cohort study. Controls were randomly selected at baseline. With the exception of GDM, other characteristics were not significantly different between the two groups. Vaginal swabs were collected at early second trimester. Using 16S amplicon sequencing, the main bioinformatics analysis was performed on the platform of QIIME 2. Vaginal microbiota traits of the sPTB group were compared with controls. Finally, the effects of binary taxa on sPTB in the GDM group and the non-GDM group were analyzed, respectively. Results. The proportion of GDM in the sPTB (19.51%) was higher than the controls (7.47%, P = 0.018). The vaginal microbiota of pregnant women with sPTB exhibited higher alpha diversity metrics (observed features, P = 0.001 ; Faith's phylogenetic diversity, P = 0.013) and different beta diversity metrics (unweighted UniFrac, P = 0.006 ; Jaccard's distance, P = 0.004), compared with controls. The presence of Lactobacillus paragasseri/gasseri (aOR: 3.12, 95% CI: 1.24-7.84), Streptococcus (aOR: 3.58, 95% CI: 1.68-7.65), or Proteobacteria (aOR: 3.39, 95% CI: 1.55-7.39) was associated with an increased risk of sPTB in the non-GDM group (P < 0.05). However, the relative abundance of novel L. mulieris (a new species of the L. delbrueckii group) was associated with a decreased risk of sPTB (false discovery rate, 0.10) in all pregnant women. Conclusion. GDM may modify the association of vaginal taxa with sPTB, suggesting that maternal GDM should be considered when using vaginal taxa to identify pregnant women at high risk of sPTB. [ABSTRACT FROM AUTHOR]

Using radiogenomics methods, the differences between tumor imaging data and genetic data in Chinese and Western breast cancer (BC) patients were analyzed, and the correlation between phenotypic data and genetic data was explored. In this paper, we analyzed BC patients' image characteristics and transcriptome data separately, then correlated the magnetic resonance imaging (MRI) phenotype with the transcriptome data through a computational method to develop a radiogenomics feature. The data was fed into the designed random forest (RF) model, which used the area under the receiver operating curve (AUC) as the evaluation index. Next, we analyzed the hub genes in the differentially expressed genes (DEGs) and obtained seven hub genes, which may cause Chinese and Western BC patients to behave differently in the clinic. We demonstrated that combining relevant genetic data and imaging features could better classify Chinese and Western patients than using genes or imaging characteristics alone. The AUC values of 0.74, 0.81, and 0.95 were obtained separately using the image characteristics, DEGs, and radiogenomics features. We screened SYT4, GABRG2, CHGA, SLC6A17, NEUROG2, COL2A1, and MATN4 and found that these genes were positively or negatively correlated with certain imaging characteristics. In addition, we found that the SLC6A17, NEUROG2, CHGA, and MATN4 genes were associated with clinical features. [ABSTRACT FROM AUTHOR]

Chinese indigenous pig breeds have unique genetic characteristics and a rich diversity; however, effective breed identification methods have not yet been well established. In this study, a genotype file of 62,822 single-nucleotide polymorphisms (SNPs), which were obtained from 1059 individuals of 18 Chinese indigenous pig breeds and 5 cosmopolitan breeds, were used to screen the discriminating SNPs for pig breed identification. After linkage disequilibrium (LD) pruning filtering, this study excluded 396 SNPs on non-constant chromosomes and retained 20.92~−27.84% of SNPs for each of the 18 autosomes, leaving a total of 14,823 SNPs. The principal component analysis (PCA) showed the largest differences between cosmopolitan and Chinese pig breeds (PC1 = 10.452%), while relatively small differences were found among the 18 indigenous pig breeds from the Yangtze River Delta region of China. Next, a random forest (RF) algorithm was used to filter these SNPs and obtain the optimal number of decision trees (ntree = 1000) using corresponding out-of-bag (OOB) error rates. By comparing two different SNP ranking methods in the RF analysis, the mean decreasing accuracy (MDA) and mean decreasing Gini index (MDG), the effects of panels with different numbers of SNPs on the assignment accuracy, and the statistics of SNP distribution on each chromosome in the panels, a panel of 1000 of the most breed-discriminative tagged SNPs were finally selected based on the MDA screening method. A high accuracy (>99.3%) was obtained by the breed prediction of 318 samples in the RF test set; thus, a machine learning classification method was established for the multi-breed identification of Chinese indigenous pigs based on a low-density panel of SNPs. [ABSTRACT FROM AUTHOR]

Acute coronary syndrome (ACS) is a serious cardiovascular disease that can lead to cardiac arrest if not diagnosed promptly. However, in the actual diagnosis and treatment of ACS, there will be a large number of redundant related features that interfere with the judgment of professionals. Further, existing methods have difficulty identifying high-quality ACS features from these data, and the interpretability work is insufficient. In response to this problem, this paper uses a hybrid feature selection method based on gradient boosting trees and recursive feature elimination with cross-validation (RFECV) to reduce ACS feature redundancy and uses interpretable feature learning for feature selection to retain the most discriminative features. While reducing the feature set search space, this method can balance model simplicity and learning performance to select the best feature subset. We leverage the interpretability of gradient boosting trees to aid in understanding key features of ACS, linking the eigenvalue meaning of instances to model risk predictions to provide interpretability for the classifier. The data set used in this paper is patient records after percutaneous coronary intervention (PCI) in a tertiary hospital in Fujian Province, China from 2016 to 2021. In this paper, we experimentally explored the impact of our method on ACS risk prediction. We extracted 25 key variables from 430 complex ACS medical features, with a feature reduction rate of 94.19%, and identified 5 key ACS factors. Compared with different baseline methods (Logistic Regression, Random Forest, Gradient Boosting, Extreme Gradient Boosting, Multilayer Perceptron, and 1D Convolutional Networks), the results show that our method achieves the highest Accuracy of 98.8%. [ABSTRACT FROM AUTHOR]

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