Search

Tree species classification using hyperspectral imagery shows incredible promise in developing a large-scale, high-resolution model for identifying tree species, providing unprecedented details on global tree species distribution. Many questions remain unanswered about the best practices for creating a global, general hyperspectral tree species classification model. This study aims to address three key issues in creating a hyperspectral species classification model. We assessed the effectiveness of three data-labeling methods to create training data, three data-splitting methods for training/validation/testing, and machine-learning and deep-learning (including semi-supervised deep-learning) models for tree species classification using hyperspectral imagery at National Ecological Observatory Network (NEON) Sites. Our analysis revealed that the existing data-labeling method using the field vegetation structure survey performed reasonably well. The random tree data-splitting technique was the most efficient method for both intra-site and inter-site classifications to overcome the impact of spatial autocorrelation to avoid the potential to create a locally overfit model. Deep learning consistently outperformed random forest classification; both semi-supervised and supervised deep-learning models displayed the most promising results in creating a general taxa-classification model. This work has demonstrated the possibility of developing tree-classification models that can identify tree species from outside their training area and that semi-supervised deep learning may potentially utilize the untapped terabytes of unlabeled forest imagery. [ABSTRACT FROM AUTHOR]

Accurate prediction of lake surface water temperature (LSWT) is essential for understanding the impacts of climate change on aquatic ecosystems and for guiding environmental management strategies. Predictions of LSWT for two prominent lakes in northern China, Qinghai Lake and Hulun Lake, under various future climate scenarios, were conducted in the present study. Utilizing historical hydrometeorological data and MODIS satellite observations (MOD11A2), we employed three advanced machine learning models—Random Forest (RF), XGBoost, and Multilayer Perceptron Neural Network (MLPNN)—to predict monthly average LSWT across three future climate scenarios (ssp119, ssp245, ssp585) from CMIP6 projections. Through the comparison of training and validation results of the three models across both lake regions, the RF model demonstrated the highest accuracy, with a mean MAE of 0.348 °C and an RMSE of 0.611 °C, making it the most optimal and suitable model for this purpose. With this model, the predicted LSWT for both lakes reveals a significant warming trend in the future, particularly under the high-emission scenario (ssp585). The rate of increase is most pronounced under ssp585, with Hulun Lake showing a rise of 0.55 °C per decade (R2 = 0.72) and Qinghai Lake 0.32 °C per decade (R2 = 0.85), surpassing trends observed under ssp119 and ssp245. These results underscore the vulnerability of lake ecosystems to future climate change and provide essential insights for proactive climate adaptation and environmental management. [ABSTRACT FROM AUTHOR]

Determining the responses of non-photosynthetic vegetation (NPV) and photosynthetic vegetation (PV) communities to climate change is crucial in illustrating the sensitivity and sustainability of these ecosystems. In this study, we evaluated the accuracy of inverting NPV and PV using Landsat imagery with random forest (RF), backpropagation neural network (BPNN), and fully connected neural network (FCNN) models. Additionally, we inverted MODIS NPV and PV time-series data using spectral unmixing. Based on this, we analyzed the responses of NPV and PV to precipitation and drought across different ecological regions. The main conclusions are as follows: (1) In NPV remote sensing inversion, the softmax activation function demonstrates greater advantages over the ReLU activation function. Specifically, the use of the softmax function results in an approximate increase of 0.35 in the R2 value. (2) Compared with a five-layer FCNN with 128 neurons and a three-layer BPNN with 12 neurons, a random forest model with over 50 trees and 5 leaf nodes provides better inversion results for NPV and PV (R2_RF-NPV = 0.843, R2_RF-PV = 0.861). (3) Long-term drought or heavy rainfall events can affect the utilization of precipitation by NPV and PV. There is a high correlation between extreme precipitation events following prolonged drought and an increase in PV coverage. (4) Under long-term drought conditions, the vegetation in the study area responded to precipitation during the last winter and growing season. This study provides an illustration of the response of semi-arid ecosystems to drought and wetting events, thereby offering a data basis for the effect evaluation of afforestation projects. [ABSTRACT FROM AUTHOR]

Machine learning methods have improved in recent years and provide increasingly powerful tools for understanding landscape evolution. In this study, we used the random forest method based on Google Earth Engine to evaluate the desertification dynamics in northern China from 1995 to 2020. We selected Landsat series image bands, remote sensing inversion data, climate baseline data, land use data, and soil type data as variables for majority voting in the random forest method. The method's average classification accuracy was 91.6% ± 5.8 [mean ± SD], and the average kappa coefficient was 0.68 ± 0.09, suggesting good classification results. The random forest classifier results were consistent with the results of visual interpretation for the spatial distribution of different levels of desertification. From 1995 to 2000, the area of aeolian desertification increased at an average rate of 9977 km2 yr−1, and from 2000 to 2005, from 2005 to 2010, from 2010 to 2015, and from 2015 to 2020, the aeolian desertification decreased at an average rate of 2535, 3462, 1487, and 4537 km2 yr−1, respectively. [ABSTRACT FROM AUTHOR]

Soil erodibility (K) refers to the inherent ability of soil to withstand erosion. Accurate estimation and spatial prediction of K values are vital for assessing soil erosion and managing land resources. However, as most K-value estimation models are empirical, they suffer from significant extrapolation uncertainty, and traditional studies on spatial prediction focusing on individual empirical K values have neglected to explore the spatial pattern differences between various empirical models. This work proposed a universal framework for selecting an optimal soil-erodibility map using empirical models enhanced by machine learning. Specifically, three empirical models, namely, the erosion-productivity impact calculator model (K_EPIC), the Shirazi model (K_Shirazi), and the Torri model (K_Torri) were used to estimate K values. Random Forest (RF) and Gradient-Boosting Decision Tree (GBDT) algorithms were employed to develop prediction models, which led to the creation of three K-value maps. The spatial distribution of K values and associated environmental covariates were also investigated across varying empirical models. Results showed that RF achieved the highest accuracy, with R2 of K_EPIC, K_Shirazi, and K_Torri increasing by 46%, 34%, and 22%, respectively, compared to GBDT. And distinctions among environmental variables that shape the spatial patterns of empirical models have been identified. The K_EPIC and K_Shirazi are influenced by soil porosity and soil moisture. The K_Torri is more sensitive to soil moisture conditions and terrain location. More importantly, our study has highlighted disparities in the spatial patterns across the three K-value maps. Considering the data distribution, spatial distribution, and measured K values, the K_Torri model outperformed others in estimating soil erodibility in the plateau lake watershed. This study proposed a framework that aimed to create optimal soil-erodibility maps and offered a scientific and accurate K-value estimation method for the assessment of soil erosion. [ABSTRACT FROM AUTHOR]

This paper focuses on large-scale landslide susceptibility modelling in NW Croatia. The objective of this research was to provide new insight into stable and unstable area sampling strategies on a representative inventory of small and shallow landslides mainly occurring in soil and soft rock. Four strategies were tested for stable area sampling (random points, stable area polygon, stable polygon buffering and stable area centroid) in combination with four strategies for unstable area sampling (landslide polygon, smoothing digital terrain model derived landslide conditioning factors, polygon buffering and landslide centroid), resulting in eight sampling scenarios. Using Logistic Regression, Neural Network, Random Forest and Support Vector Machine algorithm, 32 models were derived and analysed. The main conclusions reveal that polygon sampling of unstable areas is an imperative in large-scale modelling, as well as that subjective and/or biased stable area sampling leads to misleading models. Moreover, Random Forest and Neural Network proved to be more favourable methods (0.804 and 0.805 AUC, respectively), but also showed extreme sensitivity to the tested sampling strategies. In the comprehensive comparison, the advantages and disadvantages of 32 derived models were analysed through quantitative and qualitative parameters to highlight their application to large-scale landslide zonation. The results yielded by this research are beneficial to the susceptibility modelling step in large-scale landslide susceptibility assessments as they enable the derivation of more reliable zonation maps applicable to spatial and urban planning systems. [ABSTRACT FROM AUTHOR]

Brazil, a leading rice producer globally, faces challenges in systematically mapping its diverse rice fields due to varying cropping systems, climates, and planting calendars. Existing rice mapping methods often rely on complex techniques like deep learning or microwave imagery, posing limitations for large-scale mapping. This study proposes a novel approach utilizing Sentinel-2 spectral–temporal metrics (STMs) in conjunction with a random forest classifier for rice paddy mapping. By extracting diverse STMs and training both regional and global classifiers, we validated the method across independent areas. While regional models tended to overestimate rice areas, the global model effectively reduced discrepancies between our data and the reference maps, achieving an overall classifier accuracy exceeding 80%. Despite the need for further refinement to address confusion with other crops, STM exhibits promise for national-scale rice paddy mapping in Brazil. [ABSTRACT FROM AUTHOR]

Monitoring air pollution is important for human health and the environment. Previous studies on the prediction of air pollutants from satellite images have employed machine learning, yet there are few enhancements to the constructure of model. Moreover, the existing models have been successful in predicting pollutants like PM2.5, PM10, and O3. They have not been as effective in predicting other primary air pollutants. To improve the overall prediction performance of the existing model, a novel residual learning backpropagation model, abs. as BresNet, has been proposed in this research. This model has revealed the availability to precisely predict the ground-surface concentration of the six primary air pollutants, PM2.5, PM10, O3, NO2, CO, and SO2, based on the satellite imagery of MODIS AOD. Two of the most commonly used machine learning models so far, viz. the multilayer backpropagation neural network (MLBPN) and random forest (RF), were employed as the control. In the conducted experiments, the proposed BresNet model demonstrated significant improvements of 18.75%/31.94%, 33.82%/85.71%, 15.00%/35.29%, 39.06%/134.21%, 23.23%/68.00%, and 137.14%/260.87% in terms of R2 for the six primary air pollutants, compared to the RF/MLBPN model. Moreover, the convergence speed and loss function of the BresNet model compared to that of the MLBPN decreased by 55.15%, revealing superior convergence speed with the lower loss function. [ABSTRACT FROM AUTHOR]

Wildfire susceptibility maps play a crucial role in preemptively identifying regions at risk of future fires and informing decisions related to wildfire management, thereby aiding in mitigating the risks and potential damage posed by wildfires. This study employs eXplainable Artificial Intelligence (XAI) techniques, particularly SHapley Additive exPlanations (SHAP), to map wildfire susceptibility in Izmir Province, Türkiye. Incorporating fifteen conditioning factors spanning topography, climate, anthropogenic influences, and vegetation characteristics, machine learning (ML) models (Random Forest, XGBoost, LightGBM) were used to predict wildfire-prone areas using freely available active fire pixel data (MODIS Active Fire Collection 6 MCD14ML product). The evaluation of the trained ML models showed that the Random Forest (RF) model outperformed XGBoost and LightGBM, achieving the highest test accuracy (95.6%). All of the classifiers demonstrated a strong predictive performance, but RF excelled in sensitivity, specificity, precision, and F-1 score, making it the preferred model for generating a wildfire susceptibility map and conducting a SHAP analysis. Unlike prevailing approaches focusing solely on global feature importance, this study fills a critical gap by employing a SHAP summary and dependence plots to comprehensively assess each factor's contribution, enhancing the explainability and reliability of the results. The analysis reveals clear associations between factors such as wind speed, temperature, NDVI, slope, and distance to villages with increased fire susceptibility, while rainfall and distance to streams exhibit nuanced effects. The spatial distribution of the wildfire susceptibility classes highlights critical areas, particularly in flat and coastal regions near settlements and agricultural lands, emphasizing the need for enhanced awareness and preventive measures. These insights inform targeted fire management strategies, highlighting the importance of tailored interventions like firebreaks and vegetation management. However, challenges remain, including ensuring the selected factors' adequacy across diverse regions, addressing potential biases from resampling spatially varied data, and refining the model for broader applicability. [ABSTRACT FROM AUTHOR]

Site-specific weed management employs image data to generate maps through various methodologies that classify pixels corresponding to crop, soil, and weed. Further, many studies have focused on identifying specific weed species using spectral data. Nonetheless, the availability of open-access weed datasets remains limited. Remarkably, despite the extensive research employing hyperspectral imaging data to classify species under varying conditions, to the best of our knowledge, there are no open-access hyperspectral weed datasets. Consequently, accessible spectral weed datasets are primarily RGB or multispectral and mostly lack the temporal aspect, i.e., they contain a single measurement day. This paper introduces an open dataset for training and evaluating machine-learning methods and spectral features to classify weeds based on various biological traits. The dataset comprises 30 hyperspectral images, each containing thousands of pixels with 204 unique visible and near-infrared bands captured in a controlled environment. In addition, each scene includes a corresponding RGB image with a higher spatial resolution. We included three weed species in this dataset, representing different botanical groups and photosynthetic mechanisms. In addition, the dataset contains meticulously sampled labeled data for training and testing. The images represent a time series of the weed's growth along its early stages, critical for precise herbicide application. We conducted an experimental evaluation to test the performance of a machine-learning approach, a deep-learning approach, and Spectral Mixture Analysis (SMA) to identify the different weed traits. In addition, we analyzed the importance of features using the random forest algorithm and evaluated the performance of the selected algorithms while using different sets of features. [ABSTRACT FROM AUTHOR]

Effective planning and management strategies for restoring and conserving tropical peat swamp ecosystems require accurate and timely estimates of aboveground biomass (AGB), especially when monitoring the impacts of restoration interventions. The aim of this research is to assess changes in AGB and evaluate the effectiveness of restoration efforts in the North Selangor Peat Swamp Forest (NSPSF), one of the largest remaining peat swamp forests in Peninsular Malaysia, using advanced remote sensing techniques. A Random Forest machine learning method was employed to upscale AGB estimates, derived from a 'LiDAR AGB model', to larger landscape-scale areas with Sentinel-2 spectral and textural variables. The time period under investigation (2015–2018) marked a concentrated phase of restoration and regeneration efforts in NSPSF. The results demonstrate an overall increase in tropical peat swamp AGB during these years, where the total amount of estimated AGB stored in NSPSF increased from 19.3 Tg in 2015 to an estimated 19.8 Tg in 2018. The research found that a tailored variable selection approach improved predictions of AGB, with optimised input variables (n = 62) and parameter adjustments producing a good plausible result (R2 = 0.80; RMSE = 55.2 Mg/ha). This paper concludes by emphasizing the importance of long-term studies (>5 years) for analyzing the success of tropical peat swamp restoration methods, with a potential for integrating remote sensing technology. [ABSTRACT FROM AUTHOR]

Salinization is a major soil degradation process threatening ecosystems and posing a great challenge to sustainable agriculture and food security worldwide. This study aimed to evaluate the potential of state-of-the-art machine learning algorithms in soil salinity (EC1:5) mapping. Further, we predicted the distribution patterns of soil salinity under different future scenarios in the Yellow River Delta. A geodatabase comprising 201 soil samples and 19 conditioning factors (containing data based on remote sensing images such as Landsat, SPOT/VEGETATION PROBA-V, SRTMDEMUTM, Sentinel-1, and Sentinel-2) was used to compare the predictive performance of empirical bayesian kriging regression, random forest, and CatBoost models. The CatBoost model exhibited the highest performance with both training and testing datasets, with an average MAE of 1.86, an average RMSE of 3.11, and an average R2 of 0.59 in the testing datasets. Among explanatory factors, soil Na was the most important for predicting EC1:5, followed by the normalized difference vegetation index and soil organic carbon. Soil EC1:5 predictions suggested that the Yellow River Delta region faces severe salinization, particularly in coastal zones. Among three scenarios with increases in soil organic carbon content (1, 2, and 3 g/kg), the 2 g/kg scenario resulted in the best improvement effect on saline–alkali soils with EC1:5 > 2 ds/m. Our results provide valuable insights for policymakers to improve saline–alkali land quality and plan regional agricultural development. [ABSTRACT FROM AUTHOR]

Rocky coastlines are characterised by steep cliffs, which frequently experience a variety of natural processes that often exhibit intricate interdependencies, such as rainfall, ice and water run-off, and marine actions. The advent of high temporal and spatial resolution data, that can be acquired through remote sensing and geomatics techniques, has facilitated the safe exploration of otherwise inaccessible areas. The datasets that can be gathered from these techniques, typically combined with data from fieldwork, can subsequently undergo analyses employing/applying machine learning algorithms and/or numerical modeling, in order to identify/discern the predominant influencing factors affecting cliff top erosion. This study focuses on a specific case situated at the Conero promontory of the Adriatic Sea in the Marche region. The research methodology entails several steps. Initially, the morphological, geological and geomechanical characteristics of the areas were determined through unmanned aerial vehicle (UAV) and conventional geological/geomechanical surveys. Subsequently, cliff top retreat was determined within a GIS environment by comparing orthophotos taken in 1978 and 2022 using the DSAS tool (Digital Shoreline Analysis System), highlighting cliff top retreat up to 50 m in some sectors. Further analysis was conducted via the use of two Machine Learning (ML) algorithms, namely Random Forest (RF) and eXtreme Gradient Boosting (XGB). The Mean Decrease in Impurity (MDI) methodology was employed to assess the significance of each factor. Both algorithms yielded congruent results, emphasising that cliff top erosion rates are primarily influenced by slope height. Finally, a validation of the ML algorithm results was conducted using 2D Limit Equilibrium Method (LEM) codes. Ten sections extracted from the sector experiencing the most substantial cliff top retreat, as identified by DSAS, were utilised for 2D LEM analysis. Factor of Safety (FS) values were identified and compared with the cliff height of each section. The results from the 2D LEM analyses corroborated the outputs of the ML algorithms, showing a strong correlation between the slope instability and slope height (R2 of 0.84), with FS decreasing with slope height. [ABSTRACT FROM AUTHOR]

In recent years, the semantic segmentation model has been widely applied in fields such as the extraction of crops due to its advantages such as strong discrimination ability, high accuracy, etc. Currently, there is no standard set of ground true label data for major crops in China, and the visual interpretation process is usually time-consuming and laborious. The sample size also makes it difficult to support the model to learn enough ground features, resulting in poor generalisation ability of the model, which in turn makes the model difficult to apply in fine extraction tasks of large-area crops. In this study, a method to establish a pseudo-label sample set based on the random forest algorithm to train a semantic segmentation model (U-Net) was proposed to perform winter wheat extraction. With the help of the GEE platform, Winter Wheat Canopy Index (WCI) indicators were employed in this method to initially extract winter wheat, and training samples (i.e., pseudo labels) were built for the semantic segmentation model through the iterative process of "generating random sample points—random forest model training—winter wheat extraction"; on this basis, the U-net model was trained with multi-time series remote sensing images; finally, the U-Net model was employed to obtain the spatial distribution map of winter wheat in Henan Province in 2022. The results illustrated that: (1) Pseudo-label data were constructed using the random forest model in typical regions, achieving an overall accuracy of 97.53% under validation with manual samples, proving that its accuracy meets the requirements for U-Net model training. (2) Utilizing the U-Net model, U-Net++ model, and random forest model constructed based on pseudo-label data for 2022, winter wheat mapping was conducted in Henan Province. The extraction accuracy of the three models is in the order of U-Net model > U-Net++ model > random forest model. (3) Using the U-Net model to predict the winter wheat planting areas in Henan Province in 2019, although the extraction accuracy decreased compared to 2022, it still exceeded that of the random forest model. Additionally, the U-Net++ model did not achieve higher classification accuracy. (4) Experimental results demonstrate that deep learning models constructed based on pseudo-labels exhibit higher classification accuracy. Compared to traditional machine learning models like random forest, they have higher spatiotemporal adaptability and robustness, further validating the scientific and practical feasibility of pseudo-labels and their generation strategies, which are expected to provide a feasible technical pathway for intelligent extraction of winter wheat spatial distribution information in the future. [ABSTRACT FROM AUTHOR]

The distribution of forest-dominant tree species is crucial for ecosystem assessment. Remote sensing monitoring requires annual ground sample data, but consistent field surveys are challenging. This study addresses this by combining sample migration learning and machine learning for multi-year tree species classification in the Three Gorges Reservoir area in China. Using the continuous change detection and classification (CCDC) algorithm, sample data from 2023 were successfully migrated to 2018–2022, achieving high migration accuracy ( R 2 = 0.8303, RMSE = 4.64). Based on migrated samples, random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGB) algorithms classified forest tree species with overall accuracies above 70% and Kappa coefficients above 0.6. XGB. They outperformed other algorithms, with classification accuracy of over 80% and Kappa above 0.75 in almost all years. The final map indicates stable distribution from 2018 to 2023, with eucalyptus covering over 40% of the forest area, followed by horsetail pine, fir, cypress, and wetland pine. [ABSTRACT FROM AUTHOR]

A timely and accurately predicted grain yield can ensure regional and global food security. The scientific community is gradually advancing the prediction of regional-scale maize yield. However, the combination of various datasets while predicting the regional-scale maize yield using simple and accurate methods is still relatively rare. Here, we have used multi-source datasets (climate dataset, satellite dataset, and soil dataset), lasso algorithm, and machine learning methods (random forest, support vector, extreme gradient boosting, BP neural network, long short-term memory network, and K-nearest neighbor regression) to predict China's county-level maize yield. The use of multi-sourced datasets advanced the predicting accuracy of maize yield significantly compared to the single-sourced dataset. We found that the machine learning methods were superior to the lasso algorithm, while random forest, extreme gradient boosting, and support vector machine represented the most preferable methods for maize yield prediction in China (R2 ≥ 0.75, RMSE = 824–875 kg/ha, MAE = 626–651 kg/ha). The climate dataset contributed more to the prediction of maize yield, while the satellite dataset contributed to tracking the maize growth process. However, the methods' accuracies and the dominant variables affecting maize growth varied with agricultural regions across different geographic locations. Our research serves as an important effort to examine the feasibility of multi-source datasets and machine learning techniques for regional-scale maize yield prediction. In addition, the methodology we have proposed here provides guidance for reliable yield prediction of different crops. [ABSTRACT FROM AUTHOR]

Chromophoric dissolved organic matter (CDOM) is a mixture of various types of organic matter and a useful parameter for monitoring complex inland surface waters. Remote sensing has been widely utilized to detect CDOM in various studies; however, in many cases, the dataset is relatively imbalanced in a single region. To address these concerns, data were acquired from hyperspectral images, field reflection spectra, and field monitoring data, and the imbalance problem was solved using a synthetic minority oversampling technique (SMOTE). Using the on-site reflectance ratio of the hyperspectral images, the input variables Rrs (452/497), Rrs (497/580), Rrs (497/618), and Rrs (684/618), which had the highest correlation with the CDOM absorption coefficient aCDOM (355), were extracted. Random forest and light gradient boosting machine algorithms were applied to create a CDOM prediction algorithm via machine learning, and to apply SMOTE, low-concentration and high-concentration datasets of CDOM were distinguished by 5 m−1. The training and testing datasets were distinguished at a 75%:25% ratio at low and high concentrations, and SMOTE was applied to generate synthetic data based on the training dataset, which is a sub-dataset of the original dataset. Datasets using SMOTE resulted in an overall improvement in the algorithmic accuracy of the training and test step. The random forest model was selected as the optimal model for CDOM prediction. In the best-case scenario of the random forest model, the SMOTE algorithm showed superior performance, with testing R2, absolute error (MAE), and root mean square error (RMSE) values of 0.838, 0.566, and 0.777 m−1, respectively, compared to the original algorithm's test values of 0.722, 0.493, and 0.802 m−1. This study is anticipated to resolve imbalance problems using SMOTE when predicting remote sensing-based CDOM. It is expected to produce and implement a machine learning model with improved reliable performance. [ABSTRACT FROM AUTHOR]

Landslide susceptibility denotes the likelihood of a disaster event under specific conditions. The assessment of landslide susceptibility has transitioned from qualitative to quantitative methods. With the integration of information technology in geological hazard analysis, a range of quantitative models for assessing landslide susceptibility has emerged and is now widely used. To compare and evaluate the accuracy of these models, this study focuses on Xupu County in Hunan Province, applying several models, including the CF model, FR model, CF-LR coupled model, FR-LR coupled model, SVM model, and RF model, to assess regional landslide susceptibility. ROC curves are used to evaluate the reliability of the model's predictions. The evaluation results reveal that the CF model (AUC = 0.756), FR model (AUC = 0.764), CF-LR model (AUC = 0.776), FR-LR model (AUC = 0.781), SVM model (AUC = 0.814), and RF model (AUC = 0.912) all have AUC values within the range of 0.7–0.9, indicating that the overall accuracy of the models is good and can provide a reference for landslide susceptibility zoning in the study area. Among these, the Random Forest model demonstrates the best accuracy for landslide susceptibility zoning in the study area. By extracting the extremely high susceptibility zones from the landslide susceptibility zonings obtained by six models, a comparative analysis of model adaptability was conducted. The results indicate that the Random Forest model has the best adaptability under specific conditions in Xupu County. [ABSTRACT FROM AUTHOR]

The seasonal temperature trends and ice phenology in the Great Slave Lake (GSL) are significantly influenced by inflow from the Slave River. The river undergoes a sequence of mechanical break-ups all the way to the GSL, initiating the GSL break-up process. Additionally, upstream water management practices impact the discharge of the Slave River and, consequently, the ice break-up of the GSL. Therefore, monitoring the break-up process at the Slave River Delta (SRD), where the river meets the lake, is crucial for understanding the cascading effects of upstream activities on GSL ice break-up. This research aimed to use Random Forest (RF) models to monitor the ice break-up processes at the SRD using a combination of satellite images with relatively high spatial resolution, including Landsat-5, Landsat-8, Sentinel-2a, and Sentinel-2b. The RF models were trained using selected training pixels to classify ice, open water, and cloud. The onset of break-up was determined by data-driven thresholds on the ice fraction in images with less than 20% cloud coverage. Analysis of break-up timing from 1984 to 2023 revealed a significant earlier trend using the Mann–Kendall test with a p-value of 0.05. Furthermore, break-up data in recent years show a high degree of variability in the break-up rate using images in recent years with better temporal resolution. [ABSTRACT FROM AUTHOR]

During the summer, substantial freshwater discharge from the Changjiang River into the East China Sea (ECS) results in extensive low-salinity water (LSW) plumes that significantly affect regions along the southern Korean Peninsula and near Jeju Island. Previous research developed an empirical equation to estimate sea surface salinity (SSS) in the ECS during the summer season using remote-sensing reflectance (Rrs) data from bands 3–6 (490, 555, 660, and 680 nm) of the Geostationary Ocean Color Imager (GOCI). With the conclusion of the GOCI mission in March 2021, this study aims to ensure the continuity of SSS estimation in the ECS by transitioning to its successor, the GOCI-II. This transition was facilitated through two approaches: applying the existing GOCI-based equation and introducing a new machine learning method using a random forest model. Our analysis demonstrated a high correlation between SSS estimates derived from the GOCI and GOCI-II when applying the equation developed for the GOCI to both satellites, as indicated by a robust R2 value of 0.984 and a low RMSD of 0.8465 psu. This study successfully addressed the challenge of maintaining continuous SSS estimation in the ECS post-GOCI mission and evaluated the accuracy and limitations of the GOCI-II-derived SSS, proposing future strategies to enhance its effectiveness. [ABSTRACT FROM AUTHOR]

In addressing the challenges of quantitative precipitation estimation (QPE) using weather radar, the importance of enhancing the rainfall estimates for applications such as flash flood forecasting and hydropower generation management is recognized. This study employed dual-polarization weather radar data to refine the traditional Z–R relationship, which often needs higher accuracy in areas with complex meteorological phenomena. Utilizing tree-based machine learning algorithms, such as random forest and gradient boosting, this research analyzed polarimetric variables to capture the intricate patterns within the Z–R relationship. The results highlight machine learning's potential to improve the precision of precipitation estimation, especially under challenging weather conditions. Integrating meteorological insights with advanced machine learning techniques is a remarkable achievement toward a more precise and adaptable precipitation estimation method. [ABSTRACT FROM AUTHOR]

The Yellow River Delta wetlands in China belong to the coastal wetland ecosystem, which is one of the youngest and most characteristic wetlands in the world. The Yellow River Delta wetlands are constantly changed by inland sediment and the influence of waves and storm surges, so the accurate classification of the coastal wetlands in the Yellow River Delta is of great significance for the rational utilization, development and protection of wetland resources. In this study, the Yellow River Delta sentinel-2 multispectral data were processed by super-resolution synthesis, and the feature bands were optimized. The optimal feature-band combination scheme was screened using the OIF algorithm. A deep learning model attention mechanism ResNet based on feature optimization with attention mechanism integration into the ResNet network is proposed. Compared with the classical machine learning model, the AM_ResNet model can effectively improve the classification accuracy of the wetlands in the Yellow River Delta. The overall accuracy was 94.61% with a Kappa of 0.93, and they were improved by about 6.99% and 0.1, respectively, compared with the best-performing Random Forest Classification in machine learning. The results show that the method can effectively improve the classification accuracy of the wetlands in the Yellow River Delta. [ABSTRACT FROM AUTHOR]

Exploring soil organic carbon (SOC) mapping is crucial for addressing critical challenges in environmental sustainability and food security. This study evaluates the suitability of the synergistic use of multi-temporal and high-resolution radar and optical remote sensing data for SOC prediction in the Kaffrine region of Senegal, covering over 1.1 million hectares. For this purpose, various scenarios were developed: Scenario 1 (Sentinel-1 data), Scenario 2 (Sentinel-2 data), Scenario 3 (Sentinel-1 and Sentinel-2 combination), Scenario 4 (topographic features), and Scenario 5 (Sentinel-1 and -2 with topographic features). The findings from comparing three different algorithms (Random Forest (RF), XGBoost, and Support Vector Regression (SVR)) with 671 soil samples for training and 281 samples for model evaluation highlight that RF outperformed the other models across different scenarios. Moreover, using Sentinel-2 data alone yielded better results than using only Sentinel-1 data. However, combining Sentinel-1 and Sentinel-2 data (Scenario 3) further improved the performance by 6% to 11%. Including topographic features (Scenario 5) achieved the highest accuracy, reaching an R2 of 0.7, an RMSE of 0.012%, and an RPIQ of 5.754 for the RF model. Applying the RF and XGBoost models under Scenario 5 for SOC mapping showed that both models tended to predict low SOC values across the study area, which is consistent with the predominantly low SOC content observed in most of the training data. This limitation constrains the ability of ML models to capture the full range of SOC variability, particularly for less frequent, slightly higher SOC values. [ABSTRACT FROM AUTHOR]

It is crucial to monitor algal blooms in freshwater reservoirs through an examination of chlorophyll-a (Chla) concentrations, as they indicate the trophic condition of these waterbodies. Traditional monitoring methods, however, are expensive and time-consuming. Addressing this hindrance, we conducted a comprehensive investigation using several machine learning models for Chla modeling. To this end, we used in situ collected water sample data and remote sensing data from the Sentinel-2 satellite, including spectral bands and indices, for large-scale coverage. This approach allowed us to conduct a comprehensive analysis and characterization of the Chla concentrations across 149 freshwater reservoirs in Ceará, a semi-arid region of Brazil. The implemented machine learning models included k-nearest neighbors, random forest, extreme gradient boosting, the least absolute shrinkage, and the group method of data handling (GMDH); in particular, the GMDH approach has not been previously explored in this context. The forward stepwise approach was used to determine the best subset of input parameters. Using a 70/30 split for the training and testing datasets, the best-performing model was the GMDH model, achieving an R2 of 0.91, an MAPE of 102.34%, and an RMSE of 20.4 μg/L, which were values consistent with the ones found in the literature. Nevertheless, the predicted Chla concentration values were most sensitive to the red, green, and near-infrared bands. [ABSTRACT FROM AUTHOR]

Accurately estimating forest aboveground biomass (AGB) is imperative for comprehending carbon cycling, calculating carbon budgets, and formulating sustainable forest management plans. Currently, random forest (RF) and other machine learning models are widely used to estimate forest AGB, as they can effectively handle nonlinear relationships. However, by constructing a global model using all the samples collected from a study area, these models fail to account for the spatial heterogeneity in the AGB and cannot correct the prediction biases, thereby constraining the estimation accuracy. To overcome these limitations, we proposed a novel approach termed geographical random forest and empirical Bayesian kriging (GRFEBK). This hybrid model combines the localized modeling capability of geographical random forest (GRF) with the bias correction strength of empirical Bayesian kriging (EBK). GRF adapts RF to account for the spatial heterogeneity of the AGB, while EBK utilizes the spatial autocorrelation of residuals to correct the prediction deviations. This study was conducted in Hainan Island, utilizing spectral bands, vegetation indices, tasseled cap components derived from Landsat-8 imagery, backscattering coefficients from ALOS-2 synthetic aperture radar, topographic features, and the forest canopy height as the explanatory variables. A total of 195 forest aboveground biomass (AGB) samples were collected for modeling and assessing the predictive accuracy. The results demonstrate that, among the tested models, including GRFEBK, RF, support vector machine (SVM), k-nearest neighbor (KNN), geographically weighted regression (GWR), GRF, and EBK, GRFEBK attains the highest R 2 (0.78) and the lowest RMSE (36.04 Mg/ha) and RRMSE (22.87%), significantly outperforming the conventional models and using GRF or EBK alone. These results demonstrate that by accounting for local non-stationarity in AGB and correcting prediction biases, GRFEBK achieves significantly higher accuracy than conventional RF and other models. While the results are promising, the computational cost of GRFEBK and its performance under varying geographical conditions warrant further investigation at larger scales to assess its broader applicability. Nevertheless, GRFEBK provides an innovative and more reliable approach for accurate forest AGB estimation with great potential to support global forest resource monitoring. [ABSTRACT FROM AUTHOR]

In shallow water, Sentinel-2 multispectral imagery has only four visible bands and limited quantization levels, which easily leads to the occurrence of the same spectral profile but different depth (SSPBDD) phenomenon, resulting in a one-to-many relationship between water depth and spectral profile. Investigating the impact of this relationship on water depth inversion models is the main objective of this paper. The Stumpf model and three machine learning models (Random Forest, Support Vector Machine, and Mixture Density Network) are employed, and the performance of these models is analysed based on the spatial distribution of the training dataset and the input information composition of these models. The results show that the root mean square errors (RMSEs) of the depth inversion of Random Forest and Support Vector Machine are significantly affected by the spatial distribution of the training dataset, while minimal effects are observed for the Stumpf model and the Mixture Density Network model. The SSPBDD phenomenon is widespread in Sentinel-2 images at all depths, particularly between 5 m and 15 m, with most of the depth maximum difference of the SSPBDD pixels ranging from 0 to 5 m. The SSPBDDs phenomenon can significantly reduce the inversion accuracy of any model. The number and the depth maximum difference of the SSPBDDs pixels are the main influencing factors. However, by increasing the visible spectral information and the spatial neighbourhood information in the input layer of machine learning models, the inversion accuracy and stability of the models can be improved to a certain extent. Among the models, the Mixture Density Network achieves the best inversion accuracy and stability. [ABSTRACT FROM AUTHOR]

Comprehensive vegetation surveys are crucial for species selection and layout during the restoration of degraded island ecosystems. However, due to the poor accessibility of uninhabited islands, traditional quadrat surveys are time-consuming and labor-intensive, and it is challenging to fully identify the specific species and their spatial distributions. With miniaturized sensors and strong accessibility, high spatial and temporal resolution, Unmanned Aerial Vehicles (UAVs) have been extensively implemented for vegetation surveys. By collecting UAVs multispectral images and conducting field quadrat surveys on Anyu Island, we employ four machine learning models, namely Gradient Boosting Decision Tree (GBDT), Support Vector Machine (SVM), Random Forest (RF) and Multiple Classifier Systems (MCS). We aim to identify the dominant species and analyze their spatial distributions according to spectral characteristics, vegetation index, topographic factors, texture features, and canopy heights. The results indicate that SVM model achieves the highest (88.55%) overall accuracy (OA) (kappa coefficient = 0.87), while MCS model does not significantly improve it as expected. Acacia confusa has the highest OA among 7 dominant species, reaching 97.67%. Besides the spectral characteristics, the inclusion of topographic factors and texture features in the SVM model can significantly improve the OA of dominant species. By contrast, the vegetation index, particularly the canopy height even reduces it. The dominant species exhibit significant zonal distributions with distance from the coastline on the Anyu Island (p < 0.001). Our study provides an effective and universal path to identify and map the dominant species and is helpful to manage and restore the degraded vegetation on uninhabited islands. [ABSTRACT FROM AUTHOR]

Powdery mildew significantly impacts the yield of natural rubber by being one of the predominant diseases that affect rubber trees. Accurate, non-destructive recognition of powdery mildew in the early stage is essential for the cultivation management of rubber trees. The objective of this study is to establish a technique for the early detection of powdery mildew in rubber trees by combining spectral and physicochemical parameter features. At three field experiment sites and in the laboratory, a spectroradiometer and a hand-held optical leaf-clip meter were utilized, respectively, to measure the hyperspectral reflectance data (350–2500 nm) and physicochemical parameter data of both healthy and early-stage powdery-mildew-infected leaves. Initially, vegetation indices were extracted from hyperspectral reflectance data, and wavelet energy coefficients were obtained through continuous wavelet transform (CWT). Subsequently, significant vegetation indices (VIs) were selected using the ReliefF algorithm, and the optimal wavelengths (OWs) were chosen via competitive adaptive reweighted sampling. Principal component analysis was used for the dimensionality reduction of significant wavelet energy coefficients, resulting in wavelet features (WFs). To evaluate the detection capability of the aforementioned features, the three spectral features extracted above, along with their combinations with physicochemical parameter features (PFs) (VIs + PFs, OWs + PFs, WFs + PFs), were used to construct six classes of features. In turn, these features were input into support vector machine (SVM), random forest (RF), and logistic regression (LR), respectively, to build early detection models for powdery mildew in rubber trees. The results revealed that models based on WFs perform well, markedly outperforming those constructed using VIs and OWs as inputs. Moreover, models incorporating combined features surpass those relying on single features, with an overall accuracy (OA) improvement of over 1.9% and an increase in F1-Score of over 0.012. The model that combines WFs and PFs shows superior performance over all the other models, achieving OAs of 94.3%, 90.6%, and 93.4%, and F1-Scores of 0.952, 0.917, and 0.941 on SVM, RF, and LR, respectively. Compared to using WFs alone, the OAs improved by 1.9%, 2.8%, and 1.9%, and the F1-Scores increased by 0.017, 0.017, and 0.016, respectively. This study showcases the viability of early detection of powdery mildew in rubber trees. [ABSTRACT FROM AUTHOR]

Crop mapping using satellite imagery is crucial for agriculture applications. However, a fundamental challenge that hinders crop mapping progress is the scarcity of samples. The latest foundation model, Segment Anything Model (SAM), provides an opportunity to address this issue, yet few studies have been conducted in this area. This study investigated the parcel segmentation performance of SAM on commonly used medium-resolution satellite imagery (i.e., Sentinel-2 and Landsat-8) and proposed a novel automated sample generation framework based on SAM. The framework comprises three steps. First, an image optimization automatically selects high-quality images as the inputs for SAM. Then, potential samples are generated based on the masks produced by SAM. Finally, the potential samples are subsequently subjected to a sample cleaning procedure to acquire the most reliable samples. Experiments were conducted in Henan Province, China, and southern Ontario, Canada, using six proven effective classifiers. The effectiveness of our method is demonstrated through the combination of field-survey-collected samples and differently proportioned generated samples. Our results indicated that directly using SAM for parcel segmentation remains challenging, unless the parcels are large, regular in shape, and have distinct color differences from surroundings. Additionally, the proposed approach significantly improved the performance of classifiers and alleviated the sample scarcity problem. Compared to classifiers trained only by field-survey-collected samples, our method resulted in an average improvement of 16% and 78.5% in Henan and Ontario, respectively. The random forest achieved relatively good performance, with weighted-average F1 of 0.97 and 0.996 obtained using Sentinel-2 imagery in the two study areas, respectively. Our study contributes insights into solutions for sample scarcity in crop mapping and highlights the promising application of foundation models like SAM. [ABSTRACT FROM AUTHOR]

Red needle cast (RNC), mainly caused by Phytophthora pluvialis, is a very damaging disease of the widely grown species radiata pine within New Zealand. Using a combination of satellite imagery and weather data, a novel methodology was developed to pre-visually predict the incidence of RNC on radiata pine within the Gisborne region of New Zealand over a five-year period from 2019 to 2023. Sentinel-2 satellite imagery was used to classify areas within the region as being disease-free or showing RNC expression from the difference in the red/green index (R/Gdiff) during a disease-free time of the year and the time of maximum disease expression in the upper canopy (early spring–September). Within these two classes, 1976 plots were extracted, and a classification model was used to predict disease incidence from mean monthly weather data for key variables during the 11 months prior to disease expression. The variables in the final random forest model included solar radiation, relative humidity, rainfall, and the maximum air temperature recorded during mid–late summer, which provided a pre-visual prediction of the disease 7–8 months before its peak expression. Using a hold-out test dataset, the final random forest model had an accuracy of 89% and an F1 score of 0.89. This approach can be used to mitigate the impact of RNC by focusing on early surveillance and treatment measures. [ABSTRACT FROM AUTHOR]

Accurate mapping of forest habitats, especially in NATURA sites, is essential information for forest monitoring and sustainable management but also for habitat characterisation and ecosystem functioning. Remote sensing data and spatial modelling allow accurate mapping of the presence and distribution of tree species and habitats and are valuable tools for the long-term assessment of habitat status required by the European Commission. In order to serve the above, the present study aims to propose a methodology to accurately map the spatial distribution of forest habitats in three NATURA2000 sites of Cyprus by employing Sentinel-1 and Sentinel-2 data as well as topographic features using the Google Earth Engine (GEE). A pivotal aspect of the methodology identified was that the best band combination of the Random Forest (RF) classifier achieves the highest performance for mapping the dominant habitats in the three case studies. Specifically, in the Akamas region, eight habitat types have been mapped, in Paphos nine and six in Troodos. These habitat types are included in three of the nine habitat groups based on the EU's Habitat Directive: the sclerophyllous scrub, rocky habitats and caves and forests. The results show that using the RF algorithm achieves the highest performance, especially using Dataset 6, which is based on S2 bands, spectral indices and topographical features, and Dataset 13, which includes S2, S1, spectral indices and topographical features. These datasets achieve an overall accuracy (OA) of approximately 91–94%. In contrast, Dataset 7, which includes only S1 bands and Dataset 9, which combines S1 bands and spectral indices, achieve the lowest performance with an OA of approximately 25–43%. [ABSTRACT FROM AUTHOR]

Southern China, one of the traditional rice production bases, has experienced significant declines in the area of rice paddy since the beginning of this century. Monitoring the rice cropping area is becoming an urgent need for food security policy decisions. One of the main challenges for mapping rice in this area is the quantity of cloud-free observations that are vulnerable to frequent cloud cover. Another relevant issue that needs to be addressed is determining how to select the appropriate classifier for mapping paddy rice based on the cloud-masked observations. Therefore, this study was organized to quickly find a strategy for rice mapping by evaluating cloud-mask algorithms and machine-learning methods for Sentinel-2 imagery. Specifically, we compared four GEE-embedded cloud-mask algorithms (QA60, S2cloudless, CloudScore, and CDI (Cloud Displacement Index)) and analyzed the appropriateness of widely accepted machine-learning classifiers (random forest, support vector machine, classification and regression tree, gradient tree boost) for cloud-masked imagery. The S2cloudless algorithm had a clear edge over the other three algorithms based on its overall accuracy in evaluation and visual inspection. The findings showed that the algorithm with a combination of S2cloudless and random forest showed the best performance when comparing mapping results with field survey data, referenced rice maps, and statistical yearbooks. In general, the research highlighted the potential of using Sentinel-2 imagery to map paddy rice with multiple combinations of cloud-mask algorithms and machine-learning methods in a cloud-prone area, which has the potential to broaden our rice mapping strategies. [ABSTRACT FROM AUTHOR]

Mapping benthic habitats with bathymetric, acoustic, and spectral data requires georeferenced ground-truth information about habitat types and characteristics. New technologies like autonomous underwater vehicles (AUVs) collect tens of thousands of images per mission making image-based ground truthing particularly attractive. Two types of machine learning (ML) models, random forest (RF) and deep neural network (DNN), were tested to determine whether ML models could serve as an accurate substitute for manual classification of AUV images for substrate type interpretation. RF models were trained to predict substrate class as a function of texture, edge, and intensity metrics (i.e., features) calculated for each image. Models were tested using a manually classified image dataset with 9-, 6-, and 2-class schemes based on the Coastal and Marine Ecological Classification Standard (CMECS). Results suggest that both RF and DNN models achieve comparable accuracies, with the 9-class models being least accurate (~73–78%) and the 2-class models being the most accurate (~95–96%). However, the DNN models were more efficient to train and apply because they did not require feature estimation before training or classification. Integrating ML models into benthic habitat mapping process can improve our ability to efficiently and accurately ground-truth large areas of benthic habitat using AUV or similar images. [ABSTRACT FROM AUTHOR]

The TROPOMI XCH4 data product requires rigorous cloud filtering to achieve a product accuracy of <1%. To this end, operational XCH4 data processing has been based on SUOMI-NPP VIIRS cloud observations. However, SUOMI-NPP is nearing the end of its operational life and has encountered malfunctions in 2022 and 2023. In this study, we introduce a novel machine learning cloud-clearing approach based on a random forest classifier (RFC). The RFC is trained on collocated TROPOMI and SUOMI-NPP VIIRS data to emulate VIIRS-like cloud clearing. After training, cloud masking requires only TROPOMI data, and so becomes operationally independent of SUOMI-NPP. We demonstrate the RFC approach by applying cloud clearing to operational TROPOMI XCH4 data for August 2022, a period in which VIIRS was not operational. For validation, we analyze the TROPOMI XCH4 data at 12 TCCON stations. Comparison of cloud clearing using the RFC and the original VIIRS method reveals excellent agreement with a similar station-to-station bias (−7.4 ppb versus −5.6 ppb), a similar standard deviation of the station-to-station bias (11.6 ppb versus 12 ppb), and the same Pearson correlation coefficient of 0.9. Remarkably, the RFC cloud clearing provides a slightly higher volume of data (2182 versus 2035 daily means) and appears to have fewer outliers. Since 21 November 2023, the RFC approach is part of the operational processing chain of the European Space Agency (ESA). For now, the default practice is to utilize SNPP-VIIRS when accessible. Only in cases where VIIRS data are unavailable do we resort to the RFC cloud mask. [ABSTRACT FROM AUTHOR]

The aim of this study is to predict and map winter wheat yield in the Parvomay municipality, situated in the Upper Thracian Lowland of Bulgaria, utilizing satellite data from Sentinel-2. The main crops grown in the research area are winter wheat, rapeseed, sunflower, and maize. To distinguish winter wheat fields accurately, we evaluated classification methods such as Support Vector Machines (SVM) and Random Forest (RF). These methods were applied to satellite multispectral data acquired by the Sentinel-2 satellites during the growing season of 2020–2021. In accordance with their development cycles, temporal image composites were developed to identify suitable moments when each crop is most accurately distinguished from others. Ground truth data obtained from the integrated administration and control system (IACS) were used for training the classifiers and assessing the accuracy of the final maps. Winter wheat fields were masked using the crop mask created from the best-performing classification algorithm. Yields were predicted with regression models calibrated with in situ data collected in the Parvomay study area. Both SVM and RF algorithms performed well in classifying winter wheat fields, with SVM slightly outperforming RF. The produced crop maps enable the application of crop-specific yield models on a regional scale. The best predictor of yield was the green NDVI index (GNDVI) from the April monthly composite image. [ABSTRACT FROM AUTHOR]

The identification of wetland vegetation is essential for environmental protection and management as well as for monitoring wetlands' health and assessing ecosystem services. However, some limitations on vegetation classification may be related to remote sensing technology, confusion between plant species, and challenges related to inadequate data accuracy. In this paper, vegetation classification in the Yancheng Coastal Wetlands is studied and evaluated from Sentinel-2 images based on a random forest algorithm. Based on consistent time series from remote sensing observations, the characteristic patterns of the Yancheng Coastal Wetlands were better captured. Firstly, the spectral features, vegetation indices, and phenological characteristics were extracted from remote sensing images, and classification products were obtained by constructing a dense time series using a dataset based on Sentinel-2 images in Google Earth Engine (GEE). Then, remote sensing classification products based on the random forest machine learning algorithm were obtained, with an overall accuracy of 95.64% and kappa coefficient of 0.94. Four indicators (POP, SOS, NDVIre, and B12) were the main contributors to the importance of the weight analysis for all features. Comparative experiments were conducted with different classification features. The results show that the method proposed in this paper has better classification. [ABSTRACT FROM AUTHOR]

Wetlands provide vital ecological and socioeconomic services but face escalating pressures worldwide. This study undertakes an integrated spatiotemporal assessment of the multifaceted vulnerabilities shaping Khinjhir Lake, an ecologically significant wetland ecosystem in Pakistan, using advanced geospatial and machine learning techniques. Multi-temporal optical remote sensing data from 2000 to 2020 was analyzed through spectral water indices, land cover classification, change detection and risk mapping to examine moisture variability, land cover modifications, area changes and proximity-based threats over two decades. The random forest algorithm attained the highest accuracy (89.5%) for land cover classification based on rigorous k-fold cross-validation, with a training accuracy of 91.2% and a testing accuracy of 87.3%. This demonstrates the model's effectiveness and robustness for wetland vulnerability modeling in the study area, showing 11% shrinkage in open water bodies since 2000. Inventory risk zoning revealed 30% of present-day wetland areas under moderate to high vulnerability. The cellular automata–Markov (CA–Markov) model predicted continued long-term declines driven by swelling anthropogenic pressures like the 29 million population growth surrounding Khinjhir Lake. The research demonstrates the effectiveness of integrating satellite data analytics, machine learning algorithms and spatial modeling to generate actionable insights into wetland vulnerability to guide conservation planning. The findings provide a robust baseline to inform policies aimed at ensuring the health and sustainable management and conservation of Khinjhir Lake wetlands in the face of escalating human and climatic pressures that threaten the ecological health and functioning of these vital ecosystems. [ABSTRACT FROM AUTHOR]

Automated built-up infrastructure classification is a global need for planning. However, individual indices have weaknesses, including spectral confusion with bare ground, and computational requirements for deep learning are intensive. We present a computationally lightweight method to classify built-up infrastructure. We use an ensemble of spectral indices and a novel red-band texture layer with global thresholds determined from 12 diverse sites (two seasonally varied images per site). Multiple spectral indexes were evaluated using Sentinel-2 imagery. Our texture metric uses the red band to separate built-up infrastructure from spectrally similar bare ground. Our evaluation produced global thresholds by evaluating ground truth points against a range of site-specific optimal index thresholds across the 24 images. These were used to classify an ensemble, and then spectral indexes, texture, and stratified random sampling guided training data selection. The training data fit a random forest classifier to create final binary maps. Validation found an average overall accuracy of 79.95% (±4%) and an F1 score of 0.5304 (±0.07). The inclusion of the texture metric improved overall accuracy by 14–21%. A comparison to site-specific thresholds and a deep learning-derived layer is provided. This automated built-up infrastructure mapping framework requires only public imagery to support time-sensitive land management workflows. [ABSTRACT FROM AUTHOR]

Soil particle size fractions (PSFs) are important properties for understanding the physical and chemical processes in soil systems. Knowledge about the distribution of soil PSFs is critical for sustainable soil management. Although log-ratio transformations have been widely applied to soil PSFs prediction, the statistical distribution of original data and the transformed data given by log-ratio transformations is different, resulting in biased estimates of soil PSFs. Therefore, multivariate random forest (MRF) was utilized for the simultaneous prediction of soil PSFs, as it is able to capture dependencies and internal relations among the three components. Specifically, 243 soil samples collected across the Loess Plateau were used. Meanwhile, Landsat data, terrain attributes, and climatic variables were employed as environmental variables for spatial prediction of soil PSFs. The results depicted that MRF gave satisfactory soil PSF prediction performance, where the R2 values were 0.62, 0.53, and 0.73 for sand, silt, and clay, respectively. Among the environmental variables, nighttime land surface temperature (LST_N) presented the highest importance in predicting soil PSFs in the Loess Plateau, China. Maps of soil PSFs and texture were generated at a 30 m resolution, which can be utilized as alternative data for soil erosion management and ecosystem conservation. [ABSTRACT FROM AUTHOR]

Generating orchards spatial distribution maps within a heterogeneous landscape is challenging and requires fine spatial and temporal resolution images. This study examines the effectiveness of Sentinel-1 (S1) and Sentinel-2 (S2) satellite data of relatively high spatial and temporal resolutions for discriminating major orchards in the Khairpur district of the Sindh province, Pakistan using machine learning methods such as random forest (RF) and a support vector machine. A Multicollinearity test (MCT) was performed among the multi-temporal S1 and S2 variables to remove those with high correlations. Six different feature combination schemes were tested, with the fusion of multi-temporal S1 and S2 (scheme-6) outperforming all other combination schemes. The spectral separability between orchards pairs was assessed using Jeffries-Matusita (JM) distance, revealing that orchard pairs were completely separable in the multi-temporal fusion of both sensors, especially the indistinguishable pair of dates-mango. The performance difference between RF and SVM was not significant, SVM showed a slightly higher accuracy, except for scheme-4 where RF performed better. This study concludes that multi-temporal fusion of S1 and S2 data, coupled with robust ML methods, offers a reliable approach for orchard classification. Prospectively, these findings will be helpful for orchard monitoring, improvement of yield estimation and precision based agricultural practices. [ABSTRACT FROM AUTHOR]

The assessment of solar resources involves the utilization of physical or satellite models for the determination of solar radiation on the Earth's surface. However, a critical aspect of model validation necessitates comparisons against ground-truth measurements obtained from surface radiometers. Given the inherent challenges associated with establishing and maintaining solar radiation measurement networks—characterized by their expense, logistical complexities, limited station availability and the imperative consideration of climatic criteria for siting—countries endowed with substantial climatic diversity face difficulties in station placement. In this investigation, the measurements of annual solar irradiation, from meteorological stations of the National Weather Service in Mexico, were compared in different regions clustered by similarities in altitude, TL Linke, albedo and cloudiness index derived from satellite images; the main objective is to find the best ratio of annual solar irradiation in a set of clusters. Employing machine learning algorithms, this research endeavors to identify the most suitable model for predicting the ratio of annual solar irradiation and to determine the optimal number of clusters. The findings underscore the efficacy of the L-method as a robust technique for regionalization. Notably, the cloudiness index emerges as a pivotal feature, with the Random Forest algorithm yielding superior performance with a R2 score of 0.94, clustering Mexico into 17 regions. [ABSTRACT FROM AUTHOR]

Improved agricultural production systems, together with increased grain yield, are essential to feed the growing global population in the 21st century. Global gridded crop models (GGCMs) have been extensively used to assess crop production and yield simulation on a large geographical scale. However, GGCMs are less effective when they are used on a finer scale, significantly limiting the precision in capturing the yearly maize yield. To address this issue, we propose a relatively more advanced approach that downsizes GGCMs by combining machine learning and crop modeling to enhance the accuracy of maize yield simulations on a regional scale. In this study, we combined the random forest algorithm with multiple data sources, trained the algorithm on low-resolution maize yield simulations from GGCMs, and applied it to a finer spatial resolution on a regional scale in China. We evaluated the performance of the eight GGCMs by utilizing a total of 1046 county-level maize yield data available over a 30-year period (1980–2010). Our findings reveal that the downscaled models created for maize yield simulations exhibited a remarkable level of accuracy (R2 ≥ 0.9, MAE < 0.5 t/ha, RMSE < 0.75 t/ha). The original GGCMs performed poorly in simulating county-level maize yields in China, and the improved GGCMs in our study captured an additional 17% variability in the county-level maize yields in China. Additionally, by optimizing nitrogen management strategies, we identified an average maize yield gap at the county level in China ranging from 0.47 to 1.82 t/ha, with the south maize region exhibiting the highest yield gap. Our study demonstrates the high effectiveness of machine learning methods for the spatial downscaling of crop models, significantly improving GGCMs' performance in county-level maize yield simulations. [ABSTRACT FROM AUTHOR]

The presence of deep convective clouds is directly related to potential convective hazards, such as lightning strikes, hail, severe storms, flash floods, and tornadoes. On the other hand, Mexico has a limited and heterogeneous network of instruments that allow for efficient and reliable monitoring and forecasting of such events. In this study, a quasi-real-time framework for deep convective cloud identification and modeling based on machine learning (ML) models was developed. Eight different ML models and model assembly approaches were fed with Interest Fields estimated from Advanced Baseline Imager (ABI) sensor data on the Geostationary Operational Environmental Satellite-R Series (GOES-R) for one region in central Mexico and another in northeastern Mexico, both selected for their intense convective activity and high levels of vulnerability to severe weather. The results indicate that a simple approach such as Logistic Regression (LR) or Random Forest (RF) can be a good alternative for the identification and simulation of deep convective clouds in both study areas, with a probability of detection of (POD) ≈ 0.84 for Los Mochis and POD of ≈ 0.72 for Mexico City. Similarly, the false alarm ratio (FAR) ≈ 0.2 and FAR ≈ 0.4 values were obtained for Los Mochis and Mexico City, respectively. Finally, a post-processing filter based on lightning incidence (Lightning Filter) was applied with data from the Geostationary Lightning Mapper (GLM) of the GOES-16 satellite, showed great potential to improve the probability of detection (POD) of the ML models. This work sets a precedent for the implementation of an early-warning system for hazards associated with intense convective activity in Mexico. [ABSTRACT FROM AUTHOR]

The integration of optical and SAR datasets through ensemble machine learning models shows promising results in urban remote sensing applications. The integration of multi-sensor datasets enhances the accuracy of information extraction. This research presents a comparison of two ensemble machine learning classifiers (random forest and extreme gradient boost (XGBoost)) classifiers using an integration of optical and SAR features and simple layer stacking (SLS) techniques. Therefore, Sentinel-1 (SAR) and Landsat 8 (optical) datasets were used with SAR textures and enhanced modified indices to extract features for the year 2023. The classification process utilized two machine learning algorithms, random forest and XGBoost, for urban impervious surface extraction. The study focused on three significant East Asian cities with diverse urban dynamics: Jakarta, Manila, and Seoul. This research proposed a novel index called the Normalized Blue Water Index (NBWI), which distinguishes water from other features and was utilized as an optical feature. Results showed an overall accuracy of 81% for UIS classification using XGBoost and 77% with RF while classifying land use land cover into four major classes (water, vegetation, bare soil, and urban impervious). However, the proposed framework with the XGBoost classifier outperformed the RF algorithm and Dynamic World (DW) data product and comparatively showed higher classification accuracy. Still, all three results show poor separability with bare soil class compared to ground truth data. XGBoost outperformed random forest and Dynamic World in classification accuracy, highlighting its potential use in urban remote sensing applications. [ABSTRACT FROM AUTHOR]

The mapping of invasive plant species is essential for effective ecosystem control and planning, especially in protected areas. One of the widespread invasive plants that threatens the species richness of Natura 2000 habitats in Europe is the large-leaved lupine (Lupinus polyphyllus). In our study, this species was identified at two Natura 2000 sites in southern Poland using airborne HySpex hyperspectral images, and support vector machine (SVM) and random forest (RF) classifiers. Aerial and field campaigns were conducted three times during the 2016 growing season (May, August, and September). An iterative accuracy assessment was performed, and the influence of the number of minimum noise fraction (MNF) bands on the obtained accuracy of lupine identification was analyzed. The highest accuracies were obtained for the August campaign using 30 MNF bands as input data (median F1 score for lupine was 0.82–0.85), with lower accuracies for the May (F1 score: 0.77–0.81) and September (F1 score: 0.78–0.80) campaigns. The use of more than 30 MNF bands did not significantly increase the classification accuracy. The SVM and RF algorithms allowed us to obtain comparable results in both research areas (OA: 89–94%). The method of the multiple classification and thresholding of frequency images allowed the results of many predictions to be included in the final map. [ABSTRACT FROM AUTHOR]

We present an evaluation of different deep learning and machine learning approaches for tree health classification in the Black Forest, the Harz Mountains, and the Göttinger Forest on a unique, highly accurate tree-level dataset. The multispectral UAV data were collected from eight forest plots with diverse tree species, mostly conifers. As ground truth data (GTD), nearly 1500 tree polygons with related attribute information on the health status of the trees were used. This data were collected during extensive fieldwork using a mobile application and subsequent individual tree segmentation. Extensive preprocessing included normalization, NDVI calculations, data augmentation to deal with the underrepresented classes, and splitting the data into training, validation, and test sets. We conducted several experiments using a classical machine learning approach (random forests), as well as different convolutional neural networks (CNNs)—ResNet50, ResNet101, VGG16, and Inception-v3—on different datasets and classes to evaluate the potential of these algorithms for tree health classification. Our first experiment was a binary classifier of healthy and damaged trees, which did not consider the degree of damage or tree species. The best results of a 0.99 test accuracy and an F1 score of 0.99 were obtained with ResNet50 on four band composites using the red, green, blue, and infrared bands (RGBI images), while VGG16 had the worst performance, with an F1 score of only 0.78. In a second experiment, we also distinguished between coniferous and deciduous trees. The F1 scores ranged from 0.62 to 0.99, with the highest results obtained using ResNet101 on derived vegetation indices using the red edge band of the camera (NDVIre images). Finally, in a third experiment, we aimed at evaluating the degree of damage: healthy, slightly damaged, and medium or heavily damaged trees. Again, ResNet101 had the best performance, this time on RGBI images with a test accuracy of 0.98 and an average F1 score of 0.97. These results highlight the potential of CNNs to handle high-resolution multispectral UAV data for the early detection of damaged trees when good training data are available. [ABSTRACT FROM AUTHOR]

Water parameter estimation based on remote sensing is one of the common water quality evaluation methods. However, it is difficult to describe the relationship between the reflectance and the concentration of non-optically active substances due to their weak optical characteristics, and machine learning has become a viable solution for this problem. Therefore, based on machine learning methods, this study estimated four non-optically active water quality parameters including the permanganate index (CODMn), dissolved oxygen (DO), total nitrogen (TN), and total phosphorus (TP). Specifically, four machine learning models including Support Vector Machine Regression (SVR), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and K-Nearest Neighbor (KNN) were constructed for each parameter and their performances were assessed. The results showed that the optimal models of CODMn, DO, TN, and TP were RF (R2 = 0.52), SVR (R2 = 0.36), XGBoost (R2 = 0.45), and RF (R2 = 0.39), respectively. The seasonal 10 m water quality over the Zhejiang Province was measured using these optimal models based on Sentinel-2 images, and the spatiotemporal distribution was analyzed. The results indicated that the annual mean values of CODMn, DO, TN, and TP in 2022 were 2.3 mg/L, 6.6 mg/L, 1.85 mg/L, and 0.063 mg/L, respectively, and the water quality in the western Zhejiang region was better than that in the northeastern Zhejiang region. The seasonal variations in water quality and possible causes were further discussed with some regions as examples. It was found that DO would decrease and CODMn would increase in summer due to the higher temperature and other factors. The results of this study helped understand the water quality in Zhejiang Province and can also be applied to the integrated management of the water environment. The models constructed in this study can also provide references for related research. [ABSTRACT FROM AUTHOR]

Remote sensing-based models usually have difficulty in generating spatio-temporally continuous terrestrial evapotranspiration (ET) due to cloud cover and model failures. To overcome this problem, machine learning methods have been widely used to reconstruct ET. Therefore, studies comparing and evaluating the accuracy and effectiveness of reconstruction among different machine learning methods at the basin scale are necessary. In this study, four popular machine learning methods, including deep forest (DF), deep neural network (DNN), random forest (RF) and extreme gradient boosting (XGB), were used to reconstruct the ET product, addressing gaps resulting from cloud cover and model failure. The ET reconstructed by the four methods was evaluated and compared for Heihe River Basin. The results showed that the four methods performed well for Heihe River Basin, but the RF method was particularly robust. It not only performed well compared with ground measurements (R = 0.73) but also demonstrated the ability to fully reconstruct gaps generated by the TSEB model across the entire basin. Validation based on ground measurements showed that the DNN and XGB models performed well (R > 0.70). However, some gaps still existed in the desert after reconstruction using the DNN and XGB models, especially for the XGB model. The DF model filled these gaps throughout the basin, but this model had lower consistency compared with ground measurements (R = 0.66) and yielded many low values. The results of this study suggest that machine learning methods have considerable potential in the reconstruction of ET at the basin scale. [ABSTRACT FROM AUTHOR]

This study demonstrates a framework for using high-resolution satellite imagery to automatically map and monitor outbreaks of red needle cast (Phytophthora pluvialis) in planted pine forests. This methodology was tested on five WorldView satellite scenes collected over two sites in the Gisborne Region of New Zealand's North Island. All scenes were acquired in September: four scenes were acquired yearly (2018–2020 and 2022) for Wharerata, while one more was obtained in 2019 for Tauwhareparae. Training areas were selected for each scene using manual delineation combined with pixel-level thresholding rules based on band reflectance values and vegetation indices (selected empirically) to produce 'pure' training pixels for the different classes. A leave-one-scene-out, pixel-based random forest classification approach was then used to classify all images into (i) healthy pine forest, (ii) unhealthy pine forest or (iii) background. The overall accuracy of the models on the internal validation dataset ranged between 92.1% and 93.6%. Overall accuracies calculated for the left-out scenes ranged between 76.3% and 91.1% (mean overall accuracy of 83.8%), while user's and producer's accuracies across the three classes were 60.2–99.0% (71.4–91.8% for unhealthy pine forest) and 54.4–100% (71.9–97.2% for unhealthy pine forest), respectively. This work demonstrates the possibility of using a random forest classifier trained on a set of satellite scenes for the classification of healthy and unhealthy pine forest in new and completely independent scenes. This paves the way for a scalable and largely autonomous forest health monitoring system based on annual acquisitions of high-resolution satellite imagery at the time of peak disease expression, while greatly reducing the need for manual interpretation and delineation. [ABSTRACT FROM AUTHOR]

Several satellite missions are currently available to provide thermal infrared data at different spatial resolutions and revisit time. Furthermore, new missions are planned thus enabling to keep a nearly continuous 'eye' on thermal volcanic activity around the world. This massive volume of data requires the development of artificial intelligence (AI) techniques for the automatic processing of satellite data in order to extract significant information about volcano conditions in a short time. Here, we propose a robust machine learning approach to accurately detect, recognize and quantify high-temperature volcanic features using Sentinel-2 MultiSpectral Instrument (S2-MSI) imagery. We use the entire archive of high spatial resolution satellite data containing more than 6000 S2-MSI scenes at ten different volcanoes around the world. Combining a 'top-down' cascading architecture, two different machine learning models, a scene classifier (SqueezeNet) and a pixel-based segmentation model (random forest), we achieved a very high accuracy, namely 95%. These results show that the cascading approach can be applied in near-real time to any available satellite image, providing a full description of the scene, with an important contribution to the monitoring, mapping and characterization of volcanic thermal features. [ABSTRACT FROM AUTHOR]