(1) Background: Most hemodialysis patients may experience physiological and psychological stress. Exposure to nature has been previously reported to reduce the measures of psychological and physiological stress, and immune function. This study aimed to investigate psychological and physiological effects of integrated indirect forest therapy on chronic renal failure patients undergoing hemodialysis. (2) Methods: As a quasi-experiment, this study employed a nonequivalent control group, repeated measurements, and a non-synchronized design. A total of 54 participants were included: 26 and 28 in the experimental and control groups, respectively. During hemodialysis, five types of forest therapy stimuli (visual, auditory, olfactory, tactile, and motor) were applied 3 times per week for 4 weeks during 15-minute sessions. (3) Results: Positive but not negative emotion measures differed between the groups after the intervention. Fatigue and physiological stress levels were significantly reduced in the experimental group, whereas no significant difference was found between the groups on the measures of psychological stress. Activation of both the parasympathetic and sympathetic nervous systems was similar in both groups, as was the number of natural killer cells. (4) Conclusion: Integrated indirect forest therapy may help increase positive emotions and reduce fatigue and stress levels during hemodialysis in patients with chronic renal failure.

ClinVar is a web platform that stores around 774k curated entries, which allows exploring genetic variants and their associations with complex phenotypes. A partial set of ClinVar’s genetic associations were reported with conflict of interpretation or uncertain clinical impact significance, which currently challenges clinicians and geneticists. Here, we evaluate the performance of data pre-processing methods combined with classical prediction methods, such as Naive Bayes, Random Forest, and Support Vector Machine to build a meta-prediction model aiming to improve genetic pathogenicity interpretation. Models were trained with ClinVar data (September 2020), and genetic variants were annotated with eight functional impact predictors catalogued with SnpEff/SnpSift (v4.3). A 10-fold cross-validation strategy was performed for evaluation by accuracy, F1-Score, Receiver Operating Characteristic, Area Under Curve. The best meta-prediction model raises by combining one-hot encoding with tree-based classifiers as Random Forest, which shows Area Under Curve ≥ 0,93. We predict pathogenicity for 109k genetic variants, which were found labeled as uncertain significance or conflict of interpretation. Additionally, we implemented AmazonForest (https://www.lghm.ufpa.br/amazonforest), a web tool to query data for a set of 5k variants that were predicted with high pathogenic probability (RFprob >= 0.9).

Wildfire is the main disturbance in forested ecosystems of southern Europe and is due to complex interactions between climate-weather, fuels and people. Warmer and drier conditions projected in this region are expected to profoundly affect wildfires, which will impact ecosystems and humans. We review the scientific literature addressing the assessment of climate change impacts on wildfires in southern Europe, with a twofold objective: (i) report the trends in wildfire danger and activity projected under warming climate in southern Europe and (ii) discuss the limitations of wildfire projections under the specific biogeographical context of southern Europe.We identified 22 projection studies that examined future wildfire danger or wildfire activity at local, regional or continental scale. Under the scenario with the highest greenhouse gas emissions, we found that projections studies estimate an increase in future fire danger and burnt areas varying, on average, from 2 to 4 % and from 15 to 25 % per decade, respectively. Fire-prone area expansion to the north and to Mediterranean mountains is a concern, while climate-induced burnt area increase might be limited by fuel availability in the most arid areas. While all studies agreed on the direction of changes, further comparisons on the magnitude of increase remained challenging because of heterogeneous methodological choices between projections studies (climate models, projection period, spatial scale and fire metrics). We then described three main sources of uncertainty that may affect the reliability of wildfire projections: climate projections, climate-fire models, and the influences of fuel load/structure and human related factors on the climate-fire relationships. We finally suggest research directions to address some of these issues for the purpose of refining fire danger and fire activity projections in southern Europe.

The research objectives was to determine the plants vegetation profile in forest revegetation and Mount Arjuna springs. The method used was descriptive method. While based on the techniques and tools used to research, the author uses the survey method to obtain facts that occur in the research area, namely in the area around the springs of Arjuna mount, Pasuruan, East Java, Indonesia. Data obtained in the field are then processed and analyzed using GIS, ArcView 3.3 and Google Earth programs. The results of the study explain the profile of plant vegetation, namely plant stratification, bird wealth, taxonomic wealth and plant density have a significant effect on forest revegetation and springs in Arjuna mount. The forest vegetation profile has a positive and significant effect on the preservation of springs. The better the profile of forest vegetation, the preservation of springs will be better, and vice versa.

Remote sensing technologies may play a fundamental role in the environmental assessment of open-cast mining and the accurate quantification of mine land rehabilitation efforts. Here, we developed a systematic geographic object-based image analysis (GEOBIA) approach to map the amount of revegetated area and to quantify the land-use changes in open-cast mines in the Carajás region situated in the eastern Amazon. Based on high-resolution satellite images from 2011 to 2015 from different sensors (GeoEye, WorldView-3 and Ikonos), we quantified forests, cangas (natural metalliferous savanna ecosystems), mine land, revegetated areas and water bodies. Based on the GEOBIA approach, threshold values were established to discriminate land cover classes using spectral bands, and the NDVI and NDWI indices and LiDAR digital ground and slope models. The overall accuracy was higher than 90%, and the Kappa indices varied between 0.82 and 0.88. During the observation period, the mining complex expanded; for that, canga and forest vegetation was converted to mine land. At the same time, the amount of revegetated area increased. Thus, we conclude that our approach is capable of providing consistent information regarding land cover changes in mines, with a special focus on the amount of revegetation necessary to fulfill environmental liabilities.

Data on the seasonally dry tropical forests of Mexico have been examined in the light of statistical mechanics. The results suggest a division into two classes of species. There are drifting populations of a cosmopolitan class capable of existing in most dry forest sites; these have a statistical distribution previously only observed (globally) for populations of alien species. A high proportion of species found only at a single site are endemic and these prefer sites comparatively low in species richness.

The study was conducted in coniferous and deciduous old growth forests in two forest complexes located in: i) the fenced area of the Popielno Research Station of the Polish Academy of Sciences, with free-living Polish pony [Polish primitive horse (Equus ferus caballus)], and ii) open Maskulińskie Forest District managed (harvested) forest, without horses. The impact of forest animals on ground cover layer as well as on understory shrub layer and undergrowth, in i) area (horses and other forest animals) was compared with the results in ii) area (forest animals without horses). Very significant differences in the understory and undergrowth (above 0,5 m) layer vegetation communities structure between both areas and type of stands were found. The results suggest that the presence of the Polish horse substantially changed the species composition and increased the species diversity of the ground layer and shrub layer both on coniferous forest and in the deciduous forest habitats. The height of the shrub layer trees was lower by 30% in the area with the Polish horse.

While lidar-based forest inventory methods have been widely demonstrated, prediction of tree diameters with lidar is not well understood. The performance metrics typically used in studies for prediction of diameters can be difficult to interpret and may not support comparative inferences between sampling designs or study areas. We evaluate a variety of lidar and k nearest neighbor (k-NN) strategies for prediction of tree diameter distributions using two indices which are easier to interpret and compare. The indices are based on the coefficient of determination (R²), and root mean square deviation (RMSD). These indices facilitate comparisons with alternative (non-lidar) inventory strategies, and with other project areas. We evaluate k nearest neighbors (k-NN) dbh density (relative frequency by dbh class) prediction strategies with lidar for 190 training plots distribute across the 800 km² Savannah River Site in South Carolina, USA. We evaluate the performance of k-NN with respect to distance metrics, number of neighbors, predictor sets, and response sets. Amongst the examined strategies we found Mahalanobis distance with k = 3 neighbors performed best according to a number of criteria.

Many agonists for the estrogen receptor are known to disrupt endocrine functioning. We have developed a computational model that predicts agonists for the estrogen receptor ligand-binding domain in an assay system. Our model was entered into the Tox21 Data Challenge 2014, a computational toxicology competition organized by the National Center for Advancing Translational Sciences. This competition aims to find high-performance predictive models for various adverse-outcome pathways, including the estrogen receptor. Our predictive model, which is based on the random forest method, delivered the best performance in its competition category. In the current study, the predictive performance of the random forest models was improved by strictly adjusting the hyperparameters to avoid overfitting. The random forest models were optimized from 4,000 descriptors simultaneously applied to 10,000 activity assay results for the estrogen receptor ligand-binding domain, which have been measured and compiled by Tox21. At this time, our model delivers the highest predictive power on estrogen receptor agonists in the world. Furthermore, analysis of the optimized model revealed some important features of the agonists, such as the number of hydroxyl groups in the molecules.

The Gandoca-Manzanillo Wildlife Refuge in Costa Rica is facing significant threats from deforestation, raising concerns about its impact on forest health and carbon sequestration. This study utilized high-resolution satellite data from Sentinel-2 and Planet NICFI to assess forest degradation and quantify deforestation from January 2023 to August 2024. Multiple vegetation indices, including NDVI, GNDVI, EVI, SAVI, and MSI, were applied to monitor changes in forest health. Deforestation was calculated using relative NDVI change detection and thresholding techniques, while CO₂ emissions were estimated based on biomass loss derived from NDVI. The results show a 5% decrease in NDVI from Planet NICFI and a 2-3% decline from Sentinel-2, leading to the loss of 3.5 hectares of forest cover. Carbon dynamics revealed a CO₂ loss of 33,322.64 tons between January 2023 and January 2024, followed by a gain of 100,834.54 tons by August 2024, indicating partial recovery. The integration of Sentinel-2 and Planet NICFI data detected 13.22% deforestation with Sentinel-2 and 16.76% with Planet NICFI, emphasizing the importance of multi-source satellite imagery for accurate monitoring. These findings highlight the need for continuous monitoring and effective management to sustain carbon storage and combat deforestation.

The Suspended soils (which are also called canopy soil), associated with epiphytic plants in tropical forest canopies, have a very special type of biological cycle, morphological organization and ecological functional role. To date, only some features of these soils have been described for certain sectors of the tropics, in particular Southeast Asia. It was studies out whether the chemical composition and hydrological parameters depend significantly on the genesis of suspended soil, or they represent a single soil phenomenon with a certain supposed ecosystem function. Suspended soils are characterized by average values of hygroscopic moisture comparable with mineral soils. At the same time the maximum water holding capacity and field moisture capacity reaches hundreds and even thousands of percent which cannot but lead to mesophilization of the microclimate of the near-stem environments. In any case suspended soils are characterized by 4-10 higher values of hydrologic properties than mineral terrestrial soils of the same landscapes. The carbon content in suspended soil materials varied widely and was comparable with that of tropical forest litter and boreal peatlands . At the same time, tropical suspended soils are characterized by very low nitrogen content, which is reflected in very high carbon-to-nitrogen ratios. Such a values are typical of peats and forest litter and indicates a very low degree of humification of organic matter. Soil pH values tends to the acidic, and more often very acidic range. This corresponds well to the low content of labile forms of calcium and magnesium and corresponds to the oligotrophy of the soils. The content of mobile forms of phosphorus, potassium and ammonium nitrogen is extremely high in suspended soils. The high content of nutrients in tropical suspended soils is enormously high compared to forest soils of other climatic zones and should be noted as their most important attributive characteristic. Taking into account the high accumulative-sorption capacity of suspended soils, the low content of heavy metals in them indicates the absence of modern aerotechnogenic pollution of all the studied ecosystems. Thus, we are talking about a certain deficiency of heavy metals in the eluvial ecosystem of the tropical rain forest. This once again confirms the hypothesis that the essence of epiphytic soil formation was and is that the biotic cycle breaks away from the ground and moves evolutionarily into the middle and upper tiers of the tropical rainforest ecosystem. Tree trunks in this context represent vertically and subhorizontally oriented phase interfaces, similar to parent materials, and epiphytic soils are analogues of peat accumulations with extremely high water-holding capacity, which is of extremely important environmental significance.

The monitoring of fog density is of great importance in meteorology and its applications in environment, aviation and transportation. Nowadays, vision-based fog estimation from images taken with surveillance cameras has made a great supplementary contribution to the scarcely traditional meteorological fog observation. In this paper, we propose a new Random Forest (RF) approach for image-based fog estimation. In order to reduce the impact of data imbalance on recognition, the StyleGAN2-ADA (generative adversarial network with adaptive discriminator augmentation) algorithm is used to generate virtual images to expand the data of low proportions. Key image features related to fog are extracted, and an RF method, integrated with the hierarchical and k-medoids clustering, is deployed to estimate the fog density. The experiment conducted in Sichuan in February 2024 shows that the improved RF model has achieved an average accuracy of fog density observation to 93%, 6.4% higher than the RF model without data expansion, 3%-6% higher than the VGG16, the VGG19, the ResNet50, and the DenseNet169 with or without data expansion. What is more, the improved RF method exhibits a very good convergence as a cost-effective solution.

Starting from the objectives of the Nature Restoration Law and the concept of Forest Landscape Restoration, the objective of the following study is to present a methodological approach aimed at identifying areas that are best suited for the implementation of NbS for the improvement of landscape and habitat status in the study area identified by the Urban Atlas of the city of Campobasso (1028.64 km2). Using open data (ISPRA ecosystem services and Regional land use capability), an expert based approach (questionnaire) and a multi-criteria analysis (AHP), the Total Ecosystem Services Value (TESV) index was determined as a weighted additive sum of the criteria (Data) considered. The index was then classified into 8 clusters and the land use “Cropland” was extracted, matching Cluster 1 (740.09 Ha) with agricultural areas with less ES provision to be allocated to NbS and Cluster 8 (482.88 Ha) with valuable areas to be preserved, for a total of 2.34% of the study area. It was then possible to compare the “Forest” areas currently present in the study area with those of a possible future scenario, represented by the areas occupied today by forest with the addition of cluster 1 croplands. A landscape analysis was then conducted: it showed greater dispersion and fragmentation of forest patches in the future scenario but also greater connectivity and thus greater ecological functionality of the patches.

The interplay between wildfires and drought stress has significantly limited forest recovery in the Mediterranean ecosystem of central Chile. Since 2010, the region has faced a prolonged 'Mega Drought,' which intensified into a 'Hyper Drought' in 2019, with record-low precipitation and high temperatures, further constraining forest recovery amidst increasing wildfire frequency. This study evaluates short-term vegetation recovery across drought gradients in three sclerophyllous evergreen forests, analyzing Landsat time series (1987–2022) for 42 wildfires. Using the LandTrendr algorithm, we assessed post-fire forest recovery based on NDVI changes between pre-fire values and subsequent years. The results revealed significant differences in recovery across drought gradients and fire severity interactions. After five years, the xeric forest dominated by Quillaja saponaria and Lithrea caustica showed 89% recovery during the Mega Drought and 34% during the Hyper Drought. In contrast, the mesic forest dominated by Cryptocarya alba and Peumus boldus exhibited 28% recovery during the Mega Drought and 43% during the Hyper Drought. Degraded thorny shrublands showed the lowest recovery, averaging 23% under Hyper Drought and medium fire severity. These findings underscore the critical role of prolonged, severe drought in shaping forest recovery dynamics and highlight the need to understand these patterns to improve future forest resilience under increasingly arid conditions.

Objective: This study aims to construct an early risk warning model to predict the risk of treatment failure in patients undergoing sequential High-Flow Nasal Cannula (HFNC) oxygen therapy following mechanical ventilation weaning. Methods: A retrospective analysis was conducted on clinical data from 145 patients who received HFNC treatment in the Emergency Intensive Care Unit of the Third People's Hospital of Bengbu City from June 2018 to June 2023. A wide range of indicators including general information, comorbidities, laboratory test results, vital signs, disease-related scores, and oxygenation data were collected. Data analysis was performed using R software, starting with the Lasso regression to filter features related to treatment outcomes. Patients were divided into a training set (70%) and a validation set (30%) using random grouping software. The Random Forest algorithm was then employed to evaluate and rank the important features related to outcome indicators. The predictive efficiency and stability of the model were assessed through the Receiver Operating Characteristic (ROC) curve, calibration curve, and decision curve analysis, culminating in the construction of an early risk warning nomogram. Results: The study found that the Random Forest model exhibited high predictive accuracy in the training set (AUC=0.98) and good stability in the validation set (AUC=0.84). Key feature variables identified by the model, such as APACHE II score, BNP, NLR, mROX, and SOFA, were found to significantly impact the prediction of HFNC treatment failure. Based on these variables, an early warning nomogram was further developed, providing clinicians with a convenient and effective risk assessment tool. Conclusion: This study has constructed an early risk warning model based on the Random Forest algorithm, capable of effectively predicting the risk of HFNC treatment failure. The model's high predictive efficiency and stability offer strong support for clinical decision-making, contributing to personalized treatment and improved patient outcomes. Future research should aim to expand the sample size, perform multicenter validation, and explore the integration of the model into Clinical Decision Support Systems for real-time risk assessment and intervention.

Road safety and the effectiveness of the transportation system as a whole are significantly impacted by driver comfort. Road surface quality can play a significant part in the driver’s comfort experienced on roads in any country. This study employs a Random Forest technique to examine the association between road roughness and drivers' comfort during long-distance driving. Using Random Forest, a dependable machine learning technique that can handle big datasets and detect nonlinear correlations, this work aims to shed light on the complex dynamics between road conditions and driver’s comfort. 1,048,576 rows of data from MIRANDA, an application developed at the University of Gustave Eiffel, were used in this study as part of the data collected from a probe vehicle. The data collected includes an International Roughness Index (IRI). The IRI thresholds offer a simple method for assessing driver comfort and road irregularity. While highlighting how uneven and uncomfortable the road is, the research's findings (Road Roughness: SD – 0.73; Driver's Comfort: - Mean, 10.01, SD – 0.64) also contribute to the standardization of road condition evaluation and maintenance communication. This finding is anticipated to aid in the development of strategies for improving the welfare of long-haul drivers and fixing road infrastructure to comply with standard road index, ultimately leading to the creation of more efficient and sustainable transportation systems.

In order to reduce the environmental impact of energy enterprises, it is important to promptly identify cases of possible changes in the quality of wastewater generated at power plants, that is cases of exceeding the maximum permissible concentrations of pollutants in wastewater. The goal of the method for detecting complex anomalies in multidimensional time series obtained from smart energy station’s sensor channels is to increase the accuracy of detecting pollution levels in industrial wastewater. To achieve this goal, the following tasks were solved: methods for detecting time series anomalies were analyzed, the method for detecting complex anomalies was developed, software implementation of the algorithm was carried out and experiments were conducted. The developed method is proposed to be used in an smart energy monitoring system.

Liposomal amphotericin B (L-AmB) has become the cornerstone for the treatment of severe invasive fungal infections. However, renal toxicity may be an associated risk. Our aim was to evaluate the incidence of acute kidney injury (AKI) in critically ill patients receiving L-AmB for more than 48 hours. Clinical, demographic and laboratory variables were obtained retrospectively from electronic health record. Follow-up of the patients was for 7 days after L-AmB infusion. Patients were analyzed globally and differentiated between patients with "low" and "high" risk of developing AKI with machine learning techniques. Sixty-seven patients were included: 61 years, 67% male, with mean SOFA score of 4 (3-6.5) and a crude intensive care unit mortality of 34.3%. No significant variations in serum creatinine were observed during the observation period, except for the decrease in the high-risk subgroup. Overall, 26.8% of patients developed AKI-I, while 25% and 13% of patients at low and high risk of AKI developed AKI-I, respectively. The random forest models performed in overall and low-risk subgroup identified several major contributor’s factors to the development of AKI other than L-AmB administration. The development of AKI is multifactorial and the administration of L-AmB appears to be safe in this group of patients.

Over the last fifty years, management plans have become more descriptive with regard to potential sustainability of forest systems, raising questions about the feasibility of implementing management activities and ensuring sustainability of a wide variety of ecosystem services. To assess this issue, we conducted a survey among forest planning and operation research communities to understand their perceptions regarding the potential of a sample of sustainability statements currently used in forest plans to be incorporated into optimization models or other mathematical operations. The results revealed that only a few statements from the sample were deemed to have relatively mature or firm methodologies and data to enable inclusion in modern mathematical models for land use optimization. These statements were mostly related to economic sustainability, offering quantifiable information such as a non-declining flow of wood products over time and limits on the amount of timber harvested per decade. In contrast, sociocultural and, to some extent, ecological statements regarding sustainability were generally perceived to be more difficult to translate into mathematical modeling efforts. Particularly challenging were statements corresponding to sustaining natural or scenic characteristics of a forest. These findings may be attributed to various factors, including a lack of measurable indicators for sustainability and a potential lack of understanding about the modeling components and their interactions with planned management activities.

Forests have multifunctional roles. They not only increase forest carbon sinks through afforestation and management, but also indirectly realize multiple benefits of "economy + ecology + society". Therefore, we construct a multiple output-benefit system of forest carbon sinks based on the development of forest carbon sinks in provinces of China. This system is also able to distinguish the importance of forest carbon sinks from other outputs. For the efficiency evaluation of multi-level output systems, this paper constructs a cross-evaluation model based on evidential reasoning method, the original efficiency information by conducting efficiency evaluation before information fusion. Further, the spillover effects and influencing factors of forest carbon sink efficiency are analyzed through a spatial regression model to explore the spatial characteristics of forest carbon sink efficiency in China. The results show that: (1) there is a large gap between the cross-efficiency levels of forest carbon sinks in various regions of China, with significant spatial correlation. For the perspective of direct output benefits, the cross-efficiency of the eastern region is higher, while the cross-efficiency of the northwestern region is lower; for the perspective of indirect output benefits, the cross-efficiency of the southeastern region is higher. (2) China's forest carbon sink has a significant positive spillover effect, and the indirect output cross-efficiency has a significant negative impact on the forest carbon sink in its own region. Further analyzing the impact of indirect output cross-efficiency, it is found that the cross-efficiency of the ecological dimension and the cross-efficiency of the social dimension have a significant negative impact on the forest carbon sink in its own region, and the economic dimension has a significant positive impact on the forest carbon sink in its own region. In conclusion, corresponding policy recommendations are proposed for the development of forest carbon sinks to promote the realization of the dual-carbon goal.

In the context of the Lubumbashi Coal Production Basin (LCPB), the socio-economic repercussions of the COVID-19 pandemic have exacerbated pressures on populations dependent on forest resources for their subsistence. This study employs a comprehensive methodological approach, integrating advanced remote sensing techniques, including image classification, mapping, and detailed landscape analysis, to quantify alterations in forest cover within the LCPB during the pandemic period. Our findings reveal a consistent trend of declining forested area, characterized by processes of attrition and fragmentation observed throughout various study phases, spanning from May 2019 to November 2023. This reduction in forest cover, notably more pronounced in the vicinity of Lubumbashi city and the northern zone of the LCPB, proved to be less pronounced between November 2019 and September 2020, underscoring the influence of COVID-19 pandemic-induced confinement measures on forest management practices in the region. However, subsequent to this period of restriction, deforestation activity intensified, leading to significant landscape transformations within the LCPB, primarily attributable to expanded human activities, consequently resulting in a notable decrease in the proportion of land occupied by these natural ecosystems. Consequently, the size of the largest forest patch declined substantially, decreasing from 14.62% to 8.20% between May 2019 and November 2023, thereby fostering a heightened density of forest edges over time. Our findings provide a significant contribution to understanding the complex interactions between the COVID-19 pandemic and deforestation phenomena, emphasizing the urgent need to adopt adaptive management strategies and appropriate conservation measures in response to current economic challenges.

The silver fir (Abies alba Mill.) is among the most valuable conifers in Europe for ecological and economic reasons. In the course of history, primarily in the 20th century, its share in stands has been declining due to ill-suited management practices, especially clear-cut management, air pollution, and wildlife-induced damage. Based on recent knowledge of fir ecology and population dynamics, small-scale shelterwood and selection management have been introduced in fir stands, which have also stabilized them. Fir is an essential species for maintaining high stability and biodiversity, especially on planosols and in waterlogged habitats. Due to its shade tolerance and environmental flexibility, particularly at higher altitudes, it can coexist very well with many tree species in mixtures, which can increase the productive potential of stands. It can form stands of heterogeneous structure, ranging from single to multi-layer to selection. For its successful natural regeneration, it is essential to reduce cloven-hoofed game and thus prevent bud browsing damage. On the other hand, fir is a species relatively resistant to bark stripping and the spread of secondary rot compared to Norway spruce (Picea abies [L.] Karst.). During global climate change, fir is expected to shift to higher elevations with sufficient precipitation, while in the southern part of its natural range or at lower elevations, outside water-influenced habitats, it is likely to decline. This paper essentially reviews the description and distribution of the species, its ecological requirements, threats and diseases, habitat and stand conditions, and close-to-nature forest management practices with emphasis on ongoing climate change.

Fraudulent transactions pose a significant threat to financial institutions and e-commerce platforms.Machine learning models, trained on historical labeled data (fraudulent vs. legitimate transactions), are often employed to identify and prevent fraud. However, real-world datasets frequently exhibit class imbalance, where fraudulent transactions (minority class) are significantly outnumbered by legitimate transactions (majority class). Machine learning models may perform poorly as a result of this imbalance, underestimating fraud and favouring the majority class. This paper proposes a novel approach to address class imbalance and improve fraud detection accuracy. We explore the implementation of FROST (Feature space RObust Synthetic saTuration) oversampling, a technique specifically designed to generate synthetic samples for the minority class. The FROST function leverages the k-nearest neighbors (KNN) algorithm and a user-defined amplification factor (m) to create synthetic data points that closely resemble existing minority class instances. We integrate the FROST-enhanced oversampling technique into the machine learning pipeline for fraud detection. The paper evaluates the effectiveness of this approach compared to traditional oversampling methods and analyzes its impact on classification accuracy metrics.

This paper explores the capability of various machine learning algorithms, including Random Forests and advanced gradient boosting techniques such as XGBoost, LightGBM, and CatBoost, to predict customer churn in the telecommunications sector. For this analysis, a dataset available to the public was employed. The performance of these algorithms was assessed using recognized metrics, including Accuracy, Precision, Recall, F1-score, and the Receiver Operating Characteristic Area Under Curve (ROC AUC). These metrics were evaluated at different phases: subsequent to data preprocessing and feature selection; following the application of SMOTE and ADASYN sampling methods; and after conducting hyperparameter tuning on the data that had been adjusted by SMOTE and ADASYN. The outcomes underscore the notable efficiency of upsampling techniques such as SMOTE and ADASYN in addressing the imbalance inherent in customer churn prediction. Notably, the application of random grid search for hyperparameter optimization did not significantly alter the results, which remained comparatively unchanged. The algorithms' performance post- ADASYN application marginally surpassed that observed after SMOTE application. Remarkably, LightGBM achieved an exceptional F1-score of 89% and an ROC AUC of 95% subsequent to the ADASYN sampling technique. This underlines the effectiveness of advanced boosting algorithms and upsampling methods like SMOTE and ADASYN in navigating the complexities of imbalanced datasets and intricate feature interdependencies.

Soil, geochemical, microbiological and archaeological studies were conducted at eight settlements dating from the Paleolithic to Late Medieval and New Times near the Southern Trans-Urals mountains, Russia. The forest-steppe landscapes, rivers and abundance of mineral resources have attracted people to region since ancient times. Properties of cultural layer (CLs): close to neutral pH, well structured, absence of salinity, enrichment with exchangeable calcium and anthropogenic phosphorus (0.2–0.4%). The majority of CLs start from 3–25 cm and last up to 40–60 cm and contain 6–10% organic carbon (Corg), in a layer of 0–20 cm because modern processes are developing here, at Ishkulovo site (around 14th century) Corg decreases to 1.3%, because CL is below 80 cm and in the absence of fresh organic material it is mineralized. The content of microbial biomass of the 0–10 cm layer is 520–680 μg С/g of soil, microbes cause the emission of 0.2–1.0 μg C–CO2/g soil per h. The proximity of the studied sites to deposits of copper, chromium, zinc, manganese and Magnitogorsk Metallurgical and Uchaly processing plants creates geochemical anomalies of soils. Pollutants are also released into the CLs from bronze products, ore, ceramic fragments and raw materials from their manufacture. It leads to the concentration of pollutants of the 1st (arsenic, lead, zinc) and 2nd (cobalt, copper, nickel, chromium) hazard classes in the studied CLs, often in quantities higher than the clarke and less often – more than the allowable content for human health.

Background: Atypical teratoid/rhabdoid tumor (AT/RT) is an uncommon and aggressive pediatric central nervous system neoplasm. However, a universal clinical consensus or reliable prognostic evaluation system for this malignancy is lacking. Our study aimed to develop a risk model based on comprehensive clinical data to assist in clinical decision-making. Methods: We conducted a retrospective study by examining data from the Surveillance, Epidemiology, and End Results (SEER) repository, spanning 2000 to 2019. The external validation cohort was sourced from the Children's Hospital Affiliated to Chongqing Medical University, China. To discern independent factors affecting overall survival (OS) and cancer-specific survival (CSS), we applied Least Absolute Shrinkage and Selection Operator (LASSO) and Random Forest (RF) regression analyses. Based on these factors, we structured nomogram survival predictions and initiated a dynamic online risk-evaluation system. To contrast survival outcomes among diverse treatments, we used propensity score matching (PSM) methodology. Molecular data with the most common mutations in AT/RT were extracted from the Catalogue of Somatic Mutations in Cancer (COSMIC) database. Results: The annual incidence of AT/RT showed an increasing trend (APC, 2.86%; 95% CI:0.75-5.01). Our prognostic study included 316 SEER database participants and 27 external validation patients. The entire group had a median OS of 18 months (range 11.5 to 24 months) and median CSS of 21 months (range 11.7 to 29.2). Evaluations involving C-statistics, DCA, and ROC analysis underscored the distinctive capabilities of our prediction model. An analysis via PSM highlighted that individuals undergoing triple therapy (integrating surgery, radiotherapy, and chemotherapy) had discernibly enhanced OS and CSS. The most common mutations of AT/RT identified in the COSMIC database were SMARCB1, BRAF, SMARCA4, NF2, and NRAS. Conclusions: In this study, we devised a predictive model that effectively gauges the prognosis of AT/RT and briefly analyzed its genomic features, which might offer a valuable tool to address existing clinical challenges.

Abstract: Wildfires in forest ecosystems exert substantial ecological, economic, and social impacts. The effectiveness of fire management hinges on precise pre‐fire risk assessments to inform mitigation efforts. This study aims to investigate the relationship between predictions from pre‐fire risk assessments and outcomes observed through post‐fire burn severity analyses. In this study, forest fire risk was assessed with the Fuzzy Analytical Hierarchy Process, in which fire behavior factors were used as input. The degree of burn was determined using the Random Forest method using 11519 training points and 400 test points on Sentinel‐2 satellite images in three different classes. According to the results obtained from 266 selected test points located within the forest boundaries, all primary factors showed an increased effect in areas with high burn severity. Climate, in particular, emerged as the most influential factor, accounting for 52% of the overall impact. However, in cases of high fire severity, climate proved to be the most effective risk factor, ac‐counting for 67%. It followed by topography with 50% accuracy at high fire intensity. In the risk assessment is based on the FAHP method, climate was assigned the highest weight value among other factors with 32.2%. Topography emerged as the second most effective risk factor with 27 %. To evaluate the results more comprehensively both visually and statistically, two regions with different stand canopy characteristics were selected within the study area. While high burn severity had the highest accuracy in Case 1 area, moderate burn severity had the highest in Case 2 area. During the days when the fire continued, the direction of spread was obtained from MODIS images. In this way, the fire severity depending on the direction of fire progression was also interpreted. Through an analysis of various case studies and existing literature, the research underscores both the strengths and limitations inherent in predicting forest fire behavior‐based pre‐fire risk assessments. Furthermore, it emphasizes the necessity of continuous improvement to increase the success of forest fire management.

Fossil body parts of the two cold-adapted ground beetle species, Elaphrus lapponicus and Diacheila polita, were found from a deposit dated in the late Last Glacial period in Hokkaido, Japan. The paleoenvironment was estimated from their modern distribution and habitat, along with the results of the paleobotanical studies at the site. The temperature was at least 6 degrees lower than modern in summer, and the area around the site was mostly covered with forest tundra, which can currently be seen in areas further north of Hokkaido.

Prescriptions for fuels management are universally applied across forest types in British Columbia, Canada, to reduce fire behaviour potential in the wildland-urban interface. Fuel thinning treatments are assumed to reduce the potential for sustained ignition and crown fire initiation by reducing surface fuel loading. We hypothesized that these prescriptions are not appropriate for the coastal wet forests in the Whistler region of the province. Our study measured the efficacy of fuel thinning treatments in four stands located in the Whistler community forest. We examined several in-stand microclimate variables during snow melt in the spring and at the height of fire danger in late summer, at thinned and unthinned locations paired using GIS- analysis. We found that thinning increased the wildfire risk based on the differences between unthinned and thinned areas in the same forest stand.

A land cover map of two arctic catchments, nearby the Abisko Scientific Research Station, was obtained from a classification of a Sentinel-2 satellite image and a ground survey performed in July 2022. The two contiguous catchments, Miellajokka and Stordalen, are covered by various ecotypes, from boreal forest to alpine tundra and peatland. Two classification algorithms, support vector machine and random forest, were tested and gave very similar results. The percentage of correctly classified pixels was over 88% in both cases. The developed workflow relies solely on open source software and acquired ground observations. Space organization was directed by the altitude as demonstrated by the intersection of the land cover with the topography. Comparison between this new land cover map and previous ones based on data acquired between 2008 and 2011 shows some trends of vegetation cover evolution in response to climate change in the considered area. This land cover map is key input data for permafrost modeling, and hence for the quantification of climate change impacts in the studied area.

Nowadays, there is extensive production and use of plastic materials for different industrial activities. These plastics, either from their primary production sources or through degradation processes of the plastics themselves, can contaminate the ecosystem with micro and nanoplastics. Once in the aquatic environment, these microplastics can be the basis for the adsorption of chemical pollutants, favoring that these chemical pollutants disperse more quickly in the environment and can affect living beings. Due to the lack of information on adsorption, three machine learning models (random forest, support vector machine, and artificial neural network) to predict different microplastic/water partition coefficients (log Kd) were developed using two different approximations (based on the number of input variables). The best-selected machine learning models present, in general, correlation coefficients upper than 0.92 in the query phase, which indicate that these type of models could be used for a rapid estimation of the absorption of organic contaminants on microplastics.

The long-term resilience of Pyrenean forests in the face of historical anthropogenic clearing remains largely unknown. The palynological study of the varved sediments from a mid-elevation (1027 m) karstic lake provides a high-resolution record of three major century-scale deforestation/recovery (DR) cycles that occurred in the last two millennia, during Roman, Medieval and Modern times. Each DR cycle is characterized considering three different levels: overall forest trends, by forest type and by individual taxa. Overall, the studied forests exhibited high resilience, as they recovered almost completely after each deforestation event (bulk resilience). The critical point of no return (tipping point) beyond which forests would have irreversibly disappeared from the region was never reached, even after deforestation magnitudes above 60%. The different forest types identified (conifer, sclerophyll and deciduous) persisted over time, showing similar heterogeneous patterns with minor spatial reorganizations (mosaic resilience). Individually, the main forest taxa underwent minor variations in their relative abundances, always within the same attraction domains (community resilience). The high levels of resilience documented in these Pyrenean forests are attributed to the action of metapopulation and metacommunity processes and mechanisms in a highly dynamic patchy environment. Conservation actions should be focused on the maintenance of these spatial patterns and the associated ecological dynamics.

Non-woody forest products play a vital role in the economic uplifts of the local communities living in forest areas. This study was conducted in the Shishikoh Valley, district Chital to identify the economic importance of the Chilghoza nuts. The study was based on interviews and questionnaires, and a total of 50 questionnaires were collected from the whole valley. The results showed that before 2003, the people were unaware of the high economic value of the Chilghoza nuts, they used to sell the Chilghoza nuts at a very low price (US$ 1.7 to 2.4 per bag: approximately 5 Kg) and harvest it for fuelwood purposes. After the IPRP interventions, training and awareness, the local people started conservation of Chilghoza trees and sustainable cone harvesting. The total collection and sell out of Chilghoza nuts in the Shishikoh valley jumped to 87.9 tons and US$ 7.64/kg in 2005 from 32.7 tons and US$ 0.34-0.48/kg, respectively. This study concludes that IPRP interventions improved the economic values of Chilghoza nuts as well as the living standards of the local people.

: Emaravirus (Order Bunyavirales; Family Fimoviridae) is a genus comprising over 20 emerging plant viruses with a worldwide distribution and economic impact. Emaraviruses infect a variety of host plants and have especially become prevalent in important long-living woody plants. These viruses are enveloped, with a segmented, single-stranded, negative-sense RNA genome and are transmitted by eriophyid mites or mechanical transmission. Emaraviruses have four core genome segments encoding an RNA-dependent RNA polymerase, a glycoprotein precursor, a nucleocapsid protein, and a movement protein. They also have additional genome segments, whose number varies widely. We report here that the proteins encoded by these segments form three main homology groups: a homolog of the sadwavirus Glu2 Pro glutamic protease; a protein involved in pathogenicity, which we named “ABC”; and a protein of unknown function, which we named “P55”. The distribution of these proteins parallels the emaravirus phylogeny and suggests, with other analyses, that emaraviruses should be split into at least two genera. Reliable diagnosis systems are urgently needed to detect emaraviruses, assess their economic and ecological importance, and take appropriate measures to prevent their spread (such as routine testing, hygiene measures, and control of mite vectors). Additional research needs include understanding the function of emaravirus proteins, breeding resistant plants and clarifying transmission modes.

Uganda possesses natural rainforests that serve enormous environmental ecosystems and biodiversity services. Moreover, the country is known for its various tropical rainforest hardwoods, birds, and animal species. Over the years, the trend in the natural forest land has declined at an alarming rate; hence need to investigate the possible drivers. The loss of such biodiversity and ecosystems risks desertification and extreme climatic condition. As the world moves towards Zero Deforestation 2030, understanding the determinants of deforestation and forest degradation is paramount. Therefore, the main objective of this study was to understand the impact and relationships between net forest conversion, energy emission, agriculture, and forest production of Roundwood. We used data from FAO for the period 2004-2016. Using the ADF and KPSS test, we checked for the unit root presence in the variables. Also, the study used two different regression models: multiple linear regression and dynamic linear model. To analyze the determinants of deforestation, we used net forest conversion in Uganda. There was 94 % variation in the dependent variable (Net Forest conversion). The outcome of the dynamic linear regression showed that agriculture and energy emission positively impact net forest conversion. Based on our findings, this study recommended the modernization of agriculture by the government of Uganda to stop cutting down the forests on a big scale. Also, the study suggested that the government strictly legislate Roundwood and wood fuels/charcoal and firewood to reduce huge dependency on forests toward Zero-Deforestation by 2030. If well-structured and implemented, government policies could solve the unnecessary over dependency on the rainforest, the heart of the region's climatic conditions.

Fossil-based emissions can be avoided through using wood in place of non-renewable raw materials as energy and materials. However, increasing wood harvest influences forest carbon stocks. This effect may reduce the overall climate benefit of wood use significantly but is widely overlooked. We carried out a systematic review of simulation studies and compared differences in forest carbon and amount of wood harvested between more and less intensive wood harvest scenarios for three different time perspectives; short (1-30 years), mid (31-70 years) and long (71-100 years). Out of more than 450 reviewed studies 44 provided adequate data. Our results show that increased harvesting reduced carbon stocks over 100 years in temperate and boreal forests roughly 1.6 (stdev 0.9) t_C per t_C harvested. The value proved to be robust when outliers explicitly influenced by other factors than change in harvest rate, such as increase in fertilization or forest area, were removed. Interestingly, no significant difference in carbon impacts was found for average values of boreal and temperate forests or between short and long time-horizons. However, impacts tend to be greatest in the mid-term. This carbon balance indicator that we estimated can be interpreted as carbon debit of wood harvest in forests. It is significant compared with the typical GHG credits in technosphere generated by avoiding fossil emissions in substitution and increase in carbon storage in harvested wood products, and should not be ignored. Our estimates provide default values that can directly be included in GHG balances of products or assessment of mitigation policies and measures related to wood use. However, more systematic scenarios and transparent data in which different factors influencing forest carbon stocks are separately studied are clearly required to provide better constrained estimates for specific forest types.

Incident solar radiation (insolation) passing through the forest canopy to the ground surface is either absorbed or scattered. This phenomenon, known as radiation attenuation, is measured using the extinction coefficient (K). The amount of radiation at the ground surface of a given site is effectively controlled by the canopy’s surface and structure, determining its suitability for plant species.Menhinick’s and Simpson biodiversity indexes were selected as spatially explicit response variables for the regression equation using canopy structure metrics as predictors. Independent variables include modeled area solar radiation, LiDAR derived canopy height, effective leaf area index data derived from multi-spectral imagery, and canopy strata metrics derived from LiDAR point-cloud data. The results support the hypothesis that, 1.) canopy surface and strata variability may be associated with understory species diversity due to habitat partitioning and radiation attenuation, and that, 2.) such a model can predict both this relationship and biodiversity clustering.The study data yielded significant correlations between predictor and response variables and was used to produce a multiple-linear model comprising canopy relief, texture of heights, and vegetation density to predict understory plant diversity. When analyzed for spatial autocorrelation, the predicted biodiversity data exhibited non-random spatial continuity.

The construction of sustainable urban forests follows the principle that well-being in people is promoted when exposed to tree population. Facial expression is the direct representation of inner emotion that can be used to assess real-time perception in urban forests. The emergence and change of facial expressions for forest visitors are in an implicit process. The reserved character of oriental races strengthens the requirement for the accuracy to recognize expressions through instrument rating. In this study, a dataset was established with 2,886 randomly photographed faces from visitors in a constructed urban forest park and a promenade at summertime in Shenyang City at Northeast China. Six experts were invited to choose 160 photos in total with 20 images representing one of eight typical expressions as angry, contempt, disgusted, happy, neutral, sad, scared, and surprised emotions. The FireFACE ver. 3.0 software was used to test hit-ratio validation as the accuracy (ac.) to match machine-recognized photos with those identified by experts. According to Kruskal-Wallis test on the difference from averaged scores in 20 recently published papers, contempt (ac.=0.40%, P=0.0038) and scared (ac.=25.23%, P=0.0018) expressions cannot pass the validation test. Therefore, the FireFACE can be used as an instrument to analyze facial expression from oriental people in urban forests but contempt and scared expressions cannot be identified.

Trees in residential landscapes provide many benefits, but can injure persons and damage property when they fail. In hurricane-prone regions like Florida, USA, the regular occurrence of hurricanes has provided an opportunity to assess factors that influence the likelihood of wind-induced tree failure and develop species failure profiles. We assessed open-grown trees in Naples, Florida, following the passage of Hurricane Irma in September 2017 to determine the effect of relevant factors on the degree of damage sustained by individual trees. Of 4,034 assessed individuals (n = 15 species), 74% sustained no damage, 4% sustained only minor damage (i.e., minimal corrective pruning needed), 6% sustained significant damage (i.e., major corrective pruning needed), and 15% were whole tree failures (i.e., overturned trees or trees requiring removal). The proportion of individuals in each damage category varied among species, stem diameter at 1.4 m above ground, and the presence of utility lines, which was a proxy for maintenance. We compared our results with the findings of seven previous hurricanes in the region to explore species’ resilience in hurricanes.

Reliable estimates of poplar plantations area are not available at the French national scale due to the unsuitability and low update rate of existing forest databases for this short-rotation species. While supervised classification methods have been shown to be highly accurate in mapping forest cover from remotely sensed images, their performance depends to a great extent on the labelled samples used to build the models. In addition to their high acquisition cost, such samples are often scarce and not fully representative of the variability in class distributions. Consequently, when classification models are applied to large areas with high intra-class variance, they generally yield poor accuracies. In this paper, we propose the use of active learning (AL) to efficiently adapt a classifier trained on a source image to spatially distinct target images with minimal labelling effort and without sacrificing classification performance. The adaptation consists in actively adding to the initial local model, new relevant training samples from other areas, in a cascade that iteratively improves the generalisation capabilities of the classifier, leading to a global model tailored to different areas. This active selection relies on uncertainty sampling to directly focus on the most informative pixels for which the algorithm is the least certain of their class labels. Experiments conducted on Sentinel-2 time series showed that when the same number of training samples was used, active learning outperformed passive learning (random sampling) by up to 5% of overall accuracy and up to 12% of class F-score. In addition, and depending on the class considered, the random sampling required up to 50% more samples to achieve the same performance of an active learning-based model. Moreover, the results demonstrate the suitability of the derived global model to accurately map poplar plantations among other tree species with overall accuracy values up to 14% higher than those obtained with local models. The proposed approach paves the way for national-scale mapping in an operational context.

Wildfires have impacted thousands of acres of forests in Arizona and New Mexico in recent years. The extensive damage has been partially attributed to the current forest condition where many once open stands now consist of dense stands of younger, unhealthy trees and buildups of forest fuels. In Arizona, land managers and community organizations developed the Four Forests Restoration Initiative (4FRI) to attempt to rectify the situation by thinning the forest while protecting wildlife and watershed resources. Integrated resource information and research are particularly important today with society’s increased demands for high quality water, healthy forests, sound fire management, and viable wildlife populations. However, many managers are not familiar with previous integrated watershed research in the Southwest that could provide a strong basis for current management decisions or as a basis for future research. New and modified prescriptions to manage the Southwest’s forest resources are vital to answer the threats of wildfires and insect infestations. Integrated forest resource management is necessary to meet the diverse needs of society and the land. The 4FRI plan raised many questions about the effects of silvicultural prescriptions on tree, wildlife, and water resources. The Beaver Creek, Thomas Creek, and Castle Creek watershed experiments all had goals of evaluating integrated resource management options and of providing managers and scientists with useful management information. Much of the knowledge gained in these programs can provide forest managers with a better, more holistic basis for future integrated management of the Southwest USA’s forests and woodlands. This paper reviews research on three watershed areas in Arizona where integrated resource management was successful.

The research objectives were (1) to examine the determinants of community empowerment towards the revegetation of Arjuna mount forests; (2) assessing the influence of geography and soil conditions on the revegetation of Arjuna mount forests; (3) reviewing the effect of Arjuna mount forest revegetation on the preservation of spring water sources and (4) formulate a model of community empowerment around the forest based on revegetation of Arjuna mount and preservation of the area around the springs. The method used was descriptive method. While based on the techniques and tools used to research, the author uses the survey method to obtain facts that occur in the research area, namely in the area around the spring of the Kedunglarangan and Gumandar Prigen watersheds, Pasuruan Regency. Data obtained in the field are then processed and analyzed using GIS, ArcView 3.3 and Google Earth programs. The study population covering the community around the forest and springs, who lived in three villages, namely Leduk, Jatiarjo and Dayurejo village, Pasuruan, East Java-Indonesia. Data obtained from respondents with questionnaires and analyzed using SEM (structural equation modeling), to find out a general description of respondents' responses about community participation in the preservation of Arjuna mount forest. The results showed that (1) human resources, economic, social, local institutional factors, facilities and infrastructure had a very significant effect on forest revegetation; (2) geography and land variables have a significant effect on forest revegetation; (3) forest revegetation variables have a significant effect on the preservation of springs; and (4) The development of human resources, economic, local institutional, facilities and infrastructure in the community around the forest was a relevant model for the success of forest revegetation and the preservation of springs in Arjuna mount.

The microstructures and mechanical properties of the hot-rolled Fe-6.5wt.%Si sheet are analyzed. The microstructure of the hot-rolled sheet is layered along the thickness direction. The surface exhibits fine and equiaxed grains, whereas the center part shows coarse and elongated grains with a <101> fiber texture along the rolling direction. Serrated flow behavior is observed during tensile deformation of both the hot-rolled sheet and its center samples at 350 °C; thus, the serrated flow of the hot-rolled sheet is mainly attributed to the serration of the center part. The analyses of dislocation configurations, ordered structures, and crystal orientation show that the serrated flow behavior results from the interaction of solutes with mobile dislocations. Mobile dislocations are pinned by combining parallel forest dislocations with the pipe diffusion of solution atoms. This study provides a new perspective for the deformation mechanism of the Fe-6.5wt.%Si alloy.

The black timber bark beetle Xylosandrus germanus (Blandford) is an invasive ambrosia beetle originating from Southeastern Asia that has become successfully established within Europe and North America. Herein, we provide a review of the spread and distribution of this pest of trees and timber across Europe before and after 2000, along with a review of its habitat preferences. Since the spread of X. germanus across Europe has accelerated rapidly post-2000, emphasis is placed on this period. X. germanus was first recorded in Germany in 1951 and since then in 21 European countries along with Russia. Ethanol-baited traps were deployed in oak, beech, and spruce forest ecosystems in the Western Carpathians, Central Europe, Slovakia, to characterize the distribution and habitat preference of this non-native ambrosia beetle. Captures of X. germanus within Slovakia have been rising rapidly since its first record in 2010, and now this species dominates captures of native ambrosia beetles. X. germanus has spread throughout the whole Slovakia from the south-southwest to the north-northeast over the period of 5–10 years and has also spread vertically into higher altitudes within this country. While living but weakened trees in Europe and North America are attacked by X. germanus, the greatest negative impact within Slovakia is attacks on recently felled logs of oak, beech and spruce trees providing high quality timber/lumber. We suggest that the recent rapid spread of X. germanus in Central Europe is being facilitated by environmental changes, specifically global warming, and the increasing frequency of timber trade. Recommendations for management of X. germanus in forest ecosystems are proposed and discussed, including early detection, monitoring, sanitary measures, etc.

A timely inventory of agricultural areas and crop types is an essential requirement for ensuring global food security. Satellite remote sensing has proven to be an increasingly more reliable tool to identify crop types. With the Copernicus program and its Sentinel satellites, a growing source of satellite remote sensing data is publicly available at no charge. Here we use joint Sentinel-1 radar and Sentinel-2 optical imagery to create a crop map for Belgium. To ensure homogenous radar and optical input across the country, Sentinel-1 12-day backscatter composites were created after incidence angle normalization, and Sentinel-2 NDVI images were smoothed to yield dekadal cloud-free composites. An optimized random forest classifier predicted the 8 crop types with a maximum accuracy of 82% and a kappa coefficient of 0.77. We found that a combination of radar and optical imagery always outperformed a classification based on single-sensor inputs, and that classification performance increased throughout the season until July, when differences between crop types are largest. Furthermore we showed that the concept of classification confidence derived from the random forest classifier provided insight in the reliability of the predicted class for each pixel, clearly showing that parcel borders have a lower classification confidence. We concluded that the synergistic use of radar and optical data for crop classification led to richer information increasing classification accuracies compared to optical-only classification. Further work should focus on object-level classification and crop monitoring to exploit the rich potential of combined radar and optical observations.

This paper proposes an Integrated Fire Management (IFM) framework that can be used to support communities and resource managers in finding effective and efficient approaches to prevent damaging fires, as well as maintain desirable fire regimes in Kenya. Designing and implementing an IFM approach in Kenya calls for a systematic understanding of the various uses of fire and the underlying perceptions and traditional ecological knowledge of the local people. The here proposed IFM framework allows an evaluation of the risks posed by fires, while balancing them with their beneficial ecological and economic effects, and thus developing effective fire management approaches. A case study of the proposed IFM framework was conducted in Gathiuru Forest that is part of the larger Mt. Kenya Forest Ecosystem. Focus group discussions were held with key resource persons, primary and secondary data on socio-economic activities were studied, fire and weather records were analyzed and the current fire management plans were consulted. Questionnaires were used to assess how the IFM is implemented in the Gathiuru Forest Station. The results show that the proposed IFM framework is scalable and can be applied in places with fire-dependent ecosystems as well as in places with fire-sensitive ecosystems in Kenya. The effectiveness is dependent on the active participation, formulation and implementation of the IFM activities by the main stakeholder groups (Kenya Forest Service (KFS), Kenya Wildlife Service (KWS), and the Community Forest Associations (CFA)). The proposed IFM framework helps in implementing cost-effective approaches to prevent damaging fires and maintain desirable fire regimes in Kenya.

The productivity of forests is often considered to be limited by the availability of phosphorus (P). Knowledge of the role of organic and inorganic P in humid subtropical forest soils is lacking. In this study, we used chemical fractionation and 31P nuclear magnetic resonance (NMR) spectroscopy to characterize the form of P and its distribution in undisturbed perhumid Chamaecyparis forest soils. The toposequence of transects was investigated for the humic layer from summit to footslope and lakeshore. The clay layer combined with a placic-like horizon in the subsoil may affect the distribution of soil P because both total P and organic P (Po) contents in all studied soils decreased with soil depth. In addition, Po content was negatively correlated with soil crystalline Fe oxide content, whereas inorganic P (Pi) content was positively correlated with soil crystalline Fe oxide content and slightly increased with soil depth. Thus, Pi may be mostly adsorbed by soil crystalline Fe oxides in the soils. Among all extractable P fractions, the NaOH-Po fraction appeared to be the major component, followed by NaHCO3-Po; the resin-P and HCl-Pi fractions were lowest. In addition, we found no typical trend for Pi and Po contents in soils with topographical change among the three sites. From the 31P-NMR spectra, the dominant Po form in soils from all study sites was monoesters with similar spectra. The 31P-NMR findings were basically consistent with those from chemical extraction. Soil formation processes may be the critical factor affecting the distribution of soil P. High precipitation and year-round high humidity may be important in the differentiation of the P species in this landscape.

Riparian zones represent ecotones between terrestrial and aquatic ecosystems and are of utmost importance to biodiversity and ecosystem functions. Modelling/mapping of these valuable and fragile areas is needed for an improved ecosystem management, based on an accounting of changes and on monitoring of their functioning in time. In Europe, the main legislative driver behind this goal is the European Commission’s Biodiversity Strategy to 2020, on one hand aiming at reducing biodiversity loss, on the other hand enhancing ecosystem services by 2020, and restoring them as far as feasible. A model, based on Earth Observation data, including Digital Elevation Models, hydrological, soil, land cover/land use data, and vegetation indices is employed in a multi-modular and stratified approach, based on fuzzy logic and object based image analysis, to delineate potential, observed and actual riparian zones. The approach is designed in an open modular way, allowing future modifications and repeatability. The results represent a first step of a future monitoring and assessment campaign for European riparian zones and their implications on biodiversity and on ecosystem functions and services. Considering the complexity and the enormous extent of the area, covering 39 European countries, including Turkey, the level of detail is unprecedented. Depending on the accounting modus, 0.95%–1.19% of the study area can be attributed as actual riparian area (considering Strahler’s stream orders 3-8, based on the Copernicus EU-Hydro dataset), corresponding to 55,558–69.128 km². Similarly depending on the accounting approach, the potential riparian zones are accounted for about 3-5 times larger. Land cover/land use in detected riparian areas was mainly of semi-natural characteristics, while the potential riparian areas are predominately covered by agriculture, followed by semi-natural and urban areas.

Precipitation during a specific return period plays an important role in the design of hydraulic infrastructure. The traditional approach involves collecting annual maximum precipitation data from a station, then subjecting it to statistical probability distributions (PDFs), and finally selecting the one with the lowest value in a goodness-of-fit test (e.g., Kolmogorov-Smirnov). Nevertheless, this methodology assumes current data, leaving uncertainty regarding its suitability for outdated data. The aim of this study is to compare the probability density functions (e.g., Normal, Log Normal, Pearson III) with the machine learning algorithm known as Random Forest (RF) for calculating precipitation at different return periods, using the province of Arequipa in Peru as the study area, through 5 stations located in different parts of the province. This comparison was conducted using the RMSE metric in both methods to evaluate their performance, resulting in RF having a lower RMSE than PDFs in most cases in calculating precipitation for return periods of 2, 5, 10, 20, 50 and 100 years for the studied stations.

Latent Tuberculosis Infection (LTBI) poses a significant public health challenge, especially in populations with high HIV prevalence and limited healthcare access. Early detection and targeted interventions are essential to prevent the progression of active tuberculosis. This study develops predictive models for LTBI outcomes using logistic regression and machine learning approaches and evaluates strategies to improve LTBI awareness and testing. Data from rural areas in the Eastern Cape, South Africa, were analyzed to identify key demographic, health, and knowledge-related factors influencing LTBI outcomes. Logistic regression was employed to pre-dict LTBI positivity based on factors such as age, education, and HIV status. Machine learning models, including decision trees and random forests, were also applied to compare predictive accuracy. A knowledge diffusion model was used to assess the impact of educational interventions on increasing LTBI awareness and testing rates. Logistic regression achieved an accuracy of 66.67% with high precision (80%) but low recall (33%) for LTBI-positive cases, identifying age, HIV status, and LTBI awareness as significant predictors. The random forest model outperformed logistic regression in accuracy (59.26%) and F1-score (0.63), providing a better balance between precision and recall. Feature importance analysis revealed that age, occupation, and knowledge of LTBI symptoms were the most critical factors across both models. The knowledge diffusion model demonstrated that targeted interventions significantly increased LTBI awareness and testing, particularly in high-risk groups. While logistic regression offers more interpretable results for public health interventions, machine learning models like random forests provide enhanced predictive power by capturing complex relationships between demographics and health factors. These findings highlight the need for targeted educational campaigns and increased LTBI testing in high-risk populations, particularly those with limited awareness of LTBI symptoms.

Apple proliferation is among the most important diseases in European fruit production. Early and reliable detection enables farmers to respond appropriately and to prevent further spreading of the disease. Traditional phenotyping approaches by human observers consider multiple symptoms, but these are difficult to measure automatically in the field. Therefore, we investigated the potential of hyperspectral imaging in combination with data analysis by machine learning algorithms to detect the symptoms solely based on the spectral signature of collected leaf samples. In the growing seasons 2019 and 2020, we collected a total of 1,160 leaf samples. Hyperspectral imaging with a dual camera setup in spectral bands from 400 nm to 2500 nm was accompanied with subsequent PCR analysis of the samples to provide reference data for the machine learning approaches. Data processing consists of preprocessing for segmentation of the leaf area, feature extraction, classification and subsequent analysis of relevance of spectral bands. Results show, that imaging multiple leaves of a tree enhances detection results, that spectral indices are a robust means to detect the diseased trees, and that the potentials of the full spectral range can be exploited using machine learning approaches.

This paper investigates the applicability of deep learning models for predicting the severity of forest wildfires, utilizing an innovative benchmark dataset called EO4WildFires. EO4WildFires integrates multispectral imagery from Sentinel-2, SAR data from Sentinel-1 and meteorological data from NASA Power annotated with EFFIS data for forest fire detection and size estimation. Resulting in a coverage of 45 countries (Figure \ref{Figure1-events}) with a total of 31730 wildfire events from 2018 to 2022. All these various sources of data are archived into data cubes, with the intention of assessing wildfire severity by considering both current and historical forest conditions, utilizing a broad range of data including temperature, precipitation, and soil moisture. The experimental setup has been arranged to test different deep learning architectures’ effectiveness in predicting the size and shape of wildfire burned areas. The study incorporates both Image Segmentation networks and Visual Transformers, employing a consistent experimental design across various models to ensure comparability of results. Adjustments were made to the training data, such as the exclusion of empty labels and very small events, to refine the focus on more significant wildfire events and potentially improve prediction accuracy. The models’ performance was evaluated using metrics like F1 Score, IoU Score, and Average Percentage Difference (aPD). These metrics offer a multi-faceted view of model performance, assessing aspects such as precision, sensitivity, and the accuracy of the burned area estimation. Through extensive testing with Image Segmentation networks and Visual Transformers, the research not only aims at enhancing the accuracy of estimating the burned area but also underscores the significance of quality training data and the comparative effectiveness of traditional segmentation methods over transformer-based models for this specific application. The dataset is published with an Open Access license and is hosted at: \url{10.5281/zenodo.7762564}.

This study contributes to the Sybil nodes detecting algorithm in Online Social Networks (OSNs). As major communication platforms, online social networks get significantly guarded from malicious activity. A thorough literature review identified various detection and prevention Sybil attack algorithms. Additional exploration of distinct reputation systems and their practical application led to the study’s discovery of machine learning algorithms, i.e., the KNN, Support Vector Machines, and Random Forest algorithms. The study details the data cleansing process for the employed dataset in its process for optimizing the computational demands required to train machine learning algorithms, achieved through dataset partitioning. Such a process led to explaining and analysis of conducting experiments and comparing their results. Such experiments demonstrate the algorithm’s ability to detect Sybil nodes in OSNs (100% accuracy in SVM, 99.6% in Random Forest, and 97% in KNN algorithms); thus concluding by proposing future research opportunities.

Introduction: Vitamin B1 (thiamine) deficiency (TD) after metabolic and bariatric surgery (MBS) is often insidious and, if unrecognized, can lead to irreversible damage or death. As TD symptoms are vague and overlap with other disorders, we aim to identify predictors of recurrent TD and failure to collect B1 labs. Methods: We analyzed a large sample of data from patients with MBS (n=878) to identify potential predictors of TD risk. We modeled recurrent TD and failure to collect B1 labs using classical statistical and machine learning (ML) techniques. Results: We identified clusters of labs associated with increased risk of recurrent TD: micronutrient deficiencies, abnormal blood indices, malnutrition, and fluctuating electrolyte levels (aIRR range: 1.62-4.68). Additionally, demographic variables associated with lower socioeconomic status were predictive of recurrent TD. ML models predicting characteristics associated with failure to collect B1 labs achieved 75-81% accuracy, indicating that clinicians may fail to match symptoms with the underlying condition. Conclusion: Our analysis suggests that both clinical and social factors can increase the risk of life-threatening TD episodes in some MBS patients. Identifying these indicators can help with diagnosis and treatment.

The forest regeneration phase in Sweden commonly involves mechanical soil preparation followed by planting of Scots pine or Norway spruce seedlings. The prepared soil offers planting positions with different properties, including decreased damage by pine weevil (Hylobius abietis L.). Nitrogen fertilization can be applied at the time of planting to aid establishment of the seedlings. In this study we compared the effect of organic nitrogen-fertilization for different planting positions, and seedling sizes on early survival and growth of Scots pine and Norway spruce seedlings. The main planting positions were capped mound, hinge, and mineral soil. Seedlings planted close to organic material were categorized as “low quality positions”, since proximity to organic material increase pine weevil attraction. Higher mortality rates related to pine weevil damage were recorded for seedlings planted in the low-quality positions, regardless of seedling size or N-fertilization. Pine weevil attack rates increased with increasing seedling size. Growth was in general lowest in mineral soil positions. The effect of organic N-fertilization on growth was positive for spruce regardless of planting positions and seedling size, while depending on planting position and seedling size for pine, indicating that effects of organic N-fertilization depend on seedling species, -size and planting position.

Cancer is a disease caused by the abnormal growth of cells in different parts of body is one of the top causes of death globally. Microarray gene expression data plays a critical role in the identification and classification of cancer tissues. Due to recent advancements in Machine Learning (ML) techniques, researchers are analyzing gene expression data using a variety of such techniques to model the progression rate &amp; treatment of cancer patients with great effect. But high dimensionality alongside the presence of highly correlated columns in gene expression datasets leads to computational difficulties. This paper aims to propose the use of ML classification techniques- Linear Discriminant Analysis (LDA) &amp; Random Forest (RF) for classifying five types of cancer (breast cancer, kidney cancer, colon cancer, lung cancer and prostate cancer) based on high dimensional microarray gene expression data. Principal component analysis (PCA) was used for dimensionality reduction, and principal component scores of the raw data for classification. Six distinct categorization performance measures were used to evaluate these approaches; RF method provided us with higher accuracy than LDA method. The method and results of this article should be helpful to researchers who are dealing with many genes in microarray data.

To investigate the variation patterns of spectral characteristics of main branches in fruit trees due to uneven distribution of growth, pruning, and fruit quality, as well as to determine the optimal fruit vi-bration harvesting, a vibration differential model of tree branches was constructed using the concen-trated mass method. A three-dimensional model of the fruit tree was built using finite element software. The spectral characteristics of each branch were obtained using the linear sweep frequency method. The method of changing the mass of each branch was employed to simulate the variation in branch mass. The frequency of occurrence, mean acceleration, and coefficient of variation of the inherent frequencies were used as evaluation indicators. The results revealed that when the vibration frequency of the fruit tree was at the peak frequency in the tree trunk spectrum, the frequency of occurrence and mean accel-eration were optimal. And the coefficient of variation of branch acceleration was less than 1, indicating a moderate dispersion. Additionally, reducing the branch mass resulted in an increase in the peak fre-quency of the tree trunk spectrum. Therefore, the growth, pruning, and uneven distribution of the branches in fruit trees can alter the spectral characteristics. When harvesting fruit by vibration, selecting the peak frequency in the trunk vibration spectrum as the excitation force frequency yields the best fruit detachment.

This study provides a comprehensive assessment of Derawan Island's coral reefs over two decades (2003, 2011, and 2021) from a machine learning classification perspective. Employing non-parametric algorithms like Random Forest (RF), Support Vector Machine (SVM), and Classification and Regression Tree (CART), our analysis primarily focused on spatial and temporal changes in coral habitats. RF emerged as the most accurate method, demonstrating an accuracy of 71.43% with Landsat 9, 73.68% with Sentinel 2, and 78.28% with Multispectral Aerial Photos. We found that the accuracy of our classification results was significantly influenced by the geographic resolution, as well as the quality of field and satellite/aerial image data. Through spatial clustering, the coral habitats exhibited an Nearest Neighbor Index (NNI) value of 0.8727, indicating specific patterns of distribution. The analysis revealed a decrease in coral reef extent from 2003 to 2011, shrinking to 16 hectares with varying densities, followed by a slight area increase but with more heterogeneous densities between 2011 and 2021. This study not only highlights the dynamic nature of coral reef habitats over two decades but also underscores the critical role of machine learning in environmental monitoring and conservation efforts.

This study presents a research plan that utilizes data obtained from wearable devices to identify human activities and gain insights into human behavior. We developed a model capable of classifying activities similar to human behavior and evaluated the effectiveness and generalization capabilities of this model. The data underwent initial preprocessing, including standardization and normalization. Additionally, recognizing the inherent similarities between human activity behaviors, we introduced a multi-layer classifier model. The first layer is a random forest model based on stepwise regression, which may encounter reduced accuracy for similar activities. The second layer employs a Support Vector Machine (SVM) model based on Kernel Fisher Discriminant Analysis (KFDA). KFDA is used to reduce the dimensionality of data points with potential confusion, followed by SVM for classification. The model was experimentally evaluated and applied to four benchmark datasets: UCI DSA, UCI HAR, WISDM, and IM-WSHA. The experimental results demonstrate that our approach achieved recognition accuracies of 99.71%, 98.71%, 99.12%, and 97.6% on these datasets, indicating excellent recognition performance. Furthermore, to assess the model's generalization ability, we performed K-fold cross-validation on the random forest model and utilized ROC curves for the SVM classifier. The results indicate that our multi-layer classifier model exhibits robust generalization capabilities.

Geospatial knowledge in massive academic papers can provide knowledge services such as location-based research hotspot analysis, spatio-temporal data aggregation, research results recommendation, etc. However, geospatial knowledge often exists implicitly in literature resources in unstructured form, which is difficult to be directly accessed and mined and utilized for rapid production of massive thematic maps. In this paper, we take the geospatial knowledge of the area studied as an example and introduce its extraction method in detail. An integrated feature template matching and random forest classification algorithm is proposed for accurately identifying research areas from the abstract texts of academic papers and producing thematic maps. Firstly, the precise recognition of geographical names is achieved step by step based on BiLSTM-CRF algorithm and improved heuristic disambiguation method; then, the area studied is extracted by the designed integrated feature recognition template of area studied using random forest classification algorithm, and a fast thematic map is designed for the knowledge of area studied, topic and literature. The experimental results show that the area studied recognition accuracy can reach 97%, the F-value is 96%, and the recall rate reaches 96%, achieving high accuracy and high efficiency of area studied extraction in text. Based on the geospatial knowledge, the thematic map can achieve the effect of fast map formation and accurate expression.

Mechanical engineering plays an important role in the design and manufacture of medical devices, implants, prostheses, and other medical equipment, where the machining of bio-compatible materials have a special place. There are a lot of different conventional and non-conventional types of machining of biocompatible materials. One of the most frequently used methods is milling. The first part of this paper considers the optimization of machining parameters for minimizing surface roughness in milling biocompatible alloy Ti-6Al-4V. With four factors (cutting speed, feed rate, depth of cut, and cooling/lubricating method), each having three levels, the full factorial design implies 81 experiments have to be carried out. Using the Taguchi method, the experiment number was reduced from 81 to 27 runs through an orthogonal design. According to the analysis of variance (ANOVA), the most significant parameter for surface roughness is feed rate. In the second part possibilities of different ML techniques to create a predictive model for average surface roughness using previously created small dataset are explored. Several ML technologies and techniques that can deal successfully with small dataset were explored. The best results show commonly used machine learning algorithm Random Forest that is widely used in regression problems.

This research undertakes a quantitative exploration of the impact of procurement mo-dalities on economic efficiency and inclusive procurement within the Dominican Repub-lic's public sector. Applying data science methodologies and machine learning techniques, this study analyzes a rich dataset of public procurement contracts spanning from 2018 to 2023. A focus is placed on understanding how different procurement modalities affect contract amounts and processing times—two key indicators of economic efficiency. Machine learning models, including Random Forest and Gradient Boosting, are em-ployed to predict these essential metrics across different procurement modalities. The findings indicate a significant positive correlation between certain procurement modali-ties and economic efficiency. The machine learning models reveal that procurement strategies focused on local sourcing and transparent bidding processes tend to yield high-er economic returns and more equitable outcomes. This research underscores the poten-tial of quantitative analysis to inform procurement policies, fostering a more economi-cally efficient and inclusive procurement ecosystem in the Dominican Republic's public sector. The insights garnered from this study could have significant implications for the design of more inclusive procurement strategies, contributing to socio-economic growth and corruption reduction in the Dominican Republic.

Forest damage has become more frequent in Europe in the last decades, and remote sensing offers a powerful tool for monitoring them rapidly and cost-effectively. A novel approach was developed to utilize high-resolution ESA Sentinel-2 satellite imagery and Google Earth Engine cloud computing. The processing, analysing, and visualization of vegetation and water index (NDVI, NDVIch, Z NDVI, EVI, NDWI) maps and charts derived from satellite images took place in the cloud to ensure the detection of forest disturbances in the Hungarian study site for the period 2017-2020. The index maps were classified to reveal forest disturbances, and the cloud-based method successfully showed drought and frost damage in the oak-dominated Nagyerdő forest of Debrecen. Differences in the reaction to damage between tree species were visible in the index values, therefore a Random Forest Machine Learning classifier was applied to show the spatial distribution of dominant species. Accuracy assessment was accomplished with confusion matrices that compared classified index maps to field-surveyed data, demonstrating 71% Total Accuracy for forest damage and 76% for tree species. Based on the results of this study and the resilience of Google Earth Engine, the presented method has the potential to be extended to monitor the entire Hungary.

For tropical forests to survive anthropogenic climate change, the trees that comprise them will need to tolerate the newly emerging conditions through adaptation or acclimation, or they can avoid climate change through range shifts and “species migrations”. In this review, we show that the rapid pace and extreme severity of modern climate change makes it extremely unlikely that tropical tree species can adapt (with some possible exceptions). We also show that while many tropical tree species are shifting their distributions to higher, cooler elevations, the rate of these migrations are mostly insufficient to offset ongoing changes in temperatures, especially in lowland tropical rainforests where thermal gradients are shallow or nonexistent. We argue that the best hope for tropical tree species to avoid becoming “committed to extinction” is acclimation. While several new methods are being developed to test for acclimation, we unfortunately still do not know if tropical tree species can acclimate or what factors may prevent or facilitate acclimation. Until these questions are answered, our ability to predict the fate of tropical species and tropical forests – and the many services that they provide to humanity – remains critically impaired.

Aiming at the prediction of environmental vibrations induced by elevated high-speed railway, a machine-learning method is developed by combining random forest algorithm and Bayesian optimization, which using the dataset from on-site experiments . When it comes to achieving a rapid and effective prediction of environmental vibration, there is few research on com-parisons and verifications of different algorithms, and neither on parameter tuning and optimi-zation of machine learning algorithms. In this paper, a field experiment is firstly carried out to measure the ground vibrations caused by high-speed trains running on bridge, and then the en-vironmental vibration characteristics are analyzed in view of ground accelerations and weighted vibration levels. Subsequently, three machine-learning algorithms of linear regression, support vector machine and random forest are developed by using experimental database, and their prediction performance are discussed. Finally, two optimization models for the hyperparameter set of random forest algorithm are further compared. It turns out that the integrated random forest algorithm has higher accuracy in predicting environmental vibration than linear regression and support vector machine; the Bayesian optimization has excellent performance and high efficiency in achieving efficient and in-depth optimization of parameters, and can be combined with the RF machine learning algorithm to effectively predict the environmental vibrations induced by the high-speed railway.

Tree growth is under the combined influence of abiotic and biotic factors. Trees with different sizes may respond differently to these factors, implying that tree heterogeneity in tree size may also modulate the overall growth trend. To test this hypothesis, we focused on the radial growth trends of natural subalpine forests on the Tibetan Plateau. We first extended the iterative growth model (IGM) to the tree-ring scale (IGMR) to test the applicability of the generalized metabolic growth theory to tree growth. As predicted by the IGMR, the radial growth of trees at the aggregate scale is constrained by a unimodal pattern. Using the IGMR, we reconstructed the historical best growth trajectory (HBGT) of trees within the same community based on the tree with the largest radius and/or longest age in the community. From the average difference between the HBGT and the current radial growth rate of trees with different sizes, we constructed an indicator that can measure the overall variation in tree radial growth. Based on this indicator, we found a negative effect of tree size heterogeneity on the overall variability of tree growth across elevations. Further analysis also revealed that the radial growth rate of trees on the Tibetan Plateau has increased significantly compared to the past, where growing season average temperature and annual minimum temperature were negatively and positively correlated with tree growth below and above the treeline, respectively. Our study not only confirmed that the overall variability of tree growth depends on tree size heterogeneity but also proposed an indicator that reveals net changes in tree radial growth rate relative to the past. These theoretical advances facilitate the understanding of changes in the extent of subalpine forests.

In a limestone quarry mine, fragmentation is a crucial outcome of blasting operations. The optimization of blasting operations greatly benefits from the prediction of rock fragmentation. The main factors that affect fragmentation are rock mass characteristics, blast geometry, and explosive properties. This paper is a step towards the implementation of machine learning and deep learning algorithms for predicting the extent of fragmentation (in percentage) in opencast mining. Various parameters can affect fragmentation. But, in this paper initially, ten parameters (spacing, drill hole diameter, burden, average bench height, powder factor, number of holes, charge per delay, uniaxial compressive strength, specific drilling, and stemming) are collected to train the model. However, due to a weak correlation with rock fragmentation, drill diameter, Average bench height, compressive strength, stemming, and charge per delay are eliminated to reduce model complexity. A total of 219 data sets having five input features i.e., the number of holes, spacing, burden, specific drilling, and powder factor are used to develop the models. To predict rock fragmentation due to blasting in limestone quarry mines, both machine learning models (Random Forest Regression (Bagging), Support Vector Regression, and XG Boost Regression (Boosting)), as well as a deep learning model (Neural Network Regression), are applied to develop a model that can optimize the prediction of fragmentation. The Artificial neural network model optimization showed that the model with architecture 64-32-16-1 can perform well giving MSE (mean squared error) values of 41.32 and 28.59 on training and test data respectively. The R2 value for both training and test is 0.83. Random Forest regression is also performing well compared to SVR and XG boost with the MSE value 12.37 and 9.89 on training and testing data respectively. Here, the R2 value for both sets are 94%. Based on the permutation importance and Shapely plot values, the powder factor has the highest impact, and the burden has the lowest impact on fragmentation.

Windthrow (i.e., trees broken and uprooted by wind) is a major natural disturbance in the Amazon. Images from medium-resolution optical satellites (mostly Landsat) combined with extensive field data have allowed researchers to assess patterns of tree mortality and monitor forest recovery over decades of subsequent succession in different regions. Although satellites with high spatial-resolution have become available for the Amazon in the last decade, they have not yet been employed for the mapping and quantification of windthrow tree-mortality. Here, we address how increasing the spatial resolution of satellites affects plot-to-landscape estimates of windthrow tree-mortality. We combined forest inventory data with Landsat 8 (30 m pixel), Sentinel 2 (10 m), and WorldView 2 (2 m) imagery over an old-growth forest in the Central Amazon that was disturbed by a single windthrow event in November/2015. Remote sensing estimates of tree mortality were produced with Spectral Mixture Analysis and analyzed together with forest inventory data using Generalized Linear Models. Windthrow tree-mortality measured in 3 transects (30 subplots) crossing the entire disturbance gradient was 26.9 ± 11.1% (mean ± 95% CI). Based on this ground truth, the three satellites produced reliable and statistically similar estimates (from 26.5% to 30.3% windthrow tree-mortality, p&lt;0.001). The mean-associated uncertainties decreased systematically with increasing spatial resolution (i.e., from Landsat 8 to Sentinel 2 and WorldView 2). However, the overall quality of fit of models showed the opposite pattern, which may reflect the influence of crown damage not accounted for in our field study, and fast-growing regeneration of leaf area. Among the satellites studied, Landsat 8 most accurately captured field observations of variations in tree mortality across the disturbance gradient (i.e., lower under- and/or overestimation from undisturbed to extremely damaged forest). Although satellites with high spatial-resolution can refine estimates of windthrow severity by allowing the quantification of individual tree damage and mortality, our results validate the reliability of Landsat imagery for assessing patterns of windthrow tree-mortality in dense and heterogeneous tropical forests. Although high-resolution imagery may improve estimates of tree damage and mortality, these should be validated using field data at compatible scales.

Gardens play a key role in the definition of the cultural landscape since they reflect the culture, identity and history of a people. They also contribute to the ecological balance of the city. Despite gardens have an historic and social value, they are not protected as much as the rest of the existing heritage, like architecture and archaeological sites. While methods of built-heritage mapping and monitoring are increasing and constantly improving to reduce built-heritage loss and the severe impact of natural disasters, the documentation and survey techniques for gardens are often antiquated, inventories are typically made by non-updated/updatable reports, and rarely they are on digital format and in 3D. This paper presents the preliminary results of a study on latest technology for gardens laser scanning. We compared static Terrestrial Laser Scanning and Mobile Laser Scanning point clouds, to evaluate their quality for documentation and the estimation of the tree attributes. The evaluation is based on visual observation and graphic comparison of the two point clouds acquired in different instances. Both methods produced useful outcomes for the research scope within their limitations. Terrestrial Laser Scanning is still the method that offers more accurate point clouds with a higher point density and less noise level. However, the more recent Mobile Laser Scanning is able to survey in less time, significantly reducing the costs for site activities, data post-production and registration. Both methods have their own restrictions that are amplified by site features, mainly the lack of plans for the geometric alignment of scans and for the Simultaneous Location and Mapping (SLAM) process. We also offer the results of a comparison of the functional range of the two machines, as well as for a comparison of their terrain information extraction capabilities.

The current study aims to apply and compare the performance of six machine learning algorithms, including three basic classifiers: random forest (RF), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGB), as well as their hybrid classifiers, using the logistic regression (LR) method (RF+LR, GBDT+LR, and XGB+LR), in order to map the landslide susceptibility of Zhangjiajie City, Hunan Province, China. First, a landslide inventory map was created with 206 historical landslide points and 412 non-landslide points, which was randomly divided into two datasets for model training (80%) and model testing (20%). Second, 15 landslide conditioning factors (i.e., altitude, slope, aspect, plane curvature, profile curvature, relief, roughness, rainfall, topographic wetness index (TWI), normalized difference vegetative index (NDVI), distance to roads, distance to rivers, land use/land cover (LULC), soil texture, and lithology) were initially selected to establish a landslide factor database. Thereafter, the multicollinearity test and information gain ratio (IGR) technique were applied to rank the importance of the factors. Subsequently, we used a series of metrics (e.g., accuracy, precision, recall, f-measure, area under the ROC (receiver operating characteristic) curve (AUC), kappa index, mean absolute error (MAE), and root mean square error (RMSE)) to evaluate the accuracy and performance of the six models. Based on the AUC values derived from the models, the GBDT+LR model with the highest AUC value (0.8168) was identified as the most efficient model for mapping landslide susceptibility, followed by the XGB+LR, XGB, RF+LR, GBDT, and RF models, which achieved AUC values of 0.8124, 0.8118, 0.8060, 0.7927, and 0.7883, respectively. The results from this study suggest that the stacking ensemble machine learning method is promising for use in landslide susceptibility mapping in the Zhangjiajie area and is capable of targeting the areas prone to landslides.

We assessed true bugs among above-groundand and soil inhabitants of four different biocoenosis of pine forests representing the intra-zone areas of flora and fauna of the Kazakh upland. 15 species of three families were listed for the Central Kazakhstan region for the first time. Species Eremocoris podagricus was not registered for the territory of Kazakhstan earlier. Fifteen species (50 %) were true above-ground inhabitants or ones living on grasses, whereas other species use soil surface and ground litter as temporary habitats. Drymus brunneus and Eremocoris fenestratus play a key role in the structure and function of the true bugs assemblages in the in biotopes studied. The ecological success of typically forest and boreal inhabitants in biotopes of arid region is explained by relict nature of forests and ancient connections with the taiga zone of Western Siberia and mountain forests of the Urals and Altai. The results of a comparative analysis of the similarities and differences of the dominant and ecological structures of Heteroptera complexes of aboriginal studied forests and derived type of forest are presented.

Forests play an important role in reducing greenhouse gases in the atmosphere, thereby mitigating the impact of climate change. Estimating the accumulated biomass in a forest ecosystem is important for assessing the productivity and sustainability of the forest. Allometric models for above ground biomass (AGB) are linear regression equations based on the relationships between biomass and diameter at breast height (DBH), tree height (H), and/or wood density. This study estimates AGB in the Colo-I-Suva Forest Park by applying the allometry of Chave et al (2005) and the diameter: height ratio derived from Payton & Weaver (2011) for height estimation in a plot of 20

Recent expansion in data sharing has created unprecedented opportunities to explore structure-function linkages in ecosystems across spatial and temporal scales. However, characteristics of the same data product, such as resolution, can change over time or spatial locations, as protocols are adapted to new technology or conditions, which may impact the data’s potential utility and accuracy for addressing end user scientific questions. The National Ecological Observatory Network (NEON) provides data products for users from 81 sites and over a planned 30-year time frame, including discrete return Light Detection and Range (LiDAR) from an airborne observatory platform. LiDAR is a well-established and increasingly available remote sensing technology for measuring three-dimensional (3D) characteristics of ecosystem and landscape structure, including forest structural diversity. The LiDAR product that NEON provides can vary in point density from 2 – 25+ points/m² depending on instrument and acquisition date. We used NEON LiDAR from five forested sites to (1) identify the minimum point density at which structural diversity metrics can be robustly estimated across forested sites from different ecoclimatic zones in the USA and (2) to test the effects of variable point density on the estimation of a suite of structural diversity metrics and multivariate structural complexity types within and across forested sites. Twelve out of sixteen structural diversity metrics were sensitive to LiDAR point density in at least one of the five NEON forested sites. The minimum point density to reliably estimate the metrics ranged from 2.0 to 7.5 pt/m^2, but our results indicate that point densities above 7-8 pt/m² should provide robust measurements of structural diversity in forests for temporal or spatial comparisons. The delineation of multivariate structural complexity types from a suite of 16 structural diversity metrics was robust within sites and across forest types for a LiDAR point density of 4 pt/m² and above. This study shows that different metrics of structural diversity can vary in their sensitivity to the resolution of LiDAR data and users of these open-source data products should consider the point density of their data and use caution in metric selection when making spatial or temporal comparisons from these datasets.

The most common index for representing structural condition of the pavement is the structural number. The current procedure for determining structural numbers involves utilizing falling weight deflectometer and ground-penetrating radar tests, recording pavement surface deflections, and analyzing recorded deflections by back-calculation manners. This procedure has two drawbacks: 1. falling weight deflectometer and ground-penetrating radar are expensive tests, 2. back-calculation ways has some inherent shortcomings compared to exact methods as they adopt a trial and error approach. In this study, three machine learning methods entitled Gaussian process regression, m5p model tree, and random forest used for the prediction of structural numbers in flexible pavements. Dataset of this paper is related to 759 flexible pavement sections at Semnan and Khuzestan provinces in Iran and includes “structural number” as output and “surface deflections and surface temperature” as inputs. The accuracy of results was examined based on three criteria of R, MAE, and RMSE. Among the methods employed in this paper, random forest is the most accurate as it yields the best values for above criteria (R=0.841, MAE=0.592, and RMSE=0.760). The proposed method does not require to use ground penetrating radar test, which in turn reduce costs and work difficulty. Using machine learning methods instead of back-calculation improves the calculation process quality and accuracy.

Alpine forest gaps can distribute snowfall, solar radiation and rainfall, thus inducing a heterogeneous hydrothermal microenvironment between the inside and outside areas of forest gaps. Additionally, the characteristics of the heterogeneous microenvironment could vary greatly across the gap location properties during winter and the growing season. To determine the response of total phenol loss (TPL) from the litter to alpine forest gap disturbance during decomposition, we conducted a field litterbag experiment within a representative fir (Abies faxoniana Rehd.) forest based on the gap location properties. The TPL and abundances of fungi and bacteria from two typical shrub species (willow, Salix paraplesia Schneid., and bamboo, Fargesia nitida (Mitford) Keng f.) were measured during the following periods over two years: snow formation (SF), snow cover (SC) snow melting (ST), the early growing season (EG) and the later growing season (LG). At the end of the study, we found that the snow cover depth, frequencies of the freeze-thaw cycle and the fungal copy g-1 to bacterial copy g-1 ratio had significant effects on the litter TPL. The abundances of fungi and bacteria decreased from the gap center to the closed canopy during the two SF, SC, ST and LG periods and reversed during the two EG periods. The TPL closely followed the same trend as the microbial abundance during the first year of incubation. In addition, both species had larger TPLs in the gap center during the first winter, first year and entire two years. These findings suggest that alpine forest gap formation accelerates litter TPL and plays a dual role during specific critical periods by distributing abiotic and biotic factors directly and indirectly. In conclusion, reduced snow cover depth and duration during winter warming under current climate change scenarios or as gaps vanish may slow litter TPL in alpine biomes.

During recent years UAVs have been increasingly used in agriculture and forestry research and application. Nevertheless, most of this work has been devoted to improving accuracy and explanatory power, often at the cost of usability and affordability. We tested a low-cost UAV and a simple workflow to apply four different greenness indices to the monitoring of pine (Pinus sylvestris and P. nigra) after-fire regeneration in a Mediterranean forest. We selected two sites and masured all pines within a pre-selected plot. Winter flights were carried out at each of the sites, at two flight altitudes (50 and 100 m). Automatically normalizing images entered an SfM based photogrammetric software for restitution and the obtained point cloud and orthomosaic processed to get a canopy height model and four different greenness indices. Sum of pine DBH was regressed on summary statistics of greenness indices and canopy height model. ExGI and GCC indices outperformed VARI and GRVI in estimating pine DBH, while canopy height model slightly improved the models. Flight altitude did not severely affect model performance. Our results show that low cost UAVs may improve forest monitoring after disturbance, even in those habitats and situations were resource limitation is an issue.

Hyperspectral images are an important tool to assess ecosystem biodiversity both on terrestrial and benthic habitats. To obtain more precise analysis of biodiversity indicators that agree with indicators obtained using field data, analysis of spectral diversity calculated from images have to be validated with field based diversity estimates. The plant species richness is one of the most important indicators of biodiversity. This indicator can be measured in hyperspectral images considering the Spectral Variation Hypothesis (SVH) which states that the spectral heterogeneity is related to spatial heterogeneity and thus to species richness. The goal of this research is to capture spectral heterogeneity from hyperspectral images for a terrestrial neo tropical forest site using Vector Quantization (VQ) method and then use the result for prediction of plant species richness. The results are compared with that of Hierarchical Agglomerative Clustering (HAC). The validation of the process index is done calculating the Pearson correlation coefficient between the Shannon entropy from actual field data and the Shannon entropy computed in the images. Terrestrial dry forest and marine coastal hyperspectral images with different resolutions have been used for spectral diversity feature validation.

In this paper, we propose an adaptive duty-cycled hybrid X-MAC (ADX-MAC) protocol for energy-efficient forest fire prediction. The X-MAC protocol acquires the additional environmental status collected by each forest fire monitoring sensor for a certain period. And, based on these values, the length of sleep interval of duty-cycle is changed to efficiently calculate the risk of occurrence of forest fire according to the mountain environment. The performance of the proposed ADX-MAC protocol was verified through experiments the proposed ADX-MAC protocol improves throughput by 19% and was more energy-efficient by 24% compared to X-MAC protocol. As the probability of forest fires increases, the length of the duty cycle is shortened, confirming that the forest fires are detected at a faster cycle.

Abstract: Changes in forest landscapes have been connected with human activity for centuries, which can be considered as one of the main driving forces of change in the global perspective. The spatial distribution of forests changes along with the geopolitical situation, demographic changes, intensification of agriculture, urbanization or changes in the land use policy. However, due to the limited availability of historical data, the driving forces of changes in forest landscapes are most often considered in relation to recent decades, without taking into account long-term analyzes. The aim of this paper is to determine the impact of natural and socio-economic factors on changes in forest landscapes within the protected area – Ślęża Landscape Park and its buffer zone in the aspect of long-term analyzes covering the period of 140 years (1883-2013). The comparison of historical and current maps, demographic data on 4 different periods as well as natural and location factors by using the ArcGIS software allowed analyzing selected driving forces of forest landscape transformations. We took into account natural factors like altitude, slope, exposure of the hillside and socio-economic drivers like population changes, distances to centers of municipalities, main roads and built-up areas.

In recent years, remote sensing technology has been widely used to predict forest stand parameters. In order to compare the effects of different features of remote sensing images and topographic information on the prediction of forest stand parameters, multivariate stepwise regression analysis method was used to build estimation models for important forest stand parameters by using textural and spectral features as well as topographic information of SPOT-5 satellite images in northeastern Heilongjiang Province in China as independent variables. The study results show that the optimal window to predict forest stand parameters using textural features of SPOT-5 satellite image is 9×9; the ability of textural features was better than that of spectral features in terms of predicting forest stand parameters; with the inclusion of topographic information, the accuracy of prediction of all models was improved, of which elevation has the most significant effect. The highest accuracy was achieved when predicting the stand volume (SV) (R²_adj=0.820), followed by basal area (BA) (R²_adj =0.778), accuracy of both above models exceeded 75%. The results show that models combined use of textural, spectral features and topographic information of SPOT-5 images have a good application prospect in predicting forest stand parameters.

This study is aimed at demonstrating application of vegetation spectral techniques for detection and monitoring of impact of oil spills on vegetation. Vegetation spectral reflectance from Landsat 8 data were used in the calculation of five vegetation indices (normalized difference vegetation index (NDVI), soil adjusted vegetation index (SAVI), adjusted resistant vegetation index 2 (ARVI2), green-infrared index (G/NIR) and green-shortwave infrared (G/SWIR) from the spill sites (SS) and non-spill (NSS) sites in 2013 (pre-oil spill), 2014 (oil spill date) and 2015 (post-oil spill) for statistical comparison. The result shows that NDVI, SAVI, ARVI2, G/NIR and G/SWIR indicated certain level difference between vegetation condition at the SS and the NSS were significant with p-value <0.5 in December 2013. In December 2014 vegetation conditions indicated higher level of significant difference between the vegetation at the SS and NSS as follows where NDVI, SAVI and ARVI2 with p-value 0.005, G/NIR - p-value 0.01 and GSWIR p-value 0.05. Similarly, in January 2015 a very significant difference with p-value <0.005. Three indices NDVI, ARVI2 and G/NIR indicated highly significant difference in vegetation conditions with p-value <0.005 between December 2013 and December 2014 at the same sites. Post—spill analysis show that NDVI and ARVI2 indicated low level of significance difference p-value <0.05 suggesting subtle change in vegetation conditions between December 2014 and January 2015. This technique is essential for real time detection, response and monitoring of oil spills from pipelines for mitigation of pollution at the affected sites in the mangrove forest.

Lomami National Park, located in the Democratic Republic of Congo, is renowned for the integrity of its forest ecosystems, safeguarded by the absence of agricultural activities and limited road access. However, these ecosystems remain under-researched, particularly regarding forest cover dynamics. This research gap poses a significant challenge to establishing rigorous monitoring, which is essential for ensuring the long-term preservation of these valuable ecosystems. This study evaluates land cover dynamics within Lomami National Park through supervised classification of Landsat images from 2008, 2016, and 2024. Seven spatial structure indices were calculated to highlight the dynamics of landscape composition and configuration, distinguishing between the park's core and buffer zones. The underlying spatial transformation processes were identified using a decision tree approach. The results highlight a striking contrast in forest cover stability between Lomami National Park and its surrounding periphery. Within the park, forest cover has not only been preserved but has also shown a modest increase, rising from 92.60% to 92.75%. In contrast, the peripheral zone experienced a significant decline in forest cover, decreasing from 79.32% to 70.48% over the same period. This stability within the park goes beyond simply maintaining forested areas; it includes the preservation and strengthening of the spatial structure of forest ecosystems. For instance, edge density, which is a key indicator of forest edge compactness, remained stable in the park, fluctuating between 8 m/ha and 9 m/ha. However, in the peripheral zone, edge density exceeded 35 m/ha, suggesting that forest edges in the park are considerably more cohesive and intact than in the surrounding areas. Furthermore, the spatial transformation processes within each area underscore these contrasting dynamics. In the park, the main process was aggregation of primary forest patches, which reflects a trend toward continuous and connected forest landscapes. By contrast, the peripheral zone experienced dissection, indicating fragmentation and breakdown of forest patches. Together, these findings underscore the park's role in maintaining both the extent and the structural integrity of forest ecosystems, setting it apart from the more degraded periphery. These findings highlight not only the resilience of forest ecosystems in the face of limited anthropogenic pressures but also the crucial importance of effective land management and rigorous conservation strategies in addressing the challenges of urbanization and surrounding rural expansion. They further emphasize that well-adapted conservation measures, combined with specific demographic and socio-economic conditions, can play a pivotal role in long-term forest preservation and ecological stability.

Balancing the levels of ammonia, in poultry facilities plays a role in enhancing the welfare and productivity of animals. Various environmental factors such as temperature, humidity, ammonia concentrations, carbon dioxide concentration and exposure to light have an impact on the wellbeing and efficiency of poultry. It is essential to monitor these elements to achieve outcomes necessitating practical, reliable and cost-effective solutions. Leveraging the Internet of Things (IoT) in conjunction with LoRa-WAN (Long Range Wide Area Network) technology presents an opportunity in the industry by enhancing connectivity among devices and facilitating real time data transmission. This research introduces a hardware and software setup for monitoring conditions at poultry farms through LoRa WAN and IoT technologies. More-over researchers employed Random Forest Machine Learning prediction models to tune the concentration of Ammonia (NH3) in poultry facilities across a wide area or multiple Poultry houses. Comparative assessments were conducted to evaluate the proposed system against devices revealing a correlation exceeding 0.9830 between the sensors and standard equipment while costing 28% less than traditional tools. The prototype displayed outcomes underscoring its potential, for managing poultry facilities across multiple farms with centralized oversight.

Biodiversity monitoring is a seminal key for understanding the delivery of ecosystem functions and services. Mediterranean forests and woodlands harbor the vast majority of regional vascular plant diversity, and hence, they are a good surrogate for detecting changes in biodiversity linked to global change. In this work, we present a database resulting from the study of vascular plant diversity in multi-scale plots of 0.1 ha, measured around the first decade of this century and located in Mediterranean forest environments. Diversity profiles are calculated from Hill numbers (0, 1 and 2) for local (α) and regional (ϒ) diversity, as well as multiplicative calculation of differential diversity (β). Main Mediterranean forests sampled had a medium coverage of 51 % and stand dominant height of 10.6 m being monospecific in two thirds of the cases. Local diversity reaches its highest values ​​(around 78 species per 0.1 ha) in Holm oak dehesas, with values ​​below 50 species for the set of the most productive forest stands dominated by species of the genus Pinus. As regards the contribution to regional diversity, it is the broadleaf formations that contribute the most, with stone pine forests and dehesas in an intermediate position, and pine forests contributing the lowest in species richness terms.

Machine-learning methods used for classification are challenged by the class imbalance problem, where a certain class is underrepresented. Over/under-sampling the majority/minority class observations or model selection for ensemble methods alone might not be effective if the class imbalance ratio is very high. To address this concern, a novel method is presented for generating synthetic data for minority class observations in conjunction with an optimal tree ensemble classifier (OTEC). This novel synthetic method first generates minority class instances to balance the training data. Then it applies OTEC, where models are selected based on their performance using out-of-bag observations and training subsamples. A total of 20 benchmark problems on binary classification with moderate to extreme class imbalance are used to assess the efficacy of the proposed method against other well-known methods, including optimal tree ensemble, smote random forest, under-sampling random forest, oversampling random forest, k-nearest-neighbor, support vector machine, tree, and artificial neural network, using classification error rate, sensitivity, specificity, precision, recall and F1-score as performance metrics to assess the efficacy of the proposed method. The analyses presented in this study revealed that the proposed method based on data balancing and model selection yielded better results than other methods.

: This study aimed to validate the effects of self-compassion-based forest healing therapy as an intervention for reducing stress and promoting coping resources. Therefore, we investigated the effects of self-compassion-based forest healing therapy on stress, self-compassion, and connectedness to nature. A total of 30 participants were divided into an experimental and control group. Two sessions of forest healing meditation activities were held at the Forest of Healing every two weeks between October and November 2023. As a repeated measures experiment, data were collected before the intervention, after the first intervention, and after the second intervention. Perceived stress, positive and negative affect, and salivary cortisol were measured to assess physiological and psychological stress in participants. Connectedness to nature and self-compassion were measured to evaluate stress-coping resources. The results showed that the forest healing program had significant positive effects on stress and emotional states in the experimental group after the intervention. Further, there was a significant incremental effect on connectedness to nature and self-compassion in the experimental group. Therefore, self-compassion meditation-based forest healing therapy is a feasible and efficient approach for facilitating stress management and coping resource building.

Dengue, a perilous fever-type illness transmitted by mosquitoes, remains a significant global health concern. The incidence of dengue outbreaks is primarily influenced by climate factors, contributing to fluctuating dengue cases. Therefore, this study aims to develop predictive models for dengue outbreaks employing Machine Learning (ML) techniques. Four distinct ML models, namely K-Nearest Neighbor (KNN), Random Forest (RF), Gradient Boosting Regressor (GBR), and Support Vector Regressor (SVR), were employed in this study. Prior to model analysis, thorough data preprocessing was conducted. In some instances, certain attributes exhibiting sparse correlations with dengue occurrence were eliminated based on correlation analysis. Upon finalizing the models, performance evaluation was executed through comparison based on Mean Absolute Error (MAE) metrics. The findings of this comparative analysis revealed that the KNN model exhibited significantly superior performance compared to the other three models. This outcome underscores the potential of KNN in enhancing the accuracy of dengue outbreak predictions.

Soil secondary salinization has seriously affected the greenhouse vegetable production in China. To improve the secondary salinization greenhouse soil and enhance the soil physical-chemical properties in an eco-environmental way, different organic amendments (straw, straw biochar, Trichoderma bio-organic manure, commercial organic manure) were applied using pot experiment for 60 day to comprehensively screen the optimal remediation method. In this study, soil nutrient condition and salt movement were assessed, and predictive models using multi-linear regression (MLR) and Random Forest (RF) were developed to estimate soil salinization parameters. The findings indicated a significant decrease in soil salt content following the application of organic materials compared to the control treatment. Specifically, the addition of straw at a rate of 250g/kg resulted in a 59.38% reduction in soil salt levels after 60 days (P < 0.05). Furthermore, the main salt ions exhibited dynamic changes over the course of the experiment, with reductions observed in Na+, Ca2+, Cl-, and NO3- content under the 250 g/kg straw treatment by 83.63%, 74.67%, 63.26%, 59.07%, respectively, compared to CK (P < 0.05). In addition, the SO42- content under 125 g/kg commercial organic manure amendment reduced by 48.94%, respectively (P < 0.05). Contrary to expectations, the addition of organic materials significantly increased the levels of total nutrients (N/P/K) and available potassium (AK) and phosphorus (AP) in the soil. Specifically, the addition of straw at a rate of 250 g/kg resulted in increases of 18.02%-87.38% in total potassium (TK), 66.67%-200% in total phosphorus (TP), and 142.87%-367.8% in AK after a period of sixty days (P < 0.05). Ultimately, the treatment involving the addition of 250 g/kg of straw demonstrated the most pronounced impact on the physical and chemical properties of the soil. The random forest method shows promise in accurately predicting soil salt and soil sodium adsorption ratio (SAR)indicators, thereby providing a valuable tool for estimating soil properties.

The development of reliable predictive algorithm for house price as the housing market is a stand-out among the most involved regarding valuing the price and continues to fluctuate, is constantly a need for socio-economic advancement and welfare of citizen. In this paper, we develop machine learning algorithms for forecasting UK housing Price, and find an optimal algorithm that forecasts housing price accurately on the premises of the presence of many features or covariates. After applying correlation analysis to remove correlated variables in order to avoid multicollinearity, thereby increasing the statistical power, a novel method of using regression analysis to first of all understand and select statistically significant features for the various regions in England based on North South divide is adopted. These features are then used in the machine learning algorithm to further increase the statistical power of the algorithm, increase the level of accuracy for each of them and ultimately increase the predictive values for the algorithms. The model construction involves 3 stages: 1- correlation analysis to identify and remove correlated variables thereby avoiding multicollinearity and increasing the statistical power of the linear regression, 2 - using linear regression to determine variables that are statistically significant and 3 - building the machine learning algorithms based on the variables that are statistically significant from the linear regression. A comprehensive dataset of UK Paid housing Price from 2010 to 2019 was linked to a number of other datasets to generate a total 21 variables or features used for the models. Catboost, Gradient Boosting, Bagging, Random Forest, Extra Tree all achieved the excellent models performance result in all the regions considered. The comparison of the seven models showed that Extra Tree algorithm consistently achieved the best performance in term of level of accuracy in all the regions. K-Nearest Neighbours (KNN) is the only algorithm with less than 50% level of accuracy. Noticeably, the regions considered had varying or differing insignificant variables, implying that although many variables are common (statistically significant) to all the regions, there are regional differences and impact when modelling or predicting housing prices. This study validates the practicability of developing a machine learning methodology for the prediction of housing price. This research offers a reference for future house price prediction based on machine learning.

In-depth insights into the profound impacts of climate change and human activities on water resources are garnered through the dynamic changes in surface water, a crucial aspect of effective water resource management and the preservation of aquatic ecosystems. This paper introduces an innovative approach employing the random forest algorithm for the systematic extraction and monitoring of surface water at large regional or national scales. This method integrates spectral bands, spectral indices, and digital elevation model data, offering a novel perspective on this critical task. A data filling model is proposed to mitigate the impact of missing data due to cloud cover. Leveraging the capabilities of the Google Earth Engine (GEE), detailed information on surface water dynamics during the rainy and dry seasons in the Yangtze River Basin (YRB) from 1991 to 2021 is extracted using Landsat time series imagery. The analysis encompasses spatial-temporal variation characteristics and trends, with a specific focus on the intricate interplay between the areal extent of surface water and hydro-meteorological factors in each sub-basin of the YRB. Importantly, this includes considerations of potential groundwater contributions to surface water. Key findings from our research include: (1) Achieving a remarkable overall accuracy of 0.96 ± 0.03 in obtaining reliable surface water datasets with the support of GEE. (2) Identifying significant trends, such as a noteworthy increase in rainy season surface water bodies (+248.0 km2·yr−1) and a concerning decrease in surface ice/snow cover during both rainy and dry seasons, with change rates of −39.7 km2·yr−1 and −651.3 km2·yr−1, respectively. (3) Uncovering the driving mechanisms behind these changes, revealing positive correlations between the areal extent of rainy season surface water bodies and precipitation, as well as negative correlations between surface ice/snow cover area and average surface skin temperature. It is crucial to note that these driving factors exhibit variation among secondary river systems, underscoring the complexity of surface water dynamics. Furthermore, comparative analyses with existing surface water products are conducted, contributing to a deeper understanding of the advantages and uncertainties inherent in our proposed extraction method. The proposed method for large-scale surface water extraction not only enhances the monitoring of spatio-temporal surface water dynamics in the YRB but also provides valuable insights for the sustainable utilization and protection of water resources, considering the potential role of groundwater in supplementing surface water.

Wildfires are important natural drivers of forest stands dynamics, strongly influencing on their natural regeneration and ecosystem services. This paper presents a comprehensive analysis of spatiotemporal burnt area (BA) patterns over the period 2000–2022 in the Middle Volga region of the Russian Federation on the base of remote sensing time series, considering the impact of cli-matic factors on forest fires. The temporal trends were assessed with the Mann-Kendall nonpara-metric statistical test and Theil-Sen’s slope estimator using the LandTrendr algorithm on the Google Earth Platform (GEE). The accuracy assessment indicated a high overall accuracy (> 84%) and F-score value (> 82%) for forest burnt area detection as evaluated against 581 test sites of ref-erence data. The results revealed that the fire occurrences in the region were mainly irregular with the highest frequency of 7.3 over a 22-year period. The total forest BA was about 280 thousand ha, which equals to 1.7% of the land surface area or 4.0% of the total forested area under study in the Middle Volga region. The coniferous forest stands are the most fire-prone ecosystems accounting for 59.0 % of the total BA; deciduous stands accounts for 25.1%; and insignificant fire occurrences were registered in young forests and shrub lands. On a seasonal scale, temperature generally has a greater impact on the BA than precipitation and wind speed.

In the present work we have studied the short-term edaphic recovery of a burned pine forest of Pinus pinaster Aiton in Cerro del Castillo (Sierra de Guadarrama, Central Spain) in which a hydrological-forestry restoration was being carried out after the fire occurred in July 2022. For this purpose, variations in physicochemical properties (conductivity, relative humidity, organic matter and pH) and collembolan communities were analyzed by comparing the soils of a natural pine forest and a burned pine forest during the 24 weeks after the fire in 3 different samplings. With the data obtained, a series of diversity indices were calculated and statistical analyses such as ANOVA, hierarchical classification and principal component analysis were developed, with the aim of understanding in greater depth the recovery of collembolan populations, the changes in these physicochemical variables and the relationships between both parameters. This study shows how the fire reduced the amount of organic matter and water, and increased soil pH and conductivity, affecting edaphic fauna populations. In the short term, there was an increase in relative humidity and a decrease in electrical conductivity. The edaphic fauna communities of the burned pine forest increased their diversity and equity without reaching the values of the natural pine forest. At the specific level, however, the springtail populations matched the richness and even reached a very slightly higher Shanon-Wiener diversity index than that the natural pine forest populations in just 24 weeks. Finally, it was observed that the samples from the burned pine forest with recovered collembolan populations had the closest physicochemical parameters to those of the natural pine forest.

Gesture recognition technology has been quickly developed in the field of human-computer interaction. The multiple-input multiple-output (MIMO) radar has been widely adopted in gesture recognition because of its notable spatial resolution. In this work, a highly accurate MIMO radar-based hand gesture recognition algorithm with very low complexity is proposed. To make the proposed system applicable in the industry and work well even when the size of the training data is limited, we applied several low-complexity adaptive signal processing methods to extract features and reduce the noise effect. First, spectrum analysis is applied to range-Doppler maps (RDMs) and a cell-averaging constant false alarm rate (CA-CFAR) with mirror filters is applied to improve the robustness to noise. Afterward, the features related to the distance, speed, direction, and the elevation angle of the moving object are determined by the proposed adaptive signal analysis techniques and the random forest is applied for classification. The proposed system has the capability to precisely distinguish and identify eight motions, including waving, moving to the left or right, patting, pushing, pulling, and rotating clockwise or anticlockwise, with an accuracy of 95%. Experiments demonstrate the capability of the proposed hand gesture recognition system to classify different movements precisely.

When the formation flying consists of satellites that are independent from each other and forms a virtual synthetic aperture telescope that replaces a single large telescope with all of its optical elements it is required mutual coordination between the satellites with high accuracy in order to achieve good optical observations. In this paper we consider the problem of developing a control system for a tetrahedral satellite formation flying representing a synthetic aperture telescope in the geostationary orbit for high resolution monitoring of fire in the forest area in infrared spectrum. Synthesizing the image with good quality requires to keep the configuration of formation with µm-class accuracy. Study of dynamics of passive uncontrolled motion of formation flying showed a significant deviation of the configuration from the required shape with a high frequency due to the action of gravitational forces. To keep the configuration in the required form with µm accuracy the analysis of efficiency of various controllers was carried out in the process of numerical simulation. The simulation results made it possible to highlight the features of using various approaches to the development of control system for the satellite formation flying representing a synthetic aperture telescope.

The emergence of new plant diseases is an increasingly important concern. Climate change is likely to be among the factors causing most of the emerging diseases endangering forest and tree heritage around the world. Such diseases may be caused by latent pathogens or microorganisms cryptically associated with plants. The shift from a non-pathogenic to a pathogenic stage may depend on physiological alterations of the host, environmental changes and/or stress factors. In some woods of the Salento peninsula (Apulia region, Italy) sudden declines of holm oak plants (Quercus ilex L.) have been observed since 2016. The morphological and molecular characterization of representative fungal isolates associated with canker and necrosis in declining plants indicated that these isolates belong to the Botryosphaeriaceae family, and the most frequent species were Diplodia corticola and Diplodia quercivora, followed by Neofusicoccum vitifusiforme. In artificially inoculated young holm oak plants, both D. corticola and D. quercivora species produced intense and severe subcortical and leaf margin necrosis. N. vitifusiforme, although less aggressive, induced the same symptoms. Our research in addition to confirm the involvement of D. corticola in olm oak decline, represents the first report of D. quercivora as a new pathogen on Quercus ilex in Italy. Furthermore, to the best of our knowledge, we also found N. vitifusiforme as a new pathogen on Q. ilex for the first time.

As a consequence of applying advanced maintenance practices, the theoretical probability of failures is relatively low. However, observation of low market intelligence and maintenance management has been reported. The experimental investigation is supported by findings from a survey targeting asset-intensive companies applying hydraulic power systems. Next, the study incorporates multidimensional data analysis using CA-AHC (Correspondance Analysis with Agglomerative Hierarchical Clustering) approach. The non-parametric machine learning models are used from generated feature subspace to extract features affecting maintenance performance indicators. The results support empirical evidence that equipment age increases the probability of failures. However, the novel findings show that number of maintenance personnel, equipment size measured by nominal working energy consumption, and activities dedicated to maintaining fluid cleanliness impact regression results of companies utilising hydraulic applications.

Knowledge on the health of banana trees is critical for farmers to profit from banana cultivation. Fusarium wilt and banana blood diseases (BDB), two significant diseases infecting banana trees, are caused by Fusarium oxysporum and Ralstonia syzygii, respectively. They have successfully caused a decline in crop yield as they destroy the trees, starting sequentially from the pseudostem to the fruits. The entire distribution of BDB and Fusarium on a plantation can be understood using advanced geospatial information obtained from multispectral aerial photographs taken using an unmanned aerial vehicle (UAV), combined with the reliable data field of infected trees. Vegetation and soil indices derived from a multispectral aerial photograph, such as normalized difference vegetation index, modified chlorophyll absorption ratio index, normalized difference water index (NDWI) and soil pH, may have to be relied on to explain the precise location of these two diseases. In this study, a random forest algorithm was used to handle a large dataset consisting of two models: the banana diseases multispectral model and the banana diseases spectral model. The results show that the soil indices, soil pH and NDWI are the most important variables for predicting the spatial distribution of these two diseases. Simultaneously, the plantation area affected by BDB is more extensive than that affected by Fusarium, if the variation of planted banana cultivars is not considered.

Little is known about how differences in water availability within the “super humid” tropics can influence the distribution of understory plant species and the composition of understory plant communities. We investigated variation in the physiological drought tolerances of understory plants and plant communities across a large elevation and precipitation gradient. We established 58 understory plots along a gradient of 400 – 3600 m asl elevation and 1000 – 6000 mm yr-1 rainfall in and around Manu National Park in southeastern Peru. Within the plots, we sampled all understory woody plants and measured three metrics of physiological leaf drought tolerance - turgor loss point (TLP), cuticular conductance (Gmin), and solute leakage (SL) - and assessed how the community-level means of these traits related to mean annual precipitation (MAP) and elevation (in the study gradient temperature decreases linearly and vapor pressure deficit increases monotonically with elevation). We did not find any correlations between the three metrics of leaf drought tolerance, suggesting that they represent independent strategies for coping with low water availability. Despite being widely used metrics of leaf drought tolerance, neither TLP nor Gmin showed any significant relationships with elevation or MAP. In contrast, SL, which has only recently been developed for use in ecological field studies, increased significantly at higher precipitations and at lower elevations (i.e., plants in colder and drier habitats have lower SL, indicating greater drought tolerances). Our results illustrate that differences in water availability may play a strong role in structuring plant communities even in in the wet tropics and highlight the potential for SL assays to be an efficient and effective tool for measuring drought tolerances in the field.

The 2030 global agenda of the United Nations emphasizes peace, human rights, gender equality, partnership, and women's empowerment. It balances the social, economic, and environmental dimensions of sustainable development. The SDG framework was designed to be integrated and indivisible, making the interconnections of 17 Goals and 169 Sustainable Development Goals Targets (SDTs) more complex that need extensive and intensive investigative research. Recent studies have increased on the interrelationships of SDGs, but none of these have focused on integrating Climate Change (CC), Sustainable Forest Management (SFM), and Information and Communication Technology (ICT), also known as CSI Nexus. This study aims to investigate and identify the 169 SDTs linked to CC, SFM, and ICT and assesses the significance of the relationships between these variables. The alignment of SDTs to CSI Nexus was identified through clustering and mapping techniques. The result argued that 56 SDTs are directly connected within CC+SFM+ICT, 16 within CC+SFM, one within SFM+ICT, and 51 within ICT+CC. Our result suggests that CC is significantly associated with SFM; however, ICT has no significant association with CC and SFM. It further asserts that the ICT and SFM have minimal influence on the SDG 2030 framework. The proposed CSI Nexus and SDTs Integration Framework was a science-informed guide for priority-setting, policy coherence, and decision-making supporting the 2030 Sustainable Development Goals. This study does not include interactions, network analysis and visualization, simulations, and modeling between SDG targets.

Lettuce is a salt-sensitive crop and has a threshold electrical conductivity of 1.3–2.0 mS cm-1 and above that is considered detrimental. As there has been very little information on the physio-logical response of different critical stages of lettuce under different salt stress (SS), the current study is focused on investigating the effects of SS on the critical physiological traits influencing the carbon assimilation in different growth stages of lettuce. The experiment was conducted in deep-water culture hydroponic system in a greenhouse condition. Four levels of sodium chloride salt treatments (EC: 20, 16, 8, and 1.8 mS cm-1) were applied. During both growth stages (day 11 (GS1) and day 19 (GS2) after salt treatment), the leaf transpiration rate, stomatal conductance, and intercellular CO2 were severely decreased. However, the carbon assimilation rate remained unchanged under SS. Similarly, the water use efficiency increased under the SS. It is concluded that the increasing SS increased stomatal and non-stomatal limiting factors during GS1 suggesting the enhanced limitation in photosynthetic activity while no such trend was observed during GS2. The decreased gm with increased SS at GS1 and GS2 suggested that SS induced the irreversible decrease of gm, which in turn can be responsible for the transient reduction in the Vcmax and Jmax during GS2. Taken together, the evidence from this research recommends that varying the SS levels can significantly affect the physiological performance of lettuce at both growth stages.