Search

Machine learning (ML) algorithms are frequently used in landslide susceptibility modeling. Different data handling strategies may generate variations in landslide susceptibility modeling, even when using the same ML algorithm. This research aims to compare the combinations of inventory data handling, cross validation (CV), and hyperparameter tuning strategies to generate landslide susceptibility maps. The results are expected to provide a general strategy for landslide susceptibility modeling using ML techniques. The authors employed eight landslide inventory data handling scenarios to convert a landslide polygon into a landslide point, i.e., the landslide point is located on the toe (minimum height), on the scarp (maximum height), at the center of the landslide, randomly inside the polygon (1 point), randomly inside the polygon (3 points), randomly inside the polygon (5 points), randomly inside the polygon (10 points), and 15 m grid sampling. Random forest models using CV–nonspatial hyperparameter tuning, spatial CV–spatial hyperparameter tuning, and spatial CV–forward feature selection–no hyperparameter tuning were applied for each data handling strategy. The combination generated 24 random forest ML workflows, which are applied using a complete inventory of 743 landslides triggered by Tropical Cyclone Cempaka (2017) in Pacitan Regency, Indonesia, and 11 landslide controlling factors. The results show that grid sampling with spatial CV and spatial hyperparameter tuning is favorable because the strategy can minimize overfitting, generate a relatively high-performance predictive model, and reduce the appearance of susceptibility artifacts in the landslide area. Careful data inventory handling, CV, and hyperparameter tuning strategies should be considered in landslide susceptibility modeling to increase the applicability of landslide susceptibility maps in practical application.

Despite over two decades since the EU Water Framework Directive have passed, achieving the desired water quality in German surface waters remains challenging, regardless of efforts to reduce phosphorus inputs and associated environmental impacts. This study aims at analyzing the characteristics governing the concentrations of four key water quality parameters (total phosphorus, orthophosphate, particulate phosphate, and suspended solids) in two lowland catchments: the 50 km2 catchment of the Kielstau, Germany, and its 7 km2 tributary, the Moorau, which are dominated by agricultural land use. To this end, different sampling methods, particularly high-resolution precipitation event-based sampling and daily mixed samples, are conducted and evaluated, and their effectiveness is compared. The identification of sources and characteristics that affect phosphorus and suspended sediment dynamics, both in general and specifically during heavy precipitation events, is one focus of the study. Over a 15-year period, increasing concentrations of these parameters were observed in daily mixed samples, exhibiting distinct seasonal patterns—higher in summer and lower in winter—consistent with lowland catchment behavior. Particularly during heavy precipitation events, the smaller catchment exhibits a more complex and less predictable response to chemical concentrations compared with the dilution effect observed in the larger catchment. The results underline the complexity of phosphorus dynamics in small catchments and emphasize the importance of event-based sampling for capturing short-term concentration peaks for all four parameters, particularly beneficial regarding measuring suspended solids. While daily mixed samples capture average phosphorus concentrations, event-based sampling is crucial for detecting short-term spikes, providing a more comprehensive understanding of phosphorus dynamics. [ABSTRACT FROM AUTHOR]

Land Use/ Land Cover change detection is an inspiring, interesting task to be performed worldwide. The real-time satellite data of the earth's surface and its different findings can be studied with the assistance of Remote Sensing and Geographic Information Systems. This work builds the novel Fuzzy Swin Transformer-based LU/LC classification model using the LISS-III satellite data of Yelagiri Hills. In the proposed work, the LU/LC features extracted from fuzzy clustering were used as the input training patches when executing the Swin Transformer model. The training patches assist the transformer in finding the LU/LC classification map with good computational complexity and accuracy. The Simple Random, Cluster, Systematic, and Stratified Random sampling methods were used to validate the accuracy of the acquired LU/LC classification map. The proposed Fuzzy Swin Transformer model attains good results with an average classification accuracy of 98.43% using Simple Random Sampling, 97.45% using Stratified Random Sampling, 97.36% using Systematic Sampling, and 96.97% using Cluster Sampling. The LU/LC change detected in this work was considered an important source of information to support the concerned land resource planners in taking necessary action to preserve the land cover, exclusively for the forest-covered areas of the hill stations. [ABSTRACT FROM AUTHOR]

Mobile applications have been used for multiple purposes from entertainment to critical domains. As a consequence, the quality of mobile applications has become a crucial aspect, for instance, by promoting the use of testing as a quality assurance practice. The diversity of mobile platforms is pervaded by several resources related to communication capabilities, sensors, and user-controlled options. As a result, applications can present unexpected behaviors and resource interactions can introduce failures that manifest themselves in specific resource combinations. These failures can compromise the mobile application quality and harm the user experience. We evaluate the failure-detection capability and effectiveness of five sampling testing strategies (Random, One-Enabled, One-Disabled, Most-Enabled-Disabled, and Pairwise) in the context of resource related failures in 15 mobile applications. We focus on 14 common resources of the Android platform and analyze the resource interactions more likely to cause failures. Random had great percentages of failing test cases, followed by One-Enabled, Most-Enabled-Disabled, and Pairwise. We observe that One-Enabled and One-Disabled are the most effective testing strategies for six and four applications, respectively. Surprisingly, resource pairs have more influence on failures than other resource combinations, varying widely among the applications. [ABSTRACT FROM AUTHOR]

The chemical burden on the environment and human population is increasing. Consequently, regulatory risk assessment must keep pace to manage, reduce, and prevent adverse impacts on human and environmental health associated with hazardous chemicals. Surveillance of chemicals of known, emerging, or potential future concern, entering the environment-food-human continuum is needed to document the reality of risks posed by chemicals on ecosystem and human health from a one health perspective, feed into early warning systems and support public policies for exposure mitigation provisions and safe and sustainable by design strategies. The use of less-conventional sampling strategies and integration of full-scan, high-resolution mass spectrometry and effect-directed analysis in environmental and human monitoring programmes have the potential to enhance the screening and identification of a wider range of chemicals of known, emerging or potential future concern. Here, we outline the key needs and recommendations identified within the European Partnership for Assessment of Risks from Chemicals (PARC) project for leveraging these innovative methodologies to support the development of next-generation chemical risk assessment.

Object detection (OD) coupled with active learning (AL) has emerged as a powerful synergy in the field of computer vision, harnessing the capabilities of machine learning (ML) to automatically identify and perform image-based objects localisation while actively engaging human expertise to iteratively enhance model performance and foster machine-based knowledge expansion. Their prior success, demonstrated in a wide range of fields (e.g., industry and medicine), motivated this work, in which a comprehensive and systematic review of OD and AL techniques was carried out, considering reputed technical/scientific publication databases—such as ScienceDirect, IEEE, PubMed, and arXiv—and a temporal range between 2010 and December 2022. The primary inclusion criterion for papers in this review was the application of AL techniques for OD tasks, regardless of the field of application. A total of 852 articles were analysed, and 60 articles were included after full screening. Among the remaining ones, relevant topics such as AL sampling strategies used for OD tasks and groups categorisation can be found, along with details regarding the deep neural network architectures employed, application domains, and approaches used to blend learning techniques with those sampling strategies. Furthermore, an analysis of the geographical distribution of OD researchers across the globe and their affiliated organisations was conducted, providing a comprehensive overview of the research landscape in this field. Finally, promising research opportunities to enhance the AL process were identified, including the development of novel sampling strategies and their integration with different learning techniques. [ABSTRACT FROM AUTHOR]

With recent technological advancements, production systems have become more susceptible to cyber‐physical attacks. Such attacks can be intelligently designed to both alter product dimensions and avoid detection by current Quality Control (QC) tools. The objective of this work is to assess the performance of random sampling strategies for control charting in detecting cyber‐physical attacks. More specifically, the methodology adopts traditional random sampling approaches used for inspection and applies them to univariate control charts. Different random sampling strategies are discussed and their performances under varying attack scenarios are evaluated. In addition, new control chart performance metrics were developed for a more in‐depth analysis, to guide practitioners in assessing how well these control charts increase their robustness to cyber‐physical attacks. [ABSTRACT FROM AUTHOR]

In this work, we tested different combinations of sampling strategies, random sampling and conditioned Latin Hypercube sampling (cLHS)] and sample ratios (10% = 147 and 25% = 368) to predict soil phosphorus and potassium contents, previously estimated using standard laboratory protocols. Other environmental covariates, used as input data for prediction, were obtained from different sources (multispectral Landsat-OLI 8 image, WorldClim database, ISRIC soil database, and ASTER-GDEM). Our findings showed that random sampling was suitable for predicting phosphorus, while the conditioned Latin Hypercube sampling was suitable for predicting potassium. Furthermore, we observed that when the sample ratio increased from 10 to 25%, model accuracy improved in random sampling and cLHS for phosphorus and potassium prediction. However, before generalizing these findings, we recommend that further studies be conducted under different conditions (climate, soil types and parent materials) and testing other sample ratios to determine the best sampling strategy with the optimum ratio to predict soil nutrients better. [ABSTRACT FROM AUTHOR]

Simple Summary: The livestock sector is considered a source of negative impacts on the environment. The measurement of gas concentrations inside dairy barns has a relevant role in the implementation of reduction strategies as well as in improving the indoor air quality for both workers and animals in the barn. The aim of this research study was to contribute to assessing the methodological choices in measurement campaigns of pollutant gases in open dairy barns. Based on the assessment of data variability, the location and the number of sampling points influenced the measurement of the gas in an open dairy barn. In the livestock sector, the monitoring of gas concentrations plays a relevant role in the implementation of mitigation strategies, as well as the improvement in the indoor air quality for both workers and animals in barns. In the present study, measurements of gas concentrations (NH3, CH4, and CO2) at different locations were carried out in an open dairy barn using a commercial photo-acoustic spectroscopy system. Measurement variability for different datasets was computed to contribute to the evaluation of the sampling strategy in the measurement campaign. The results showed that the position of sampling locations (SLs) significantly influenced (p < 0.001) the variability in the gas distribution. Specifically, the SLs located on the perimeter of the building had the highest variability. The number of SLs utilised for determining the mean value of gas concentration in the barn had a significant influence on NH3 (p < 0.001), CO2 (p < 0.001), and CH4 (p < 0.001) in both the central zone and the perimeter one. At least three SLs were necessary to obtain a mean value of gas concentration that reduced the variability to under the 10% in the central zone and 16% in the perimeter one. Moreover, the NH3 concentrations obtained as the mean value of the gas measurements at two SLs with a 10 m distance had a lower variability compared to those recorded at two SLs located at 5 m. [ABSTRACT FROM AUTHOR]

Nitrogen dioxide (NO 2) is a critical air pollutant affecting health and the environment. However, existing monitoring station networks often fail to adequately capture the regional distribution of NO 2 , indicating a need for enhanced sampling strategies. This study focuses on Southwest Fujian in China and introduces a high-precision NO 2 background map to delineate spatial stratified heterogeneity. Subsequently, the Mean of Surface with Non-homogeneity (MSN) method was employed to optimize the design of the NO 2 monitoring network, proposing the addition of 125 new stations. The Bayesian Kriging analysis, utilized to evaluate the optimized network, resulted in a coefficient of determination (R2) of 0.87, a mean absolute percentage error (MAPE) of 9.61%, a mean absolute error (MAE) of 2.75 μg/m3 and a root mean square error (RMSE) of 2.47 μg/m3. The improved accuracy and efficiency of the NO 2 monitoring network were validated against the background map. This research underscores the effectiveness of integrating precise mapping techniques with strategic network optimization for superior environmental monitoring outcomes. • NO 2 s Monitoring stations deployment based on background map is vital to assessing health exposure. • Our method uses a NO 2 background map for spatial variations and stratified area stratification. • Improved NO 2 Monitoring stations sampling efficiency through accurate stratification of spatial heterogeneity. [ABSTRACT FROM AUTHOR]

Object detection (OD) coupled with active learning (AL) has emerged as a powerful synergy in the field of computer vision, harnessing the capabilities of machine learning (ML) to automatically identify and perform image-based objects localisation while actively engaging human expertise to iteratively enhance model performance and foster machine-based knowledge expansion. Their prior success, demonstrated in a wide range of fields (e.g., industry and medicine), motivated this work, in which a comprehensive and systematic review of OD and AL techniques was carried out, considering reputed technical/scientific publication databases—such as ScienceDirect, IEEE, PubMed, and arXiv—and a temporal range between 2010 and December 2022. The primary inclusion criterion for papers in this review was the application of AL techniques for OD tasks, regardless of the field of application. A total of 852 articles were analysed, and 60 articles were included after full screening. Among the remaining ones, relevant topics such as AL sampling strategies used for OD tasks and groups categorisation can be found, along with details regarding the deep neural network architectures employed, application domains, and approaches used to blend learning techniques with those sampling strategies. Furthermore, an analysis of the geographical distribution of OD researchers across the globe and their affiliated organisations was conducted, providing a comprehensive overview of the research landscape in this field. Finally, promising research opportunities to enhance the AL process were identified, including the development of novel sampling strategies and their integration with different learning techniques.

Abstract Background Early detection and prediction of type two diabetes mellitus incidence by baseline measurements could reduce associated complications in the future. The low incidence rate of diabetes in comparison with non-diabetes makes accurate prediction of minority diabetes class more challenging. Methods Deep neural network (DNN), extremely gradient boosting (XGBoost), and random forest (RF) performance is compared in predicting minority diabetes class in Tehran Lipid and Glucose Study (TLGS) cohort data. The impact of changing threshold, cost-sensitive learning, over and under-sampling strategies as solutions to class imbalance have been compared in improving algorithms performance. Results DNN with the highest accuracy in predicting diabetes, 54.8%, outperformed XGBoost and RF in terms of AUROC, g-mean, and f1-measure in original imbalanced data. Changing threshold based on the maximum of f1-measure improved performance in g-mean, and f1-measure in three algorithms. Repeated edited nearest neighbors (RENN) under-sampling in DNN and cost-sensitive learning in tree-based algorithms were the best solutions to tackle the imbalance issue. RENN increased ROC and Precision-Recall AUCs, g-mean and f1-measure from 0.857, 0.603, 0.713, 0.575 to 0.862, 0.608, 0.773, 0.583, respectively in DNN. Weighing improved g-mean and f1-measure from 0.667, 0.554 to 0.776, 0.588 in XGBoost, and from 0.659, 0.543 to 0.775, 0.566 in RF, respectively. Also, ROC and Precision-Recall AUCs in RF increased from 0.840, 0.578 to 0.846, 0.591, respectively. Conclusion G-mean experienced the most increase by all imbalance solutions. Weighing and changing threshold as efficient strategies, in comparison with resampling methods are faster solutions to handle class imbalance. Among sampling strategies, under-sampling methods had better performance than others.

Fecal cortisol is a good indicator to measure the stress intensity of animals,but its detection accuracy is affected by many factors.In order to evaluate the effects of different sampling strategies on fecal cortisol of giant panda(Ailuropoda melanoleuca),feces was divided into four equal parts along the horizontally and vertically.Each part of feces was treated by freeze-drying screening method and its cortisol was measured by enzyme-linked immunosorbent assay(ELISA).The aim was to analyze the distribution of fecal hormone and further explore the sampling strategies.The results showed that there was no significant difference in cortisol content of each part of feces(df=3,F=0.033,P=0.992).Indicating that the hormone distribution in feces was relatively uniform under this four equal treatment.In other words,hormone at each part of feces can represent the overall level,to some extent.In addition,the coefficient of variation generated by 1/2 sampling was smaller than that generated by 1/4 sampling,indicating that 1/2 sampling had higher detection accuracy.There was a significant positive correlation between wet weight and size of feces(r=0.897,P<0.001)and the linear regression equation was y=2.3x-29.3(y meaning wet weight of feces,x meaning size of feces,R²=0.804,F=418.846,P<0.001).Dry weight of feces is determined by fecal type,size and moisture content.In the field,longitudinal 1/2 sampling is recommended. [ABSTRACT FROM AUTHOR]

The exchange between estuaries and the coastal ocean is a key dynamical driver impacting nutrient and phytoplankton concentrations and regulating estuarine residence time, hypoxia, and acidification. Estuarine exchange flows can be particularly challenging to monitor because many systems have strong vertical and lateral velocity shear and sharp gradients in water properties that vary over space and time, requiring high‐resolution measurements in order to accurately constrain the flux. The total exchange flow (TEF) method provides detailed information about the salinity structure of the exchange, but requires observations (or model resolution) that resolve the time and spatial co‐variability of salinity and currents. The goal of this analysis is to provide recommendations for measuring TEF with the most efficient spatial sampling resolution. Results from three realistic hydrodynamic models were investigated. These model domains included three estuary types: a bay (San Diego Bay), a salt‐wedge (Columbia River), and a fjord (Salish Sea). Model fields were sampled using three different mooring strategies, varying the number of mooring locations (lateral resolution) and sample depths (vertical resolution) with each method. The exchange volume transport was more sensitive than salinity to the sampling resolution. Most (>90%) of the exchange flow magnitude was captured by three to four moorings evenly distributed across the estuarine channel with a minimum threshold of 1–5 sample depths, which varied depending on the vertical stratification. These results can improve our ability to observe and monitor the exchange and transport of water masses efficiently with limited resources. Plain Language Summary: The two‐way exchange of water and properties such as heat and salinity as well as other suspended material between estuaries and the coastal ocean is important to regulating these marine habitats. This exchange can be challenging to measure. The total exchange flow (TEF) method provides a way to organize the complexity of this exchange into distinct layers based on a given water property. This method has primarily been applied in numerical models that provide high resolution output in space and time. The goal here is to identify the minimum horizontal and vertical sampling resolutions needed to measure TEF depending on estuary type. Results from three realistic hydrodynamic models were investigated. These models included three estuary types: bay (San Diego Bay), salt‐wedge (Columbia River), and fjord (Salish Sea). The models were sampled using three different mooring strategies, varying the number of mooring locations and sample depths with each method. Most of the exchange magnitude was captured by three to four moorings evenly distributed across the estuarine channel with a minimum threshold of 1–5 sample depths, which varied by estuary. These results can improve our ability to observe and monitor the exchange and transport of water masses efficiently with limited resources. Key Points: A threshold of 1–5 sample depths was needed to measure the exchange flowMost (>90%) of the exchange flow magnitude was captured by 3–4 moorings spread across the channelThe exchange volume transport was more sensitive than salinity to the sampling resolution [ABSTRACT FROM AUTHOR]

A proper-orthogonal-decomposition (POD)-based procedure has been developed to identify the tests/experiments strictly needed for modelling a complex system. The case here studied concerns the aerodynamic efficiency response of a low-pressure-turbine cascade to the variation of the most influencing flow parameters. A large number of experimental tests have been carried out to sample the design space. POD has been applied to reduce data dimensionality, identifying the rank of the problem and defining a vector subspace where data are as compact as possible. Lasso and Gaussian process regressions, complemented by cross-validation and marginal-likelihood criteria, have been used to learn parsimonious efficiency trends in the POD subspace. It will be shown that a model educated with the most orthogonal combination of data projected in the POD space performs as well as the model educated with the entire dataset. Then, it will be shown that the strictly needed tests can be iteratively localized (a priori) in the design space by means of the data-driven procedure here proposed, thus allowing for an active application. This will provide significant time reduction in the execution of tests and can be generalized to other complex systems, either of an experimental or numerical nature, even pertaining to other disciplines. [ABSTRACT FROM AUTHOR]

This paper assesses the challenges and the benefits of 2 methodological approaches for improving the study of transnational mobilities of older adults: mixed methods and inclusiveness. The first approach refers to a mixed-methods research design based on surveys and qualitative interviews. We share our experience of conducting a "fully mixed-concurrent-equal-status-design" research project, for which we collected data through quantitative surveys on transnational practices of people aged 55+. Furthermore, we conducted semistructured interviews with adults aged 64+ living or who have lived in Switzerland and spend part of the year in Spain. The second approach relates to the use of an inclusive sample diversified in terms of nationality and migration backgrounds. The inclusive design affects the formulation of questions asked to all participants (and the response options provided in the survey), as well as decisions related to language choices and translations. These elements have to be considered to ensure that older adults from different backgrounds feel included. In a research project on transnational mobility of older adults, conducting an inclusive plus mixed-methods research project pushes researchers to find strategies to balance research objectives with available resources. Maximizing the research team's methodological background appears to be a suitable approach to address different population groups while working within a budget. [ABSTRACT FROM AUTHOR]

Stable water isotopes are invaluable in helping understand catchment functioning and are widely used in experimental catchments, with higher frequency data becoming increasingly common. Such datasets incur substantial logistical costs, reducing their feasibility for use by decision makers needing to understand multi-catchment, landscape-scale functioning over a relatively short period to assess the impact of proposed land use change. Instead, reconnaissance style surveys (high spatial resolution across the landscape at a lower temporal frequency, over a relatively short period) offer an alternative, complementary approach. To test if such sampling could identify heterogeneities in hydrological functioning, and associated landscape controls, we sampled 27 stream sites fortnightly for one year within a peri-urban landscape undergoing land use change. Visual examination of raw data and application of mean transit time and young water fraction models indicated urbanisation, agriculture and responsive soils caused more rapid cycling of precipitation to stream water, whereas mature forestry provided attenuation. We were also able to identify contiguous catchments which functioned fundamentally differently, meaning their response to land use alteration would also be different. This study demonstrated how stable water isotopes can be a valuable, low-cost addition to tools available for environmental decision makers by providing local, process-based information. [ABSTRACT FROM AUTHOR]

Abstract Background We evaluated whether occupancy modeling, an approach developed for detecting rare wildlife species, could overcome inherent accuracy limitations associated with rapid disease tests to generate fast, accurate, and affordable SARS-CoV-2 prevalence estimates. Occupancy modeling uses repeated sampling to estimate probability of false negative results, like those linked to rapid tests, for generating unbiased prevalence estimates. Methods We developed a simulation study to estimate SARS-CoV-2 prevalence using rapid, low-sensitivity, low-cost tests and slower, high-sensitivity, higher cost tests across a range of disease prevalence and sampling strategies. Results Occupancy modeling overcame the low sensitivity of rapid tests to generate prevalence estimates comparable to more accurate, slower tests. Moreover, minimal repeated sampling was required to offset low test sensitivity at low disease prevalence (0.1%), when rapid testing is most critical for informing disease management. Conclusions Occupancy modeling enables the use of rapid tests to provide accurate, affordable, real-time estimates of the prevalence of emerging infectious diseases like SARS-CoV-2.

Environmental DNA (eDNA) integrated with metabarcoding is a promising and powerful tool for species composition and biodiversity assessment in aquatic ecosystems and is increasingly applied to evaluate fish diversity. To date, however, no standardized eDNA-based protocol has been established to monitor fish diversity. In this study, we investigated and compared two filtration methods and three DNA extraction methods using three filtration water volumes to determine a suitable approach for eDNA-based fish diversity monitoring in the Pearl River Estuary (PRE), a highly anthropogenically disturbed estuarine ecosystem. Compared to filtration-based precipitation, direct filtration was a more suitable method for eDNA metabarcoding in the PRE. The combined use of DNeasy Blood and Tissue Kit (BT) and traditional phenol/chloroform (PC) extraction produced higher DNA yields, amplicon sequence variants (ASVs), and Shannon diversity indices, and generated more homogeneous and consistent community composition among replicates. Compared to the other combined protocols, the PC and BT methods obtained better species detection, higher fish diversity, and greater consistency for the filtration water volumes of 1 000 and 2 000 mL, respectively. All eDNA metabarcoding protocols were more sensitive than bottom trawling in the PRE fish surveys and combining two techniques yielded greater taxonomic diversity. Furthermore, combining traditional methods with eDNA analysis enhanced accuracy. These results indicate that methodological decisions related to eDNA metabarcoding should be made with caution for fish community monitoring in estuarine ecosystems. [ABSTRACT FROM AUTHOR]

Background: Early detection and prediction of type two diabetes mellitus incidence by baseline measurements could reduce associated complications in the future. The low incidence rate of diabetes in comparison with non-diabetes makes accurate prediction of minority diabetes class more challenging.Methods: Deep neural network (DNN), extremely gradient boosting (XGBoost), and random forest (RF) performance is compared in predicting minority diabetes class in Tehran Lipid and Glucose Study (TLGS) cohort data. The impact of changing threshold, cost-sensitive learning, over and under-sampling strategies as solutions to class imbalance have been compared in improving algorithms performance.Results: DNN with the highest accuracy in predicting diabetes, 54.8%, outperformed XGBoost and RF in terms of AUROC, g-mean, and f1-measure in original imbalanced data. Changing threshold based on the maximum of f1-measure improved performance in g-mean, and f1-measure in three algorithms. Repeated edited nearest neighbors (RENN) under-sampling in DNN and cost-sensitive learning in tree-based algorithms were the best solutions to tackle the imbalance issue. RENN increased ROC and Precision-Recall AUCs, g-mean and f1-measure from 0.857, 0.603, 0.713, 0.575 to 0.862, 0.608, 0.773, 0.583, respectively in DNN. Weighing improved g-mean and f1-measure from 0.667, 0.554 to 0.776, 0.588 in XGBoost, and from 0.659, 0.543 to 0.775, 0.566 in RF, respectively. Also, ROC and Precision-Recall AUCs in RF increased from 0.840, 0.578 to 0.846, 0.591, respectively.Conclusion: G-mean experienced the most increase by all imbalance solutions. Weighing and changing threshold as efficient strategies, in comparison with resampling methods are faster solutions to handle class imbalance. Among sampling strategies, under-sampling methods had better performance than others. [ABSTRACT FROM AUTHOR]

Abstract Background Mate choice is a taxonomically wide-spread phenomenon, mostly exerted by females although male mate choice occurs as well. While costs and benefits of choosiness have been well studied, the underlying mechanisms are largely unclear. Different models exist, namely best-of-n or threshold rules, based on sequential or simultaneous sampling, which differ in the required cognitive demands. We applied an experimental approach to shed light on the underlying mechanisms of male mate choice in the sexually cannibalistic spider Argiope bruennichi. Males are limited to two copulations and preferentially monopolise large females, while they may leave smaller females after a single copulation and resume mate search. Here, we utilised significant size-differences between females from Northern and Southern populations and presented males with three different-sized females that were matched for origin: all three females originated either from the same Northern European population as the males or from Southern populations where the smallest female was about the same size as the largest Northern female. This allowed testing the hypothesis that males base their mating tactic on a fixed local size threshold. We predicted Northern males to be choosy among Northern females, but to accept all Southern females since they would all be above that threshold. Results Males copulated with the first female they encountered, which was independent of her body size. Regardless of the females’ origins, males chose a monogynous tactic with the largest female in the trio, while they left the smallest female after one copulation. The same pattern applied to Southern females even though the smallest females in the trio were of a similar size as monopolised Northern females. Since males have poor eyesight and did not actively sample all females, they likely have gained information about relative size differences between females based on volatile chemical cues only. Conclusions Our findings suggest that male A. bruennichi can assess relative differences in mate quality and adjust their mating tactic to the prevailing conditions (Northern vs. Southern). We reject the presence of a locally-adapted fixed threshold and argue that our results are best explained by an adjustable threshold that was raised under Southern conditions.

In this paper, we propose the use of theT2chart with the mixed sampling strategy (MS) to monitor the mean vector of bivariate processes with observations that fit to a first-order vector autoregressive model. With the MS, rational subgroups of size n are taken from the process and the selected units are regrouped to form the mixed samples. The units of the mixed samples are units selected from the last two rational subgroups. The aim of the proposed sampling strategy is to reduce the negative effect of the autocorrelation on the performance of theT2chart. When the two variables are autocorrelated, the MS always enhances theT2chart performance, however, the mixed samples are not recommended for bivariate processes with only one autocorrelated variable which is rarely affected by the assignable cause. [ABSTRACT FROM AUTHOR]

In the livestock sector, the monitoring of gas concentrations plays a relevant role in the implementation of mitigation strategies, as well as the improvement in the indoor air quality for both workers and animals in barns. In the present study, measurements of gas concentrations (NH3, CH4, and CO2) at different locations were carried out in an open dairy barn using a commercial photo-acoustic spectroscopy system. Measurement variability for different datasets was computed to contribute to the evaluation of the sampling strategy in the measurement campaign. The results showed that the position of sampling locations (SLs) significantly influenced (p < 0.001) the variability in the gas distribution. Specifically, the SLs located on the perimeter of the building had the highest variability. The number of SLs utilised for determining the mean value of gas concentration in the barn had a significant influence on NH3 (p < 0.001), CO2 (p < 0.001), and CH4 (p < 0.001) in both the central zone and the perimeter one. At least three SLs were necessary to obtain a mean value of gas concentration that reduced the variability to under the 10% in the central zone and 16% in the perimeter one. Moreover, the NH3 concentrations obtained as the mean value of the gas measurements at two SLs with a 10 m distance had a lower variability compared to those recorded at two SLs located at 5 m.

In this paper, we consider the use of total variation (TV) minimization for compressive imaging, that is, image reconstruction from subsampled measurements. Focusing on two important imaging modalities--namely, Fourier imaging and structured binary imaging via the Walsh--Hadamard transform--we derive uniform recovery guarantees asserting stable and robust recovery for arbitrary random sampling strategies. Using this, we then derive a class of sampling strategies which are theoretically near-optimal for recovery of approximately gradient-sparse images. For Fourier sampling, we show recovery of such an image from m ≳ s · log²(s) · log4(N) measurements, in d ≥ 1 dimensions. When d = 2, this improves the current state-of-the-art result by a factor of log(s) · log(N). It also extends it to arbitrary dimensions d ≥ 1. For Walsh sampling, we prove that m ≳ds · log²(s) · log²(N/s) · log³(N) measurements suffice in d ≥ 2 dimensions. To the best of our knowledge, this is the first recovery guarantee for structured binary sampling with TV minimization. [ABSTRACT FROM AUTHOR]

Isotopes are increasingly used in rainfall‐runoff models to constrain conceptualisations of internal catchment functioning and reduce model uncertainty. However, there is little guidance on how much tracer data is required to adequately do this, and different studies use data from different sampling strategies. Here, we used a 7‐year time series of daily stable water isotope samples of precipitation and streamflow to derive a range of typical stream sampling regimes and investigate how this impacts calibration of a semi‐distributed tracer‐aided model in terms of flow, deuterium and flux age simulations. Over the 7 years weekly sampling facilitated an almost identical model performance as daily, and there were only slight deteriorations in performance for fortnightly sampling. Monthly sampling resulted in poorer deuterium simulations and greater uncertainty in the derived parameter sets ability to accurately represent catchment functioning, evidenced by unrealistic reductions in the volumes of water available for mixing in the saturation area causing simulated water age decreases. Reducing sampling effort and restricting data collection to 3 years caused reductions in the accuracy of deuterium simulation, though the deterioration did not occur if sampling continued for 5 years. Analysis was also undertaken to consider the effects of reduced sampling effort over the driest and wettest hydrological years to evaluate effects of more extreme conditions. This showed that the model was particularly sensitive to changes in sampling during dry conditions, when the catchment hydrological response is most non‐linear. Across all dataset durations, sampling in relation to flow conditions, rather than time, revealed that samples collected at flows >Q50 could provide calibration results comparable to daily sampling. Targeting only extreme high flows resulted in poor deuterium and low flow simulations. This study suggests sufficient characterization of catchment functioning can be obtained through reduced sampling effort over longer timescales and the targeting of flows >Q50. [ABSTRACT FROM AUTHOR]

This article outlines a research design for a qualitative comparative study of communication across platforms, media and contexts – in China, the US and Denmark. After addressing the limitations in previous research on digital media in everyday life, we argue in favour of a comparative ethnography of communication that emphasizes the study of intermediality by taking a people-centred approach. The methodological design combines network sampling and maximum variation sampling with communication diaries and elicitation interviews. This design makes it possible to collect small and deep communicative trace data, to capture individuals’ unique linking of all the communication tools and channels available to them and, in turn, to identify the role of the internet as it interacts and intersects with other forms of communication.

This study evaluated the utility of concurrent water signal acquisition as part of the water suppression in MR spectroscopic imaging (MRSI), to allow simultaneous water referencing for metabolite quantification, and to concurrently acquire functional MRI (fMRI) data. We integrated a spatial‐spectral binomial water excitation RF pulse and a short spatial‐spectral echo‐planar readout into the water suppression module of 2D and 3D proton‐echo‐planar‐spectroscopic‐imaging (PEPSI) with a voxel size as small as 4 x 4 x 6 mm3. Metabolite quantification in reference to tissue water was validated in healthy controls for different prelocalization methods (spin‐echo, PRESS and semi‐LASER) and the clinical feasibility of a 3‐minute 3D semi‐Laser PEPSI scan (TR/TE: 1250/32 ms) with water referencing in patients with brain tumors was demonstrated. Spectral quality, SNR, Cramer‐Rao‐lower‐bounds and water suppression efficiency were comparable with conventional PEPSI. Metabolite concentration values in reference to tissue water, using custom LCModel‐based spectral fitting with relaxation correction, were in the range of previous studies and independent of the prelocalization method used. Next, we added a phase‐encoding undersampled echo‐volumar imaging (EVI) module during water suppression to concurrently acquire metabolite maps with water referencing and fMRI data during task execution and resting state in healthy controls. Integration of multimodal signal acquisition prolongated minimum TR by less than 50 ms on average. Visual and motor activation in concurrent fMRI/MRSI (TR: 1250–1500 ms, voxel size: 4 x 4 x 6 mm3) was readily detectable in single‐task blocks with percent signal change comparable with conventional fMRI. Resting‐state connectivity in sensory and motor networks was detectable in 4 minutes. This hybrid water suppression approach for multimodal imaging has the potential to significantly reduce scan time and extend neuroscience research and clinical applications through concurrent quantitative MRSI and fMRI acquisitions. [ABSTRACT FROM AUTHOR]

Background: We evaluated whether occupancy modeling, an approach developed for detecting rare wildlife species, could overcome inherent accuracy limitations associated with rapid disease tests to generate fast, accurate, and affordable SARS-CoV-2 prevalence estimates. Occupancy modeling uses repeated sampling to estimate probability of false negative results, like those linked to rapid tests, for generating unbiased prevalence estimates.Methods: We developed a simulation study to estimate SARS-CoV-2 prevalence using rapid, low-sensitivity, low-cost tests and slower, high-sensitivity, higher cost tests across a range of disease prevalence and sampling strategies.Results: Occupancy modeling overcame the low sensitivity of rapid tests to generate prevalence estimates comparable to more accurate, slower tests. Moreover, minimal repeated sampling was required to offset low test sensitivity at low disease prevalence (0.1%), when rapid testing is most critical for informing disease management.Conclusions: Occupancy modeling enables the use of rapid tests to provide accurate, affordable, real-time estimates of the prevalence of emerging infectious diseases like SARS-CoV-2. [ABSTRACT FROM AUTHOR]

The ubiquity of organophosphate esters (OPEs) in various environmental matrices inevitably pose human exposure risks. Numerous studies have investigated human exposure pathways to OPEs, including air inhalation, dust ingestion, dermal contact, and dietary and drinking water intake, and have indicated that indoor dust and indoor air routes are frequently the two main human exposure pathways. This article reviews the literature on OPE contamination in indoor air and dust from various microenvironments and on OPE particle size distributions and bioavailability in dust conducted over the past 10 years. Ways in which sampling strategies are related to the uncertainty of exposure assessment results and comparability among different studies in terms of sampling tools, sampling sites, and sample types are addressed. Also, the associations of OPEs in indoor dust/air with human biological samples were summarized. Studies on two emerging matrices, hand wipes and silicone wristbands, are demonstrated to be more comprehensive and accurate in reflecting personal human exposure to OPEs in microenvironments and are summarized. Given the direct application of some diester OPEs (di-OPEs) in numerous products, research on their existence in indoor dust and food and on their effects on human urine are also discussed. Finally, related research trends and avenues for future research are prospected.

Surface-sourced organic compounds in infiltrating waters and percolates are transformed during their belowground passage. Biotic and abiotic processes thereby lead to continuously changing chemical environments in subsurface compartments. The investigation of such transformations of organic compounds aims for tracing subsurface fluxes as well as biotic and abiotic activity. To collect samples of soil solution, different kinds of lysimeters are available, spanning simple free-draining devices that sample water based on gravimetric flow and tension lysimeters allowing for approximating natural hydraulic conditions. Protocols for untargeted analytical profiling of organic soil solution constituents are scarce. We report here a solid phase extraction followed by GC–MS analysis, utilizing two long-term sampling devices in the Hainich Critical Zone Exploratory in Thuringia, Germany. In addition, we introduce a new lysimeter constructed exclusively from inert materials that allows for obtaining samples with little background signals in GC–MS. Polyvinylchloride (PVC)-based lysimeters introduce substantial background signals from plasticizers. We show how signals from these contaminants can be lowered during data analysis using chemometric background removal. Applying multivariate statistics for data analysis, we demonstrate the ability for monitoring of several sugars, fatty acids and phenolic acids at the topsoil-subsoil boundary and even beyond, via an untargeted analytical approach. Statistical tools facilitated the detection of differences in chemical signatures at three different land use sites. Data mining methods for metabolomics led to the identification of 3-carboxyphenylalanin as marker for a pasture site. The combined approach is suitable for the collection and extraction of topsoil and subsoil solution for untargeted metabolomics under near-natural flow conditions.

Choosing the appropriate sampling strategy is significant while estimating the pollutant loads in a snow pile and assessing environmental impacts of dumping snow into water bodies. This paper compares different snow pile sampling strategies, looking for the most efficient way to estimate the pollutant loads in a snow pile. For this purpose, 177 snow samples were collected from nine snow piles (average pile area − 30 m2, height − 2 m) during four sampling occasions at Frihamnen, Ports of Stockholm's port area. The measured concentrations of TSS, LOI, pH, conductivity, and heavy metals (Zn, Cu, Cd, Cr, Pb, and V) in the collected samples indicated that pollutants are not uniformly distributed in the snow piles. Pollutant loads calculated from different sampling strategies were compared against the load calculated using all samples collected for each pile (best estimate of mass load, BEML). The results/study showed that systematic grid sampling is the best choice when the objective of sampling is to estimate the pollutant loads accurately. Estimating pollutant loads from single snow column samples (collected at a point from the snow pile through the entire depth of the pile) produced up to 400% variation from BEML, whereas samples composed by mixing volume-proportional subsamples from all samples (horizontal composite samples) produced only up to 50% variation. Around nine samples were required to estimate the pollutant loads within 50% deviation from BEML for the studied snow piles. Converting pollutant concentrations in snow to equivalent concentrations in snowmelt and comparing it with available guideline values for receiving water, Zn was identified as the critical pollutant. [ABSTRACT FROM AUTHOR]

The chemical burden on the environment and human population is increasing. Consequently, regulatory risk assessment must keep pace to manage, reduce, and prevent adverse impacts on human and environmental health associated with hazardous chemicals. Surveillance of chemicals of known, emerging, or potential future concern, entering the environment-food-human continuum is needed to document the reality of risks posed by chemicals on ecosystem and human health from a one health perspective, feed into early warning systems and support public policies for exposure mitigation provisions and safe and sustainable by design strategies. The use of less-conventional sampling strategies and integration of full-scan, high-resolution mass spectrometry and effect-directed analysis in environmental and human monitoring programmes have the potential to enhance the screening and identification of a wider range of chemicals of known, emerging or potential future concern. Here, we outline the key needs and recommendations identified within the European Partnership for Assessment of Risks from Chemicals (PARC) project for leveraging these innovative methodologies to support the development of next-generation chemical risk assessment., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024. Published by Elsevier Ltd.)

Multidimensional MR experiments of relaxation and diffusion have been successful for material characterization and have attracted attention recently for biomedical applications. However, such experiments typically require many scans of data acquisition and are time‐consuming. This work discusses a method for systematic optimization of the pulse‐sequence parameters to obtain optimal resolution within the experimental conditions, such as the number of acquisitions. Other optimization goals can also be incorporated in this framework. [ABSTRACT FROM AUTHOR]

Tropical forests are predicted to harbor most of the insect diversity on earth, but few studies have been conducted to characterize insect communities in tropical forests. One major limitation is the lack of consensus on methods for insect collection. Deciding which insect trap to use is an important consideration for ecologists and entomologists, yet to date few study has presented a quantitative comparison of the results generated by standardized methods in tropical insect communities. Here, we investigate the relative performance of two flight interception traps, the windowpane trap, and the more widely used malaise trap, across a broad gradient of lowland forest types in French Guiana. The windowpane trap consistently collected significantly more Coleoptera and Blattaria than the malaise trap, which proved most effective for Diptera, Hymenoptera, and Hemiptera. Orthoptera and Lepidoptera were not well represented using either trap, suggesting the need for additional methods such as bait traps and light traps. Our results of contrasting trap performance among insect orders underscore the need for complementary trapping strategies using multiple methods for community surveys in tropical forests.

Tropical forests are predicted to harbor most of the insect diversity on earth, but few studies have been conducted to characterize insect communities in tropical forests. One major limitation is the lack of consensus on methods for insect collection. Deciding which insect trap to use is an important consideration for ecologists and entomologists, yet to date few study has presented a quantitative comparison of the results generated by standardized methods in tropical insect communities. Here, we investigate the relative performance of two flight interception traps, the windowpane trap, and the more widely used malaise trap, across a broad gradient of lowland forest types in French Guiana. The windowpane trap consistently collected significantly more Coleoptera and Blattaria than the malaise trap, which proved most effective for Diptera, Hymenoptera, and Hemiptera. Orthoptera and Lepidoptera were not well represented using either trap, suggesting the need for additional methods such as bait traps and light traps. Our results of contrasting trap performance among insect orders underscore the need for complementary trapping strategies using multiple methods for community surveys in tropical forests.

The sampling frequency of temperature data is examined. The range of sampling from hourly to twice daily are explored to determine the uncertainty that is introduced by sampling less frequently than hourly in the determination of daily temperature and daily temperature extrema. The standards for comparison for the daily average temperature are the average of hourly data and the average of the daily maximum and minimum temperature. Hourly temperature data from 12 Canadian climate stations are examined for several decades leading up to 2017. Daily average temperatures were calculated using data sampled 24 times a day (hourly), 12 times, 8 times, 6 times, 4 times and twice daily. Two triad algorithms from the literature and an experimental one are assessed relative to these sampling frequencies. The sampling frequency analysis was remarkably consistent across all climate stations. The departure from the hourly estimate ranged from 0.1°C for the bi‐hourly sampling to ~1°C for the twice daily sampling. The uncertainty associated with the min/max method consistently fell within that of three and four samples per day. Comparison of triad algorithms, based on a quantitative criterion for determination of best sampling hours, revealed a station specific triad that outperforms algorithms from the literature and thrice daily evenly spaced sampling. Minimum and maximum estimates were compared across the different sampling frequencies for all stations as well. The accuracy of estimating temperature extrema decreases with lower sampling rates with the exception of the 8 hr sampling where hour of sampling influences accuracy. The results demonstrate that the local climate characteristics needs to be considered when choosing the optimal sampling frequency and calculation method for daily means and extrema. [ABSTRACT FROM AUTHOR]

Background: Mate choice is a taxonomically wide-spread phenomenon, mostly exerted by females although male mate choice occurs as well. While costs and benefits of choosiness have been well studied, the underlying mechanisms are largely unclear. Different models exist, namely best-of-n or threshold rules, based on sequential or simultaneous sampling, which differ in the required cognitive demands. We applied an experimental approach to shed light on the underlying mechanisms of male mate choice in the sexually cannibalistic spider Argiope bruennichi. Males are limited to two copulations and preferentially monopolise large females, while they may leave smaller females after a single copulation and resume mate search. Here, we utilised significant size-differences between females from Northern and Southern populations and presented males with three different-sized females that were matched for origin: all three females originated either from the same Northern European population as the males or from Southern populations where the smallest female was about the same size as the largest Northern female. This allowed testing the hypothesis that males base their mating tactic on a fixed local size threshold. We predicted Northern males to be choosy among Northern females, but to accept all Southern females since they would all be above that threshold. Results: Males copulated with the first female they encountered, which was independent of her body size. Regardless of the females' origins, males chose a monogynous tactic with the largest female in the trio, while they left the smallest female after one copulation. The same pattern applied to Southern females even though the smallest females in the trio were of a similar size as monopolised Northern females. Since males have poor eyesight and did not actively sample all females, they likely have gained information about relative size differences between females based on volatile chemical cues only. Conclusions: Our findings suggest that male A. bruennichi can assess relative differences in mate quality and adjust their mating tactic to the prevailing conditions (Northern vs. Southern). We reject the presence of a locally-adapted fixed threshold and argue that our results are best explained by an adjustable threshold that was raised under Southern conditions. [ABSTRACT FROM AUTHOR]

We are concerned with the efficiency of stochastic gradient estimation methods for large-scale nonlinear optimization in the presence of uncertainty. These methods aim to estimate an approximate gradient from a limited number of random input vector samples and corresponding objective function values. Ensemble methods usually employ Gaussian sampling to generate the input samples. It is known from the optimal design theory that the quality of sample-based approximations is affected by the distribution of the samples. We therefore evaluate six different sampling strategies to optimization of a high-dimensional analytical benchmark optimization problem, and, in a second example, to optimization of oil reservoir management strategies with and without geological uncertainty. The effectiveness of the sampling strategies is analyzed based on the quality of the estimated gradient, the final objective function value, the rate of the convergence, and the robustness of the gradient estimate. Based on the results, an improved version of the stochastic simplex approximate gradient method is proposed based on UE(s2) sampling designs for supersaturated cases that outperforms all alternative approaches. We additionally introduce two new strategies that outperform the UE(s2) designs previously suggested in the literature. [ABSTRACT FROM AUTHOR]

Mathematical modelling is a widely used technique for describing the temporal behaviour of biological systems. One of the most challenging topics in computational systems biology is the calibration of non‐linear models; i.e. the estimation of their unknown parameters. The state‐of‐the‐art methods in this field are the frequentist and Bayesian approaches. For both of them, the performance and accuracy of results greatly depend on the sampling technique employed. Here, the authors test a novel Bayesian procedure for parameter estimation, called conditional robust calibration (CRC), comparing two different sampling techniques: uniform and logarithmic Latin hypercube sampling. CRC is an iterative algorithm based on parameter space sampling and on the estimation of parameter density functions. They apply CRC with both sampling strategies to the three ordinary differential equations (ODEs) models of increasing complexity. They obtain a more precise and reliable solution through logarithmically spaced samples. [ABSTRACT FROM AUTHOR]