Additive Model Research Articles

Numerical investigation of scramjet inlet models for side spillage reduction

This study introduces and evaluates five models of Mach 7 scramjet inlets designed to reduce side spillage. The baseline model is characterized as a 2D-wedge type inlet without sidewalls. To mitigate interference from expansion waves, an alternative model featuring a converging angle identical to the freestream Mach angle is proposed. Another inlet model presents each ramp's converging angle matched to the local Mach angle of its respective ramp. Subsequently, an inlet model with sidewalls attached to the baseline model is introduced, and an additional inlet model is presented with reduced-height sidewalls. Three-dimensional Reynolds-Averaged Navier-Stokes simulations are conducted for each inlet model, and the performance of the inlets is compared. The mass flow averaged properties at the isolator exit for each inlet model are computed. To evaluate the performance of engines equipped with each inlet model, thrust and specific impulse are determined using thermodynamic cycle analysis. Ultimately, net thrust is calculated by subtracting drag from thrust, and the results are compared. The inlet with converging ramps, which has the largest captured area, indicates the lowest net thrust, amounting to 38% of that of the baseline model. In contrast, the inlet with reduced height sidewalls represents the highest net thrust, which is 3.5 times that of the inlet with converging ramps. These results demonstrate that for scramjet inlet design, using sidewalls to reduce drag is a more effective method for reducing side spillage compared to employing converging ramps with a large captured area.

SPATIAL REGRESSION APPROACH TO MODELLING POVERTY IN JAVA ISLAND 2022

Geographically Weighted Regression (GWR) model is a powerful tool for analyzing spatial patterns in data. However, the standard form of a spatial model that uses a single bandwidth calibration may be unrealistic because the response-predictor relationship may be either linear or nonlinear. To address this issue, the Multiscale GWR (MSGWR) model offers improved model performance by employing Generalized Additive Model (GAM) with varying bandwidth or smoothing function for each covariate in the model. This research aims to analyze the Percentage of Poor Population (PPP) on Java Island in 2022 using the geospatial models and related socioeconomic and demographic attributes, such as Open Unemployment Rate, Human Development Index, Labor Force Participation Rate, and GRDP Per capita to identify the best model in explaining the spatial pattern and to find out the determinant of PPP on Java Island in 2022. This study uses secondary data from Statistics Indonesia. The findings reveal that the MSGWR model provides the highest R2 and smallest AICc value compared to single bandwidth models, specifically the GWR and MXGWR models. Furthermore, the MSGWR model indicates that HDI has a significant negative effect on PPP, whereas LFPR has a significant positive effect on PPP across all districts in Java Island in 2022.

Electric vehicle and supply equipment adoption dynamics in the United States

In the present era of climate change concerns, plug-in electric vehicles (PEV) and their associated infrastructure are strategic investments by many governments. Investment discussions often include the dynamic supply-demand relationship between PEV adoption and charging station deployment, but rarely is it given an empirical treatment. We make use of publicly available charging station installation and PEV vehicle registration data by U.S. County between the years 2012 and 2021 to examine the phasing and causal relationship between adoption trends in these two markets. Non-linear Granger causality tests suggest a feedback relationship between PEV vehicle adoption and charging station installations. It can be further explained that when individuals adopt PEV, the demand for the charging infrastructure increases to meet the demand for PEV. Conversely, when more charging station installations become available, it encourages adoption of this technology making it convenient for individuals to have access to these charging facilities. It is observed in aggregate that PEVs are penetrating across the U.S., but the penetration rate differs across states and counties. We examine this interaction using a generalized propensity score (GPS) identification strategy for the effect of charging station installations on PEV registrations, coupled with a generalized additive model (GAM) to construct a causal dose-response curve. We observe statistically significant effects on per capita charging station rates for income and total population. We suggest a policy package that includes charging infrastructure investments but also provides wider adoption of subsidies for low-income households across all states.

Performance of unmarked abundance models with data from machine‐learning classification of passive acoustic recordings

AbstractThe ability to conduct cost‐effective wildlife monitoring at scale is rapidly increasing due to the availability of inexpensive autonomous recording units (ARUs) and automated species recognition, presenting a variety of advantages over human‐based surveys. However, estimating abundance with such data collection techniques remains challenging because most abundance models require data that are difficult for low‐cost monoaural ARUs to gather (e.g., counts of individuals, distance to individuals), especially when using the output of automated species recognition. Statistical models that do not require counting or measuring distances to target individuals in combination with low‐cost ARUs provide a promising way of obtaining abundance estimates for large‐scale wildlife monitoring projects but remain untested. We present a case study using avian field data collected in the forests of Pennsylvania during the spring of 2020 and 2021 using both traditional point counts and passive acoustic monitoring at the same locations. We tested the ability of the Royle–Nichols and time‐to‐detection models to estimate the abundance of two species from detection histories generated by applying a machine‐learning classifier to ARU‐gathered data. We compared abundance estimates from these models with estimates from the same models fit using point‐count data and to two additional models appropriate for point counts, the N‐mixture model and distance models. We found that the Royle–Nichols and time‐to‐detection models can be used with ARU data to produce abundance estimates similar to those generated by a point‐count‐based study but with greater precision. ARU‐based models produced confidence or credible intervals that were on average 31.9% (±11.9 SE) smaller than their point‐count counterpart. Our findings were consistent across two species with differing relative abundance and habitat use patterns. The higher precision of models fit using ARU data is likely due to higher cumulative detection probability, which itself may be the result of greater survey effort using ARUs and machine‐learning classifiers to sample significantly more time for focal species at any given point. Our results provide preliminary support for the use of ARUs in abundance‐based study applications, and thus may afford researchers a better understanding of habitat quality and population trends, while allowing them to make more informed conservation recommendations and actions.

Glycated Hemoglobin A1c Time in Range and Dementia in Older Adults With Diabetes

Individuals with diabetes commonly experience Alzheimer disease and related dementias (ADRD). Factors such as hypoglycemia, hyperglycemia, and glycemic variability have been associated with increased risk of ADRD. Traditional glycemic measures, such as mean glycated hemoglobin A1c (HbA1c), may not identify the dynamic and complex pathophysiologic factors in the association between diabetes and ADRD. The HbA1c time in range (TIR) is a previously developed measure of glycemic control that expresses HbA1c stability over time within specific ranges. This measure may inform the current understanding of the association between glucose levels over time and ADRD incidence. To examine the association between HbA1c TIR and incidence of ADRD in older veterans with diabetes. The study sample for this cohort study was obtained from administrative and health care utilization data from the Veterans Health Administration and Medicare from January 1, 2004, to December 31, 2018. Veterans 65 years or older with diabetes were assessed. Participants were required to have at least 4 HbA1c tests during the 3-year baseline period, which could start between January 1, 2005, and December 31, 2014. Data analysis was conducted between July and December 2023. Hemoglobin A1c TIR was calculated as the percentage of days during baseline in which HbA1c was in individualized target ranges based on clinical characteristics and life expectancy, with higher HbA1c TIR viewed as more favorable. The association between HbA1c TIR and ADRD incidence was estimated. Additional models considered ADRD incidence in participants who were above or below HbA1c target ranges most of the time. The study included 374 021 veterans with diabetes (mean [SD] age, 73.2 [5.8] years; 369 059 [99%] male). During follow-up of up to 10 years, 41 424 (11%) developed ADRD. Adjusted Cox proportional hazards regression models showed that lower HbA1c TIR was associated with increased risk of incident ADRD (HbA1c TIR of 0 to <20% compared with ≥80%: hazard ratio, 1.19; 95% CI, 1.16-1.23). Furthermore, the direction of out-of-range HbA1c levels was associated with incident ADRD. Having greater time below range (≥60%, compared with ≥60% TIR) was associated with significantly increased risk (hazard ratio, 1.23; 95% CI, 1.19-1.27). Findings remained significant after excluding individuals with baseline use of medications associated with hypoglycemia risk (ie, insulin and sulfonylureas) or with hypoglycemia events. In this study of older adults with diabetes, increased HbA1c stability within patient-specific target ranges was associated with a lower risk of ADRD. Lower HbA1c TIR may identify patients at increased risk of ADRD.

Role and applications of advanced digital technologies in achieving sustainability in multimodal logistics operations: A systematic literature review

Integration of advanced digital technologies along with ensuring environmental sustainability in the logistics sector is becoming the need of the hour novel multimodal transportation concept. The rapid technological development in the era of industry 4.0 can not only create new opportunities in achieving highly efficient, intelligent, and smart logistics systems but also enable new business models for value addition, additionally, this novel system will also avoid environmental degradation and use of scarce natural resources that contribute to climate change. This paper summarizes the research trends of these technologies in multimodal logistics systems since 2005 through a systematic literature review. The results revealed a mixed picture of the situation, starting from emerging technologies like artificial intelligence, machine learning, and blockchain becoming key technological concepts for multimodal logistics systems to government policy hurdles and lack of impact breakthroughs in the industry. However, the results show that adaptation to these technologies improves the overall environmental, economic, and social sustainability of the system.

Thermo-economic Analysis and Comparison of Novel Desiccant Air Conditioning Systems

Desiccant air conditioning (DAC) systems are considered an effective alternative to traditional vapor compression systems due to their ability to independently control temperature and humidity, coupled with their environmentally friendly design. Consequently, this paper deals with experimental and simulation study on DAC systems. The experiments were conducted on a reference system (RS) equipped with all necessary control and measurement devices. This reference system includes a traditional desiccant evaporative air-conditioning system with an additional heat exchanger placed after the desiccant wheel, functioning as an after-cooler. In addition, steady state and transient simulation models were developed using TRNSYS software to predict the performance of four different configurations of DAC systems and the reference system, under hot and humid climatic conditions.The experimental data verify the precision and relevance of the simulation results, with an average error of 2.57 % for the air temperatures. The steady state findings showed that Config-4 exhibited the best thermal performance compared to the other configurations. Also, Config-4 demonstrates a stable thermal cooling capacity despite fluctuations in outdoor air dry-bulb temperature. Transient results revealed that, Config-4 has levelized costs of CO2 emissions that are 64.2% lower than RS, followed by Config-3, Config-2, and Config-1. Lower CO2 emissions result in a positive impact on mitigating climate change and improving air quality. Economic analysis revealed that the total levelized cost of cooling (LCOC) for Config-4 is 70.5% lower than RS, demonstrating the most significant cost reduction. Configurations 3, 2, and 1 also showed lower LCOC compared to the reference system.

The associations of early-life health and performance with subsequent dairy cow longevity, productivity, and profitability

Selecting and raising dairy animals that are more likely to reach their potential is a strategy to increase milk production efficiency and overall profitability. However, indicators are necessary for the early identification of animals that are less likely to perform well, allowing for their early culling and ensuring that resources are allocated to those with the highest potential. The objective of this study was to analyze the association between early-life animal health and performance with longevity, production, and profitability. After data cleaning, the following early-life measures (i.e., predictors) were available for 363 female calves born between June 2014 and November 2015 in eight dairy herds from New Brunswick, Canada (average: 45 calves/farm; standard deviation: 26.1 calves/farm; median: 42 calves/farm; range: 15 to 95 calves/farm): birth weight, weaning weight, weaning age, weaning average daily gain (weaning ADG), immunoglobulin G (IgG) serum concentration, the occurrence of navel infection, diarrhea, and pneumonia, and if animals received antibiotic treatment between birth and weaning. Their subsequent length of life (LL), length of productive life (LPL), lifetime cumulative energy-corrected milk (ECM), and lifetime cumulative milk value (i.e., response variables) were provided by the Canadian dairy herd improvement agency. Bayesian Additive Regression Tree models were trained for each response variable using 5-fold cross-validation. Models were evaluated using the RMSE and R2. The three most important predictors were identified using permutation, and the relationship between response variables and important predictors was assessed using accumulated local effect plots. The RMSE for LL, LPL, ECM, and milk value were 1.43 years, 1.37 years, 16 314.94 kg, and $CAD 11 525.68, respectively, whereas the R2 values were 0.30, 0.25, 0.29, and 0.29, respectively, indicating a moderate relationship between predictors and response variables. Non-linear relationships were found between the response variables and important predictors. Animals born with low or high birth weights were associated with decreased LL, LPL, ECM, and milk value. The highest LL, LPL, and milk value was observed for calves weaned between 1.9 and 2.0 months old, followed by a decline for calves weaned at older ages. The lowest LL and ECM were associated with weaning ADG of 0.786 kg/day, while 0.787 kg/day was associated with the lowest LPL. Lastly, both ECM and milk value were highest when serum IgG values were 1 659 mg/dL. These findings provide valuable insights for optimizing early culling decisions and enhancing the productivity and profitability of dairy farms.

Enhancing hydrochar production and proprieties from biogenic waste: Merging response surface methodology and machine learning for organic pollutant remediation

The valorization of biogenic waste by hydrothermal carbonization is widely discussed in research. However, to our knowledge, no study has combined almond shells and olive pomace to synthesize a solid carbon material. The purpose of this study is to enhance the hydrochar process from AS and OP using RSM methodology and machine learning models: ANN, SVM and XG-Boost. Subsequently, a study was carried out on the removal of organic pollutants by the synthesized material. The optimum Co-HTC operating conditions obtained at 180 C, 90 min with acid catalyst corresponding to 71.51 % and 87.13 % for mass yield and carbon retention rate respectively according to RSM-CCD. The comparison between RSM-CCD and ML in terms of prediction concludes that RSM remains more efficient in terms of planning and optimization. However, ANN is more suitable for modeling and predicting mass yields and carbon retention rates. Hydrochar’s physicochemical properties were evaluated by the use of spectroscopic methods like FTIR, SEM, XRD, and CHNO. To conclude, we studied the performance of HCop in methylene blue adsorption, varying the following parameters: pH, contact time, initial dye concentration, adsorbent dose and temperature. In addition, kinetic and isothermal models were studied to describe the dominant mechanisms in the MB adsorption process. The MB maximal adsorption using HCop obtained at pH 9, with an initial dye concentration of 100 mg. L−1, 40-min contact time, 0.1 g/L adsorbent dose, and a temperature between 25 and 30 °C. In conclusion, these results provide important information on the use of Co-HTC to convert biogenic wastes into high-performance carbon materials for the appropriate removal of organic pollutants. More studies are needed to use the material in other fields of application.

Associations between short-term exposure to airborne carbonaceous particles and mortality: A time-series study in London during 2010–2019

Exposure to ambient particulate matter (PM) has been identified as a major global health concern; however, the importance of specific chemical PM components remains uncertain. Recent studies have suggested that carbonaceous aerosols are important detrimental components of the particle mixture. Using time-series methods, we investigated associations between short-term exposure to carbonaceous particles and mortality in London, UK. Daily counts of non-accidental, respiratory, and cardiovascular deaths were obtained between 2010 and 2019. For the same period, daily concentrations of carbonaceous particles: organic (OC), elemental (EC), wood-burning (WC), total carbon (TC) and equivalent black carbon (eBC) were sourced from two centrally located monitoring sites (one urban-traffic and one urban-background). Generalized additive models were used to estimate the percentage change in mortality risk associated with interquartile range increases in particulate concentrations. Lagged effects up to 3 days were examined. Stratified analyses were conducted by age, sex, and season, separate analyses were also performed by site-type. For non-accidental mortality, positive associations were observed for all particle species at lag1, including statistically significant percentage risk changes in WC (0.51% (95%CI: 0.19%, 0.82%) per IQR (0.68 μg/m3)) and OC (0.45% (95%CI: 0.04%, 0.87% per IQR (2.36 μg/m3)). For respiratory deaths, associations were greatest for particulate concentrations averaged over the current and previous 3 days, with increases in risk of 1.70% (95%CI: 0.64%, 2.77%) for WC and 1.31% (95%CI: −0.08%, 2.71%) for OC. No associations were found with cardiovascular mortality. Results were robust to adjustment for particle mass concentrations. Stratified analyses suggested particulate effects were greatest in the summer and respiratory associations more pronounced in females. Our findings are supportive of an association between carbonaceous particles and non-accidental and respiratory mortality. The strongest evidence of an effect was for WC; this is of significance given the rising popularity of wood-burning for residential space heating and energy production across Europe.

Estimating the short-term effect of PM2.5 on the mortality of cardiovascular diseases based on instrumental variables

BackgroundPM2.5 can induce and aggravate the occurrence and development of cardiovascular diseases (CVDs). The objective of our study is to estimate the causal effect of PM2.5 on mortality rates associated with CVDs using the instrumental variables (IVs) method.MethodsWe extracted daily meteorological, PM2.5 and CVDs death data from 2016 to 2020 in Binzhou. Subsequently, we employed the general additive model (GAM), two-stage predictor substitution (2SPS), and control function (CFN) to analyze the association between PM2.5 and daily CVDs mortality.ResultsThe 2SPS estimated the association between PM2.5 and daily CVDs mortality as 1.14% (95% CI: 1.04%, 1.14%) for every 10 µg/m3 increase in PM2.5. Meanwhile, the CFN estimated this association to be 1.05% (95% CI: 1.02%, 1.10%). The GAM estimated it as 0.85% (95% CI: 0.77%, 1.05%). PM2.5 also exhibited a statistically significant effect on the mortality rate of patients with ischaemic heart disease, myocardial infarction, or cerebrovascular accidents (P < 0.05). However, no significant association was observed between PM2.5 and hypertension.ConclusionPM2.5 was significantly associated with daily CVDs deaths (excluding hypertension). The estimates from the IVs method were slightly higher than those from the GAM. Previous studies based on GAM may have underestimated the impact of PM2.5 on CVDs.

Association between short-term ambient particulate matters and acute liver injury outpatient visits: A hospital-based study

Acute liver injury (ALI) manifests as a sudden liver malfunction, posing a serious, sometimes fatal, threat. This condition is of global medical significance due to its high prevalence and mortality rates. The link between ALI and environmental air contamination remains unstudied. This research delves into the potential connections between brief exposures to particulate matters (PMs) and daily ALI-related outpatient visits in China. Our study encompassed an analysis of 32,261 cases of ALI from January 1, 2013, to December 31, 2019, covering 2,549 days in total. We applied over-dispersed Poisson generalized additive models, incorporating both meteorological factors and day of the week. Notably, a 10 μg/m3 uptick in PM2.5 and PM10 led to 0.41% (95% Cl: 0.06%, 0.75%) and 0.34% (95% Cl: 0.13%, 0.56%) rises in ALI visits at lag 01, PMc exhibited a clear positive trend with ALI across cumulative exposure days, most notably at lag 02, where an increase of 10 μg/m3 in PMc corresponded to a 0.64% (95% CI: 0.23%, 1.05%) rise in outpatient attendances, respectively. The stratified analyses indicated that PMs exhibited stronger effects in females, individuals aged 36–59, and during the warm seasons. This study unveils initial findings suggesting that short-lived PM exposures could indirectly enhance the ALI risk.

Speaking Deads: Sea Turtle Mortality Areas and Fisheries Overlaps Identified Through Backtracking of Stranded Carcasses in the Adriatic Sea

ABSTRACT Interaction with fishing gears represents the main anthropogenic threat at sea for sea turtles worldwide, and identifying the hotspots of turtle bycatch is a priority knowledge gap. Turtle stranding data represent a source of information about mortality areas at sea that are not fully exploited. This study aims to infer turtle mortality areas of turtles stranded along the Italian Adriatic Coast in the period 2019–2021 (1432 records), through backtrack modelling of carcasses. Specifically, the decomposition process of eight loggerhead carcasses was monitored, and the relationship between floating period (FP), turtle size and sea temperature was modelled through a generalized additive model. Oceanographic information was then used to track the routes of floating carcasses back, knowing their size and decomposition stage, and finally estimate the likely area of mortality. A complementary numerical experiment of connectivity between coastal and offshore areas gave indication that areas of potential mortality are relatively close to the coast, particularly in the northern Adriatic. Stranded turtles probably represent just a small fraction (17%–25%) of total at‐sea mortalities in the study area (Italian Adriatic waters), with decomposition rates, season and distance from shore influencing their stranding likelihood. Hence, strandings can inform only about spatio‐temporal variability of coastal mortality hotspots. Inferred areas of turtle mortality were most likely located in the North and Central Adriatic all over the year and overlap with heatmaps of fishing effort, obtained from vessel monitoring system (VMS) and automatic identification system (AIS) data, in the Gulf of Manfredonia and in the North‐West Adriatic in the cold (September–December) and warm (May–August) periods, respectively.

Extreme flash flood susceptibility mapping using a novel PCA-based model stacking approach

This study introduces an efficient methodology for model stacking, incorporating six diverse machine learning and statistical models alongside principal component analysis (PCA). The approach is applied for the flash flood susceptibility mapping within the Karkheh Basin in Iran. The selected models include random forest (RF), boosted regression trees (BRT), support vector machine (SVM), artificial neural networks (ANN), generalized additive model (GAM), and the least absolute shrinkage and selection operator (Lasso), with RF also serving as the meta-model for the stacking. The results revealed significant correlations among the predictions of the individual models, which could potentially impact the meta-model’s efficacy. To address this, PCA was applied to the model predictions to generate de-correlated components as inputs for the meta-model, thereby enhancing prediction accuracy and robustness. Evaluation based on the area under the receiver operating characteristic (AUROC) curve demonstrated that the GAM outperformed all other individual models with the highest accuracy score of 0.924. In contrast, the RF and ANN models had the lowest accuracy, both registering at 0.872. However, the performance disparity across models was minimal. Notably, the PCA-based stacking approach (0.936) surpassed both traditional model stacking (0.912) and the performances of all individual models, advocating for its use in enhancing predictive accuracy. These findings endorse the PCA-stacking method over conventional stacking techniques. Nonetheless, further research across varied applications is warranted to generalize its efficacy.

Constructing an AQHI as a health risk communication tool for Bangkok, Thailand

In this study, we established an air quality health index (AQHI) based on the associations between multiple air pollutants and respiratory and cardiovascular outpatient department (OPD) visits to communicate the health risks from air pollution in Bangkok, Thailand. The associations between various air pollutants, namely, suspended particulate matter (PM) with an aerodynamic diameter smaller than 2.5 µm and 10 µm (PM2.5 and PM10, respectively), sulphur dioxide (SO2), and ozone (O3) and the number of OPD visits for respiratory and cardiovascular diseases in Bangkok from 2016 to 2019 were assessed using generalised additive models with a Poisson link function. Significant associations were established between most cases of cardiovascular and respiratory diseases and these pollutants with a lag time of 0–7 days. The total excess risk was calculated to construct the AQHI, which was then adjusted to an arbitrary scale and banded into four groups based on the calculated score, where 1–3, 4–6, 7–10, and 10+ represented low risk, moderate risk, high risk, and very high risk, respectively. We found that the AQHI captured both high and very high risk levels during the day for most stations. The constructed AQHI also recorded a greater number of high and very high risk days than the currently used AQI but fewer than the WHO-based AQI. Our findings suggest that the AQHI can capture the combined effects of multiple air pollutants, which makes it an effective tool for communicating air pollution-related health risks.

Predicting within-city spatiotemporal variations in daily median outdoor ultrafine particle number concentrations and size in Montreal and Toronto, Canada.

Epidemiological evidence suggests that long-term exposure to outdoor ultrafine particles (UFPs, <0.1 μm) may have important human health impacts. However, less is known about the acute health impacts of these pollutants as few models are available to estimate daily within-city spatiotemporal variations in outdoor UFPs. Several machine learning approaches (i.e., generalized additive models, random forest models, and extreme gradient boosting) were used to predict daily spatiotemporal variations in outdoor UFPs (number concentration and size) across Montreal and Toronto, Canada using a large database of mobile monitoring measurements. Separate models were developed for each city and all models were evaluated using a 10-fold cross-validation procedure. In total, our models were based on measurements from 12,705 road segments in Montreal and 10,929 road segments in Toronto. Daily median outdoor UFP number concentrations varied substantially across both cities with 1st-99th percentiles ranging from 1389 to 181,672 in Montreal and 2472 to 118,544 in Toronto. Outdoor UFP size tended to be smaller in Montreal (mean [SD]: 34 nm [15]) than in Toronto (mean [SD]: 44 nm [25]). Extreme gradient boosting models performed best and explained the majority of spatiotemporal variations in outdoor UFP number concentrations (Montreal, R 2: 0.727; Toronto, R 2: 0.723) and UFP size (Montreal, R 2: 0.823; Toronto, R 2: 0.898) with slopes close to one and intercepts close to zero for relationships between measured and predicted values. These new models will be applied in future epidemiological studies examining the acute health impacts of outdoor UFPs in Canada's two largest cities.

Investigating nearshore spatial and temporal trends in nutrient concentrations along an urban northern shoreline, Lake Ontario

The nearshore zone of the Laurentian Great Lakes is of significant ecological importance, providing critical biogeochemical processes supporting spatially diverse habitats for a variety of species. Extreme plant and algal growth are common due to excessive anthropogenic eutrophication, with exacerbations from dreissenid establishment and lake mixing (e.g., coastal upwelling events and nearshore-offshore water exchanges), resulting in nuisance algal blooms across the nearshore area adjacent to Western Durham, Ontario, Canada. Thus, our main goal was to characterize the trends in essential nutrients (i.e., phosphorus and nitrogen) within the nearshore by applying general additive models on irregular time-series from 2011 to 2022 and to identify plausible contributing factors using exploratory principal component analysis. Increasing trends for total phosphorus were observed in surface (0.5 m below the surface) and benthic (0.5 m above the bottom) waters despite decreases in point source loading to the area. Most notably, the local water pollution control plant (WPCP) outfall did not seem to drive lake phosphorus concentrations across the study area, depicted by an orthogonal relationship within the principal component analysis. While the WPCP is a point source to the nearshore, it does not appear to be the primary driver of temporal lake phosphorus trends. Our results suggest that weather, inertial forcings and lake hydrodynamics in combination with traditional point-sources across Western Durham are contributing to the increasing total phosphorus trends observed as the peak period of wave spectra, wave height, alongshore wind speeds, watershed loadings, and total phosphorus concentrations within Lake Ontario were positively associated.

Evaporating temperature estimation of refrigeration systems based on vibration data-driven soft sensors

The evaluation of the operating conditions of refrigeration compressors once installed in household appliances is challenging due to the need to install pressure transducers, a process which requires system evacuation and refrigerant reintroduction. In addition, changes in the piping modify the characteristics of the original product. This paper proposes a soft-sensing technique based on vibration measurements of the compressor surface to predict the evaporating temperature. Different machine learning (ML) techniques are evaluated as data-driven prediction models, namely multilayer perceptron (MLP) neural networks, least squares boosting, generalized additive model, random forest, extreme learning machine, and random vector functional link neural networks. These techniques were applied to data obtained from a test rig designed to emulate compressor operation in a refrigeration system, with an operating envelope from -30°C to -10°C for the evaporating temperature and from 34°C to 54°C for the condensing temperature. The results showed that, with a single vibration measurement point, it was possible to use an MLP technique to estimate the evaporating temperature with a root mean squared error of 1.74°C in a non-intrusive way. For the other prediction techniques, the errors were a bit higher than for the MLP, but the maximum error value was about 2.5°C in all cases.

A matter of scale: Identifying the best spatial and temporal scale of environmental variables to model the distribution of a small cetacean.

The importance of scale when investigating ecological patterns and processes is recognised across many species. In marine ecosystems, the processes that drive species distribution have a hierarchical structure over multiple nested spatial and temporal scales. Hence, multi-scale approaches should be considered when developing accurate distribution models to identify key habitats, particularly for populations of conservation concern. Here, we propose a modelling procedure to identify the best spatial and temporal scale for each modelled and remotely sensed oceanographic variable to model harbour porpoise (Phocoena phocoena) distribution within the Irish Exclusive Economic Zone. Harbour porpoise sightings were recorded during dedicated line-transect aerial surveys conducted in the summers of 2016, 2021 and 2022. Binary generalised additive models were used to assess the relationships between porpoise presence and oceanographic variables at different spatial (5-40 km) and temporal (daily, monthly and across survey period) scales. Selected variables included sea surface temperature, thermal fronts, chlorophyll-a, sea surface height, mixed layer depth and salinity. A total of 30,514 km was covered on-effort with 216 harbour porpoise sightings recorded. Overall, the best spatial scale corresponded to the coarsest resolution considered in this study (40 km), while porpoise presence showed stronger association with oceanographic variables summarised at a longer temporal scale. Habitat models including covariates at coarse spatial and temporal scales may better reflect the processes driving availability and abundance of resources at these large scales. These findings support the hypothesis that a multi-scale approach should be applied when investigating species distribution. Identifying suitable spatial and temporal scale would improve the functional interpretation of the underlying relationships, particularly when studying how a small marine predator interacts with its environment and responds to climate and ecosystem changes.

A New Field Line Tracer for the Study of Coronal Magnetic Topologies

We present a new code for the tracing of magnetic field lines and calculation of related quantities such as the squashing factor in the solar corona. The Universal Fieldline Tracer (UFiT) is an open-source package that can currently take inputs directly from four well-established coronal models, with additional models planned to be made directly accessible in the future. This package contains tools to make use of large-scale three-dimensional field line maps to calculate volumetric quantities, such as the total volume of the open corona, or the fraction that maps to regions on the solar surface within some distance of a coronal hole boundary, which may be relevant to phenomenological models of solar wind speed such as the Wang–Sheeley–Arge model. Synthetic coronagraphs can also be produced rapidly by this package. We have postprocessed long-term magnetofrictional simulations to demonstrate that the separatrix web occupies a larger fraction of the corona during solar maximum than solar minimum.