Your search

Results 128 resources

  • The key challenge of Unsupervised Domain Adaptation (UDA) for analyzing time series data is to learn domain-invariant representations by capturing complex temporal dependencies. In addition, existing unsupervised domain adaptation methods for time series data are designed to align marginal distribution between source and target domains. However, existing UDA methods (e.g. R-DANN Purushotham et al. (2017), VRADA Purushotham et al. (2017), CoDATS Wilson et al. (2020)) neglect the conditional distribution discrepancy between two domains, leading to misclassification of the target domain. Therefore, to learn domain-invariant representations by capturing the temporal dependencies and to reduce the conditional distribution discrepancy between two domains, a novel Attentive Recurrent Adversarial Domain Adaptation with Top-k time series pseudo-labeling method called ARADA-TK is proposed in this paper. In the experiments, our proposed method was compared with the state-of-the-art UDA methods (R-DANN, VRADA and CoDATS). Experimental results on four benchmark datasets revealed that ARADA-TK achieves superior classification accuracy when it is compared to the competing methods.

  • Stock movement prediction is one of the most challenging problems in time series analysis due to the stochastic nature of financial markets. In recent years, a plethora of statistical methods and machine learning algorithms were proposed for stock movement prediction. Specifically, deep learning models are increasingly applied for the prediction of stock movement. The success of deep learning models relies on the assumption that massive training data are available. However, this assumption is impractical for stock movement prediction. In stock markets, a large number of stocks do not have enough historical data, especially for the companies which underwent initial public offering in recent years. In these situations, the accuracy of deep learning models to predict the stock movement could be affected. To address this problem, in this paper, we propose novel instance-based deep transfer learning models with attention mechanism. In the experiments, we compare our proposed methods with state-of-the-art prediction models. Experimental results on three public datasets reveal that our proposed methods significantly improve the performance of deep learning models when limited training data are available.

  • The extraction of 21 insecticides and 5 metabolites was performed using an optimized and validated QuEChERS protocol that was further used for the quantification (GC–MS/MS) in several seafood matrices (crustaceans, bivalves, and fish-mudskippers). Seven species, acquired from Hong Kong and Macao wet markets (a region so far poorly monitored), were selected based on their commercial importance in the Indo-Pacific region, market abundance, and affordable price. Among them, mussels from Hong Kong, together with mudskippers from Macao, presented the highest insecticide concentrations (median values of 30.33 and 23.90 ng/g WW, respectively). Residual levels of fenobucarb, DDTs, HCHs, and heptachlors were above the established threshold (10 ng/g WW) for human consumption according to the European and Chinese legislations: for example, in fish-mudskippers, DDTs, fenobucarb, and heptachlors (5-, 20- and tenfold, respectively), and in bivalves, HCHs (fourfold) had higher levels than the threshold. Risk assessment revealed potential human health effects (e.g., neurotoxicity), especially through fish and bivalve consumption (non-carcinogenic risk; ΣHQLT > 1), and a potential concern of lifetime cancer risk development through the consumption of fish, bivalves, and crustaceans collected from these markets (carcinogenic risk; ΣTCR > 10–4). Since these results indicate polluted regions, where the seafood is collected/produced, a strict monitoring framework should be implemented in those areas to improve food quality and safety of seafood products.

  • Mangroves are a unique group of plants growing along tropical and sub-tropical coastlines, with the ability to remove several types of contaminants such as heavy metals and other persistent organic compounds in coastal waters. However, little attention has been given to the possible role of mangroves in the removal of organochlorinated pesticides (OCPs) from the environment. Used worldwide, these pesticides were banned in the late 80s, withal they can still be quantified in aquatic environments due to their high stability. Moreover, as persistent and lipophilic compounds, OCPs are known for their tendency to bioaccumulate and biomagnify through the food chain, affecting local ecosystems, and potentially human health. This work aimed to investigate the potential benefits of mangrove ecosystems as OCP phytoremediators. For this purpose, a total of seventy-three articles from non-mangrove and mangrove areas around the world were gathered, integrated and re-analysed as a whole. These data include information from four different matrices (water, sediment, benthic fauna and mangrove plants). A common trend of less pesticide contamination in mangrove areas was observed for all the selected matrices. As a complement, average concentrations were discussed considering International Directives, such as the European legislation 2013/39/EU for water policy and the Dutch List together with the International Sediment Quality Guideline, for sediments. Additionally, theoretical risk assessments were also included. Since information regarding OCPs in mangroves ecosystem is very scarce compared to non-mangrove areas, this review provides valuable insights regarding these environments, and the importance of preserving them as a relevant remediation unit.

  • Uptake and depuration kinetics of 4,4′-dichlorobenzophenone (main metabolite of dicofol) in the edible clam Meretrix meretrix were evaluated through a mesocosm experiment. M. meretrix was exposed to different dicofol concentrations (environmental concentration, D1 = 50 ng/L; supra-environmental concentration, D2 = 500 ng/L) for 15 days, followed by the same depuration period. To accomplish this goal, an analytical method was successfully optimized for 4,4′-DCBP using QuEChERS as extraction method with a range of concentrations 0.3–76.8 ng/g ww quantified by gas chromatography coupled to tandem mass spectrometry. Our results demonstrated different kinetics of accumulation depending on the two dicofol treatments. For D1, the uptake kinetic was best fitted using a plateau followed by one phase association kinetic model, while for D2 a one phase association kinetic model suited better. Similar bioconcentration factors were obtained for both concentrations but only animals exposed to D2, showed 4,4′-DCBP levels above the limits of quantification after 24 h exposure. These animals also showed lower uptake rate (ku) than organisms exposed to D1. During the depuration period, only organisms exposed to D1 successfully depurated after 24 h. On the other hand, although animals exposed to D2 presented higher elimination factor, they did not reach the original levels after depuration. Moreover, values detected in these clams were higher than the Maximum Residue Level (10 ng/g) established by the European legislation. This indicates that longer periods of depuration time than the ones used in this study, may be needed in order to reach safe levels for human consumption. This work also demonstrated that studies on metabolite kinetics during uptake/depuration experiments, could be a new alternative to understand the impact and metabolism of pesticides in the marine environment.

  • Genotoxic effects of dicofol on the edible clam Meretrix meretrix were investigated through a mesocosm experiment. Individuals of M. meretrix, were exposed to environmental concentration (D1 = 50 ng/L) and supra-environmental concentration (D2 = 500 ng/L) of dicofol for 15 days, followed by the same depuration period. DNA damage (i.e., strand breaks and alkali-labile sites) was evaluated at day 1, 7 and 15, during uptake and depuration, using Comet assay (alkaline version) and nuclear abnormalities (NAs) as genotoxicity biomarkers. The protective effects of dicofol against DNA damage induced by ex vivo hydrogen peroxide (H2O2) exposure were also assessed. Comet assay results revealed no significant DNA damages under dicofol exposure, indicating 1) apparent lack of genotoxicity of dicofol to the tested conditions and/or 2) resistance of the animals due to optimal adaptation to stress conditions. Moreover, ex vivo H2O2 exposure showed an increase in the DNA damage in all the treatments without significant differences between them. However, considering only the DNA damage induced by H2O2 during uptake phase, D1 animals had significantly lower DNA damage than those from other treatments, revealing higher protection against a second stressor. NAs data showed a decrease in the % of cells with polymorphic, kidney shape, notched or lobbed nucleus, along the experiment. The combination of these results supports the idea that the clams used in the experiment were probably collected from a stressful environment (in this case Pearl River Delta region) which could have triggered some degree of adaptation to those environmental conditions, explaining the lack of DNA damages and highlighting the importance of organisms’ origin and the conditions that they were exposed during their lives.

  • Anthropogenic noise can be hazardous for the auditory system and wellbeing of animals, including humans. However, very limited information is known on how this global environmental pollutant affects auditory function and inner ear sensory receptors in early ontogeny. The zebrafish (Danio rerio) is a valuable model in hearing research, including investigations of developmental processes of the vertebrate inner ear. We tested the effects of chronic exposure to white noise in larval zebrafish on inner ear saccular sensitivity and morphology at 3 and 5 days post-fertilization (dpf), as well as on auditory-evoked swimming responses using the prepulse inhibition (PPI) paradigm at 5 dpf. Noise-exposed larvae showed a significant increase in microphonic potential thresholds at low frequencies, 100 and 200 Hz, while the PPI revealed a hypersensitization effect and a similar threshold shift at 200 Hz. Auditory sensitivity changes were accompanied by a decrease in saccular hair cell number and epithelium area. In aggregate, the results reveal noise-induced effects on inner ear structure–function in a larval fish paralleled by a decrease in auditory-evoked sensorimotor responses. More broadly, this study highlights the importance of investigating the impact of environmental noise on early development of sensory and behavioural responsiveness to acoustic stimuli.

  • Zebrafish is a well-established model organism in hearing research. Although the acoustic environment is known to shape the structure and sensitivity of auditory systems, there is no information on the natural soundscape of this species. Moreover, zebrafish are typically reared in large-scale housing systems (HS), although their acoustic properties and potential effects on hearing remain unknown. We characterized the soundscape of both zebrafish natural habitats and laboratory captive conditions, and discussed possible impact on auditory sensitivity. Sound recordings were conducted in five distinct zebrafish habitats (Southwest India), from quieter stagnant environments with diverse biological/abiotic sounds to louder watercourses characterized by current and moving substrate sounds. Sound pressure level (SPL) varied between 98 and 126 dB re 1 μPa. Sound spectra presented most energy below 3000 Hz and quieter noise windows were found in the noisiest habitats matching the species best hearing range. Contrastingly, recordings from three zebrafish HS revealed higher SPL (122-143 dB) and most energy below 1000 Hz with more spectral peaks, which might cause significant auditory masking. This study establishes an important ground for future research on the adaptation of zebrafish auditory system to the natural soundscapes, and highlights the importance of controlling noise conditions in captivity.

  • Noise pollution is increasingly present in aquatic ecosystems, causing detrimental effects on growth, physiology and behaviour of organisms. However, limited information exists on how this stressor affects animals in early ontogeny, a critical period for development and establishment of phenotypic traits. We tested the effects of chronic noise exposure to increasing levels (130 and 150 dB re 1 μPa, continuous white noise) and different temporal regimes on larval zebrafish (Danio rerio), an important vertebrate model in ecotoxicology. The acoustic treatments did not affect general development or hatching but higher noise levels led to increased mortality. The cardiac rate, yolk sac consumption and cortisol levels increased significantly with increasing noise level at both 3 and 5 dpf (days post fertilization). Variation in noise temporal patterns (different random noise periods to simulate shipping activity) suggested that the time regime is more important than the total duration of noise exposure to down-regulate physiological stress. Moreover, 5 dpf larvae exposed to 150 dB continuous noise displayed increased dark avoidance in anxiety-related dark/light preference test and impaired spontaneous alternation behaviour. We provide first evidence of noise-induced physiological stress and behavioural disturbance in larval zebrafish, showing that both noise amplitude and timing negatively impact key developmental endpoints in early ontogeny.

  • The levels of air pollution in Macao often exceeded the levels recommended by WHO. In order for the population to take precautionary measures and avoid further health risks under high pollutant exposure, it is important to develop a reliable air quality forecast. Statistical models based on multiple regression (MR) analysis were developed successfully for Macao to predict the next day concentrations of PM10, PM2.5, and NO2. All the developed models were statistically significantly valid with a 95% confidence level with high coefficients of determination (from 0.89 to 0.92) for all pollutants. The models utilized meteorological and air quality variables based on five years of historical data, from 2013 to 2017. The data from 2013 to 2016 were used to develop the statistical models and data from 2017 were used for validation purposes. A wide range of meteorological and air quality variables were identified, and only some were selected as significant dependent variables. Meteorological variables were selected from an extensive list of variables, including geopotential height, relative humidity, atmospheric stability, and air temperature at different vertical levels. Air quality variables translate the resilience of the recent past concentrations of each pollutant and usually are maximum and/or the average of latest 24-hour levels. The models were applied in forecasting the next day average daily concentrations for PM10, PM2.5, and NO2 for the air quality monitoring stations. The results are expected to be an operational air quality forecast for Macao.

  • The levels of air pollution in Macao often exceeded the levels recommended by WHO. In order for the population to take precautionary measures and avoid further health risks under high pollutant exposure, it is important to develop a reliable air quality forecast. Statistical models based on linear multiple regression (MR) and classification and regression trees (CART) analysis were developed successfully, for Macao, to predict the next day concentrations of NO2, PM10, PM2.5, and O3. All the developed models were statistically significantly valid with a 95% confidence level with high coefficients of determination (from 0.78 to 0.93) for all pollutants. The models utilized meteorological and air quality variables based on 5 years of historical data, from 2013 to 2017. Data from 2013 to 2016 were used to develop the statistical models and data from 2017 was used for validation purposes. A wide range of meteorological and air quality variables was identified, and only some were selected as significant independent variables. Meteorological variables were selected from an extensive list of variables, including geopotential height, relative humidity, atmospheric stability, and air temperature at different vertical levels. Air quality variables translate the resilience of the recent past concentrations of each pollutant and usually are maximum and/or the average of latest 24-h levels. The models were applied in forecasting the next day average daily concentrations for NO2 and PM and maximum hourly O3 levels for five air quality monitoring stations. The results are expected to be an operational air quality forecast for Macao.

  • Statistical methods such as multiple linear regression (MLR) and classification and regression tree (CART) analysis were used to build prediction models for the levels of pollutant concentrations in Macao using meteorological and air quality historical data to three periods: (i) from 2013 to 2016, (ii) from 2015 to 2018, and (iii) from 2013 to 2018. The variables retained by the models were identical for nitrogen dioxide (NO2), particulate matter (PM10), PM2.5, but not for ozone (O3) Air pollution data from 2019 was used for validation purposes. The model for the 2013 to 2018 period was the one that performed best in prediction of the next-day concentrations levels in 2019, with high coefficient of determination (R2), between predicted and observed daily average concentrations (between 0.78 and 0.89 for all pollutants), and low root mean square error (RMSE), mean absolute error (MAE), and biases (BIAS). To understand if the prediction model was robust to extreme variations in pollutants concentration, a test was performed under the circumstances of a high pollution episode for PM2.5 and O3 during 2019, and the low pollution episode during the period of implementation of the preventive measures for COVID-19 pandemic. Regarding the high pollution episode, the period of the Chinese National Holiday of 2019 was selected, in which high concentration levels were identified for PM2.5 and O3, with peaks of daily concentration exceeding 55 μg/m3 and 400 μg/m3, respectively. The 2013 to 2018 model successfully predicted this high pollution episode with high coefficients of determination (of 0.92 for PM2.5 and 0.82 for O3). The low pollution episode for PM2.5 and O3 was identified during the 2020 COVID-19 pandemic period, with a low record of daily concentration for PM2.5 levels at 2 μg/m3 and O3 levels at 50 μg/m3, respectively. The 2013 to 2018 model successfully predicted the low pollution episode for PM2.5 and O3 with a high coefficient of determination (0.86 and 0.84, respectively). Overall, the results demonstrate that the statistical forecast model is robust and able to correctly reproduce extreme air pollution events of both high and low concentration levels.

  • Air pollution is a major concern issue on Macao since the concentration levels of several of the most common pollutants are frequently above the internationally recommended values. The low air quality episodes impacts on human health paired with highly populated urban areas are important motivations to develop forecast methodologies in order to anticipate pollution episodes, allowing establishing warnings to the local community to take precautionary measures and avoid outdoor activities during this period. Using statistical methods (multiple linear regression (MLR) and classification and regression tree (CART) analysis) we were able to develop forecasting models for the main pollutants (NO2, PM2.5, and O3) enabling us to know the next day concentrations with a good skill, translated by high coefficients of determination (0.82–0.90) on a 95% confidence level. The model development was based on six years of historical data, 2013 to 2018, consisting of surface and upper-air meteorological observations and surface air quality observations. The year of 2019 was used for model validation. From an initially large group of meteorological and air quality variables only a few were identified as significant dependent variables in the model. The selected meteorological variables included geopotential height, relative humidity and air temperature at different altitude levels and atmospheric stability characterization parameters. The air quality predictors used included recent past hourly levels of mean concentrations for NO2 and PM2.5 and maximum concentrations for O3. The application of the obtained models provides the expected daily mean concentrations for NO2 and PM2.5 and maximum hourly concentrations O3 for the next day in Taipa Ambient air quality monitoring stations. The described methodology is now operational, in Macao, since 2020.

  • Employees work long hours in an environment where the ambient air quality is poor, directly affecting their work efficiency. The concentration of particulate matters (PM) produced by the interior renovation of shopping malls has not received particular attention in Macao. Therefore, this study will investigate the indoor air quality (IAQ), in particular of PM2.5, in large-scale shopping mall renovation projects. This study collected on-site PM data with low-cost portable monitoring equipment placed temporarily at specific locations to examine whether the current control measures are appropriate and propose some improvements. Prior to this study, there were no measures being implemented, and on-site monitoring to assess the levels of PM2.5 concentrations was non-existent. The results show the highest level of PM2.5 recorded in this study was 559.00 μg/m3. Moreover, this study may provide a reference for decision-makers, management, construction teams, design consultant teams, and renovation teams of large-scale projects. In addition, the monitoring of IAQ can ensure a comfortable environment for employees and customers. This study concluded that the levels of PM2.5 concentration have no correlation with the number of on-site workers, but rather were largely influenced by the processes being performed on-site.

  • Air pollution in Macau has become a serious problem following the Pearl River Delta’s (PRD) rapid industrialization that began in the 1990s. With this in mind, Macau needs an air quality forecast system that accurately predicts pollutant concentration during the occurrence of pollution episodes to warn the public ahead of time. Five different state-of-the-art machine learning (ML) algorithms were applied to create predictive models to forecast PM2.5, PM10, and CO concentrations for the next 24 and 48 h, which included artificial neural networks (ANN), random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM), and multiple linear regression (MLR), to determine the best ML algorithms for the respective pollutants and time scale. The diurnal measurements of air quality data in Macau from 2016 to 2021 were obtained for this work. The 2020 and 2021 datasets were used for model testing, while the four-year data before 2020 and 2021 were used to build and train the ML models. Results show that the ANN, RF, XGBoost, SVM, and MLR models were able to provide good performance in building up a 24-h forecast with a higher coefficient of determination (R2) and lower root mean square error (RMSE), mean absolute error (MAE), and biases (BIAS). Meanwhile, all the ML models in the 48-h forecasting performance were satisfactory enough to be accepted as a two-day continuous forecast even if the R2 value was lower than the 24-h forecast. The 48-h forecasting model could be further improved by proper feature selection based on the 24-h dataset, using the Shapley Additive Explanations (SHAP) value test and the adjusted R2 value of the 48-h forecasting model. In conclusion, the above five ML algorithms were able to successfully forecast the 24 and 48 h of pollutant concentration in Macau, with the RF and SVM models performing the best in the prediction of PM2.5 and PM10, and CO in both 24 and 48-h forecasts.

  • Despite the levels of air pollution in Macao continuing to improve over recent years, there are still days with high-pollution episodes that cause great health concerns to the local community. Therefore, it is very important to accurately forecast air quality in Macao. Machine learning methods such as random forest (RF), gradient boosting (GB), support vector regression (SVR), and multiple linear regression (MLR) were applied to predict the levels of particulate matter (PM10 and PM2.5) concentrations in Macao. The forecast models were built and trained using the meteorological and air quality data from 2013 to 2018, and the air quality data from 2019 to 2021 were used for validation. Our results show that there is no significant difference between the performance of the four methods in predicting the air quality data for 2019 (before the COVID-19 pandemic) and 2021 (the new normal period). However, RF performed significantly better than the other methods for 2020 (amid the pandemic) with a higher coefficient of determination (R2) and lower RMSE, MAE, and BIAS. The reduced performance of the statistical MLR and other ML models was presumably due to the unprecedented low levels of PM10 and PM2.5 concentrations in 2020. Therefore, this study suggests that RF is the most reliable prediction method for pollutant concentrations, especially in the event of drastic air quality changes due to unexpected circumstances, such as a lockdown caused by a widespread infectious disease.

  • Hydrothermal activity on mid-ocean ridges is an important mechanism for the delivery of Zn from the mantle to the surface environment. Zinc isotopic fractionation during hydrothermal activity is mainly controlled by the precipitation of Zn-bearing sulfide minerals, in which isotopically light Zn is preferentially retained in solid phases rather than in solution during mineral precipitation. Thus, seafloor hydrothermal activity is expected to supply isotopically heavy Zn to the ocean. Here, we studied sulfide-rich samples from the Duanqiao-1 hydrothermal field, located on the Southwest Indian Ridge. We report that, at the hand-specimen scale, late-stage conduit sulfide material has lower δ66Zn values (−0.05 ± 0.15 ‰; n = 19) than early-stage material (+0.13 ± 0.15 ‰; n = 10). These lower values correlate with enrichments in Pb, As, Cd, and Ag, and elevated δ34S values. We attribute the low δ66Zn values to the remobilization of earlier sub-seafloor Zn-rich mineralization. Based on endmember mass balance calculations, and an assumption of a fractionation factor (αZnS-Sol.) of about 0.9997 between sphalerite and its parent solution, the remobilized Zn was found consist of about 1/3 to 2/3 of the total Zn in the fluid that formed the conduit samples. Our study suggests that late-stage subsurface hydrothermal remobilization may release isotopically-light Zn to the ocean, and that this process may be common along mid-ocean ridges, thus increasing the size of the previously identified isotopically light Zn sink in the ocean.

Last update from database: 3/28/24, 11:45 AM (UTC)