Local Meteorological Variables Research Articles

High dimensional variable selection through group Lasso for multiple function‐on‐function linear regression: A case study in PM10 monitoring

SummaryAnalyzing the effect of chemical and local meteorological variables over the behaviour in concentrations in the Abruzzo region (Italy), with the objective of forecasting and controlling air quality, motivates the current work. Given that the available data are curves that represent the day‐to‐day variations, a multiple function‐on‐function linear regression (MFFLR) model is considered. By assuming the Karhunen‐Loève expansion, MFFLR model can be reduced to a classical linear regression model for each principal component of the functional response in terms of all principal components (PCs) of the functional predictors. In this sense, a regularization approach for functional principal component regression based on the merge of functional data analysis with group Lasso is proposed. This novel methodology allows to estimate the model and, simultaneously, select those relevant functional predictors with the functional response, where each functional independent variable is represented by a group of input variables derived by the PCs.

Time-delayed causal network analysis of meteorological variables and air pollutants in Baguio city

Air pollution contributes significantly to climate change and has recently become a significant concern. Previous studies on the interplay between air quality and the environment, such as meteorological variables, have hinted at their relevance in predicting air quality. While air quality prediction has been extensively explored, only a few studies have investigated the causal interactions, and no study has looked into identifying time-delayed causality between individual meteorological factors and air pollutants. In this work, we have deployed sensors in Baguio City, renowned as the “Summer Capital of the Philippines”, and collected data from September 2022 to April 2023. Convergent Cross Mapping (CCM) is then used to infer the time-delayed causation network for PM1, PM2.5, PM10, VOC, and local meteorological variables. Baguio City faces environmental challenges from urbanization and influx of tourists. Here, the top meteorological variables are identified and characterized according to the strength and delay of influences on air pollutants. Wind direction emerges as a leading influencer driving air quality instantly, while pollutants reciprocate with delayed feedback. Temperature holds sway after a day, with pollutants impacting it after eight days on average. Furthermore, links from wind speed to PM concentrations are observed as mediators in causal pathways. Here, the maximum lag delay is only nine days. This work highlights the intricate interdependence of local meteorological factors and air quality, pioneering the identification of time delays that optimize causal influences between these variables. The results of this work offer valuable insights into environmental management and the formulation of mitigation measures.

The influence of meteorological parameters on PM2.5 and PM10 values in Ümraniye and Silivri districts of İstanbul

In this study, spatiotemporal relationship between PM2.5 and PM10 concentrations and meteorological parameters were investigated for Silivri and Umraniye districts in Istanbul for 2014-2020. For this purpose, hourly PM2.5 and PM10 concentrations values of two air quality monitoring stations and meteorological data (wind speed, wind direction, relative humidity, total precipitation, minimum and maximum temperature) were examined. In all seasons, while PM concentrations were lowest at 06:00 local time (LT), PM2.5 and PM10 have peak values around 09:00 and 19:00 LT both in Silivri and Umraniye mainly due to anthropogenic activities such as vehicle exhaust emissions. In daily perspective, highest PM values were observed on Sundays in winter at Silivri. On the other side, peak PM values are shown on Fridays at Umraniye. It was found that local emission sources during low wind speeds cause the highest PM2.5 concentrations during winter months and southerly winds exceeding 8 m/s increase the PM10 levels at Silivri and Umraniye. The statistical analysis showed that PM2.5 and PM10 concentrations at Silivri were negatively correlated with wind speed with correlation coefficients of -0.56 (winter), -0.47 (autumn), respectively. Wind speed is negatively associated with PM2.5 (r=-0.48) and PM10 (r=-0.38) in winter season at Umraniye. In addition to this, relative humidity showed negative relationship with PM10 (r=-0.43) in spring at Silivri, while a positive correlation was found between PM10 (r=0.40) and PM2.5 (r=0.38) measured in the summer season and the maximum temperature. In addition to the anthropogenic factors (e.g. urbanization, transportation, and industrialization) that decrease air quality of Istanbul, local meteorological variables and atmospheric transport of pollution are observed to be the other factors that contribute to air pollution.

Hymenaea stignocarpa Mart. ex Hayne growth–climate relationships are regulated by soil water saturation in Cerrado-Pantanal ecotone

The expected intensification of the dry season, and concentration of rainfall during the wet season, can disrupt tree growth and regional biodiversity in the Cerrado-Pantanal transition zone. Thus, this study aims to assess the climatic responses of Hymenaea stigonocarpa tree growth, a common tree species in this region. Incremental cores were collected at breast height (ca. 1.3 m) from 67 trees to construct a dendrochronological series to correlate annual growth with local meteorological variables and large-scale atmospheric circulation indices from the Pacific and Atlantic Oceans. Significant positive correlations were observed between tree growth and the Pacific Ocean indices. Maximum and average air temperatures of the previous dry season negatively influenced growth, while precipitation at the beginning of the growth season (November) positively influenced growth. Tree growth was not correlated with temperature or rainfall during the wet season. However, at the end of the wet season (February), tree growth was negatively correlated with air temperature and positively correlated with rainfall, but the relationship shifted in the next month (March) suggesting that soil water saturation reduced growth. Our results indicate that dendrochronological studies of H. stignocarpa are useful for assessing the environmental change on the Cerrado-Pantanal ecotone. Furthermore, variations in water availability and temperature associated with changes in large-scale oceanic circulation and local meteorological conditions will impact the growth dynamics of this important Cerrado tree species.

Comparison of data-driven methods for linking extreme precipitation events to local and large-scale meteorological variables

Extreme precipitation events can lead to severe negative consequences for society, the economy, and the environment. It is therefore crucial to understand when such events occur. In the literature, there are a vast number of methods for analyzing their connection to meteorological drivers. However, there has been recent interest in using machine learning methods instead of classic statistical models. While a few studies in climate research have compared the performance of these two approaches, their conclusions are inconsistent. To determine whether an extreme event occurred locally, we trained models using logistic regression and three commonly used supervised machine learning algorithms tailored for discrete outcomes: random forests, neural networks, and support vector machines. We used five explanatory variables (geopotential height at 500 hPa, convective available potential energy, total column water, sea surface temperature, and air surface temperature) from ERA5, and local data from the Danish Meteorological Institute. During the variable selection process, we found that convective available potential energy has the strongest relationship with extreme events. Our results showed that logistic regression performs similarly to more complex machine learning algorithms regarding discrimination as measured by the area under the receiver operating characteristic curve (ROC AUC) and other performance metrics specialized for unbalanced datasets. Specifically, the ROC AUC for logistic regression was 0.86, while the best-performing machine learning algorithm achieved a ROC AUC of 0.87. This study emphasizes the value of comparing machine learning and classical regression modeling, especially when employing a limited set of well-established explanatory variables.

Local mechanisms for global daytime, nighttime, and compound heatwaves

Heatwaves impose serious impacts on ecosystems, human health, agriculture, and energy consumption. Previous studies have classified heatwaves into independent daytime, independent nighttime, and compound daytime-nighttime types, and examined the long-term changes in the three types. However, the underlying mechanisms associated with the variations in different heatwave types remain poorly understood. Here we present the first investigation of the local physical processes associated with the daytime, nighttime, and compound heatwaves over the global land during 1979–2020. The results show that three heatwave types occur frequently and increasingly in most regions worldwide. Nighttime and compound heatwaves exhibit stronger increases in both frequency (the yearly number of the events) and fraction (the ratio of the yearly number of one heatwave type to the total yearly number of all types) than daytime heatwaves. Composite diagnostic analyses of local meteorological variables suggest that daytime heatwaves are associated with increased solar radiation under dry conditions and reduced cloud cover and humidity under a clear sky. In contrast, nighttime heatwaves are typically accompanied by moist conditions with increases in cloud fraction, humidity, and longwave radiation at night. These synoptic conditions for daytime and nighttime heatwaves are combined to contribute to compound heatwaves. Local divergences and moisture fluxes responsible for different heatwaves are further revealed. Positive moisture divergence anomalies are seen in most land areas for daytime and compound heatwaves, while they mainly appear in low latitudes for nighttime heatwaves. Our research provides a comprehensive understanding of the local mechanisms of different heatwave types, informing future risks and impact assessments.

Hybrid Forecasting of Wind for Air Pollution Dispersion over Complex Terrain

In case of an unplanned emission event from a nuclear power plant, the local population can be protected more efficiently when valid atmospheric dispersion model results are available. Atmospheric dispersion models use local meteorological variables as inputs. When atmospheric dispersion in the future is being predicted, a forecast of the local meteorological variables is needed. The most important variable in atmospheric dispersion modelling is wind, and accurately predicting ground level winds presents a challenge to numerical weather prediction models. We therefore develop hybrid models for forecasting local ground level wind at a single location where the terrain is complex and the average wind is weak and fluctuating. Wind speed and direction are modelled as west-east and south-north wind components. Each model is composed of a numerical weather prediction model and a Gaussian process statistical model that uses numerical weather predictions as some of its inputs and is trained on historical data to predict the output component. The most advanced Gaussian process models studied are of Gaussian process nonlinear autoregressive model with exogenous input (GPNARX) type. In addition to numerical weather predictions, they also use local meteorological variables, including the output variable, as their inputs. Numerical weather prediction results based on large scale information and fundamental knowledge of the system are thus supplemented by local measurements that better reflect the effects of local topography and land use. The models are tested by prediction and simulation. The wind components predicted by more advanced models are more accurate than raw or post-processed numerical weather prediction results. As an example, a model predicting 2D wind vector 1.5 h in advance achieves a NRMSE of 0.214 if it uses all the immediately available information. This is better than both the persistence model with NRMSE of 0.188 and post-processed NWP with NRMSE of 0.164. This demonstrates that hybrid modelling provides the best weather information for short-term and medium-term atmospheric dispersion forecasting. While the method is motivated by nuclear emission sources, it could also be applied to other pollution.

A Stacking Ensemble Learning Model for Monthly Rainfall Prediction in the Taihu Basin, China

The prediction of monthly rainfall is greatly beneficial for water resources management and flood control projects. Machine learning (ML) techniques, as an increasingly popular approach, have been applied in diverse climatic regions, showing their respective superiority. On top of that, the ensemble learning model that synthesizes the advantages of different ML models deserves more attention. In this study, an ensemble learning model based on stacking approach was proposed. Four prevalent ML models, namely k-nearest neighbors (KNN), extreme gradient boosting (XGB), support vector regression (SVR), and artificial neural networks (ANN) are taken as base models. To combine the outputs from the base models, the weighting algorithm is used as second-layer learner to generate predictions. Large-scale climate indices, large-scale atmospheric variables, and local meteorological variables were used as predictors. R2, RMSE and MAE, were used as evaluation metrics. The results show that the performance of base models varied among the nine stations in the Taihu Basin, while the stacking approach generally performed better than the four base models. The stacking model showed better performance in spring and winter than in summer and autumn. During wet months, the accuracy of model prediction varied more significantly. On the whole, based on performance evaluation measures, it is concluded that the proposed stacking ensemble multi-ML model can provide a flexible and reasonable prediction framework applicable to other regions.

Evaluation of the Influence between Local Meteorology and Air Quality in Beijing Using Generalized Additive Models

Previous studies have confirmed the inextricable connection between meteorological factors and air pollutants. This study presents the complex nonlinear relationship between meteorological variables and four major air pollutants under high-concentration air pollution in Beijing. The generalized additive model combined with marginal effects is used for quantitative analysis. After controlling the confounding factors such as long-term trends, seasonality and spatio-temporal deviation, the final fitting results exhibit that temperature, relative humidity and visibility are the most significant meteorological variables associating with PM2.5 concentration, and the marginal effect reaches 80%, −23% and 270%, respectively. Temperature and relative humidity are the most significant variables for SO2, and the marginal effect reaches 15% and 7%. The most significant variables for O3 are temperature and solar radiation, with marginal effect of up to 70% and 8%. Atmospheric pressure and temperature results in a positive effect on CO, and the marginal effect can reach 18% and 80%. All these indicate that local meteorological variables are a significant driving factor for air quality in Beijing. Other variables, such as wind speed, visibility, and precipitation, display some influence on air pollutants, but have less explanatory power in the model. Overall, our study provides a better understanding of the relationship between local meteorological variables and air quality, as well as an insight into how climate change affects air quality.

Spatial distribution of meteorological factors controlling stable isotopes in precipitation in Northern Chile

A knowledge of the evolution of isotopic composition of air masses humidity and precipitation in the Western Cordilleras of the Central Andes is still incomplete. This study contributes to a better understanding of the factors that control the δ18O and δ2H contents in precipitation water in the north of Chile, above 2000 m a.s.l.. This paper deals with: (1) the relevant effects and processes that control the spatial (longitude, latitude and altitude) distribution of stable isotope contents of precipitation events in northern Chile, (2) the influence of local meteorological variables: temperature, precipitation and relative humidity on the δ18O and δ2H of precipitation, and (3) the estimation of these meteorological and isotopic variables at specific sites. To achieve these objectives, the relationships between geospatial and meteorological values are identified and analysed, followed by the estimation with empirical models. These estimation models (linear and non-linear) are obtained after examining, validating and calibrating techniques to find the best fit. This results in models for temperature, relative humidity and precipitation for each month of the year. In the same way, three isotopic models are derived from the spatial and meteorological variables (summer, winter and annual). Temperature has been shown to be controlled to a greater extent by altitude and latitude, while relative humidity is by latitude and precipitation in summer is by altitude and latitude. Monthly meteorological variables have been estimated throughout the study area. Precipitation δ18O and δ2H are controlled mainly by temperature and altitude and to a lesser extent by latitude, longitude and precipitation. In the same way, three isotopic models are derived from the spatial and meteorological variables: summer, winter and annual. This opens a new perspective of precipitation and its isotopic contents, but also allows the calculation of runoff and aquifer recharge and the path for linking future precipitation and aquifer recharge through their isotopic composition.

Statistical Modeling for PM10, PM2.5 and PM1 at Gangneung Affected by Local Meteorological Variables and PM10 and PM2.5 at Beijing for Non- and Dust Periods

Multiple statistical prediction modeling of PM10, PM2.5 and PM1 at Gangneung city, Korea, was performed in association with local meteorological parameters (air temperature, wind speed and relative humidity) and PM10 and PM2.5 concentrations of an upwind site in Beijing, China, in the transport route of Chinese yellow dusts which originated from the Gobi Desert and passed through Beijing to the city from 18 March to 27 March 2015. Before and after the dust periods, the PM10, PM2.5 and PM1 concentrations showed as being very high at 09:00 LST (the morning rush hour) by the increasing emitted pollutants from vehicles and flying dust from the road and their maxima occurred at 20:00 to 22:00 LST (the evening departure time) from the additional pollutants from resident heating boilers. During the dust period, these peak trends were not found due to the persistent accumulation of dust in the city from the Gobi Desert through Beijing, China, as shown in real-time COMS-AI satellite images. Multiple correlation coefficients among PM10, PM2.5 and PM1 at Gangneung were in the range of 0.916 to 0.998. Multiple statistical models were devised to predict each PM concentration, and the significant levels through multi-regression analyses were p < 0.001, showing all the coefficients to be significant. The observed and calculated PM concentrations were compared, and new linear regression models were sequentially suggested to reproduce the original observed PM values with improved correlation coefficients, to some extent.

Relationship between precipitation isotopic compositions and synoptic atmospheric circulation patterns in the lower reach of the Yangtze River

In this study, the stable isotopic compositions (δ18O, δ2H and d-excess) in daily precipitation from October 2017 to May 2020 in the lower reach of the Yangtze River (LYR) were examined to understand the relationship between stable isotopic compositions in precipitation and synoptic atmospheric circulation patterns (SACPs). Atmospheric circulations over the LYR were classified into seven patterns according to the Kirchhofer method with three, one and three SACPs in the East Asian winter monsoon period (EAWM), the monsoon transition period and the East Asian summer monsoon period (EASM), respectively. Precipitation events between most SACPs are distinguishable for either δ18O or d-excess, except for two SACPs in the EAWM, and two SACPs in the EASM. Local meteorological variables are primary factors determining the variation of isotopic compositions in precipitation for three SACPs in the EAWM. Though the precipitation-producing air masses come from different seas under two different SACPs in the EASM, i.e., the South China Sea for one and the western Pacific for the other, neither δ18O nor d-excess is distinguishable. Nevertheless, the relationships between isotopic composition and meteorological variables appear to be distinguishable between these two SACPs. The SACP dominant in the monsoon transition tends to have higher δ18O (mean value) and δ2H (mean value) in precipitation than all other SACPs. The d-excess in precipitation of this SACP is more likely altered by subcloud evaporation, and the retrieval of moisture source d-excess information should be treated with cautions. Similarly, d-excess in Meiyu rain, classified in one of the SACPs in the EASM period, is very likely altered by moisture recycling along the atmospheric moisture transport pathway, evident from the significant negative relationship between d-excess in precipitation and mean specific humidity along the backward trajectories. Thus d-exess of Meiyu rain is very likely positively deviated from its oceanic moisture source. Overall, the findings can provide valuable insights for understanding the relationship between stable isotopes in precipitation and SACPs in the East Asian monsoon region.

A comparative study of extensive machine learning models for predicting long‐term monthly rainfall with an ensemble of climatic and meteorological predictors

AbstractRainfall prediction is of vital importance in water resources management. Accurate long‐term rainfall prediction remains an open and challenging problem. Machine learning techniques, as an increasingly popular approach, provide an attractive alternative to traditional methods. The main objective of this study was to improve the prediction accuracy of machine learning‐based methods for monthly rainfall, and to improve the understanding of the role of large‐scale climatic variables and local meteorological variables in rainfall prediction. One regression model autoregressive integrated moving average model (ARIMA) and five state‐of‐the‐art machine learning algorithms, including artificial neural networks, support vector machine, random forest (RF), gradient boosting regression, and dual‐stage attention‐based recurrent neural network, were implemented for monthly rainfall prediction over 25 stations in the East China region. The results showed that the ML models outperformed ARIMA model, and RF relatively outperformed other models. Local meteorological variables, humidity, and sunshine duration, were the most important predictors in improving prediction accuracy. 4‐month lagged Western North Pacific Monsoon had higher importance than other large‐scale climatic variables. The overall output of rainfall prediction was scalable and could be readily generalized to other regions.

Applying ensemble neural networks to analyze industrial maintenance: Influence of Saharan dust transport on gas turbine axial compressor fouling

The planning of industrial maintenance associated with the production of electricity is vital, as it yields a current and future snapshot of an industrial component in order to optimize the human, technical and economic resources of the installation. This study focuses on the degradation due to fouling of a gas turbine in the Canary Islands, and analyzes fouling levels over time based on the operating regime and local meteorological variables. In particular, we study the relationship between degradation and the suspended dust that originates in the Sahara Desert. To this end, we use a computational procedure that relies on a set of artificial neural networks to build an ensemble, using a cross-validated committees approach, to yield the compressor efficiency. The use of trained models makes it possible to know in advance how the local fouling of an industrial rotating component will evolve, which is useful for maintenance planning and for calculating the relative importance of the variables that make up the system

Abundance of Ganoderma sp. in Europe and SW Asia: modelling the pathogen infection levels in local trees using the proxy of airborne fungal spore concentrations

Ganoderma comprises a common bracket fungal genus that causes basal stem rot in deciduous and coniferous trees and palms, thus having a large economic impact on forestry production. We estimated pathogen abundance using long-term, daily spore concentration data collected in five biogeographic regions in Europe and SW Asia. We hypothesized that pathogen abundance in the air depends on the density of potential hosts (trees) in the surrounding area, and that its spores originate locally. We tested this hypothesis by (1) calculating tree cover density, (2) assessing the impact of local meteorological variables on spore concentration, (3) computing back trajectories, (4) developing random forest models predicting daily spore concentration. The area covered by trees was calculated based on Tree Density Datasets within a 30 km radius from sampling sites. Variations in daily and seasonal spore concentrations were cross-examined between sites using a selection of statistical tools including HYSPLIT and random forest models.Our results showed that spore concentrations were higher in Northern and Central Europe than in South Europe and SW Asia. High and unusually high spore concentrations (> 90th and > 98th percentile, respectively) were partially associated with long distance transported spores: at least 33% of Ganoderma spores recorded in Madeira during days with high concentrations originated from the Iberian Peninsula located >900 km away. Random forest models developed on local meteorological data performed better in sites where the contribution of long distance transported spores was lower. We found that high concentrations were recorded in sites with low host density (Leicester, Worcester), and low concentrations in Kastamonu with high host density. This suggests that south European and SW Asian forests may be less severely affected by Ganoderma. This study highlights the effectiveness of monitoring airborne Ganoderma spore concentrations as a tool for assessing local Ganoderma pathogen infection levels.

The impact of teleconnections on the temporal dynamics in aboveground net primary productivity of the Mongolian Plateau grasslands

AbstractTemporal dynamics in aboveground net primary productivity (ANPP) are significant for environmental protection and economic development in the Mongolian Plateau grasslands (MPGs). Climate control of ANPP's temporal variations has intensified in response to local meteorological conditions, while the contributions of large‐scale atmospheric circulations to ANPP variations are poorly understood in the MPGs. First, we explored the temporal dynamics in annual ANPP, and then, we analyzed correlations among annual ANPP, local meteorological variables, and three different teleconnections during 1982–2015 in this study. Our analytical results indicated that (a) ANPP exhibited an increasing trend during the entire study period. However, the increasing trend did not persist, and two reversals occurred in 1998 and 2007. (b) ANPP variations were primarily influenced by summer precipitation in most regions and by summer precipitation and temperature together in other regions of the MPGs. (c) The Atlantic Multidecadal Oscillation (AMO) in winter negatively dominated annual ANPP variations by influencing summer precipitation and summer temperature in central Mongolia. The North Atlantic Oscillation (NAO) in winter exerted strong positive impacts on ANPP variations by influencing summer precipitation in Inner Mongolia, China. The East Atlantic Western Russia (EAWR) in spring controlled annual ANPP variations by influencing summer precipitation in western and eastern Mongolia. (d) The winter AMO, winter NAO and spring EAWR significantly dominated the annual ANPP variations in 49.78% of the MPGs, with values of 18.90, 9.81 and 21.07%, respectively. (e) Reversals of the winter AMO, winter NAO and spring EAWR between negative and positive phases resulted in two reversals of ANPP trends. This study will improve our understanding of the role of large‐scale atmospheric circulations in regulating ANPP variations and will be valuable in guiding the development of grazing land management in Mongolia and Inner Mongolia, China.

Local- and Regional-Scale Forcing of Glacier Mass Balance Changes in the Swiss Alps

Glacier mass variations are climate indicators. Therefore, it is essential to examine both winter and summer mass balance variability over a long period of time to address climate-related ice mass fluctuations. In this study we analyze glacier mass balance components and hypsometric characteristics with respect to their interactions with local meteorological variables and remote large-scale atmospheric and oceanic patterns. The results show that all selected glaciers have lost their equilibrium condition in recent decades, with persistent negative annual mass balance trends and decreasing accumulation area ratios (AARs), accompanied by increasing air temperatures of ≥+0.45 °C decade−1. The controlling factor of annual mass balance is mainly attributed to summer mass losses, which are correlated with (warming) June to September air temperatures. In addition, the interannual variability of summer and winter mass balances is primarily associated to the Atlantic Multidecadal Oscillation (AMO), Greenland Blocking Index (GBI), and East Atlantic (EA) teleconnections. Although climate parameters are playing a significant role in determining the glacier mass balance in the region, the observed correlations and mass balance trends are in agreement with the hypsometric distribution and morphology of the glaciers. The analysis of decadal frontal retreat using Landsat images from 1984 to 2014 also supports the findings of this research, highlighting the impact of lake formation at terminus areas on rapid glacier retreat and mass loss in the Swiss Alps.

The Diamond League athletic series: does the air quality sparkle?

Urban air pollution can have negative short- and long-term impacts on health, including cardiovascular, neurological, immune system and developmental damage. The irritant qualities of pollutants such as ozone (O3), nitrogen dioxide (NO2) and particulate matter (PM) can cause respiratory and cardiovascular distress, which can be heightened during physical activity and particularly so for those with respiratory conditions such as asthma. Previously, research has only examined marathon run outcomes or running under laboratory settings. This study focuses on elite 5-km athletes performing in international events at nine locations. Local meteorological and air quality data are used in conjunction with race performance metrics from the Diamond League Athletics series to determine the extent to which elite competitors are influenced during maximal sustained efforts in real-world conditions. The findings from this study suggest that local meteorological variables (temperature, wind speed and relative humidity) and air quality (ozone and particulate matter) have an impact on athletic performance. Variation between finishing times at different race locations can also be explained by the local meteorology and air quality conditions seen during races.

A robust deep learning framework for short-term wind power forecast of a full-scale wind farm using atmospheric variables

Short-term (less than 1 h) forecast of the power generated by wind turbines in a wind farm is extremely challenging due to the lack of reliable data from meteorological towers and numerical weather model outputs at these timescales. A robust deep learning model is developed for short-term forecasts of wind turbine generated power in a wind farm using the state-of-the-art temporal convolutional networks (TCN) to simultaneously capture the temporal dynamics of the wind turbine power and relationship among the local meteorological variables. An orthogonal array tuning method based on the Taguchi design of experiments is utilized to optimize the hyperparameters of the proposed TCN model. The proposed TCN model is validated using twelve months of data from a 130 MW utility-scale wind farm with 86 wind turbines in comparison with some of the existing methods in the literature. The power curves obtained from the proposed TCN model show consistent improvements over existing methods at all wind speeds.

The Effect of COVID-19 Lockdowns on the Air Pollution of Urban Areas of Central and Southern Chile

We present the effects of the confinement and physical distancing policies applied during the COVID-19 pandemic on the concentrations of PM10, PM2.5, NO, NO2 and O3 in 16 cities in central and southern Chile. The period between March and May in 2020 was compared with the corresponding months during 2017–2019, using surface data and satellite information. The relative percent changes in the concentration of atmospheric pollutants, and the meteorological variables observed between these two periods were used to quantify the effects of the lockdowns on the local air quality of the urban areas studied. The results showed statistically significant changes in 11 of the 16 cities. Significant relative changes between +14% and –33% were observed for PM10 in 9 cities; while statistically significant changes between –6% and –48% were evident for PM2.5 in 10 cities. Significant decreases between –27% and –55%, were observed in 4 cities in which NO2 data were available; while significant increases in O3, between 18% and 43%, were found in 4 of the 5 cities with available data. The local meteorological variables did not show significant changes between both periods. In all the cities studied, one of the main PM sources is wood burning for residential heating. Although the quarantine imposed during the health emergency could have induced an increase in residential emissions, these were compensated with the reductions in vehicular and/or industrial emissions. Therefore, these results should be carefully interpreted and should inspire new research considering the social, cultural, and economic factors that could alter the common emission patterns and air quality of urban centers.