Time Series Of Variables Research Articles

Species habitat suitability increased during COVID-19 lockdowns

COVID-19 lockdowns had strong positive and negative effects on the environment and biodiversity. The most evident effects were the occurrence of wildlife in the middle of urban settlements, the reduction of noise, and the improvement in air quality. However, other effects on species are less well known. Our main question is: did species habitat suitability increase during lockdowns? Here, we analysed trends in species habitat suitability over time by modelling the species distributions with a time series of environmental variables obtained from satellite remote sensing. We modelled the habitat suitability of 381 species of vascular plants, amphibians, reptiles, birds, and mammals, before, during, and after the lockdowns in the Iberian Peninsula (Portugal and Spain), from July 2017 to August 2022 with the Maxent presence-background algorithm and five MODIS variables (Albedo, Evapotranspiration, LST, NDVI, and Surface reflectance) with a periodicity of 8 days, and analysed their trends with the Mann-Kendall test. We compared our results with air quality data (PM10, CO, and NO2) from ground stations and Sentinel 5 P and with mobility data from Google mobility reports. Habitat suitability increased during both lockdowns when considering all species together. The peak in species' habitat suitability trends during the first lockdown was coincident with air quality and mobility trends: air pollution decreased when people stayed at home. Reduction in human mobility during COVID-19 lockdowns revealed positive effects on species’ habitat quality, highlighting the importance of reducing human activities as the only solution for nature conservation.

Efficient global sensitivity analysis method for dynamic models in high dimensions

AbstractDynamic models generating time‐dependent model predictions are typically associated with high‐dimensional input spaces and high‐dimensional output spaces, in particular if time is discretized. It is computationally prohibitive to apply traditional global sensitivity analysis (SA) separately on each time output, as is common in the literature on multivariate SA. As an alternative, we propose a novel method for efficient global SA of dynamic models with high‐dimensional inputs by combining a new polynomial chaos expansion (PCE)‐driven partial least squares (PLS) algorithm with the analysis of variance. PLS is used to simultaneously reduce the dimensionality of the input and output variables spaces, by identifying the input and output latent variables that account for most of their joint variability. PCE is incorporated into the PLS algorithm to capture the non‐linear behavior of the physical system. We derive the sensitivity indices associated with each output latent variable, based on which we propose generalized sensitivity indices that synthesize the influence of each input on the variance of entire output time series. All sensitivities can be computed analytically by post‐processing the coefficients of the PLS‐PCE representation. Hence, the computational cost of global SA for dynamic models essentially reduces to the cost for estimating these coefficients. We numerically compare the proposed method with existing methods by several dynamic models with high‐dimensional inputs. The results show that the PLS‐PCE method can obtain accurate sensitivity indices at low computational cost, even for models with strong interaction among the inputs.

Impact of Carbon Emission Factors on Economic Agents Based on the Decision Modeling in Complex Systems

This article presents a methodology for modeling the impact of both internal and external environmental carbon emission factors on the resulting indicators of an international company. This research uses picture fuzzy rough sets to model the impact of factors on the resulting indicators as a research method. The proposed model is based on a dataset that includes the company’s profit, revenue, valuation, share price, and market share from 2012 through 2022. This empirical period is optimal for such a type of modeling. An approach of picture fuzzy rough sets based on the time series of endogenous and exogenous variables can provide an opportunity to analyze and consider the consequences of feedback changes in the systems of which they are a part. The article proposes a valuable framework for understanding the complex relationship between carbon emissions, economic factors, and the performance of international companies. The researchers of this study recommend a discussion that attempts to gain a deeper understanding of the challenges and opportunities that lie ahead for international companies in the context of climate change and technological innovation.

The effect of education on economic institutions

ABSTRACT This research is the first to use the Heritage Foundation economic freedom index as a measure of economic institutions in order to examine the relationship between education and economic institutions. It presents a variety of empirical evidence to show that education predicts economic institutions whereas the inverse is not true. It uses 2SLS, with the level of education in 1950 as the instrument for the level of education in 2015, and finds that the effect of education on economic institutions is positive and statistically significant. Also, this paper uses fixed effects in order to control for country-level omitted variables that may affect both economic institutions and the instrument. Finally, it uses difference GMM, and system GMM. These regressions exploit both the cross sectional and the time series variation in the data, address endogeneity, and confirm that the effect of education on economic institutions is positive and statistically significant.

Bayesian modelling of time series data (BayModTS)-a FAIR workflow to process sparse and highly variable data.

Systems biology aims to better understand living systems through mathematical modelling of experimental and clinical data. A pervasive challenge in quantitative dynamical modelling is the integration of time series measurements, which often have high variability and low sampling resolution. Approaches are required to utilize such information while consistently handling uncertainties. We present BayModTS (Bayesian modelling of time series data), a new FAIR (findable, accessible, interoperable, and reusable) workflow for processing and analysing sparse and highly variable time series data. BayModTS consistently transfers uncertainties from data to model predictions, including process knowledge via parameterized models. Further, credible differences in the dynamics of different conditions can be identified by filtering noise. To demonstrate the power and versatility of BayModTS, we applied it to three hepatic datasets gathered from three different species and with different measurement techniques: (i) blood perfusion measurements by magnetic resonance imaging in rat livers after portal vein ligation, (ii) pharmacokinetic time series of different drugs in normal and steatotic mice, and (iii) CT-based volumetric assessment of human liver remnants after clinical liver resection. The BayModTS codebase is available on GitHub at https://github.com/Systems-Theory-in-Systems-Biology/BayModTS. The repository contains a Python script for the executable BayModTS workflow and a widely applicable SBML (systems biology markup language) model for retarded transient functions. In addition, all examples from the paper are included in the repository. Data and code of the application examples are stored on DaRUS: https://doi.org/10.18419/darus-3876. The raw MRI ROI voxel data were uploaded to DaRUS: https://doi.org/10.18419/darus-3878. The steatosis metabolite data are published on FairdomHub: 10.15490/fairdomhub.1.study.1070.1.

Spatial Sensitivity of River Flooding to Changes in Climate and Land Cover Through Explainable AI

AbstractExplaining the spatially variable impacts of flood‐generating mechanisms is a longstanding challenge in hydrology, with increasing and decreasing temporal flood trends often found in close regional proximity. Here, we develop a machine learning‐informed approach to unravel the drivers of seasonal flood magnitude and explain the spatial variability of their effects in a temperate climate. We employ 11 observed meteorological and land cover (LC) time series variables alongside 8 static catchment attributes to model flood magnitude in 1,268 catchments across Great Britain over four decades. We then perform a sensitivity analysis to assess how a 10% increase in precipitation, a 1°C rise in air temperature, or a 10 percentage point increase in urban or forest LC may affect flood magnitude in catchments with varying characteristics. Our simulations show that increasing precipitation and urbanization both tend to amplify flood magnitude significantly more in catchments with high baseflow contribution and low runoff ratio, which tend to have lower values of specific discharge on average. In contrast, rising air temperature (in the absence of changing precipitation) decreases flood magnitudes, with the largest effects in dry catchments with low baseflow index. Afforestation also tends to decrease floods more in catchments with low groundwater contribution, and in dry catchments in the summer. Our approach may be used to further disentangle the joint effects of multiple flood drivers in individual catchments.

Multi-Method Comparative Analysis of Hydroclimatic Trends and Variability in Dry Creek Catchment, South Australia

AbstractThe trend and variability of hydroclimatic variables over time are apparent in seasonal creeks, especially those located in urbanized areas. Understanding hydro-climatic trends in urban areas is crucial for the sustainable management of water resources and the environment. This study aimed to explore the spatiotemporal variability and trends of hydroclimate variables as well as the potential connection between rainfall and streamflow in Dry Creek catchment, South Australia. The trend-free pre-whitening Mann–Kendall (TFPW-MK) test and Innovative Trend Analysis (ITA) were utilized to examine the monotonic and nonmonotonic trends, respectively, and multiple statistical tests were employed to examine the change points in the hydroclimatic time series. Sen’s slope, Simple Linear Regression (SLR), and ITA were used as alternative approaches to assess the magnitudes of change and overcome the limitations in the underlying assumptions of the various methodologies. The variability in the hydroclimate time series was estimated using several indices, such as the coefficient of variation, seasonality indices, flashiness index, and mean zero flow index. The analyses revealed important findings, notably the high variability of rainfall and streamflow during dry periods. Streamflow displayed greater variability compared to rainfall, with high CV values recorded both seasonally and annually. Furthermore, there was a significant upward trend in seasonal rainfall during winter. Additionally, the maximum and mean temperatures demonstrated a statistically significant increase, which can be attributed to global warming and significant urbanization in the catchment area. Comparative analysis has confirmed that the ITA has superior detection capabilities for nonmonotonic trends, outperforming other methods. It excels at presenting graphical representations that accurately depict trends, effectively differentiating between low, medium, and high values. The strong relationship between rainfall and streamflow demonstrated by the Tanh curve suggests that rainfall is the most reliable predictor of streamflow. The outcomes of this investigation are expected to support local governmental organizations and decision-makers in comprehending the spatial and temporal features of rainfall, as well as its correlation with streamflow. This information will further assist in developing flood and drought mitigation strategies backed by empirical evidence. Graphical Abstract

Innovative Decision Fusion for Accurate Crop/Vegetation Classification with Multiple Classifiers and Multisource Remote Sensing Data

Obtaining accurate and real-time spatial distribution information regarding crops is critical for enabling effective smart agricultural management. In this study, innovative decision fusion strategies, including Enhanced Overall Accuracy Index (E-OAI) voting and the Overall Accuracy Index-based Majority Voting (OAI-MV), were introduced to optimize the use of diverse remote sensing data and various classifiers, thereby improving the accuracy of crop/vegetation identification. These strategies were utilized to integrate crop/vegetation classification outcomes from distinct feature sets (including Gaofen-6 reflectance, Sentinel-2 time series of vegetation indices, Sentinel-2 time series of biophysical variables, Sentinel-1 time series of backscatter coefficients, and their combinations) using distinct classifiers (Random Forests (RFs), Support Vector Machines (SVMs), Maximum Likelihood (ML), and U-Net), taking two grain-producing areas (Site #1 and Site #2) in Haixi Prefecture, Qinghai Province, China, as the research area. The results indicate that employing U-Net on feature-combined sets yielded the highest overall accuracy (OA) of 81.23% and 91.49% for Site #1 and Site #2, respectively, in the single classifier experiments. The E-OAI strategy, compared to the original OAI strategy, boosted the OA by 0.17% to 6.28%. Furthermore, the OAI-MV strategy achieved the highest OA of 86.02% and 95.67% for the respective study sites. This study highlights the distinct strengths of various remote sensing features and classifiers in discerning different crop and vegetation types. Additionally, the proposed OAI-MV and E-OAI strategies effectively harness the benefits of diverse classifiers and multisource remote sensing features, significantly enhancing the accuracy of crop/vegetation classification.

Glucose Color Index: Development and Validation of a Novel Measure of the Shape of Glycemic Variability.

Standard continuous glucose monitoring (CGM) metrics: mean glucose, standard deviation, coefficient of variation, and time in range, fail to capture the shape of variability in the CGM time series. This information could facilitate improved diabetes management. We analyzed CGM data from 141 adults with type 2 diabetes in the Hyperglycemic Profiles in Obstructive Sleep Apnea (HYPNOS) trial. Participants in HYPNOS wore CGM sensors for up to two weeks at two time points, three months apart. We calculated the log-periodogram for each time period, summarizing using disjoint linear models. These summaries were combined into a single value, termed the Glucose Color Index (GCI), using canonical correlation analysis. We compared the between-wear correlation of GCI with those of standard CGM metrics and assessed associations between GCI and diabetes comorbidities in 398 older adults with type 2 diabetes from the Atherosclerosis Risk in Communities (ARIC) study. The GCI achieved a test-retest correlation of R = .75. Adjusting for standard CGM metrics, the GCI test-retest correlation was R = .55. Glucose Color Index was significantly associated (p < .05) with impaired physical functioning, frailty/pre-frailty, cardiovascular disease, chronic kidney disease, and dementia/mild cognitive impairment after adjustment for confounders. We developed and validated the GCI, a novel CGM metric that captures the shape of glucose variability using the periodogram signal decomposition. Glucose Color Index was reliable within participants over a three-month period and associated with diabetes comorbidities. The GCI suggests a promising avenue toward the development of CGM metrics which more fully incorporate time series information.

Short-Term Load Forecasting Based on Optimized Random Forest and Optimal Feature Selection

Short-term load forecasting (STLF) plays a vital role in ensuring the safe, efficient, and economical operation of power systems. Accurate load forecasting provides numerous benefits for power suppliers, such as cost reduction, increased reliability, and informed decision-making. However, STLF is a complex task due to various factors, including non-linear trends, multiple seasonality, variable variance, and significant random interruptions in electricity demand time series. To address these challenges, advanced techniques and models are required. This study focuses on the development of an efficient short-term power load forecasting model using the random forest (RF) algorithm. RF combines regression trees through bagging and random subspace techniques to improve prediction accuracy and reduce model variability. The algorithm constructs a forest of trees using bootstrap samples and selects random feature subsets at each node to enhance diversity. Hyperparameters such as the number of trees, minimum sample leaf size, and maximum features for each split are tuned to optimize forecasting results. The proposed model was tested using historical hourly load data from four transformer substations supplying different campus areas of the University of Beira Interior, Portugal. The training data were from January 2018 to December 2021, while the data from 2022 were used for testing. The results demonstrate the effectiveness of the RF model in forecasting short-term hourly and one day ahead load and its potential to enhance decision-making processes in smart grid operations.

Unlocking Weather Observations at the End of the World: Late-XIX and Early-XX Century Monthly Mean Temperature Climatology for Southern Patagonia

A climate analysis of the monthly mean temperatures of Southern Patagonia during the late-XIXth and early-XXth centuries was carried out as part of the international data rescue Atmospheric Circulation Reconstructions over the Earth (ACRE) program partnership in Argentina, together with other data sources with regional and global records. The data from these diverse sources were combined to carry out a study in the coastal region of Patagonia, including Tierra del Fuego, between 42° S and 55° S for 11 locations. Furthermore, HadSST monthly/seasonal fields during the period 1880–1920 were also used. Both mean monthly and seasonal temperature values and timeseries variability were considered. Their analysis shows consistent behavior within the study region and compared to Southern Hemisphere mean results, which are characterized by a warm late-XIX century and a cooler early-XX century. This is also in agreement with SST variability along the coasts of Patagonia and hemispheric records. A comparison with present-day observations, where available, also yields consistent behavior. Low-frequency variability, i.e., in periods longer than 3 years, during the study period is consistent with present variability. Trend estimates at Trelew and Rio Gallegos for the period 1901–2020 yield significant trends, consistent with hemispheric warming at their latitudes.

Assessing climate impacts on slow-moving landslides in the western Alps of Piemonte: integration of monitoring techniques for detecting displacements

The influence of climatic factors on landslides triggers and displacement rates is a crucial research topic, especially due to the growing need to understand the evolution of climate change in historical periods of intense precipitation and anomalous temperature increases. Italy, highly prone to hydrogeological instability, extremely its mountainous regions such as the Alps, stands as a pertinent subject area for instability scenarios. However, the interpretation of climate effects on landslides is still an open issue. This work proposed a simplified methodology for investigating the displacements of three slow-moving landslides located in the Western Alps of Piemonte region, in response to significant meteorological events evaluated from reference normal of precipitation and temperature trends over the reference period 1991–2020. Another purpose is to emphasize the advantages of using different monitoring techniques by comparing displacement time series measured with in situ and remote sensing instruments, to detect ground deformation processes of these gravitational phenomena. The existence of a robust monitoring network, coupled with InSAR dataset support, has allowed detecting climatic factors’ impact on displacement rates for the outlined case studies. The results have demonstrated the relationships between the identified climatic events and variations in displacement time series, as well as the potential of integrating field observations and InSAR techniques to improve the interpretation of landslide dynamics. Although this study has laid the basis for understanding the influence of climatic factors on landslide displacements, there is still much to investigate and refine. The proposed preliminary analysis will further improve the ability to predict, monitor and mitigate landslide risk under changing climate conditions.

Improving the accuracy of credit scoring models using an innovative Bayesian informative prior specification method

A new Bayesian informative prior specification method (BAF method – Bayesian priors using ARIMA forecasts) is proposed to introduce additional information into credit risk modelling and improve model predictive performance. We use logistic regression to model the probability of default of mortgage loans comparing the Bayesian approach with various priors and the frequentist approach. But unlike previous literature, we treat coefficient estimates in the probability of default model as stochastic time series variables. We build ARIMA models to forecast the coefficient values in future time periods and use these ARIMA forecasts as Bayesian informative priors. We find that the Bayesian models using this prior specification method produce more accurate predictions for the probability of default as compared to frequentist models and Bayesian models with other priors.

Snow depth time series Generation: Effective simulation at multiple time scales

Snow depth (SD) is a crucial variable of the water, energy, and nutrient cycles, impacting water quantity and quality, the occurrence of floods and droughts, snow-related hazards, and sub-surface ecological functions. As a result, quantifying SD dynamics is crucial for several scientific and practical applications. Ground measurements of SD provide information at sparse locations, and physical global model simulations provide information at relatively coarse spatial resolutions. An approach to complement this information is using stochastic models that generate time series of hydroclimatic variables, preserving their statistical properties in a computationally-effective manner. However, stochastic generation methods to produce SD time series exclusively do not exist in the literature. Here, we apply a stochastic model to produce synthetic daily SD time series trained by 448 stations in Canada. We show that the model captures key statistical properties of the observed records, including the daily distributions of zero and non-zero SD, temporal clustering (i.e., autocorrelation), and seasonal patterns. The model also excelled in capturing the observed higher-order L-moments at multiple temporal scales, with biases between simulated and observed L-skewness and L-kurtosis within (-0.1, +0.1) for 93.0 % and 98.3 % of the stations, respectively. The stochastic modelling approach introduced here advances the generation of SD time series, which are needed to develope Earth-system models and assess the risk of snowmelt flooding that lead to severe damage and fatalities.

PyPPG: a Python toolbox for comprehensive photoplethysmography signal analysis

Objective. Photoplethysmography is a non-invasive optical technique that measures changes in blood volume within tissues. It is commonly and being increasingly used for a variety of research and clinical applications to assess vascular dynamics and physiological parameters. Yet, contrary to heart rate variability measures, a field which has seen the development of stable standards and advanced toolboxes and software, no such standards and limited open tools exist for continuous photoplethysmogram (PPG) analysis. Consequently, the primary objective of this research was to identify, standardize, implement and validate key digital PPG biomarkers. Approach. This work describes the creation of a standard Python toolbox, denoted pyPPG, for long-term continuous PPG time-series analysis and demonstrates the detection and computation of a high number of fiducial points and digital biomarkers using a standard fingerbased transmission pulse oximeter. Main results. The improved PPG peak detector had an F1-score of 88.19% for the state-of-the-art benchmark when evaluated on 2054 adult polysomnography recordings totaling over 91 million reference beats. The algorithm outperformed the open-source original Matlab implementation by ∼5% when benchmarked on a subset of 100 randomly selected MESA recordings. More than 3000 fiducial points were manually annotated by two annotators in order to validate the fiducial points detector. The detector consistently demonstrated high performance, with a mean absolute error of less than 10 ms for all fiducial points. Significance. Based on these fiducial points, pyPPG engineered a set of 74 PPG biomarkers. Studying PPG time-series variability using pyPPG can enhance our understanding of the manifestations and etiology of diseases. This toolbox can also be used for biomarker engineering in training data-driven models. pyPPG is available on https://physiozoo.com/.

Analysis and modeling of detours in pedestrian operational navigation

Traditional operational navigation models for pedestrian dynamics demonstrate limitations in reproducing the circle antipode experiment, an artificially designed multi-directional flow scenario. In the experiment, pedestrians take detours to avoid the congestion caused by others taking straight paths. Although the pedestrian detour action is commonly incorporated into the route choice model, it frequently goes unaddressed in the operational navigation model, resulting in a disparity between simulation outcomes and empirical observations. To reveal the mechanism underlying pedestrian detours in the circle antipode experiment, this study employed the K-means clustering method rather than using a threshold approach to categorize the experimental participants into groups taking direct or detour routes. Following this, a heuristic function is formulated to determine the desired direction of agents in a collision-free velocity model, reflecting the trade-off between shorter routes and faster speeds. The parameter of the proposed function is calibrated using the proportion of agents choosing detours, where the route types of agents are identified by a classifier based on the random forest. Compared to two traditional models that do not consider detours, the proposed model can more realistically reproduce the trajectory distributions, the travel time, the route length, and the time series of relevant variables in the circle antipode experiment. The study offers insights into employing machine learning methodologies for analyzing pedestrian flow, validating pedestrian dynamics models, and providing an accurate simulation tool for designing transportation facilities and crowd management at large events.

Temporal and vertical variations in carbon flux and export of zooplankton fecal pellets in the western South China Sea

The enhancement of particulate organic carbon transfer to the deep sea by zooplankton fecal pellets constitutes a crucial component of the marine biological carbon pump. Here, we investigated time-series variations in characteristic and flux of zooplankton fecal pellets at different water depths of two sediment-trap mooring stations from May 2021 to May 2022 in the western South China Sea. The results show that numerical fluxes of fecal pellets were 0.75 ± 0.60 × 104, 0.63 ± 0.63 × 104, and 0.38 ± 0.29 × 104 pellets m−2 d−1 at 500 m, 1170 m, and 1380 m water depth, respectively, corresponding to their carbon fluxes of 0.10 ± 0.06, 0.09 ± 0.04, and 0.10 ± 0.04 mg C m−2 d−1. Both numerical and carbon fluxes of fecal pellets exhibited clear seasonal variations, with two peaks occurred in August and early November at all water depths. The fecal pellets have distinct morphological types (spherical, cylindrical, and ellipsoidal) and their contributions to the numerical and carbon fluxes were different. At all water depths, ellipsoidal and spherical pellets accounted for 96.0% of the numerical flux and 72.1% of the carbon flux. Cylindrical pellets were rare in quantity, accounting for 4.0% of numerical flux, but their carbon contribution accounted for 27.9% of the total fecal pellet carbon flux. The proportional fractions of zooplankton fecal pellets contributed to the overall particulate organic carbon from 0.7% to 28.2% in the western South China Sea, and fell in a reasonable range around the world. Multiple mechanisms, including East Asian monsoon climate and zooplankton community structure may be responsible for the production and fate of fecal pellets as well as their contribution to the settling particulate organic carbon.

A novel time-varying grey Fourier model for variable amplitude seasonal fluctuation sequences

PurposeA time-varying grey Fourier model (TVGFM(1,1,N)) is proposed for the simulation of variable amplitude seasonal fluctuation time series, as the performance of traditional grey models can't catch the time-varying trend well.Design/methodology/approachThe proposed model couples Fourier series and linear time-varying terms as the grey action, to describe the characteristics of variable amplitude and seasonality. The truncated Fourier order N is preselected from the alternative order set by Nyquist-Shannon sampling theorem and the principle of simplicity, then the optimal Fourier order is determined by hold-out method to improve the robustness of the proposed model. Initial value correction and the multiple transformation are also studied to improve the precision.FindingsThe new model has a broader applicability range as a result of the new grey action, attaining higher fitting and forecasting accuracy. The numerical experiment of a generated monthly time series indicates the proposed model can accurately fit the variable amplitude seasonal sequence, in which the mean absolute percentage error (MAPE) is only 0.01%, and the complex simulations based on Monte-Carlo method testify the validity of the proposed model. The results of monthly electricity consumption in China's primary industry, demonstrate the proposed model catches the time-varying trend and has good performances, where MAPEF and MAPET are below 5%. Moreover, the proposed TVGFM(1,1,N) model is superior to the benchmark models, grey polynomial model (GMP(1,1,N)), grey Fourier model (GFM(1,1,N)), seasonal grey model (SGM(1,1)), seasonal ARIMA model seasonal autoregressive integrated moving average model (SARIMA) and support vector regression (SVR).Originality/valueThe parameter estimates and forecasting of the new proposed TVGFM are studied, and the good fitting and forecasting accuracy of time-varying amplitude seasonal fluctuation series are testified by numerical simulations and a case study.

Short term forecasting of persistent air quality deterioration events in the metropolis of Sao Paulo

Air pollution is one of the main environmental problems in metropolitan areas. Negative impacts to human health are intensified when poor air quality conditions persist for many consecutive days, with gradual accumulation of pollutants in the atmosphere. Persistent air quality deterioration events are typically associated with occurrence of stagnant atmospheric conditions and reach the regional scale. In this study, data-driven models were developed to forecast the occurrence of persistent air pollution events of inhalable particulate matter (PM10) and ozone (O3) in the metropolis of Sao Paulo, Brazil. On average, 8 events per year were observed between 2005 and 2022, comprising 73 event days per year. The logistic regression method was used in a supervised learning framework. Daily timeseries of surface weather variables were used as predictors. In the case of PM10, a consistent long-term decrease in the number events impacted the model performance. The PM10 model benefited from the restriction of the training set to recent years, with a significant increase in the model accuracy despite the reduction in the volume of data. The final models correctly reproduced the seasonal distribution of events, with overall accuracies of 0.92 and 0.87 for O3 and PM10, respectively, in 2022. Despite the fact that persistent exceedance events are relatively rare, the models were able to detect 81% and 97% of the event days in 2022, respectively for O3 and PM10. Daily maximum temperature was an important predictor, increasing the event odds by 483% (O3) and 84% (PM10). The classification models developed in this study can successfully forecast the occurrence of regional air pollution events concerning both primary and secondary air pollutants, which have different drivers for accumulation in the atmosphere. The models require simple input data and low computational resources, aiming to stimulate future usage by the general public and decision-makers, in order to mitigate exposure to harmful air pollutant concentrations.

Semi-real-time decision tree ensemble algorithms for very short-term solar irradiance forecasting

Industrial activities are transitioning towards decarbonization, focusing on renewable energy sources, particularly photovoltaic solar energy. However, the inherent high variability of photovoltaic energy poses challenges. Some of them can be partially addressed by predicting electricity production, which in the case of photovoltaic solar energy is heavily based on solar irradiance prediction. Although extensive research has been conducted in this field, there is a noticeable gap in research regarding very short-term (intra-minute) forecasting under high-variability scenarios. In this proposal, real data from a photovoltaic solar plant in Alderville (Canada) were used to predict irradiance with a horizon of 15 and 30 s. The objective is to make this prediction in near-real time. To achieve this, we propose the use of machine learning algorithms based on decision tree ensembles, due to their low computational training cost and known effectiveness. On the other hand, we propose pre-processing the data through a temporal and spatial correlation analysis between measurements from different sensors. Feature selection analysis allows us to determine the direction of the wind and consequently identify the most relevant panels for model training. This preprocessing enhances the model retraining without the need for external information such as sky images or wind speed and direction on days with highly variable cloud cover. The presented methodology offers promising results with significantly reduced training times, demonstrating the suitability of this semi-online training approach for highly variable time series forecasting.