Prediction of ozone concentration in tropospheric levels using artificial neural networks and support vector machine at Rio de Janeiro, Brazil

Luna, Aderval S.; Paredes, Márcio L.L.; Oliveira, G.C.G. de; Corrêa, Sérgio Machado

doi:10.1016/j.atmosenv.2014.08.060

Cited by 81 publications

(52 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Godoy et al (2009), using receptor modeling, found that vehicle emissions constitute the main source of pollutants in the city of Rio de Janeiro. Luna et al (2014), using neural network analysis, developed a system for ozone prediction. In all such initiatives, one important objective is to provide a tool that can be used by governmental agencies to predict events of high pollutant concentrations, which will make it possible to issue alerts and prepare emergency rooms to receive cases related to the poor air quality.…”

Section: Wwwfrontiersinorgmentioning

confidence: 99%

Air quality forecasting system for Southeastern Brazil

Andrade

Ynoue

Freitas

et al. 2015

Front. Environ. Sci.

View full text Add to dashboard Cite

Southeastern Brazil, the most populous and developed region of the country, faces various environmental problems associated with the growth of its population in urban areas. It is the most industrialized area in the country, comprising the metropolitan areas of São Paulo, Rio de Janeiro, Belo Horizonte, and other major cities. Air quality is a major concern, because the reported concentrations of certain regulated pollutants, typically ozone and fine particulate, have exceeded national standards. Due to the difficulty in taking measurements over many different areas, air quality modeling is a useful tool to estimate air pollutant concentrations. For southeastern Brazil, air quality modeling has been performed mostly with the Brazilian Regional Atmospheric Modeling System with Simplified Photochemical Module and the Weather Research and Forecast with Chemistry models. One of the main objectives was to study the evolution of air quality associated with improved vehicle emission factors in urban areas, the impact of climate change on air quality, and the relationship between pollutant concentrations and health. Knowledge of mobile source emission factors has been continuously expanded by in-tunnel measurements and dynamometer protocols, which provide accurate data as inputs to photochemical air quality models. The spatial distribution of the mobile source emissions was constructed based on open access data related to the streets and traffic distribution. The mobile emission module was combined to the chemistry modeling and this implementation can be an example to be applied to other places that do not have a spatial distribution of this source. Forecasts of pollutant concentrations can inform public policies, including those addressing the effects of pollutants on health of the general population, and studies of the impacts of using different fuels and implementation of emissions regulations programs.

show abstract

Section: Wwwfrontiersinorgmentioning

confidence: 99%

Air quality forecasting system for Southeastern Brazil

Andrade

Ynoue

Freitas

et al. 2015

Front. Environ. Sci.

View full text Add to dashboard Cite

show abstract

“…Clearly the inversion of H 2N complicates whole situation. Nevertheless, we have computed mean variances for specific models (8), (12) and (17); see Table 7. Surprisingly the nearest estimation has model (17).…”

Section: Simulation Studymentioning

confidence: 99%

“…We have calculated estimates for different parameter values / and x 0 (for fixed value of r e ) for data generated by model (8) in the following way. For each realization, we have computed (conditional) MLE for the original data (here we have used fixed times t 2 f1; .…”

Section: Simulation Studymentioning

confidence: 99%

“…Duenas et al [1] propose ARIMA models for estimating the ground-level ozone concentrations in air at an urban and rural sampling points in Southeastern Spain, [2] analyzes hourly ozone concentrations with multiple regression and multilayer perceptron models on observations of an urban area of the east coast of the Iberian Peninsula and [3] use ARIMA models for forecasting daily maximum surface ozone concentrations at the airport in Brunei Darussalam. A neural network approach has been used by [4][5][6][7][8] for predicting the intraday ozone concentrations. A comparison of neural network models with ARIMA and regression models has been provided by [9] for prediction of Houston's daily maximum ozone concentrations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting hourly ozone concentrations using wavelets and ARIMA models

Salazar

Nicolis

Ruggeri

et al. 2018

Neural Comput & Applic

View full text Add to dashboard Cite

In recent years, air pollution has been a major concern for its implications on human health. Specifically, ozone (O 3 ) pollution is causing common respiratory diseases. In this paper, we illustrate the process of modeling and prediction hourly O 3 pollution measurements using wavelet transforms. We split the time series of O 3 in daily intervals and estimate scale and wavelet coefficients for each interval by the discrete wavelet transform (DWT) with Haar filter. Subsequently we apply cumulated autoregressive integrated moving average (ARIMA) to estimate the coefficients and forecast their evolution in future intervals. Then the inverse discrete wavelet transform is implemented for the reconstruction of the time series and the forecast in the near future. In order to assess the performance of the proposed methodology, we compare the predictions obtained by the DWT-ARIMA with those obtained by the ARIMA model. Several theoretical results are shown through a simulation study.

show abstract

“…

…”

mentioning

confidence: 99%

Source Identification of Emission Sources for Hydrocarbon with Backward Trajectory Model and Statistical Methods

Yu¹

2017

Pol. J. Environ. Stud.

View full text Add to dashboard Cite

The formation of ground-level ozone, or secondary air pollutants [1], can be simulated with photochemical [2][3][4][5], artificial neural network [6][7], trajectory [8][9] and receptor models [10][11]. For making rapid strategic decisions when assessing abatement strategies of air-quality management programs for ground-level ozone, an approach linked with trajectory models and statistical methods could be a more effective approach than photochemical models. Trajectory models are widely used in air pollution-related studies to determine the cause-effect relationship between sources and receptors, to identify potential sources in accordance with elevated concentrations of air pollutants [12][13][14], and Pol. J. Environ. Stud. Vol. 26, No. 2 (2017), 893-902 Original Research AbstractSeveral statistical techniques were combined with a backward trajectory model and emission inventory to locate sources of total hydrocarbon (THC) emissions and to calculate contributed ratios of emission sources. Emission attraction, a novel method of combining emission inventory and residence time, was introduced to confirm respective contributions of specific emission sources with detailed meteorological and emissions data. This research combined four techniques -residence time, conditional probability function, emission inventory and principal component analysis -to locate possible regions and sources on severe surface ozone episodes and non-episode days. Temporal and spatial interpolation manners were performed on surface and rawinsonde meteorological stations, and complex terrain effects were corrected with a variation-kinematic model. Emission inventory of THC and maximum incremental reactivity (MIR) scales were utilized to calculate the accounted contributions of distinct emission sources from various jurisdictions. Conditional probability function combined with emission attraction could reveal potential regions that emitted high THC emissions and MIR scales during ozone episodes.The ratios of emission attraction for 11-h backward trajectories indicated that 68% of THC emissions and 74% of MIR scales were from the target air-quality basin during non-episode days; the respective figures during ozone episodes were 81% and 75%. The combination of emission attraction and conditional probability function could identify specific locations that cause severe ground ozone pollution and provide more detailed information about source regions compared to traditional RTA and PSCF approaches.

show abstract

Prediction of ozone concentration in tropospheric levels using artificial neural networks and support vector machine at Rio de Janeiro, Brazil

Cited by 81 publications

References 14 publications

Air quality forecasting system for Southeastern Brazil

Air quality forecasting system for Southeastern Brazil

Predicting hourly ozone concentrations using wavelets and ARIMA models

Source Identification of Emission Sources for Hydrocarbon with Backward Trajectory Model and Statistical Methods

Contact Info

Product

Resources

About