Correlation Analysis of EnvironmentalPollutants and MeteorologicalVariables ApplyingNeuronalNetworks
In order to develop an environmental contingency forecasting tool for decision making, a Pattern Recognition method applying Neural Networks is presented. For this purpose, SO2 and PM10 time series concentrations are analyzed every hour and daily, as well as a variety of meteorological variables. These pollutant concentrations and meteorological variables are self-organized by means of a SOM Neural Network in different classes. Classes are used in training fase of a General Regression Neural classifier (GRNN) to provide an air quality forecast. In this case a time series set obtained from Environmental Monitoring Network (EMN) of the city of Salamanca, Guanajuato, México is used. Results verify the potential of this method versus other statistical classification methods and also variables correlation is solved.
1. What atmospheric variables were used as the inputs for the SOM?
2. Were these atmospheric variables considered all together, or different combinations of them were used as inputs to the SOM?
3. If different combinations were used, what variables showed more correlation with each pollutant concentrations?
USMC
1.- We consider three variables, Temperature, Relative Humidity and SO2 concentration in an experiment and Temperature, relative Humidity and PM10 particles concentration in other experiment.
2.- Only three variables in each experiment, we have two types of pollutant and the meteorological variables, then we consider only trhee variables as input SOM one for PM10 and other for SO2.
3.- In each experiment the Relative Humidity (HR) is an important factor in the correlation among pollutant concentrations and meteorological variables it is because more HR produces smaller pollutant concentration measures and therefore smaller AQI levels. But the problem is the secondary pollutants like acid rain. In the case of PM10 particles, the relative humidity diminishes the measured concentration and the AQI is better.
Thanks for your question...
Antonio Vega-Corona
Emmanuel,
Yes, it is possible. In this specific paper it wasn't done, but this work is part of a project for AQI Estimation and Prediction for the city of Salamanca, Mexico. Also the paper "SO2 Concentrations Forecasting..." is part of the project. Let's say this paper is the first phase of the project, Prediction is the second one. Currently prediction was done for SO2 concentrations, but the goal is to take the outputs of the AQI Estimation phase and Predict the AQI not pollutant concentrations.
Regards,
Ulises S. Mendoza
Is it possible to estimate the AQI only on the day when of the observations are made or also for the forthcoming days?
Emmanuel Brousseau