Selection and validation of parameters in multiple linear and principal component regressions
Environmental Modelling & Software
Environmental Modelling & Software
Modelling the load curve of aggregate electricity consumption using principal components
Environmental Modelling & Software
A recursive estimation approach to the spatio-temporal analysis and modelling of air quality data
Environmental Modelling & Software
Environmental Modelling & Software
Modelling the distribution of solar spectral irradiance using data mining techniques
Environmental Modelling & Software
Hi-index | 0.00 |
Data on the concentrations of seven environmental pollutants (CH"4, NMHC, CO, CO"2, NO, NO"2 and SO"2) and meteorological variables (wind speed and direction, air temperature, relative humidity and solar radiation) were employed to predict the concentration of ozone in the atmosphere using both multiple linear and principal component regression methods. Separate analyses were carried out for day light and night time periods. For both periods the pollutants were highly correlated, but were all negatively correlated with ozone. Multiple regression analysis was used to fit the ozone data using the pollutant and meteorological variables as predictors. A variable selection method based on high loadings of varimax rotated principal components was used to obtain subsets of the predictor variables to be included in the regression model of the logarithm of the ozone data. It was found that while high temperature and high solar energy tended to increase the day time ozone concentrations, the pollutants NO and SO"2 being emitted to the atmosphere were being depleted. Night time ozone concentrations were influenced predominantly by the nitrogen oxides (NO+NO"2), with the meteorological variables playing no significant role. However, the model did not predict the night time ozone concentrations as accurately as it did for the day time. This could be due to other factors that were not explicitly considered in this study.