Application of Multiple Linear Regression Technique to Predict Noise Pollution Levels and Their Spatial Patterns in the Tarkwa Mining Community of Ghana
American Journal of Mathematical and Computer Modelling
Predicting and preventing intraurban noise levels in our communities are very challenging for urban planning, epidemiological studies and environmental management, especially in the developing world. Most existing noise-predicting models are limited in providing changes in noise levels during intraurban development and the corresponding noise pollution. In this study, noise levels were measured at 50 purpose-designed monitoring stations and then a land-use regression model was developed for the
... s developed for the intraurban noise prediction applying the multiple linear regression (MLR) technique. The measured and the predicted noise levels were compared. These were further compared with noise estimates from a standard noise model, Lyons Empirical model. The results from the developed MLR model did not show any significant differences in the patterns as compared with those of the Lyons Empirical model. The model performance indicators showed a standard deviation of 1.585, high correlation (R) of 0.98, R 2 of 0.961 and RMSE of 1.569. The resulting maps showed a heterogeneous distribution of the noise pollution levels in the community. This confirms the usefulness of the method for assessing the spatial pattern of noise pollution in a community. This makes it a useful tool for urban planning, epidemiological studies and environmental management. Sites LAeq L10 L90 TNI LNP dBA dBA dBA dBA dBA Where LAeq is A-weighted equivalent sound pressure level; L10 and L90 are the exceedence percentiles, average sound level, LD, the day-night average sound level, LDN, the noise pollution level, LNP and the traffic noise index, TNI.