AbstractThe impact of temporal aspects of noise data on model development and intra-urban variability on environmental noise levels are often ignored in the development of models used to predict its spatiotemporal variation within a city. Using a Land Use Regression approach, this study develops a framework which uses routine noise monitors to model the prevailing ambient noise, and to develop a noise variability map showing the variation within a city caused by land-use setting. The impact of data resolution on model development and the impact of meteorological variables on the noise level which are often ignored were also assessed. Six models were developed based on monthly, daily and hourly resolutions of both the noise and predictor data. Cross validation highlighted that only the hourly resolution model having 59%explanatory power of the observed data (adjusted R2) and a potential of explaining at least 0.47% variation of any independent dataset (cross validation R2), was a suitable candidate among all the developed models for explaining intraurban variability of noise.In the hourly model, regions with roads of high traffic volumes, with higher concentrations of heavy goods vehicles, and being close to activity centreswere found to have more impact on the prevailing ambient noise. Road lengthswere found to be the most influential predictors and identified as having an impact on the ambient noise monitors.
The Seismic Potential in the Seismic Gap Between the Wenchuan and Lushan Earthquakes Revealed by the Joint Inversion of Receiver Functions and Ambient Noise Data