Noise estimation model development using high-resolution transportation and land use regression.

Area of Science:

• Environmental Science
• Urban Planning
• Acoustics

Background:

• Noise pollution is a pervasive 21st-century issue.
• Existing noise prediction models primarily focus on road traffic.
• Accurate noise mapping is crucial for urban environmental assessment.

Purpose of the Study:

• To adapt and apply the land use regression (LUR) modeling methodology for noise pollution assessment.
• To estimate noise levels during both rush hour and off-peak periods.
• To evaluate the model's performance in major Israeli cities.

Main Methods:

• Utilized short-term noise measurements (20-minute intervals) for model development.
• Integrated Geographic Information System (GIS)-based predictors with traditional traffic predictors.
• Employed a ten-fold cross-validation approach for 'out of sample' performance evaluation.

Main Results:

• Achieved strong model fits with cross-validated R² values of 0.79 for Tel Aviv and 0.52 for Beer Sheva during rush hour.
• Demonstrated robust performance when the Tel Aviv model was validated with independent data from Bat Yam (R² of 0.93).
• The LUR models effectively estimated noise pollution across different times of day.

Conclusions:

• The land use regression (LUR) approach is a viable method for high-resolution spatial noise pollution estimation.
• This methodology enables accurate mapping of environmental noise for assessment purposes.
• The study validates the effectiveness of LUR modeling in diverse urban settings.