Urban noise pollution is a critical environmental stressor in cities, yet its drivers in dense, historic cores are
poorly quantified. This study investigates the pedestrian-level acoustic environment in Isfahan, Iran, a historic
city grappling with severe traffic congestion, to guide targeted policy interventions. We measured equivalent
sound pressure levels (LAeq), exclusively representing the overall exposure, at 120 roadside locations during three
diurnal periods, revealing a significant increase from morning (58.9 ± 2.49 dB(A)) to afternoon (71.6 ± 3.34 dB
(A)). These measurements were synthesized into a composite Noise Index (NI) using Principal Component
Analysis, where the first component explained 78.5% of the variance. A Generalized Linear Model (GLM)
quantified the influence of fourteen urban variables on the NI. The model explained 86.3% of the variance (RMSE
= 0.565), identifying Vehicle Flow Rate (VFR) and Commercial Land Use (CLU) as the dominant statistical
predictors (drivers). A sensitivity analysis confirmed their paramount influence, with VFR and CLU exhibiting the
largest effect sizes (Δ = 2.943 and 2.136, respectively). In contrast, small, discontinuous green patches, road
width, and building height were statistically insignificant (p > 0.05). Beyond quantifying noise levels, the results
provide a contextualized assessment of how mobility- and activity-related factors shape exposure in compact
historic settings. These findings underscore a strong statistical association, demonstrating that in compact historic
settings, pedestrian-level noise is most strongly linked to mobility systems and land-use function rather than
to the measured elements of physical morphology.
