Influence of Traffic Input Data Quality on Road Noise Estimates Using the CNOSSOS-EU Method

Area of Science:

• Environmental Science
• Acoustics
• Transportation Engineering

Background:

• Reliable road noise mapping, crucial for environmental and urban planning, depends heavily on accurate traffic input data within the CNOSSOS-EU framework.
• Heterogeneous traffic data sources and measurement practices across Europe introduce uncertainties, impacting noise estimate accuracy and cross-regional comparability.

Purpose of the Study:

• To evaluate the performance of three distinct traffic data collection methods: microwave radar counters, AI-based cameras, and Google API-derived flows.
• To assess the impact of these data sources on road noise simulation accuracy using the CNOSSOS-EU framework.
• To provide guidance on selecting appropriate traffic data collection methods for road noise assessment and strategic mapping.

Main Methods:

• Comparative analysis of traffic flow and vehicle category data from radar counters, AI cameras, and Google API at three test sites in Italy and Romania.
• Road noise simulations using CNOSSOS-EU power models with collected traffic data, compared against in situ noise measurements.
• Development of a secondary analytical approach combining CNOSSOS-EU models with sound propagation software for short-term noise level estimation.

Main Results:

• Microwave radar counters and AI cameras provide reliable traffic inputs for aggregated day/evening/night noise indicators.
• AI cameras may overestimate counts at high traffic volumes, while radar counters can miss flows in complex traffic scenarios.
• Google API-derived flows require careful calibration and perform best above 150 vehicles per hour.
• All tested methods exhibited traffic loss, leading to systematic underestimation of simulated noise levels, particularly evident in short-term analyses.

Conclusions:

• Radar counters and AI cameras are suitable for strategic road noise mapping, but their limitations must be considered.
• Google API data can be useful but requires specific conditions and calibration for accurate noise modeling.
• Traffic data collection method significantly influences noise mapping accuracy, especially for short-term assessments; understanding data loss is critical for reliable road noise evaluation.