Impact of the choice of road traffic speed data on modelled environmental noise
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates the impact of traffic speed data selection on modeled environmental noise levels, addressing a critical uncertainty in noise mapping where complete speed datasets are rarely available. The research was motivated by the need to quantify exposure errors for epidemiological studies, specifically within the EU-funded Equal-Life project examining the relationship between environmental noise and mental health. The authors compare three speed estimation approaches: national speed limits (NSL), half the national speed limit (HNSL) to approximate built-up area conditions, and a novel method deriving average speeds from satellite imagery (SAT). The methodology involved modeling noise exposure for 4,394 residential addresses in Barcelona using the CNOSSOS-EU algorithm implemented in the ‘NoiseModelling’ software. The satellite-derived speed dataset was generated using high-resolution WorldView 2/3 imagery and a deep learning object detection model (YOLOv3) to identify vehicles. Speeds were calculated by measuring the spatial displacement of vehicles between multi-spectral sensor captures occurring less than one second apart. This SAT data was benchmarked against Google’s Directions API, showing a Mean Gross Error of 8.4 km/h and an R² of 0.71. Noise estimates for Lden (day-evening-night level) were calculated for both the loudest and quietest façades of each residence, varying only the speed inputs while keeping traffic flow and composition constant. The results demonstrate that the choice of speed data significantly alters noise estimates, with differences up to 13 dB(A). Switching from NSL to SAT speeds resulted in a median decrease of 9.3 dB(A). Comparisons between HNSL and SAT showed smaller variations, ranging from -2.2 to 4 dB(A) with a median decrease of 1.7 dB(A). Specifically, average differences between NSL and SAT reached 12.2 dB(A) for tertiary roads, while differences between HNSL and SAT were substantially lower, often under 1 dB(A) for trunk and residential roads. For the loudest façade, 42% of addresses showed a greater than 2 dB(A) difference between HNSL and SAT estimates. The study concludes that using national speed limits leads to substantial overestimations of noise levels, whereas using half the national speed limit offers a more accurate, practical alternative when detailed speed data is unavailable. While the satellite method provides higher precision, its high cost and processing demands limit its scalability for large regions. The authors suggest that HNSL is a viable proxy for traffic speed in urban noise modeling, though they note that errors from speed assumptions can be equivalent to significant changes in traffic flow. Future work should focus on validating satellite-derived speeds against measured data and developing more granular speed profiles that account for junctions and mid-block variations.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-25 |
| archive | success | canonical_url | — | — | 1 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-26 |
| chunk | success | chunk | — | — | 1 | 2026-06-26 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-26 |
| enrich | success | openalex | — | — | 1 | 2026-06-26 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-26 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.