NoiseModelling: an open-source software for outdoor noise mapping
By Pierre Aumond, Nicolas Fortin, Gwendall Petit, Judicaël Picaut
May 2025
From left to right : Nicolas Fortin, Judicaël Picaut, Pierre Aumond and Gwendall Petit in the semi-anechoic room in University Gustave Eiffel: https://www.umrae.fr/en/equipment/semi-anechoic-room
The authors thank all the contributors, inside and outside the UMRAE laboratory, who are absent from this photo
Abstract
Since 2008, the Joint Research Unit in Environmental Acoustics of Gustave Eiffel University has led the development of the free and open-source tool NoiseModelling. This tool enables the production of noise maps at scales ranging from neighborhoods to metropolitan areas (see figure 1), and even national-scale infrastructure such as road and rail transport networks [2].
To achieve this, it integrates the standardized European calculation models CNOSSOS-EU for road and railway noise [7].
From a technical perspective, NoiseModelling is developed in Java and is designed to connect easily with spatial database systems such as H2GIS or PostGIS, facilitating integration with other spatial datasets (e.g population exposure to noise).
A simple user interface makes the tool accessible to researchers, educational institutions, and private companies, making high-quality noise mapping available to a broad audience.
Figure 1. Road noise map calculated with NoiseModelling in the Montaigu area (Vendée, France)
2. Why open-source research software is needed
Noise has important impacts on public health, quality of life or biodiversity. Prolonged exposure to high noise levels can lead to health issues such as stress, sleep disturbances, and cardiovascular disease [5]. In addition, noise can disrupt ecosystems, affect wildlife, and alter the natural behaviors of species [14]. Including noise in environmental assessments is essential to better prevent its negative effects and supports the development of more sustainable planning strategies.
It is therefore crucial to simulate and predict noise propagation in various environments. This provides a quantitative understanding of sound sources, enabling, for example, the assessment of noise mitigation measures and regulatory compliance.
In Europe, since 2019 and Directive (EU) 2015/996 [7], the CNOSSOS-UE emission and propagation method must be used to produce strategic noise maps as described in Directive 2002/49/EC [6].
To do so, there is software available that implement the CNOSSOS-EU method, but these are commercial solutions, with closed source codes; making it difficult, if not impossible, to implement new approaches or models, as research requires.
To address this, NoiseModelling libraries have been created jointly by Gustave Eiffel University and CNRS in an effort to provide engineering and scientific communities around the world with a fully free and open source software to compute and explore noise maps and experiment with the underlying models.
Developed since 2008, NoiseModelling has evolved significantly over time and is part of the Noise-Planet platform, which offers free and open-source scientific tools for environmental noise assessment.
To date, NoiseModelling has been applied in a variety of research projects and practical scenarios, including, but not limited to:
- Standard noise maps of the main french transport infrastructures [3]
- Dynamic noise maps from the coupling with several traffic models such as MATSim, SUMO and Symuvia [9, 4, 12]
- Sensitivity analysis of various aspects of the CNOSSOS-EU model [1]
- Dynamic noise map from noise sensors with data assimilation methods [11]
- Noise quantification of specific sources such as sirens [13] or drones [10]
3. How it works
NoiseModelling is structured into four modular Java libraries (figure 2): Emission (for calculating traffic-related sound power levels), Pathfinder (for determining source-receiver cut profiles), Propagation (for computing sound attenuation), and JDBC (for data management and inter-library communication – This module supports integration with spatial databases such as H2GIS and PostgreSQL/PostGIS). The tool is accessible via both a graphical user interface and command line, facilitating interactive use and process automation.
Figure 2. The NoiseModelling architecture
NoiseModelling incorporates CNOSSOS-EU emission models for both road and rail traffic. The system’s modular architecture supports the integration of additional source types and emission models. Indeed, custom noise emissions can be defined through direct input of sound power levels for point or line sources. The CNOSSOS-EU propagation method estimates noise levels at receiver locations by incorporating various noise sources and detailed environmental factors such as buildings, terrain, and meteorological conditions [7].
Acoustic propagation between a sound source and a receiver is computed by identifying possible sound paths: direct, reflected, and diffracted (see figure 3). Direct paths may include diffraction over terrain or building edges. Reflected paths result from specular reflections off walls, limited by a predefined reflection order, while diffracted paths occur around vertical building edges. Path-finding, the most computationally intensive step, uses an R-Tree structure to efficiently manage scene geometry. To overcome the exponential complexity of the image-source method for reflections, NoiseModelling employs a receiver-based view cone strategy, identifying valid reflections by intersecting source positions with precomputed view cones. This reduces invalid paths and improves efficiency. The algorithm outputs vertical cross-sections detailing obstacles like buildings and terrain.
Figure 3. Diffractions paths on horizontal and vertical planes
Although the method relies on vertical cut profiles from the path-finding algorithm, the system can also support other propagation models than CNOSSOS-EU.
The codebase is extensively tested with unit tests, including validation against official reference documents and independent implementations. The propagation and pathfinding models are verified against ISO/TR 17534-4:2020 [8], and the emission models are compared with external implementations. Discrepancies are traceable through dedicated test cases, with key validation results documented separately (see section 5).
4. Community
To support new users, NoiseModelling offers comprehensive online documentation and tutorials. Since 2021, the community has convened annually at the NoiseModelling Days (see Section 5) – one- or two-day events held online or in hybrid format – featuring training sessions for beginners and presentations by experienced users. These meetings serve as platforms for sharing progress, future directions, and fostering collaboration. The NoiseModelling team acknowledges the valuable contributions of the broader user and developer community.
Feel free to join the community. The following resources will help you take the necessary steps with NoiseModelling.
5. Resources
• Official web page: https://noise-planet.org/noisemodelling.html
• Code repository: https://github.com/Universite-Gustave-Eiffel/NoiseModelling/
• Online documentation: https://noisemodelling.readthedocs.io/en/latest/
• NoiseModelling Days: https://noise-planet.org/noisemodelling days.html
References
[1] Pierre Aumond, Arnaud Can, Viven Mallet, Benoit Gauvreau, and Gwenael Guillaume. Global sensitivity analysis for road traffic noise modelling. Applied Acoustics, 176:107899, May 2021.
[2] Pierre Aumond, Sophie Cariou, Olivier Chiello, David Ecotiere, Adrien Le Bellec, Damien Maltete, Claire Marconot, Nicolas Fortin, Sylvain Palominos, Gwendall Petit, et al. Strategic noise mapping in france to 2023: Coupling a national database with the open-source model noisemodelling. In INTER-NOISE and NOISE-CON Congress and Conference Proceedings, volume 265, pages 2617–2624. Institute of Noise Control Engineering, 2023.
[3] Pierre Aumond, Sophie Cariou, Olivier Chiello, David Ecoti`ere, Adrien Le Bellec, Damien Maltete, Claire Marconot, Nicolas Fortin, Sylvain Palominos, Gwendall Petit, and Judica¨el Picaut. Strategic Noise Mapping in France to 2023: Coupling a national database with the open-source model NoiseModelling. INTER-NOISE and NOISE-CON Congress and Conference Proceedings, 265(5):2617–2624, February 2023.
[4] Sacha Baclet, Kaveh Khoshkhah, Mozhgan Pourmoradnasseri, Romain Rumpler, and Amnir Hadachi. Near-real-time dynamic noise mapping and exposure assessment using calibrated microscopic traffic simulations. Transportation Research Part D: Transport and Environment, 124:103922, November 2023.
[5] Xia Chen, Mingliang Liu, Lei Zuo, Xiaoyi Wu, Mengshi Chen, Xingli Li, Ting An, Li Chen, Wenbin Xu, Shuang Peng, Haiyan Chen, Xiaohua Liang, and Guang Hao. Environmental noise exposure and health outcomes: an umbrella review of systematic reviews and meta-analysis. European Journal of Public Health, 33(4):725–731, August 2023.
[6] European Parliament and Council. Directive 2002/49/ec of the European parliament and of the council of 25 june 2002 relating to the assessment and management of environmental noise – declaration by the commission in the conciliation committee on the directive relating to the assessment and management of environmental noise, 2002.
[7] European Parliament and Council. Commission Directive (EU) 2015/996 of 19 May 2015 establishing common noise assessment methods according to Directive 2002/49/EC of the European Parliament and of the Council, July 2015.
[8] ISO. ISO/TR 17534-4:2020 : Acoustics — Software for the calculation of sound outdoors — Part 4: Recommendations for a quality assured implementation of the COMMISSION DIRECTIVE (EU) 2015/996 in software according to ISO 17534-1, 2020.
[9] Valentin Le Bescond, Arnaud Can, Pierre Aumond, and Pascal Gastineau. Open-source modeling chain for the dynamic assessment of road traffic noise exposure. Transportation Research Part D: Transport and Environment, 94:102793, May 2021.
[10] Ingrid LEGRIFFON and Elise RUAUD. Drone fleet noise impact calculation – a methodology. INTER-NOISE and NOISE-CON Congress and Conference Proceedings, 270(10):1675–1680, October 2024.
[11] Antoine Lesieur, Vivien Mallet, Pierre Aumond, and Arnaud Can. Data assimilation for urban noise mapping with a meta-model. Applied Acoustics, 178:107938, July 2021.
[12] G. Quintero, P. Aumond, A. Can, A. Balastegui, and J. Romeu. Statistical requirements for noise mapping based on mobile measurements using bikes. Applied Acoustics, 156:271–278, December 2019.
[13] Jonathan Siliezar, Pierre Aumond, Arnaud Can, Paul Chapron, and Matthieu Peroche. Case study on the audibility of siren-driven alert systems. Noise Mapping, 10(1), January 2023. Publisher: De Gruyter Open Access.
[14] Romain Sordello, Ophelie Ratel, Frederique Flamerie De Lachapelle, Clement Leger, Alexis Dambry, and Sylvie Vanpeene. Evidence of the impact of noise pollution on biodiversity: a systematic map. Environmental Evidence, 9(1):20, September 2020.
https://www.linkedin.com/in/gwendall-petit-2084b829
https://www.linkedin.com/in/judica%C3%ABl-picaut