Natural and bio-based construction materials such as bamboo, cork, or natural fiber composites offer a promising solution for enhancing the environmental sustainability of buildings. In this sense, the paper presents an experimental thermo-acoustic characterization of four common Ecuadorian natural fibers, abaca (Musa textilis), cabuya (Furcraea andina), coir (Cocos nucifera), and totora (Schoenoplectus californicus). Different densities were considered, from 85 kg/m³ (Cabuya) to 244 kg/m³ (totora), to thermo-acoustically characterize the samples built with these fibers, by means of the guarded-hot-plate (GHP) and impedance tube methods in-lab experimental benches. The exhaustive original characterization of the evaluated natural fiber composites showed a promising overall thermo-acoustic behavior. The thermal conductivity of the fibers was around 0.04–0.06 W/m·K and, therefore, comparable to other materials such as polystyrene, polyurethane, or aerogel that are already utilized for similar applications. On the other hand, the sound-absorption properties of the evaluated fibers are also very competitive, but strongly affected by the thickness of the sample, with noise reduction coefficient NRC ranging from 0.12 to 0.53. Consequently, the production and distribution of these materials in the Ecuadorian market for thermal insulation and acoustic conditioning constitute an alternative characterized by good technical performance, which, compared to synthetic composites used in the construction industry for similar duties, is ecological, sustainable, and has low built-in energy consumption.

Sustainability in construction is a growing concern due to the significant polluting waste generated before, during, and after a building’s life cycle. The use of natural materials can significantly reduce the environmental footprint in obtaining, manufacturing, transportation, execution, use, maintenance, and demolition of the building, especially when locally sourced. Natural fibers, in particular, can be used in room acoustics, offering good acoustic absorption while meeting sustainability goals. The objectives of this paper are to evaluate cabuya fiber, grown in Ecuador, as an acoustic absorbing material and to introduce a novel approach using machine learning to simulate the material’s acoustic properties. Eight samples of cabuya fiber, bound with a solution of water and Ca(OH)₂, were prepared with thicknesses between 12 and 30.6 mm. The sound absorption coefficients (SACs) were calculated using an impedance tube, following international standards. A Gaussian regression model was built for the predictions. The results showed that the 30.6 mm sample achieved maximum absorption coefficients of 0.91 at 2 kHz and 0.9 at 5 kHz. The model predictions are very accurate, with a mean square error of just 0.0002. These findings offer valuable insights into using cabuya fiber and advanced predictive models to enhance building acoustic performance and reduce environmental impact.

Predicting outdoor sound from large-scale sound reinforcement systems remains difficult at low frequencies, where coherent source arrays interact with the ground and generate interference patterns that are not represented by energy-summation models. This study combines a screening stage within ISO 9613-2, implemented in commercial software, with a coherent propagation model to improve band-limited forecasts between 25 and 250 Hz. A Morris one-at-a-time screening applied to an open-air context case shows that ground characterisation and array directivity dominate predicted levels in this band, whereas temperature and humidity are comparatively weak drivers. Guided by this ranking, a three-dimensional wide-angle parabolic equation (3D WAPE) solver in Python was implemented, using an alternating-direction implicit marching scheme. The model retains phase through a complex field, includes ISO 9613-1 atmospheric absorption, wind-profile refraction terms, and an impedance ground boundary based on the Miki formulation with optional surface roughness. Source radiation is initialised from measured complex directivity, provided in .xhn files, and applied via a directional starter method. Predictions are validated against one-third-octave measurements at 80, 120 and 200 m for a subwoofer-only and a full-system configuration. The model reproduces observed spectra and range-dependent attenuation to within ±4 dB for both the subwoofer and full system across 25–250 Hz, with a localised discrepancy at 125 Hz. The results highlight the need to represent coherent radiation and ground interaction when assessing low-frequency sound immission from outdoor concert systems.