Accuracy of prediction methods for rain noise levels

Determining levels of rain noise can be an important design aspect for various types of buildings, particularly education facilities. Measurements of rain noise levels can be carried out using relevant ISO standards (ISO 140-18:2006, ISO 10140-1:2010/Amd 2:2014, ISO 10140-5:2010 Amd 1:2014) which specify artificial rain types with, for example, constant drop size. While this can provide an indication of the levels of rain noise that might occur in practice, it is not necessarily trivial to estimate real or natural levels of rain noise from the artificial rain noise measurements. The relationship between natural and artificial rain noise levels can be better understood using prediction models. Predicting rain noise levels can also prove helpful in evaluating a range of alternative roof construction types, as rain noise measurement data is often only available for a limited number of construction types. This paper provides a brief overview of a prediction method for rain noise. Emphasis is given to relevant tolerances on prediction accuracy, with reference made to variation in laboratory rain noise level measurements (that is, measurement reproducibility). Prediction results are presented for a number of different constructions to demonstrate the extent of agreement between the prediction method and laboratory results.