How loud are we really? Differences in temporal and spectral parameters of calls effect sound pressure level metrics and signal propagation, as shown for four species of rail (Rallidae)

Abstract Knowledge of the range of bird vocalisations is crucial not only for estimating the range of species acoustic communication but also for the detection radius of an observer in monitoring projects and expert assessments. Using calibrated sound recordings and parallel acoustic localisation, we assessed three methods of determining the sound pressure level for four species of rail [Corn Crake ( Crex crex ), Water Rail ( Rallus aquaticus ), Spotted Crake ( Porzana porzana ), Little Crake ( Zapornia parva )], whose calls differ significantly in temporal organisation and spectral composition. We calculated sound propagation over long distances, taking into account spherical propagation losses and atmospheric absorption. Atmospheric absorption was calculated including all frequencies contained in the signal. In the case of the Corn Crake in particular, we demonstrated that all spectral components of the signal must be considered when estimating its range. To ensure comparability between studies, we recommend establishing a reference library containing calibrated sound recordings . The results were discussed in the context of their significance for the acoustic communication of the species.