I am pleased to announce the publication of this paper. The book editors and JASCO Applied Sciences are thanked for helping make this chapter - and the broader volume - available through open access, ensuring these scientific contributions are broadly accessible to researchers, regulators, stakeholders, and the public.

Gisiner, R.C., Loureiro, A., Fertl, D. (2026). Repeated Precautionary Inputs in Complex Underwater Sound Models Drives Overestimation of Predicted Risk. In: Popper, A.N., Sisneros, J.A., Lepper, P.A., Vigness-Raposa, K.J. (eds) The Effects of Noise on Aquatic Life IV. Springer, Cham. https://doi.org/10.1007/978-3-031-94229-7_89-1

Current risk models assessing the effects of anthropogenic sound on marine mammals are complex and multivariate. It is common practice to apply some degree of conservatism, or precaution, to input variables for which uncertainty exists about the most likely real-world values. However, when precautionary inputs are applied to multiple variables in the model, their interactions can lead to accumulated changes in predictions. This can lead to highly inflated risk predictions, ultimately reducing confidence in management decisions based on these outcomes.

Multivariate risk models are also used in other fields, such as meteorology and medicine, employing alternative methods to address uncertainty, methods that preserve the realism of model outputs while still supporting precautionary decision-making. Similar approaches could improve risk assessment for anthropogenic underwater sound by avoiding the overestimation that arises from compounding precaution within the model.