Abstract
The harbour porpoise (Phocoena phocoena), the only resident cetacean species in the Baltic Sea, is sensitive to a wide range of anthropogenic activities. In the western Baltic Sea, the Belt seas, the Sound and the southern Kattegat the so called “Belt Sea population” has recently been listed as Endangered (HELCOM 2024; from previous Vulnerable). A delineation of the Belt Sea (summer) harbour porpoise management unit, encompassing the core home range area of this population, has been formulated based on morphological studies (Wiemann et al., 2010; Galatius et al., 2012; Lah et al., 2016), satellite telemetry and passive acoustic monitoring (Sveegaard et al. 2015). A critical step towards understanding how sensitive the harbour porpoise is to various threats in different regions of the Baltic Sea is to predict its distribution and determine which areas can be characterised as important for the species.
Predictive modelling of marine mammal habitat is a powerful tool in marine science as it integrates heterogeneity in marine ecosystems and provides important information for ecological studies, management purposes, mitigation of anthropogenic impacts (e.g. Forney et al. 2012) and for understanding the processes that influence interannual and seasonal variability in species distributions (Gilles et al. 2011, 2016, 2025; Becker et al. 2016, 2017, 2018; Pigeault et al. 2024a). Marine mammals exhibit discernible fluctuations in their distributions. However, the drivers responsible for these variations remain unclear, in part due to the dynamic and poorly understood ecological relationships between, for instance, harbour porpoises, their environment and their prey species. In the absence of a comprehensive understanding of these ecological relationships, the here applied species distribution modelling can facilitate the elucidation of these relationships for harbour porpoises in the western Baltic Sea. The objective of this study is to predict and map the long-term summer distribution of harbour porpoises by fitting a habitat-based density model to the high-quality visual survey data collected for harbour porpoises in the “Belt Sea” assessment unit (as defined in Sveegaard et al. 2015) over the last 20 years.
This spatial modelling approach, described in Gilles et al. (2016), utilises physical and biological characteristics as proxies for prey abundance. The ecological theory of species distribution modelling assumes that distribution is at least partly related to environmental variables, and that the relationship between species occurrence and environmental parameters can therefore be used to predict the distribution of the species in question (Austin 2007). Although behavioural factors such as migration, predator avoidance and social interactions influence the distribution of cetaceans, many of the distribution patterns are determined by the foraging response of top predators in a dynamic system (Redfern et al. 2006). This is particularly evident in the case of the harbour porpoise, which exhibits a marked preference for predictable hotspots characterised by high food availability. In the absence of specific prey density data at the necessary spatial resolution for habitat prediction modelling, physical and biological characteristics of the sea are utilised as proxies.
The result of the project can be used for sensitivity mapping of harbour porpoises in the context of marine spatial planning or in assessing sensitivities towards a broad range of anthropogenic activities in the Belt Sea.