The monetary and logistical challenges of collecting biodiversity data in ocean environments has led to the under-representation of marine species in publicly available data. And with up to 50 per cent being collected only from coastal areas, novel sampling approaches are essential for scaling up surveys to represent marine communities more accurately.
Using the Mediterranean and its community of marine predators as a study case, this paper explored whether sampling along commercial ferry routes can yield survey outcomes. As a baseline representation of "perfect knowledge", the authors built stacked species distribution models of 43 species based on six environmental variables and all available occurrence records from GBIF and OBIS.
Selecting 15 Mediterranean ferry routes of varying length, the authors simulated sampling of different sizes and strategies, using the baseline occurrence dataset as the source. They then created new distribution models based on the ferry route network, assessing correlation between these and the "perfect knowledge" model.
The ferry route model was more closely related to the "perfect knowledge" model than when randomly sampling across the entire Mediterranean, showing that a geographically biased sampling strategy can work, provided that it covers wide areas and diversified habitats.
While the study does not address specific sampling methods, the findings are clearly relevant for upscaling sampling for emerging biodiversity monitoring techniques such as automated eDNA sampling. The approach could also be expanded to other commercial vessel types, potentially covering the entire global shipping network.