Introducing a method for extracting horizontal migration patterns from data storage tags

While data storage tags typically provide information about depth and temperature, the horizontal position of fish is difficult to obtain. The objective of this study is therefore to introduce a method for reconstructing horizontal migration patterns of fish tagged with DSTs. The method works by: establishing a database on bathymetry and environmental information about the target area, moving a large number of virtual fish between the release and recapture positions using a biased random walk procedure, and then terminate trajectories where the information at position is in disagreement with the tag information. For example a trajectory is terminated if the tag depth is greater than the bottom depth. The method is exemplified with two tag recordings from Northeast Arctic cod (Gadus morhua L.) in the Barents Sea. The results show that termination of impossible trajectories limits the number of potential trajectories between release and recapture positions, and in particular the usage of multiple termination criteria (depth and temperature) proved to be effective in reducing the number of possible trajectories. The method is general, simple to use, and can also be used in combination with geolocation methods.