Scott Forrest

Animal movement is complex, depends on many interacting factors, and can be hard to accurately predict. Here we present a novel approach using a deep learning step selection framework: tinyurl.com/2em9yyjb, with lots of supporting info and code! swforrest.github.io/deepSSF/

Stoked to have a cover photo for @EcographyJourna! In our paper doi.org/10.1111/ecog.0… we included temporal dynamics into SSFs, resulting in daily patterns of movement and habitat selection.
Simulating then gave us dynamic spatial predictions
Code! github.com/swforrest/dyna…

Fantastic week at @ISEC_stats_ecol in Swansea last week - such a wide array of inspiring ideas! Thanks to the excellent organising committee and @QUTDataScience for funding. Somehow I also managed to aquire a souvenir... 🥳 #ISEC2024

New paper! Here we GPS-tracked kākā @NZEcosanctuary, and found that younger kākā were more exploratory and had larger home ranges, which increased their exposure to risk outside the safety of the fence. We also present an approach for dynamic space use. conbio.onlinelibrary.wiley.com/doi/full/10.11…

The first paper from my PhD is now on bioRxiv! biorxiv.org/content/10.110…. We incorporated temporal dynamics into the movement, selection and memory processes of an SSF, resulting in dynamic simulations that replicated behaviours of water buffalo #animalmovement