As I wanted to explore how combining multiple data sources can improve marine detection, I built a simple toy model using hydrophone data and marked individual tracking methods to estimate cetacean densities. I simulate the movement of individuals with a simple Individual-Based Model (IBM), and for the density estimation, I’ve used Kernel Density Estimation (KDE), though other techniques like Kriging could also be applied.
You can explore the model through this Streamlit web app:
https://antoinebrias-ocean-detection-streamlit-main-uzzcfa.streamlit.app/
For more details on the project, check out my full article here:
https://www.briaslab.fr/blog/?action=view&url=cetacean-detection-simulation-a-toy-model-for-hydrophone-and-marked-individuals-detection