Using the multivariate Hawkes process to study interactions between multiple species from camera trap data

This article models species occurrences at camera traps as a multivariate Hawkes process (MHP). The MHP allows to derive interaction functions between species pairs, where the intensity of interaction is a function of the time elapsed since the first species’ occurrence. In our framework, these interactions can be positive or negative, are asymmetrical and non-parametric.

In this article, we present a proof-of-concept of the MHP on simulated and real data. We notably find interesting patterns of attraction-avoidance between five species of African mammals: zebras and impalas seem to avoid lions, while zebra and impala attract all other herbivores, and wildebeest attracts zebra.

This is my very first peer-reviewed article, published during my PhD. It’s the result of two years of hard work, multiple analyses and extensive discussions with all co-authors. I think the MHP is relevant to think about interactions inferred from camera trap data. It may not be widely applicable yet, because of methodological limitations and (relative) data scarcity, but I hope these limitations can be overcome in the near future.

If you’re interested in camera trap data and/or interspecific interactions, give it a read!

Recommended citation: Nicvert, L., Donnet, S., Keith, M., Peel, M., Somers, M. J., Swanepoel, L. H., Venter, J., Fritz, H., & Dray, S. (2024). Using the multivariate Hawkes process to study interactions between multiple species from camera trap data. Ecology, 105(4), e4237. https://doi.org/10.1002/ecy.4237
Download Paper | Download Slides | Download Bibtex