#159 Bayesian Occupancy Models, with Matthijs Hollanders

Support & Resources → Support the show on Patreon:   / learnbayesstats   → Bayesian Modeling Course (first 2 lessons free): https://topmate.io/alex_andorra/1011122 Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work at https://bababrinkman.com/ ! Takeaways: Q: What is a Bayesian occupancy model and what problem does it solve? A: An occupancy model accounts for the fact that you don't always detect a species when surveying for it, especially when the species is rare. A naive count of where you found it underestimates true occupancy. The model adds a repeated-measures component: you visit each site multiple times, and from the pattern of detections vs. non-detections it estimates a detection probability. Matthijs framed it as a zero-inflation structure where the zero-inflation happens at the site level rather than the observation level -- which keeps the model conceptually simple, just a standard GLM with a Bernoulli “is the species here at all?” stacked on top of a detection-rate process. Q: What are Automated Recording Units and why don't traditional occupancy models handle them well? A: ARUs are camera traps and acoustic monitors that record continuously over deployment periods of days, weeks, or months. The data they produce isn't a sequence of discrete human-led surveys; it's a continuous-time observation stream. Traditional occupancy models were designed for the discrete case -- a human visits a site, records yes or no, goes home. With ARUs, the question becomes how to bin or threshold the continuous data without losing the richer signal it actually contains. Q: Why does occARU model raw counts instead of binary detect or non-detect? A: Most occupancy software collapses each survey to a 1 or a 0 before modeling anything. Matthijs's argument is that this throws away information by construction -- if you saw 17 kangaroos in one window and 3 in the next, that variation matters and shouldn't be discarded a priori. Modeling counts directly gives you everything an occupancy model would, plus rich detection-rate dynamics over time and across species. You can still ignore the count side at interpretation time -- but the data deserves to be modeled before it's thresholded. Q: How does occARU handle autocorrelated detections like a group of animals lingering in front of the camera? A: With a thinning window. If 20 kangaroos stand in front of a camera for two hours, the raw counts inflate detections in a way that isn't really new information -- it's the same animals, repeatedly. occARU lets you set a thinning interval (30 minutes is a typical choice) and retain the maximum count observed within each window. The two-hour kangaroo episode collapses to four detections, each carrying the genuine peak group size in that interval. This preserves the count signal while avoiding pseudoreplication from a single behavioral event. Q: How do multi-species hierarchical Gaussian processes fit into the occupancy framework? A: Repeated measures at sites and surveys imply you should have random effects at both levels. Sites that are spatially close, and surveys close in time, should resemble each other -- that's a Gaussian process. occARU stacks species on top: the GP is structured as a matrix normal with a row covariance over sites and a column covariance over species. From the column covariance you can directly read off which species tend to co-occur in space and time. And computationally, you only need one Cholesky decomposition across all species rather than one per species. Full takeaways at https://learnbayesstats.com/episode/1... Chapters: 00:12:14 What is an occupancy model and what problem does it solve? 00:16:16 What are Automated Recording Units and why do they need different models? 00:18:45 What is the occARU R package and why does it exist? 00:23:55 Why does occARU model counts directly rather than binary detection? 00:26:38 What does multi-species hierarchical modeling with Gaussian processes look like? 00:32:22 How does occARU implement Gaussian processes efficiently? 00:41:01 Why are Gaussian processes such a powerful but tricky modeling tool? 00:44:11 What is variance decomposition with global-local shrinkage priors? 00:49:02 How does occARU leverage recent Stan features for zero-sum constraints? 00:57:37 When does within-chain parallelization actually help? 01:01:30 How does Monte Carlo integration reduce high Pareto-k values? 01:15:27 When does occARU underperform and what's on the roadmap? Thank you to my Patrons (https://learnbayesstats.com/#patrons) for making this episode possible! Links from the show at https://learnbayesstats.com/episode/1...