Looking for a course on Species Distribution Modelling (SDMs)?

PR Statistics has the right course for you—whether you’re just starting out, building on existing skills, or ready to explore Bayesian methods for SDMs.

Please feel free to share!

Species distribution models (SDMs) are especially valuable in marine mammal research, where direct observation is often limited and costly. These courses provide the tools to predict habitat use, identify critical areas for conservation, and assess how environmental change—such as shifting oceanographic conditions or climate change—may affect species ranges. By learning methods from introductory to advanced, including Bayesian approaches, marine mammal researchers can better integrate ecological data with spatial modelling, improve the accuracy of predictions, and strengthen the scientific basis for management and conservation decisions.

For beginners:
Start with our introduction to SDMs and ENMs (SDMR06). This course requires no prior SDM experience—just a basic understanding of R. If you don’t yet have the basics, check out our free recorded courses first.
SDMR06 covers the core foundations of SDM: how to calculate and interpret niche models, choose appropriate modelling approaches, run standard algorithms, compare model outcomes, build ensemble predictions, and apply models to your own data.
🔗 Learn more about SDMR06

For intermediate users:
Take your modelling further with our advanced course (ASDM01). This course goes beyond baseline workflows into model refinement, accuracy improvement, and the integration of physiological and environmental realism through mechanistic and simulated species models.
🔗 Learn more about ASDM01

For those looking to improve model accuracy with Bayesian methods:
Explore our Bayesian SDM course (SDMB07). This course covers the entire Bayesian modelling pipeline: from data preparation and model fitting, to cross-validation, performance evaluation, and interpreting variable importance with tools like partial dependence plots.
SDMB07 introduces Bayesian Additive Regression Trees (BART)—a modern approach that produces robust predictions, reduces overfitting, and explicitly quantifies uncertainty. Unlike traditional or mechanistic SDMs, Bayesian SDMs allow for a clear representation of uncertainty, providing posterior means, credible intervals, and uncertainty surfaces for deeper insight into prediction confidence.
🔗 Learn more about SDMB07

Email oliver@prstats.org with any questions.

Oliver Hooker PhD.

PR stats