Philip Patton

Job Opportunities | University of Montana, Helena College & UM Western

Post Doc Alert: "large mammal monitoring in northwest Montana...part of a collaborative project between the University of Montana and Montana Fish, Wildlife, and Parks." More: www.schooljobs.com/careers/ummi... 🧪

Estimating species distributions with occupancy models Estimating where species live is one of the most common tasks in ecology. Recreational birders know that every field guide includes a range map showing where a species can be found, as well as habi...

Please check out my recently published demo on fitting singe-season, single-species occupancy models with PyMC.

Ecologists will like PyMC's automatic marginalization feature (adios discrete latent states!) and variety of fast No-U-Turn Samplers (eg, nutpie).

www.pymc.io/projects/exa...

New preprint: we advertise DAFNEE, a database of academia-friendly eco-evo-archaeo journals. 1/6
#AcademicPublishing #ecoevo #archaeology #EthicalPublishing #SocietyJournals #DiamondOpenAccess
doi.org/10.32942/X24...

Gray whale numbers plummet to their lowest level in 50 years. Now far from their recovered level
www.seattletimes.com/seattle-news...

dafnee.isem-evolution.fr

Shoutout to my incredible lab mate, Martin! Such a rad paper!

Thank you to all of my wonderful co-authors, who are too numerous to shout out individually!

False negatives, i.e., an individual was resighted but erroneously marked as a first capture, dictated the optimal strategy in many cases. A 2% increase in the false negative rate translated to a 5% increase in the relative bias in abundance estimates. This rate is easy to compute in practice.

We fed the results into a custom optimization tool to discern the optimal strategy, that is, the optimal method for generating capture histories with an algorithm, for each dataset. We found that true automation was optimal when the algorithm matched images well.

Automating photo-ID creates tradeoffs between reducing effort and increasing estimation error. We explored these tradeoffs with a simulation study, informed by 39 photo-ID datasets representing 24 cetacean species.

I’m thrilled to see my dissertation’s second chapter published in Conservation Biology. In the paper, we found that researchers can use high-performance identification algorithms to reduce the cost of population assessments without biasing abundance estimates. tinyurl.com/ne8tbpw3

Plot showing the correlation between the false negative rate and parameter estimates from a capture-recapture model.

False negatives, i.e., an individual was resighted but erroneously marked as a first capture, dictated the optimal strategy in many cases. A 2% increase in the false negative rate translated to a 5% increase in the relative bias in abundance estimates. This rate is easy to compute in practice.

Results from out optimization for three example datasets.

We fed the results into a custom optimization tool to discern the optimal strategy, that is, the optimal method for generating capture histories with an algorithm, for each dataset. We found that true automation was optimal when the algorithm matched images well.

A schematic showing our methods for the simulation study.

Automating photo-ID creates tradeoffs between reducing effort and increasing estimation error. We explored these tradeoffs with a simulation study, informed by 39 photo-ID datasets representing 24 cetacean species.

I’d love to be added if there’s space!

Exactly. Most procedures like this create more waste than they would have prevented.

A deep learning approach to photo–identification demonstrates high performance on two dozen cetacean species Researchers can investigate many aspects of animal ecology through noninvasive photo–identification. Photo–identification is becoming more efficient as matching individuals between photos is incre...

Hi everyone! I'm a PhD candidate studying new tools for assessing stocks of whales and dolphins. Here is the first chapter of my dissertation, where we explore the strengths and weaknesses of a multispecies individual identification algorithm. besjournals.onlinelibrary.wiley.com/doi/full/10....