Post-Doctoral Associate in Statistical Ecology (Updated May 11, 2017)
I am currently looking to fill a position for a Post-Doctoral Associate (PDA) housed in either the Department of Biology or the Department of Statistical and Actuarial Sciences at the University of Western Ontario. The PDA will work closely with biologists at Western and other universities to develop and test new methods for anlaysing complex ecological data. The position has a term of 2 years at $50,000 (CAD) per year. Complete details are available here.
The expected start date is between August 2017 and January 2018. I will begin reviewing applications on July 1, 2017, and continue until the position is filled.
Doctoral and Masters Students
I able to supervise students in both the Department of Biology and the Department of Statistical and Actuarial Sciences at Western University and have many projects available for ecology students with a strong quantitative background or for statistics students with an interest in biological data. Here are some examples of the types of project that might be available:
PhD 1: Simple Methods for Analysing Big Mark-Recapture Data
The term Big Data refers to the tendency for data to be collected at rates that exceed our analysis and computing abilities. In ecology, big data sets arise from new technologies, like motion activated camera traps or data loggers, that are able capture data on the order of seconds or from the dedicated work of field technicians or citizen scientists who tag or photograph and follow the movements of 1000s of individuals from a population. We are currently working on methods that allow for fast analysis of big mark-recapture data sets by focusing on the subset of data that is most informative about the parameters of interest. We have applied our methods for fitting Cormack-Jolly-Seber based models to large data sets, and have shown that dramatic reductions in computing speed can be obtained with little loss of precision and the introduction of no bias. Further work is needed to extend these methods to more complicated mark-recapture data types and data from other sources.
MSc 1: Analysis and Validation of Motus Sentinels
In 2016 I began collaborating with researchers using the Motus wildlife tracking system to monitor the movements of small neo-tropical migratory birds. Our interest lay in understanding the detectability of birds near the VHF receiving towers and how modelling how detectability may be affected by the birds surrounding and environmental conditions. In the spring of 2016 we collected data from several sentinels — stationary tags affixed to model birds and place at fixed locations for several weeks. An MSc student is required to analyse the data that was collected and, further, to validate the bird models to understand how well our results will predict the detectability of live birds in the field. Projects in collaboration with researchers from the Department of Biology are also available for students wishing to use the Motus technology to monitor bird movements and to answer questions about migratory behaviour and the use of stopover locations.
MSc 2: Bayesian Doubly Hierarchical Generalized Linear Models
I am currently working with Dr. David Westneat from the Department of Biology at the University of Kentucky to model factors affecting the variability in birds behaviour and their ability to change phenotypes in response to changing conditions — so called plasticity. Data collected through Dr. Westneat’s field experiments are analysed with doubly hierarchical generalized linear (DHGLM) models which extend classical GLM by allowing for both the mean and variance of the response to depend on measured covariates and individual random effects. As part of this collaboration I have developed a (yet to be released) R package that allows facilitates fitting of DHGLM the Bayesian framework using JAGS as a backend to conduct MCMC sampling. I am looking for students who are familiar with R and can extend the package by: reimplementing the backend in STAN, developing automatic graphics with ggplot2, implementing further models in the DHGLM framework and testing these models through simulation and application to real data sets.
Please contact me at firstname.lastname@example.org if you are interested in a position with my lab.