Darryl MacKenzieBorn and bred in Dunedin, I started studying at the University of Otago in 1993 for a B.Sc. First year I took a general science course (majoring in Rugby and Drinking), and decided to switch my major to Statistics near the end of the year. Initially I was interested in the application of statistics in industrial settings (quality control etc.), but after completing my B.Sc in 1995, it was wildlife and ecological applications that I became really interested in. I decided to continue studying part-time for a Diploma in Applied Statistics (that I completed in 1998), and by that stage I was working full-time as a Junior Research Fellow for Bryan Manly, the former Professor of Statistics. Bryan had developed a method for detecting trends in environmental monitoring data (such as decreasing water quality readings). It was my role to investigate (using computer simulations) how well the method worked under a variety of scenarios.In 1999 I began my Ph.D. under Richard Barker (a mark-recapture guru), who had joined the Department the previous year. Mark-recapture studies involve capturing animals and placing a unique mark on them, then attempting to recapture them at subsequent trapping occasions. Once the data is collected, models are fit to the data and population sizes, survival and migration rates can be estimated, depending upon how the study was conducted in the field. My Ph.D. topic was to develop some new methods for assessing how well the mark-recapture models represent the real data. Assessing model fit is an important aspect in any modelling exercise, to ensure that the conclusions drawn from the model have some basis in the real world. I completed my thesis in July 2001 and graduated in May 2002. While completing my Ph.D., I spent a year in the U.S. as a Research Associate with North Carolina State University, based at Patuxent Wildlife Research Center. Primarily, I was there to assist Bill Kendall analysing band recovery data from the mid-continental mallard population to estimate non-harvest survival rates. The data consisted of 28 years worth of band returns that had been reported by hunters during the duck shooting season. The other major project I became involved with was developing a new method for estimating the proportion of area occupied by a species, given that the species may not always be detected if present. Such models are applicable to many ecological settings, such as large-scale monitoring programmes, metapopulation studies and habitat modelling. Currently I work as a biometrician for Proteus Research & Consulting Ltd.; a company I co-founded along with David Fletcher (a Senior Lecturer with the Department). We work on a host of different projects, although my main focus is continuing to develop the species occupancy models. We are currently contracted by the U.S. Geological Survey to further develop these models and create user-friendly software (Program PRESENCE). I have also just received a prestigious Marsden Grant to research some of the practical aspects required for these models, #PRC201: Comparison of study designs for the estimation of site occupancy when species are not detected with certainty. | |
