Improving estimates of uncertainty in wildlife population assessments

About the Project

Reliable estimates of wildlife population size are fundamental to effective management and conservation. One important component of reliability is the precision (or, conversely, variance) of the estimate. Systematic survey designs (where survey lines or points are laid out in a regular pattern over the study area) tend to produce estimates with lower variance than alternatives – but it can be hard to accurately estimate this variance with currently-implemented estimators tending to over-estimate it. The goal of this PhD will be to investigate and improve variance estimation for systematic survey designs. A particular focus will be on distance sampling surveys, which are widely used to monitor populations from tiny geckos to very large whales. We will test new methods using both simulation studies and real-world datasets. Within this broad topic there is considerable scope for the student to develop their own interests.

The impact of such research would lead to better variance estimation techniques being available to researchers via our distance sampling analysis software. Current distance sampling analysis methods rely on either assuming random designs or on the variance estimators of Fewster et. al. (2009), both of which can over-estimate variability to different degrees depending on the survey characteristics. Fewster (2011) has demonstrated how improved variance estimation can be achieved for some designs using a ‘striplet’ approach. However, this has yet to be incorporated into our analysis software. Doing so would allow investigation via a simulation tool within the software. There are also many designs for which the approach needs extended or other approaches developed – for example point designs, systematic line segment designs and camera trap surveys. The ability to better optimise survey design and more accurately estimate the variability of population estimates will allow for more cost efficient surveys and more effective wildlife management and conservation.

Funding Notes

Fully funded scholarship places (fees, plus stipend of approx. £19,775) are typically available for well-qualified students. UK, EU and other overseas students are all encouraged to apply. Further details of the application and selection procedure are at View Website (pdf - see last page) and View Website.

References

Fewster, R. M. (2011) Variance Estimation for Systematic Designs in Spatial Surveys, Biometrics 67(4):1518-31. https://doi.org/10.1111/j.1541-0420.2011.01604.x
Fewster, R. M., Buckland, S.T., Burnham, K.P., Borchers, D.L., Jupp, P.E., Laake, J.L. and Thomas, L. (2009) Estimating the Encounter Rate Variance in Distance Sampling, Biometrics 65(1):225-236. https://doi.org/10.1111/j.1541-0420.2008.01018.x