Ben Stevenson Senior Lecturer Department of Statistics University of Auckland Private Bag 92019 Auckland 1142 New Zealand Office: Room 303.326 E-mail: ben.stevenson@auckland.ac.nz Phone: +64 9 923 8474 |
Research interests
- Ecological statistics
- Capture-recapture methods
- Spatial statistics
- Statistical computing
Biography
I am a Senior Lecturer in the Department of Statistics at the University of Auckland, New Zealand.
I completed BSc (Hons) and MSc degrees at the University of Auckland before moving to the University of St Andrews, United Kingdom, where I graduated with a PhD in 2016. I remained at St Andrews for a twelve-month research fellowship, joining this department in January 2017.
I develop statistical methods and software to estimate ecological parameters of interest, usually animal abundance or density. My recent work has met statistical challenges that arise when instruments like microphones, video-cameras, and drones are deployed to detect animals on wildlife surveys. New technologies have the potential to collect mountains of data at a low cost, but complicate estimation of ecological parameters because less information is encoded within each detection. For example, we can easily identify an individual on a live-trapping survey via a securely fixed ID tag, but perhaps cannot do so from a fleeting glimpse in a video. Most of my current work involves capture-recapture or spatial capture-recapture models.
Along with ecological statistics, I have research interests in spatial statistics, statistical computing, and applied statistics in general. I am currently an associate editor at Biometrics, a journal that publishes statistics research relevant to the biosciences. To get more of a feel for my research, see the section on my current projects below.
I currently teach STATS 399, our statistics capstone course. In 2025 I will also be teaching STATS 10X, our introductory course in statistics.
My CV is available here.
Research grants
- co-PI, Marsden Standard Grant, NZ$712k,
2024–2027.
Fast statistical methods for enigmatic sensor data.
With Rachel Fewster, Jesse Goodman, Martin Hazelton (co-PIs), and Andrew Robinson (AI).
- PI, Marsden Fast-Start Grant, NZ$300k,
2020–2024.
Estimating animal population size in an unobservable spatial obstacle course.
With Rachel Fewster and David Borchers (AIs).
- AI, Marsden Grant, NZ$680k,
2018–2021.
Cells and whistles: Supercharging our biodiversity monitoring toolkit using genetic and acoustic records.
With Rachel Fewster (PI), David Borchers, and Stephen Marsland (AIs).
Peer-reviewed publications
My Google Scholar citations page can be found here.
- Setyawan, E., Erdmann, M. V., Mambrasar, R., Ambafen,
O., Hasan, A. W., Izuan, M., Mofu, I., Putra, M. I. H.,
Sianipar, A. B., Constantine, R., Stevenson, B. C.,
and Jaine, F. R. A. (in press) Spatial connectivity of
reef manta rays across the Raja Ampat archipelago,
Indonesia. Royal Society Open Science.
- van Dam-Bates, P., Papathomas, M., Stevenson,
B. C., Fewster, R. M., Turek, D., Stewart, F. E. C.,
and Borchers, D. L. (2024) A flexible framework for
spatial capture-recapture with unknown
identities. Biometrics, 80(1),
ujad019. (link)
- Durbach, I., Chopara, R., Borchers, D. L., Phillip, R.,
Sharma, K., and Stevenson, B. C. (2024) That's not
the Mona Lisa! How to interpret spatial capture-recpature
density surface
estimates. Biometrics, 80(1),
ujad020. (link).
- Martin, L. H., Hepinstall-Cymerman, J. H., Chandler,
R. B., Cooper, R. J., Parrish, M. C., Hao, L.,
and Stevenson, B. C. (2024) Estimating owl
population density using acoustic spatial
capture-recapture. Journal of Raptor
Research, 58(1),
1–13. (link).
- McGrath, S., Liu, J., Stevenson, B. C., and
Behie, A. M. (2023) Density and population size estimates
of the endangered northern yellow-cheeked crested
gibbon Nomascus annamensis in selectively logged
Veun Sai-Siem Pang National Park in Cambodia using
acoustic spatial capture-recapture methods. PLoS
ONE, 18(11),
e0292386. (link).
- Stevenson, B. C., Fewster, R. M., and Sharma,
K. (2022) Spatial correlation structures for detections of
individuals in spatial capture-recapture
models. Biometrics, 78(3),
963–973. (link)
- Setyawan, E., Stevenson, B. C., Erdmann, M. V.,
Hasan, A. W., Sianipar, A. B., Mofu, I., Putra, M. I. H.,
Izuan, M., Ambafen, O., Fewster, R. M., Aldridge-Sutton,
R., Mambrasar, R., and Constantine, R. (2022) Population
estimates of photo-identified individuals using a modified
POPAN model reveal that Raja Ampat's reef manta rays are
thriving. Frontiers in Marine Science, 9(1),
1014791. (link)
- Borchers, D. L., Nightingale, P., Stevenson,
B. C., and Fewster, R. M. (2022) A latent capture
history model for digital aerial
surveys. Biometrics, 78(1),
274–285. (link)
- Setyawan, E., Erdmann, M. V., Mambrasar, R., Hasan, A.,
Sianipar, A., Constantine, R., Stevenson, B. C.,
and Jaine, F. R. A. (2022) Residency and use of an
important nursery habitat, Raja Ampat's Wayag Lagoon, by
juvenile reef manta rays (Mobula
alfredi). Frontiers in Marine
Science, 9(1),
815094. (link)
- Setyawan, E., Stevenson, B. C., Izuan, M.,
Constantine, R., and Erdmann, M. V. (2022) How big is that
manta ray? A novel and non-invasive method for measuring
reef manta rays using small
drones. Drones, 6(3),
63. (link)
- Baron, H. R., Stevenson, B. C., and Phalen,
D. N. (2021) Comparison of in-clinic diagnostic testing
methods for Macrorhabdus ornithogaster. Journal
of Avian Medicine and Surgery, 35(1),
37–44. (link)
- Stevenson, B. C., van Dam-Bates, P., Young,
C. K. Y., and Measey, J. (2021) A spatial
capture-recapture model to estimate call rate and
population density from passive acoustic
surveys. Methods in Ecology and
Evolution, 12(3),
432–442. (link)
- Samaniego, A., Griffiths, R., Gronwald, M., Holmes,
N. D., Oppel, S., Stevenson, B. C., and Russell,
J. C. (2020) Risks posed by rat reproduction and diet to
eradications on tropical islands. Biological
Invasions, 22(4),
1365–1378. (link)
- Baron, H. R., Stevenson, B. C., and Phalen,
D. N. (2020) Inconsistent efficacy of water soluble
Amphotericin B for the treatment of Macrorhabdus
ornithogaster in a budgerigar (Melopsittacus
undulatus) aviary. Australian Veterinary
Journal, 98(7), 333–337. (link)
- Stevenson, B. C., Borchers, D. L., and Fewster,
R. M. (2019) Cluster capture-recapture to account for
identification uncertainty on aerial surveys of animal
populations. Biometrics, 75(1),
326–336. (link)
- Baron, H. R., Leung, K. C. L., Stevenson, B. C.,
Sabater Gonzalez, M., and Phalen, D. N. (2019)
Evidence of Amphotericin B resistance in Macrorhabdus
ornithogaster in Australian cage-birds. Medical
Mycology, 57(4),
421–428. (link)
- Jones-Todd, C. M., Caie, P., Illian,
J. B., Stevenson, B. C., Savage, A., Harrison,
D. J., and Bown, G. L. (2019) Identifying prognostic
structural features in tissue sections of colon cancer
patients using point pattern analysis. Statistics in
Medicine, 38(8),
1421–1441. (link)
- Measey, G. J., Stevenson, B. C., Scott, T.,
Altwegg, R., and Borchers, D. L. (2017) Counting chirps:
Acoustic monitoring of cryptic frogs. Journal of
Applied Ecology, 54(3),
894–902. (link)
- Kidney, D., Rawson, B. M., Borchers,
D. L., Stevenson, B. C., Thomas, L., and Marques,
T. A. (2016) An efficient acoustic density estimation
method with human detectors applied to gibbons in
Cambodia. PLoS ONE, 11(5),
e0155066. (link)
- Fewster, R. M., Stevenson, B. C., and Borchers,
D. L. (2016) Trace-contrast models for capture-recapture
without capture histories. Statistical
Science, 31(2),
245–258. (link)
- Borchers, D. L., Stevenson, B. C., Kidney, D.,
Thomas, L., and Marques, T. A. (2015) A unifying model for
capture-recapture and distance sampling surveys of
wildlife populations. Journal of the American
Statistical Association, 110(509),
195–204. (link)
- Stevenson, B. C., Borchers, D. L., Altwegg, R.,
Swift, R. J., Gillespie, D. M., and Measey, G. J. (2015) A
general framework for animal density estimation from
acoustic detections across a fixed microphone
array. Methods in Ecology and
Evolution, 6(1),
38–48. (link)
- Stevenson, B. C., and Millar, R. B. (2013) Promising the moon? Evaluation of indigenous and lunar fishing calendars using semiparametric generalized mixed models of recreational catch data. Environmental and Ecological Statistics, 20(4), 591–608. (link)
Current students
PhD:
- Rishika Chopara
Goodness-of-fit for spatial capture-recapture models
Cosupervisor: Rachel Fewster
Prospective students
If you are interested in working on a project with me as a postgraduate student at the University of Auckland (Hons/MSc/PhD), then please feel free to send me an email. It's helpful if you summarise your background (e.g., your skills and qualifications) and indicate the type of project you would like to work on. For example, take a look at the list of projects below and let me know which you think sound particulary interesting. It's possible there will be a related or adjacent topic to work on.
Current projects
Here are outlines of some current research projects I am working on.
Latent identity capture-recapture without sampling or enumerating
With Rachel Fewster, Jesse Goodman, and Martin Hazelton, I am developing methods to fit capture-recapture models when we cannot identify some (or all) detected individuals. There are existing methods to fit these models by either sampling latent identities within an MCMC algorithm, or computing a likelihood by summing over all possible detection-to-identity matchings, but these can be very computationally demanding. We are developing methods that can fit models to large data sets quickly by avoiding sampling or enumerating all possible matchings.
Penalised regression splines for spatial capture-recapture
With Andrew Seaton, I am developing methods to fit penalised regression splines to model spatial variation in animal density within spatial capture-recapture. Estimating animal distribution in spatial capture-recapture is typically achieved by using spatial covariates and assuming they fully explain how density varies over space. However, we might often expect variation due to other factors, such as unmeasured covariates or due to animals clustering for social reasons.
Goodness-of-fit, with applications to capture-recapture
Rishika Chopara, a PhD student cosupervised by myself and Rachel Fewster, is working on various aspects of goodness-of-fit. Often a model's deviance is compared to a chi-squared distribution to assess how well it fits the data. However, in many cases the chi-squared distribution is not a close approximation to the distribution of the deviance under the null hypothesis that the model is correct. Rishika is developing better approximations for the distribution of the deviance, which will be particularly useful for assessing fit of capture-recapture models.
Closed-form likelihoods for spatial capture-recapture
Jing Liu is a Research Fellow working with myself and Rachel Fewster on closed-form likelihoods for spatial capture-recapture. These models treat animals' activity centres as latent variables, which are usually dealt with either by sampling them by MCMC (for Bayesian models) or numerically approximating integrals (for maximum-likelihood models). Jing has been working on exact computations for model likelihoods, which entirely avoids having to sample or approximate.
Linear mixed-effects models under two-phase sampling
Zoe Luo is a PhD student I cosupervised with Thomas Lumley. Among other things, she has developed methods to fit linear mixed-effects models under two-phase sampling, with an application to genetic data of the kākāpō, a (very) endangered New Zealand parrot species.
Linear mixed-effects models for morphometric data
With Elizabeth Smit, Edy Setyawan, Mark Erdmann, and Michael Walker, I am developing methods to model morphometric data collected by drones. Using drones is a great way to collect body size measurements of animals, however often this technique introduces non-negliglbe measurement error. We are developing ways to handle this measurement error and make inferences on aspects of animal morphology that are not available with existing methods.
Method development for acoustic spatial capture-recapture
David Chan, whose PhD I supervised, has been developing new spatial capture-recapture methods for acoustic surveys. One such method is tailored to calling surveys of gibbons, and accounts for movement of gibbon groups from one morning to the next.
Software for acoustic spatial capture-recapture
During my PhD I spent a lot of time writing an R package, ascr, which provides functions to fit acoustic spatial capture-recapture models. Along with Lingyu Hao, Melissa Bather, Angeline Xiao, Joseph Reps, and other former students, I am creating a new package, acre, which will be both more user-friendly (especially when modelling spatial effects) and more flexible in terms of the types of model available to fit.
Applications of acoustic spatial capture-recapture
In separate projects with Sarah McGrath and Milou Groenenberg, Jing Liu and myself are fitting acoustic spatial capture-recapture models to estimate density and distribution of gibbon populations in national parks situated in Cambodia and Vietnam.
With Cornelia Oedekoven, David Borchers, and Tarin Eccleston, I am creating software and training materials to help make spatial capture-recapture methods more accessible to researchers conducting acoustic surveys of primates.
Spatial capture-recapture for fungi
With Sarah Christofides and Emile Blin, I am experimenting with fitting spatial capture-recapture models to understand the ecology of fungi. I suspect this is the first time spatial capture-recapture models have been fitted to any species in the fungi kingdom, and possibly for nonspatial capture-recapture, too.
Spatiotemporal space-use of cetaceans in the Hauraki Gulf/Tīkapa Moana
Led by Jessie Colbert, and with Melissa Bowen and Rochelle Constantine, I am working on a project to understand spatiotemporal patterns in occurrence of cetaceans in New Zealand's Hauraki Gulf/Tīkapa Moana. We have developed a generalised linear mixed-effects model, which includes spatiotemproal random effects alongside spatially varying effects of covariates such as sea surface temperature.
Lunar cycle effects on fishing catch rates in Corsican fisheries
The first paper I ever published was about estimating relationships between the lunar cycle and catch rates of fish in New Zealand. Marina Luccioni has collated data from commercial fishing operations in Corsica and is fitting generalised linear mixed-effects models to investigate similar relationships. Preliminary results suggest Corsican fish might be even more responsive to lunar illumination than New Zealand's snapper!
Links
Software
The R package ascr fits a range of acoustic spatial capture-recapture models. Find its homepage here.
The R package palm fits various point processes via maximisation of the Palm likelihood, and is available on CRAN. Find its homepage here.
Other software projects can be found on my GitHub page; see the link above.