Proteus Wildlife Research Consultants

Dunedin, New Zealand

Proteus Wildlife Research Consultants

Dunedin, New Zealand
Time filter
Source Type

Warton D.I.,University of South Australia | Stoklosa J.,University of South Australia | Guillera-Arroita G.,University of Melbourne | MacKenzie D.I.,Proteus Wildlife Research Consultants | Welsh A.H.,Australian National University
Methods in Ecology and Evolution | Year: 2017

Occupancy-detection models that account for imperfect detection have become widely used in many areas of ecology. As with any modelling exercise, it is important to assess whether the fitted model encapsulates the main sources of variation in the data, yet there have been few methods developed for occupancy-detection models that would allow practitioners to do so. In this paper, a new type of residual for occupancy-detection models is developed according to the method of Dunn & Smyth (Journal of Computational and Graphical Statistics, 5, 1996, 236–244). Residuals are separately constructed to diagnose the occupancy and detection components of the model. Because the residuals are quite noisy, we suggest fitting a smoother through plots of residuals against predictors of fitted values, with 95% confidence bands, to diagnose lack-of-fit. The method is illustrated using Swiss squirrel data, and evaluated using simulations based on that dataset. Plotting residuals against predictors or against fitted values performed reasonably well as methods for diagnosing violations of occupancy-detection model assumptions, particularly plots of residuals against a missing predictor. Relatively high false positive rates were sometimes observed, but this seems to be controlled reasonably well by fitting smoothers to these plots and being guided in interpretation by 95% confidence bands around the smoothers. © 2017 The Authors. Methods in Ecology and Evolution © 2017 British Ecological Society

Watts C.,Landcare Research | Thornburrow D.,Landcare Research | MacKenzie D.,Proteus Wildlife Research Consultants
Journal of Insect Conservation | Year: 2011

Appropriate monitoring tools are essential for assessing the effectiveness of management for all threatened insect taxa. In New Zealand the large-bodied flightless orthopterans in the genus Deinacrida have mostly been monitored by searching through habitat during the day or spotlighting at night but this is time consuming and the results depend on the skill of the searcher. Recently, footprint tracking tunnels, similar to those used for monitoring small mammals in New Zealand, were found to be effective for detecting adults of various giant weta species. In this study, we compared the abundance of Cook Strait giant weta (CSGW) in the vicinity of the tunnels, estimated by mark-recapture, with the number of tracking tunnels tracked by weta. We found strong indications that both baited and unbaited tracking tunnels can be used to estimate the number of adult weta present but that this probably depends on their responses to meteorological conditions which are not yet understood. Our results also show that footprint tracking tunnels are more effective for detecting adult CSGW than searching for these insects at night and that baiting tracking tunnels with peanut butter increases their effectiveness for detecting adult CSGW. We confirmed how far Cook Strait giant weta moved each night on Matiu-Somes Island by attaching transmitters to them and found that day roosts of three adult males were on average 8.6 m apart each day and those of adult females were on average 21.3 m apart. Both the low recapture rates of marked adult CSGW and the nightly displacements of those with transmitters suggest that adult CSGW show no site fidelity and are clearly capable of moving large distances each night. However, an individual weta is unlikely to track more than one tunnel per night if tunnels are 30 m apart. Tracking tunnels have the potential to be used with some other insects, provided their footprints are diagnostic. An advantage of using tracking tunnels is that they are non-lethal and would therefore be particularly suitable for monitoring other large threatened insect taxa. © 2010 Springer Science+Business Media B.V.

MacLeod C.J.,Landcare Research | MacKenzie D.I.,Proteus Wildlife Research Consultants | Allen R.B.,Landcare Research
New Zealand Journal of Ecology | Year: 2012

Robust monitoring systems are required to improve the ecological outcomes of management actions aimed at preventing biodiversity loss. We present a pilot study that measured assemblages of widespread and common bird species at the national scale in New Zealand. Bird surveys were undertaken at 18 sampling locations (six per land cover class: forest, shrubland and non-woody) randomly selected from a national grid. The full sampling protocol (five count stations surveyed on each of two consecutive days) was implemented at 80% of sampling locations. Each survey consisted of a ten-minute bird count, with distance sampling carried out in the initial 5-min period and any new species recorded in the second 5-min period. Most observations were based on aural cues (particularly in forest and shrubland). On average, one additional species was recorded per sampling location in ten- versus five-minute counts. Analyses highlighted spatial heterogeneity as a major factor influencing detection probabilities both for species present and for individuals of those species at sampling locations. This issue is often overlooked when estimating bird population trends through time. Most endemic species were detected in forest, while native and introduced species were most frequently detected in shrub. Potential uses of the information collected, along with recommendations for improving the sampling protocols, are highlighted. © New Zealand Ecological Society.

Norbury G.,Landcare Research | Mackenzie D.I.,Proteus Wildlife Research Consultants
New Zealand Journal of Ecology | Year: 2012

Two conservation tools have been developed over the last 10-15 years for species on the New Zealand mainland that are vulnerable to introduced mammalian predators: landscape-scale predator trapping networks, and eradication of predators within mammal-proof exclosures. We tested whether these tools would allow population growth of critically endangered grand skinks (Oligosoma grande) and Otago skinks (O. otagense) over three years. Skink populations were subjected to one of three predator treatments: (1) near-eradication inside a mammal-proof fence; (2) suppression by trapping within a 2100-ha area; and (3) unmanaged predator populations. Monitoring by non-invasive photo-resight methods, and data analysis using program MARK, showed that the greatest increase in abundance of both species occurred at the centre of the predator trapping treatment and inside a mammal-proof fence. For grand skinks, there was little or no change in population size at the trapping periphery. At the unmanaged sites, the grand skink population underwent a catastrophic decline whereas the Otago skink population was stable. A grouping analysis showed that the unmanaged grand skink treatment was clearly distinguished from the other predator treatments based on survival rate. Results suggest that: (1) predation by introduced mammals is a key driver in the decline of these skinks, and episodic predation events may be a component of this process; and (2) use of mammal-proof fences or intensive predator control over a large enough area should allow skink populations to recover. © New Zealand Ecological Society.

Chilvers B.L.,Marine Conservation | Chilvers B.L.,Proteus Wildlife Research Consultants | MacKenzie D.I.,Marine Conservation | MacKenzie D.I.,Proteus Wildlife Research Consultants
Journal of Mammalogy | Year: 2010

The estimation of life-history parameters for a threatened species is important for understanding its biology and helping to determine management options. This research investigates age- and sex-related survival estimates incorporating tag loss for New Zealand (NZ) sea lions (Phocarctos hookeri) from Sandy Bay, Enderby Island, Auckland Islands, New Zealand, using multistate markrecapture data from known-age individuals over 8 years (19971998 to 20052006). Survival estimates and tag loss rates differed significantly by sex and age class, with adult males having the lowest tag retention of any age or sex class and females ≥ 3 years old having lower survival estimates than their male counterparts. The variability and lower female survival relative to males is a critical problem for NZ sea lions, because even small changes in adult female survival significantly affect population trends for such large, long-lived mammals. © 2010 American Society of Mammalogists.

Forsyth D.M.,Arthur Rylah Institute for Environmental Research | Koehn J.D.,Arthur Rylah Institute for Environmental Research | MacKenzie D.I.,Proteus Wildlife Research Consultants | Stuart I.G.,Kingfisher Research
Biological Invasions | Year: 2013

There is much interest in managing invasive freshwater fish, but little is known about the dynamics of these populations following establishment. We used annual commercial catch-per-unit-effort data at multiple spatio-temporal scales to test hypotheses about the population dynamics of invading common carp (Cyprinus carpio) in the Murray-Darling Basin, Australia. We hypothesised that following establishment of the Boolara strain of this species in the Murray-Darling Basin in 1961/1962: (1) carp would undergo exponential or logistic-type population growth; and (2) carp population growth rates would be highest following over-bank flood events that provided extensive off-channel spawning and feeding habitats. The logistic (wi = 0.73) and delayed-logistic (wi = 0.27) models best explained the population dynamics of common carp in the Murray-Darling Basin during 1962/1963-2001/2002; there was negligible support for exponential growth (wi ≤ 0.01). Although we cannot exclude the possibility that floods may have been important in the early years of the invasion we found little evidence that carp population growth rates increased following flood events. Our logistic-type model-based estimates of the maximum annual population growth rate (rm; 0.378 and 0.384) indicate that >0.315 or 0.319 of the adult population would need to be removed annually to achieve eradication. We conclude that the rapid spread of the Boolara strain of common carp through the Murray-Darling Basin was facilitated by high initial population growth rates. More generally, we suggest that the lag period between an invader establishing and increasing to high abundances will be characterised by logistic-type population growth. We encourage others to investigate the long-term population dynamics of invading freshwater fish using time series and models such as those reported here. © 2012 Springer Science+Business Media B.V.

McClintock B.T.,U.S. Geological Survey | Nichols J.D.,U.S. Geological Survey | Bailey L.L.,Colorado State University | MacKenzie D.I.,Proteus Wildlife Research Consultants | And 2 more authors.
Ecology Letters | Year: 2010

Analytical methods accounting for imperfect detection are often used to facilitate reliable inference in population and community ecology. We contend that similar approaches are needed in disease ecology because these complicated systems are inherently difficult to observe without error. For example, wildlife disease studies often designate individuals, populations, or spatial units to states (e.g., susceptible, infected, post-infected), but the uncertainty associated with these state assignments remains largely ignored or unaccounted for. We demonstrate how recent developments incorporating observation error through repeated sampling extend quite naturally to hierarchical spatial models of disease effects, prevalence, and dynamics in natural systems. A highly pathogenic strain of avian influenza virus in migratory waterfowl and a pathogenic fungus recently implicated in the global loss of amphibian biodiversity are used as motivating examples. Both show that relatively simple modifications to study designs can greatly improve our understanding of complex spatio-temporal disease dynamics by rigorously accounting for uncertainty at each level of the hierarchy. Published 2010. This article is a US Government work and is in the public domain in the USA.

Chilvers B.L.,Marine Conservation Group | Wilkinson I.S.,Marine Conservation Group | Mackenzie D.I.,Proteus Wildlife Research Consultants
Journal of Agricultural, Biological, and Environmental Statistics | Year: 2010

The trade-off between survival and reproduction by individuals is central to understanding life-history parameters of a species. Few mammal species have life-history information from long-term research. Instead, demographic models are commonly utilized to investigate an individual's life-history strategy, species dynamics, and population trends. This research investigates age-related survival and reproductive performance of adult female New Zealand (NZ) sea lions (Phocarctos hookeri), using multi-state mark-recapture data from known-age branded individuals over five years. The mark-recapture analysis was integrated with a population model to predict the lifetime reproductive output of female NZ sea lions. The integration of an analysis of short-term datasets with population modeling allows for the prediction of life-history parameters of long lived animals when long-term information is not available. While such approaches involve some caveats, it provides a framework for investigating population dynamics and is preferential to unsubstantiated assumptions. This technique can lead to better design and implementation of conservation management for long lived species. © 2009 International Biometric Society.

Guillera-Arroita G.,University of Melbourne | Lahoz-Monfort J.J.,University of Melbourne | MacKenzie D.I.,Proteus Wildlife Research Consultants | Wintle B.A.,University of Melbourne | McCarthy M.A.,University of Melbourne
PLoS ONE | Year: 2014

In a recent paper, Welsh, Lindenmayer and Donnelly (WLD) question the usefulness of models that estimate species occupancy while accounting for detectability. WLD claim that these models are difficult to fit and argue that disregarding detectability can be better than trying to adjust for it. We think that this conclusion and subsequent recommendations are not well founded and may negatively impact the quality of statistical inference in ecology and related management decisions. Here we respond to WLD's claims, evaluating in detail their arguments, using simulations and/or theory to support our points. In particular, WLD argue that both disregarding and accounting for imperfect detection lead to the same estimator performance regardless of sample size when detectability is a function of abundance. We show that this, the key result of their paper, only holds for cases of extreme heterogeneity like the single scenario they considered. Our results illustrate the dangers of disregarding imperfect detection. When ignored, occupancy and detection are confounded: the same naïve occupancy estimates can be obtained for very different true levels of occupancy so the size of the bias is unknowable. Hierarchical occupancy models separate occupancy and detection, and imprecise estimates simply indicate that more data are required for robust inference about the system in question. As for any statistical method, when underlying assumptions of simple hierarchical models are violated, their reliability is reduced. Resorting in those instances where hierarchical occupancy models do no perform well to the naïve occupancy estimator does not provide a satisfactory solution. The aim should instead be to achieve better estimation, by minimizing the effect of these issues during design, data collection and analysis, ensuring that the right amount of data is collected and model assumptions are met, considering model extensions where appropriate. © 2014 Guillera-Arroita et al.

MacKenzie D.I.,Proteus Wildlife Research Consultants | Clement D.,Cawthron Institute
Journal of Agricultural, Biological, and Environmental Statistics | Year: 2016

Line-transect mark-recapture distance sampling methods can be used to estimate abundance when at least two observers sight and record distances to detected groups of individuals within the survey area. However, a lack of independence between the observer’s detections will cause biased abundance estimates. Studies are also typically designed such that there is complete overlap of the regions searched by the two observers, but that may not always be possible. Here we detail an intuitive approach for line-transect distance sampling applications based upon logistic regression to account for a potential lack of independence by using the detections of one observer as a covariate in the detection function of the second observer. Partial overlap of the observer survey regions can be addressed by constraining detection probability to equal 0 for the respective observer outside of the overlap zone. We show via simulation that the method provides reliable estimates of abundance and is not affected by random unmodeled heterogeneity in detection probability. The method is illustrated by estimating abundance within the covered region of an aerial line-transect survey for New Zealand’s endemic Hector’s dolphin (Cephalorhynchus hectori hectori) conducted in the austral summer of 2013, the motivating application for this work. Supplementary materials accompanying this paper appear on-line. © 2015, International Biometric Society.

Loading Proteus Wildlife Research Consultants collaborators
Loading Proteus Wildlife Research Consultants collaborators