• Home
  • David James Fletcher
David James Fletcher

David James Fletcher
David Fletcher Consulting Limited

BSc, MSc, PhD

About

119
Publications
41,025
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
4,234
Citations
Introduction
Statistical modelling, primarily in ecology and wildlife management
Skills and Expertise
Additional affiliations
November 2019 - present
David Fletcher Consulting Limited
Position
  • Managing Director
February 1991 - December 2019
University of Otago
Position
  • Professor (Associate)
February 1989 - January 1991
The University of Sydney
Position
  • Lecturer
Education
September 1981 - August 1983
University of Southampton
Field of study
  • Statistics
September 1980 - August 1981
University of Southampton
Field of study
  • Applied Statistics
September 1977 - June 1980
University of Southampton
Field of study
  • Mathematics

Publications

Publications (119)
Article
Full-text available
Scientific support invited by Indigenous peoples and local communities (IPLC) to assist with customary environmental management can improve conservation and community livelihoods. For example, demographic models can help to understand how alternative wildlife management strategies affect population dynamics and harvest sustainability. We developed...
Article
1. In many environmental and ecological settings, the analysis of data involves calculation of a confidence interval for a non‐linear function of the model parameters. In the frequentist approach, this is often achieved by using a Wald interval in which the standard error of the estimate of the function is obtained using the delta method. 2. The pu...
Article
Many factors have contributed to a decline in the tītī (sooty shearwater) Ardenna grisea population in Aotearoa New Zealand since at least the 1960s. The relative impacts of Southern Oscillation Index (SOI) variation, bycatch, predation, and traditional harvesting by Rakiura Māori were recently estimated in a study fitting population models to data...
Article
Full-text available
A suite of factors may have contributed to declines in the tītī (sooty shearwater; Ardenna grisea ) population in the New Zealand region since at least the 1960s. Recent estimation of the magnitude of most sources of non-natural mortality has presented the opportunity to quantitatively assess the relative importance of these factors. We fit a range...
Article
Full-text available
Model averaging is commonly used to allow for model uncertainty in parameter estimation. As well as providing a point estimate that is a natural compromise between the estimates from different models, it also provides confidence intervals with better coverage properties, compared to those based on a single best model. In recent years, the concept o...
Article
Multinomial data arise in many areas of the life sciences, such as mark-recapture studies and phylogenetics, and will often by overdispersed, with the variance being higher than predicted by a multinomial model. The quasi-likelihood approach to modelling this overdispersion involves the assumption that the variance is proportional to that specified...
Article
Full-text available
In many scientific studies, the underlying data-generating process is unknown and multiple statistical models are considered to describe it. For example, in a factorial experiment we might consider models involving just main effects, as well as those that include interactions. Model-averaging is a commonly-used statistical technique to allow for mo...
Data
R code and data for obtaining MATA-SBoot intervals for the hydronium ion example. (R)
Book
This book provides a concise and accessible overview of model averaging, with a focus on applications. Model averaging is a common means of allowing for model uncertainty when analysing data, and has been used in a wide range of application areas, such as ecology, econometrics, meteorology and pharmacology. The book presents an overview of the meth...
Chapter
We provide an overview of the key ideas and results in Bayesian and frequentist model averaging, and suggest directions for future research.
Chapter
We provide an overview of frequentist model averaging. For point estimation, we consider different methods for selecting the model weights, including those based on AIC, bagging, weighted AIC, stacking and focussed methods. For interval estimation, we consider Wald, MATA and percentile-bootstrap intervals. Use of the methods are illustrated by exam...
Chapter
Model averaging is a means of allowing for model uncertainty in estimation which can provide better estimates and more reliable confidence intervals than model selection. We illustrate its use via examples involving real data, discuss when it is likely to be useful, and compare the frequentist and Bayesian approaches to model averaging.
Chapter
We provide an overview of Bayesian model averaging (BMA), starting with a summary of the mathematics associated with classical BMA, including the calculation of posterior model probabilities and the choice of priors for both the models and the model parameters. We also consider prediction-based approaches to BMA and argue that these are preferable...
Article
Intrinsic population growth rate (rmax) is an important parameter for many ecological applications, such as population risk assessment and harvest management. However, rmax can be a difficult parameter to estimate, particularly for long-lived species, for which appropriate life table data or abundance time-series are typically not obtainable. We de...
Article
Intrinsic population growth rate (rmax) is an important parameter for many ecological applications, such as population risk assessment and harvest management. However, rmax can be a difficult parameter to estimate, particularly for long- lived species, for which appropriate life table data or abundance time series are typically not obtainable. We d...
Article
Full-text available
Bayesian integrated population modelling provides a natural tool for estimating immigration into a single study population when we have indices of population size, mark-recapture data and fecundity data. We consider the choice of both the parametrization of immigration and its prior. Using a simulation study for a model that is typical of those use...
Article
Full-text available
Background: Smoking is a major risk factor for periodontal disease. Conventional oral epidemiology approaches have found strong, consistent associations between chronic smoking and periodontal attachment loss (AL) through ages 26, 32, and 38 years, but those statistical methods disregarded the data's hierarchical structure. This study reexamines t...
Article
Full-text available
Using mark–recapture, we estimated the age at first return and survival of juveniles of the Sooty Shearwater (Puffinus griseus), a medium-sized burrow-breeding procellariiform, at two New Zealand breeding colonies, Taiaroa Head and The Snares. Sooty Shearwaters were first observed at the colony between 2 and 10 years after being banded as chicks, w...
Article
Full-text available
We studied habitat use by Hector's dolphins Cephalorhynchus hectori in order to quantify the influences of location, season, time and tide, and to determine how frequently dolphins use inner Akaroa Harbour where unattended gillnetting is allowed for 6 mo of the year (1 April to 30 September). T-POD acoustic dataloggers were moored in outer, mid- an...
Article
Full-text available
The process of model averaging has become increasingly popular as a method for performing inference in the presence of model uncertainty. In the frequentist setting, a model-averaged estimate of a parameter is calculated as the weighted sum of single-model estimates, often using weights derived from an information criterion such as AIC or BIC. A st...
Article
Full-text available
To examine the potential for using multiple list sources and capture-recapture methods for estimating the prevalence of diagnosed diabetes. A model-averaging procedure using an adjusted Akaike's Information Criterion (QAICc) was used to combine capture-recapture estimates from log-linear models obtained from simultaneously analysing four sources of...
Article
Full-text available
Rakiura Māori seek robust population monitoring to guide sustainable management of the annual take of muttonbird (tītī, Puffinus griseus), in southern New Zealand. Analyses of eight muttonbirder harvest records spanning 67 years and consisting of 6744 hunts show systematic commonalities in seasonal patterns of hunt tallies. General linear models (G...
Article
Full-text available
1. Mark–recapture studies are often used to estimate adult survival probability , which is an important demographic parameter for long-lived species, as it can have a large impact on the population growth rate. We consider the impact of variation in capture probability among individuals (capture heterogeneity) on the estimation of ϕ from a mark–rec...
Article
Full-text available
We consider the coverage rate of model-averaged confidence intervals for the treatment means in a factorial experiment, when we use a normal linear model in the analysis. Model-averaging provides a useful compromise between using the full model (containing all main effects and interactions) and a “best model” obtained by some model-selection proces...
Article
Full-text available
Seabirds such as albatrosses and petrels are frequently caught in longline and trawl fisheries, but limited demographic data for many species creates management challenges. A method for estimating the potential biological removal (the PBR method) for birds requires knowledge of adult survival, age at first breeding, a conservation goal, and the low...
Article
Full-text available
Model-averaging is commonly used as a means of allowing for model uncertainty in parameter estimation. In the frequentist framework, a model-averaged estimate of a parameter is the weighted mean of the estimates from each of the candidate models, the weights typically being chosen using an information criterion. Current methods for calculating a mo...
Article
Full-text available
We consider the problem of fitting a generalized linear model to overdispersed data, focussing on a quasilikelihood approach in which the variance is assumed to be proportional to that specified by the model, and the constant of proportionality, φ, is used to obtain appropriate standard errors and model comparisons. It is common practice to base an...
Article
Full-text available
The age at first breeding is an important demographic parameter in determining maximum growth rate, population size and generation time and is a key parameter in calculating the potential biological removal of birds. Albatrosses and petrels do not begin breeding for many years, with some first breeding in their teens. This means that even long-term...
Article
Full-text available
1. We consider the use of weighted regression when modelling data from different sites, times or studies. Our primary focus is on the coverage rate of the 95% confidence interval for the slope parameter when we have a single predictor variable. We use simulation to assess this coverage rate for both weighted and unweighted regression, across a rang...
Article
Full-text available
We estimated apparent annual survival of adult and young grey-faced petrels (Pterodroma macroptera gouldi) and age of first return to the natal colony of young birds from 2 colonies in the Bay of Plenty, NZL, between 1991 and 2008. We analysed the capture histories of 5844 adult birds and 928 chicks in a mark-recapture framework. The apparent adult...
Chapter
Full-text available
Article
Full-text available
Cultural evidence suggests that sooty shearwater (Puffinus griseus) chicks have been harvested by Rakiura MAori on islands in southern New Zealand since prehistoric times. Concerns exist that modern harvests may be impacting sooty shearwater abundance. We modeled human-related and ecological determinants of harvest (total no. of individuals harvest...
Article
We investigated the precision and accuracy of an infrared burrowscope for detecting sooty shearwater (Pufffinus griseus) chicks at 13 plots from 3 islands in southern New Zealand in 2003. We partially excavated burrows systems to reveal the entire burrow contents after 2 teams of observers had prospected all burrow entrances. Accuracy was similar b...
Article
Full-text available
A sample of 834 female New Zealand sea lions (Phocarctos hookeri), which were aged and measured, was obtained between 1998 and 2001. In addition, the reproductive histories of 505 marked females from the Auckland Islands were recorded from 1998 to 2005. These data sets were used to investigate growth and reproductive rates. Length and weight ranged...
Article
Full-text available
Stochastic matrix population models are often used to help guide the management of animal populations. For a long-lived species, environmental stochasticity in adult survival will play an important role in determining outcomes from the model. One of the most common methods for modelling such stochasticity is to randomly select the value of adult su...
Article
The age distribution of 865 lactating New Zealand sea lions (NZSLs; Phocarctos hookeri) was investigated over 3 yr (1999–2001) at two breeding colonies, Sandy Bay and Dundas Island, New Zealand. Lactating females were aged between 3 and 26 yr with a maximum observed age of 28 yr. The mean age of lactating females was 11.1 (SE = 0.16) yr. Age distri...
Article
Full-text available
Between 2001 and 2006, we systematically sampled the entire coast of Whenua Hou, a rugged offshore island in southern New Zealand, to estimate the population densities of sooty shearwaters (Puffinus griseus) and mottled petrels (Pterodroma inexpectata) by counting the entrances of breeding burrows. A two-step regression modelling process using bino...
Article
Full-text available
We estimated the change in abundance of sooty shearwater (titi, Puffinus griseus) at six Rakiura Titi Islands, New Zealand, by comparing historical and recent surveys of the density of entrances to breeding burrows. We found evidence that entrance density between 1994 and 2006 was lower than it was between 1961 and 1976. Our overall estimate of the...
Article
Full-text available
Sooty shearwater (Puffinus griesus, titi) abundance, harvest levels and chick mass were monitored repeatedly on Putauhinu Island, south‐west of Rakiura (Stewart Island) between 1997 and 2005. Putauhinu is the second largest of the Titi Islands and has a relatively high density of chicks distributed over most of the island, so it supports what is li...
Article
Full-text available
Adaptive co‐management and Participatory Action Research (PAR) promotes social ecological resilience by simultaneously protecting wildlife and its habitat and promoting capacity and motivation for sustainable harvest management by communities. We report here on a case study of learning through a partnership (1994–2009) between science and Tradition...
Article
Full-text available
Recent comprehensive survey data from multiple New Zealand offshore islands were combined with demographic population models to produce the first formal estimate of the total population of sooty shearwaters within New Zealand territory. We estimated the total population over 1994–2005 to be 21.3 (19.0–23.6) million individual birds in the New Zeala...
Article
Full-text available
The effects of a range of habitat variables on spatial variation of breeding burrow density of sooty shearwaters, Puffinus griseus, were measured on 5 islands near Rakiura (Stewart Is) and 1 island in The Snares Is group, during the 2000-01 breeding season. Density estimates for 4 islands where Rakiura Maori harvest chicks ranged from 0.30 to 0.47...
Chapter
Full-text available
Adult survival probability is a key parameter in any population model for a long-lived species. For many species, information on adult survival comes from a capture–recapture study involving individuals for whom age is unknown. If the species experiences senescence, the estimate of overall adult survival probability will be negatively biased. The p...
Article
Full-text available
Monitoringofbreedingsuccessinburrow-nestingseabirdsisproblematic,owingtothedifficultiesofdetecting occupants in complex burrow systems. We summarise 6 years of monitoring the breeding success of sooty shearwater (t�it�i, muttonbird, Puffinus griseus) on two southern New Zealand islands, The Snares and Whenua Hou, with a portable infrared camera sys...
Article
Full-text available
Recent declines of many seabird populations have placed increased emphasis on determining the status of potentially threatened species. However, the burrow-nesting habits and inter-annual fluctuation in breeding numbers of some species make trend detection difficult, and so knowledge of their population dynamics often remains coarse. Here we report...
Article
Full-text available
The Sooty Shearwater (Puffinus griseus), also known in New Zealand as tt or muttonbird, is an abundant, long-lived, trans-equatorial migrant that breeds in colonies in the southern hemisphere during the austral summer and migrates to and forages in the North Pacific Ocean during the boreal summer. That populations of Sooty Shearwaters have been dec...
Article
Full-text available
Many bird species are subject to human-caused mortality, either through direct harvest (e.g. game birds) or through incidental mortalities (e.g. fisheries-related bycatch of seabirds, impact with vehicles, wind turbines, or power lines). In order to assess the impact of additional mortalities on birds, both the number of birds killed and their abil...
Article
Full-text available
Data that are skewed and contain a relatively high proportion of zeros can often be modelled using a delta-lognormal distribution. We consider three methods of calculating a 95% confidence interval for the mean of this distribution, and use simulation to compare the methods, across a range of realistic scenarios. The best method, in terms of covera...
Article
Knowledge of DNA evolution is central to our understanding of biological history, but how fast does DNA change? Previously, pedigree and ancient DNA studies--focusing on evolution in the short term--have yielded molecular rate estimates substantially faster than those based on deeper phylogenies. It has recently been suggested that short-term, elev...
Article
Full-text available
We estimated the total number of burrow entrances, chicks and total population size of the Sooty Shearwater, Puffinus griseus , colony on Taukihepa (Big South Cape), the largest of the 36 Titi Islands where Titi (Sooty Shearwater chicks) are harvested. Between 1999 and 2005 we surveyed six manu (family birded areas) as well as a colony in the inter...
Article
Full-text available
Decision theory provides an organised approach to decision making in natural resource conservation. The theory requires clearly stated objectives, decision alternatives and decision-outcome utilities, and thus allows for the separation of values (conservation and other societal objectives) from beliefs. Models express belief in the likely response...
Article
Full-text available
The number of Sooty Shearwater, Puffinus griseus, burrow entrances on North East Island of The Snares, New Zealand, declined from 3 288 000 (95% CI = 3.1-3.5 million) entrances between 1969 and 1971, to 2 061 000 (95% CI = 1.9-2.2 million) between 1996 and 2001. This represents a decline of 37% over 27 years, or 1.72% per year (95% CI = 1.35%-2.12%...
Article
Full-text available
Fisheries bycatch is a threatening process for populations of procellariiform seabirds, and is of particular importance for the conservation of albatross, an especially threatened group at a global scale. There is a high level of endemism of albatross and petrels in New Zealand waters, and around one-third of the world's species of procellariiform...
Article
Full-text available
In many ecological research studies, abundance data are skewed and contain more zeros than might be expected. Often, the aim is to model abundance in terms of covariates, and to estimate expected abundance for a given set of covariate values. An approach that has been advocated recently involves the use of a conditional model. This allows one to se...
Article
Reintroductions of threatened species are increasingly common in conservation. The translocation of a small subset of individuals from a genetically diverse source population could potentially lead to substantial inbreeding depression due to the high genetic load of the parent population. We analysed 12 years of data from the reintroduced populatio...
Article
Full-text available
Distance sampling surveys are commonly used to estimate animal abundance (N). The choice of survey design has only recently received attention in the line-transect research literature, which has tended to focus more on the violation of assumptions. In this stu dy, simulation methods were used to assess the effect of line placement on the accuracy a...
Article
Full-text available
Genetic variability in the apple [Malus] fruit disorder bitter pit and fruit calcium concentrations ([Ca]) was determined in 25 seedling families at each of two sites from 1999 to 2000 and again for one site, in 2001. Most trees were free of pit or had low pit incidence, which could be approximated by an over-dispersed binomial distribution. A gene...
Article
Full-text available
A stage-structured population model was developed to predict which of nine hypothetical translocation scenarios was likely to produce the best outcome for the rare Hamilton's frog (Leiopelma hamiltoni McCulloch). Model outcome was measured in terms of population growth rate and probability of extinction. Only females were modelled. The model predic...
Article
Full-text available
We consider estimation of the magnitude of incidental fisheries ‘bycatch’ for two petrel species, sooty shearwaters (Puffinus griseus) and short-tailed shearwaters (Puffinus tenuirostris). There are clear statistical advantages in estimating bycatch for abundant species such as these, and our results may also guide the conservation and management o...
Article
Full-text available
We discuss a method for analyzing data that are positively skewed and contain a substantial proportion of zeros. Such data commonly arise in ecological applications, when the focus is on the abundance of a species. The form of the distribution is then due to the patchy nature of the environment and/or the inherent heterogeneity of the species. The...
Article
Full-text available
In a capture-recapture analysis, uncertainty in the parameter estimates is usually expressed by presenting classical Wald-type confidence intervals. This approach involves (1) the assumption that the maximum likelihood estimates are asymptotically normal and (2) numerical computation of the variance-covariance matrix of these estimates. When the sa...