54 research outputs found
A spatially explicit capture-recapture estimator for single-catch traps
No full textThis work was part-funded by EPSRC grant EP/I000917/1.1. Single-catch traps are frequently used in live-trapping studies of small mammals. Thus far a likelihood for single-catch traps has proven elusive and usually the likelihood for multi-catch traps is used for spatially explicit capture-recapture (SECR) analyses of such data. Previous work found the multi-catch likelihood to provide a robust estimator of average density. 2. We build on a recently developed continuous-time model for SECR to derive a likelihood for single-catch traps. We use this to develop an estimator based on observed capture times and compare its performance by simulation to that of the multi-catch estimator for various scenarios with non-constant density surfaces. 3. While the multi-catch estimator is found to be a surprisingly robust estimator of average density, its performance deteriorates with high trap saturation and increasing density gradients. Moreover, it is found to be a poor estimator of the height (but not range) of the detection function. By contrast, the single catch estimators of density, distribution and detection function parameters are found to be unbiased or nearly unbiased in all scenarios considered. This gain comes at the cost of higher variance, so that despite the lower bias of the single-catch estimator of the density surface over space, its root mean squared error is similar to that of the multi-catch estimator. 4. If there is no interest in interpreting the detection function parameters themselves, and if density is expected to be fairly constantover the survey region, then the multi-catch estimator performs well with single-catch traps. However if accurate estimation of the detection function is of interest, or if density is expected to vary substantially in space, then there is merit in using the single-catch estimator when trap saturation is above about 60%. The estimator’s performance is improved if care is taken to place traps so as to span the range of variables that affect animal distribution. As a single-catch likelihood with unknown capture times remains intractable for now, researchers using single-catch traps should aim to incorporate timing devices with their traps.Peer reviewe
Enumerating and estimating maternal and neonatal deaths in the Western Cape Province, South Africa
No full textMeasuring and monitoring progress towards global development goals requires valid and reliable estimates of maternal and child mortality. This thesis has aimed at enumerating and estimating maternal and neonatal deaths from 2010 to 2013 in the Western Cape Province; and determining factors associated with these outcomes during the same period. This thesis comprises nine chapters, of which six present the research findings. The first results chapter has presented the findings from a systematic review, determining trends of maternal and neonatal mortality from 1990 to 2015 in South Africa. The review found that estimates of maternal and neonatal mortality are widely divergent across data sources and estimation methods, with conflicting trends over the analysis period. The second results chapter compared the performance of the existing decision-rule based linkage approach (provincial linkage) which uses fuzzy linkage to an independent fully probabilistic record linkage (PRL) implementation for identifying mortality records across the Western Cape Provincial Health Data. The PRL was shown to be a feasible method for future implementation, while the existing linkage performed similarly to the independent linkage exercise, providing reassurance on the adequacy of the linked datasets on which the subsequent chapters were based. The third and fourth results chapters involved the applications of three-source capture-recapture methods, to estimate maternal and neonatal mortality under-reporting in the Western Cape province. Based on these models, maternal and neonatal mortality under-reporting were estimated at 45.6% and 17.7% over the full 4-year period respectively. The last two results chapters focused on determining factors associated with maternal and neonatal mortality in this setting and exploring whether the estimates of association were altered through using an expanded number of outcome events based on database linkage across multiple data sources. Most findings were consistent with known associations, as well as estimates from single-source analyses in the same setting. The thesis concludes that estimates of maternal and neonatal mortality are widely divergent in South Africa, and single-source reporting likely under-estimates the event rates. The application of capture recapture methods is a viable approach in South Africa to resolve the problems of under ascertainment in estimation of these outcomes
Camera traps as sensor networks for space-time exploration of terrestrial mammal communities
No full textMost of the conservation issues which ecologists are called on to help resolve are essentially about ecological communities. Camera trapping technology has led to a surge in the collection of large ecological datasets, which provides an unmissable opportunity to attain deeper knowledge of animal community assembly and structure. Using extensive camera trap data, this thesis examines whether camera traps can be used as sensor networks for a space-time exploration of the terrestrial mammal community that occurs in the Little Karoo of South Africa. In Chapter 1, the species-habitat relationship along a ruggedness gradient was studied. Using resource selection functions and multivariate statistics, this chapter showed that the strength of affinities, which mammals developed with specific terrain roughness, varied among species. It also enabled the recognition of subtle and continuous nuances in the spectrum of habitat preferences, providing a novel tool to explore the forces driving species coexistence in local animal communities. The theme of Chapter 2 was to consider patterns of seasonal occurrence within species circadian rhythms. Using kernel density functions with descriptive and multivariate statistics, this chapter showed that most mammal species responded to the ecological variability brought about by seasonality by adjusting their diel activity rhythms between winter and summer, resulting in a reduction of time exposure to a physiologically stressful environment caused by high temperatures in summer. It also highlighted that while some shifts only result from photoperiodism alignment, most are driven by other factors too. Chapter 3 examined temporal-partitioning as a mechanism driving sympatry. Using kernel density functions and mutivariate statistical analyses, this chapter enabled subtle nuances in the spectrum of diel activity rhythms to be visualised, highlighting the variety of temporal niche breadths and of activity onset/offset timings, which allowed diel activity rhythms to diversify and the mammal community to partition the temporal resources. Finally, in Chapter 4, topics dealing with leopard habitat preferences and leopard population density were explored. Using spatially explicit capture-recapture models, this chapter showed that leopard density remained low but varied with topographic relief; it increased with ruggedness of the terrain up to an optimum, and followed a reversed trend as the terrain roughness kept increasing. The population was composed of two groups of individuals with significantly different home range sizes, potentially explained by gender duality in movement. The chapter provided leopard density estimates ranging from 0.49 to 0.82 individual per 100 km2 . Local communities, such as that of the mammal species of the Little Karoo, are neither closed nor isolated. Therefore, it would be insightful if future studies were to embrace the metacommunity concept and explain these patterns of species distribution, abundance and interaction at multiple scales of spatio-temporal organisation
A continuous-time formulation for spatial capture-recapture models
No full textSpatial capture-recapture (SCR) models are relatively new but have become the standard approach used to estimate animal density from capture-recapture data. It has in the past been impractical to obtain sufficient data for analysis on species that are very difficult to capture such as elusive carnivores that occur at low density and range very widely. Advances in technology have led to alternative ways to virtually "capture" individuals without having to physically hold them. Some examples of these new non-invasive sampling methods include scat or hair collection for genetic analysis, acoustic detection and camera trapping. In traditional capture-recapture (CR) and SCR studies populations are sampled at discrete points in time leading to clear and well defined occasions whereas the new detector types mentioned above sample populations continuously in time. Re- searchers with data collected continuously currently need to define an appropriate occasion and aggregate their data accordingly thereby imposing an artificial construct on their data for analytical convenience. This research develops a continuous-time (CT) framework for SCR models by treating detections as a temporal non homogeneous Poisson process (NHPP) and replacing the usual SCR detection function with a continuous detection hazard func- tion. The general CT likelihood is first developed for data from passive (also called "proximity") detectors like camera traps that do not physically hold individuals. The likelihood is then modified to produce a likelihood for single-catch traps (traps that are taken out of action by capturing an animal) that has proven difficult to develop with a discrete-occasion approach. The lack of a suitable single-catch trap likelihood has led to researchers using a discrete-time (DT) multi-catch trap estimator to analyse single-catch trap data. Previous work has found the DT multi-catch estimator to be robust despite the fact that it is known to be based on the wrong model for single-catch traps (it assumes that the traps continue operating after catching an individual). Simulation studies in this work confirm that the multi-catch estimator is robust for estimating density when density is constant or does not vary much in space. However, there are scenarios with non-constant density surfaces when the multi-catch estimator is not able to correctly identify regions of high density. Furthermore, the multi-catch estimator is known to be negatively biased for the intercept parameter of SCR detection functions and there may be interest in the detection function in its own right. On the other hand the CT single-catch estimator is unbiased or nearly so for all parameters of interest including those in the detection function and those in the model for density. When one assumes that the detection hazard is constant through time there is no impact of ignoring capture times and using only the detection frequencies. This is of course a special case and in reality detection hazards will tend to vary in time. However when one assumes that the effects of time and distance in the time-varying hazard are independent, then similarly there is no information in the capture times about density and detection function parameters. The work here uses a detection hazard that assumes independence between time and distance. Different forms for the detection hazard are explored with the most exible choice being that of a cyclic regression spline. Extensive simulation studies suggest as expected that a DT proximity estimator is unbiased for the estimation of density even when the detection hazard varies though time. However there are indirect benefits of incorporating capture times because doing so will lead to a better fitting detection component of the model, and this can prevent unexplained variation being erroneously attributed to the wrong covariate. The analysis of two real datasets supports this assertion because the models with the best fitting detection hazard have different effects to the other models. In addition, modelling the detection process in continuous-time leads to a more parsimonious approach compared to using DT models when the detection hazard varies in time. The underlying process is occurring in continuous-time and so using CT models allows inferences to be drawn about the underlying process, for example the time- varying detection hazard can be viewed as a proxy for animal activity. The CT formulation is able to model the underlying detection hazard accurately and provides a formal modelling framework to explore different hypotheses about activity patterns. There is scope to integrate the CT models developed here with models for space usage and landscape connectivity to explore these processes on a finer temporal scale. SCR models are experiencing a rapid growth in both application and method development. The data generating process occurs in CT and hence a CT modelling approach is a natural fit and opens up several opportunities that are not possible with a DT formulation. The work here makes a contribution by developing and exploring the utility of such a CT SCR formulation
A continuous-time formulation for spatial capture-recapture models
No full textSpatial capture-recapture (SCR) models are relatively new but have become the
standard approach used to estimate animal density from capture-recapture data. It
has in the past been impractical to obtain sufficient data for analysis on species that
are very difficult to capture such as elusive carnivores that occur at low density and
range very widely. Advances in technology have led to alternative ways to virtually
“capture" individuals without having to physically hold them. Some examples of
these new non-invasive sampling methods include scat or hair collection for genetic
analysis, acoustic detection and camera trapping.
In traditional capture-recapture (CR) and SCR studies populations are sampled
at discrete points in time leading to clear and well defined occasions whereas the
new detector types mentioned above sample populations continuously in time. Researchers
with data collected continuously currently need to define an appropriate
occasion and aggregate their data accordingly thereby imposing an artificial construct
on their data for analytical convenience.
This research develops a continuous-time (CT) framework for SCR models by
treating detections as a temporal non homogeneous Poisson process (NHPP) and
replacing the usual SCR detection function with a continuous detection hazard function.
The general CT likelihood is first developed for data from passive (also called
“proximity") detectors like camera traps that do not physically hold individuals. The
likelihood is then modified to produce a likelihood for single-catch traps (traps that
are taken out of action by capturing an animal) that has proven difficult to develop
with a discrete-occasion approach.
The lack of a suitable single-catch trap likelihood has led to researchers using
a discrete-time (DT) multi-catch trap estimator to analyse single-catch trap data.
Previous work has found the DT multi-catch estimator to be robust despite the fact
that it is known to be based on the wrong model for single-catch traps (it assumes
that the traps continue operating after catching an individual). Simulation studies in
this work confirm that the multi-catch estimator is robust for estimating density when
density is constant or does not vary much in space. However, there are scenarios with
non-constant density surfaces when the multi-catch estimator is not able to correctly
identify regions of high density. Furthermore, the multi-catch estimator is known
to be negatively biased for the intercept parameter of SCR detection functions and
there may be interest in the detection function in its own right. On the other hand
the CT single-catch estimator is unbiased or nearly so for all parameters of interest
including those in the detection function and those in the model for density.
When one assumes that the detection hazard is constant through time there is
no impact of ignoring capture times and using only the detection frequencies. This
is of course a special case and in reality detection hazards will tend to vary in time.
However when one assumes that the effects of time and distance in the time-varying
hazard are independent, then similarly there is no information in the capture times
about density and detection function parameters. The work here uses a detection
hazard that assumes independence between time and distance. Different forms for
the detection hazard are explored with the most flexible choice being that of a cyclic
regression spline.
Extensive simulation studies suggest as expected that a DT proximity estimator is
unbiased for the estimation of density even when the detection hazard varies though
time. However there are indirect benefits of incorporating capture times because
doing so will lead to a better fitting detection component of the model, and this can
prevent unexplained variation being erroneously attributed to the wrong covariate.
The analysis of two real datasets supports this assertion because the models with the
best fitting detection hazard have different effects to the other models. In addition,
modelling the detection process in continuous-time leads to a more parsimonious
approach compared to using DT models when the detection hazard varies in time.
The underlying process is occurring in continuous-time and so using CT models
allows inferences to be drawn about the underlying process, for example the timevarying
detection hazard can be viewed as a proxy for animal activity. The CT
formulation is able to model the underlying detection hazard accurately and provides
a formal modelling framework to explore different hypotheses about activity patterns.
There is scope to integrate the CT models developed here with models for space usage
and landscape connectivity to explore these processes on a finer temporal scale.
SCR models are experiencing a rapid growth in both application and method
development. The data generating process occurs in CT and hence a CT modelling
approach is a natural fit and opens up several opportunities that are not possible
with a DT formulation. The work here makes a contribution by developing and
exploring the utility of such a CT SCR formulation
The scientific and sociocultural value of citizen science
No full textAt its best, conservation science serves as an interface between multiple disciplines, bringing together innovation and insight from the humanities, arts, business and economics, health, natural and social sciences to address the most pressing challenges facing the planet and its biodiversity. Conservation is also unique in its ability to transcend the academic. Subdisciplines such as biological citizen science involve non-academic individuals in data collection and as co-creators in research design and implementation. There is wide theoretical appeal to these initiatives, as they may simultaneously generate enough data for fine-scale biological monitoring and, through community engagement, promote the democratic generation and dissemination of knowledge. However, despite their potential within both the scientific and sociocultural realms, the regional scale impacts of citizen science initiatives in continental Africa remain poorly understood. Studying local citizen science initiatives affords opportunities to gain an understanding of their impact; within the African context, citizen science participants contribute to African Bird Atlas Projects in fifteen countries by compiling comprehensive avian species checklists at a fine geographic scale. Two projects are of particular interest for analysis: The Second Southern African Bird Atlas Project (SABAP2), and the Nigerian Bird Atlas Project (NiBAP). SABAP2 occupies a unique position as a long-term African citizen science project with ample data for statistical analyses. NiBAP employs an innovative grassroots approach to atlasing and hosts an enthusiastic and growing community of citizen science participants. In addition to impressive datasets, 'atlasers' in South Africa and Nigeria are a wealth of information regarding the personal- and community-level values surrounding participation and motivation. Furthermore, variations in project structure and cultural contexts between SABAP2 and NiBAP allow for some comparison of the values of participants between subprojects and contribute towards an understanding of the broader values of African Bird Atlas Project participants. In light of this potential, this thesis aims to examine the scientific and sociocultural contributions of SABAP2 and NiBAP to the work of conservation—namely achieving fine-scale species monitoring and advancing the democratic generation and sharing of scientific knowledge—within continental Africa. I examine the quality of data collected by a community of SABAP2 atlasers in Hessequa, South Africa, and assess their value in monitoring species population trends. I ask 1. Does systematic atlasing improve the temporal quality of atlas data for monitoring? and 2. Can these data detect trends in species populations and inform local response? Then, I explore values to nature expressed by atlasers and non-citizen science participants in both Hessequa, South Africa, and Jos, Nigeria. I ask the following two questions: 1. Does the type and frequency of values (instrumental, intrinsic, and relational) differ between participants and non-participants, and between cultural contexts?, and 2. what relational values (if any) are linked to participant motivation? Results demonstrate the ability of SABAP2 data to support biodiversity monitoring at a local scale, and show the potential for democratising research by including participants in data collection, analysis and application. The study of values and motivations emphasises the particular importance of relational values in connection with participant motivations and human-nature relationships broadly. Overall, the findings of this research demonstrate the ability of African Bird Atlas Projects and similar initiatives to meet the challenges of contemporary conservation, and lay a foundation for future research into the practical application of co-created monitoring schemes and incorporating values into the design of both citizen science projects and conservation intervention
Using mark-recapture methods to estimate population size and survival of pyjama sharks (Poroderma africanum) in Mossel Bay, South Africa
No full textSharks are vulnerable to exploitation as a result of their biological characteristics. Markrecapture models were applied to conventional tag recapture data and acoustic telemetry data to estimate abundance, apparent survival, recapture probability and temporary emigration for the pyjama shark, Poroderma africanum in Mossel Bay, South Africa over a five-year period. This study applied Pollock's robust design (with the conventional tag data) and Cormack-Jolly-Seber (CJS) models (with the acoustic tag data) to analyze the mark-recapture data. In addition, a von Bertalanffy model was fit to the data to estimate individual growth. The best-fit robust design model showed the population as having no temporary emigration, survival probability that is dependent on the length at first capture, and time-constant capture probabilities. The best-fit CJS model showed the population also having time-constant survival, but sex dependent capture probabilities. Robust design abundance estimates (with 95% C.I.) in Mossel Bay varied from 279 (102-787) sharks to 733 (320-1777) sharks, although confidence intervals were quite large. CJS apparent annual survival (95% C.I.; CJS) was estimated to be 0.271 year⁻¹ (0.04 to 0.56) and annual recapture probability (95% C.I.) was estimated to be 0.008 year⁻¹ (0.003-0.20), indicating that survival and recaptures for this endemic species are relatively low. Annual somatic growth rate (k) was estimated to be 0.213 year⁻¹, indicating that this population is slow growing, a characteristic common in most shark species. Overall, the results in this study provide baseline knowledge on this population in Mossel Bay and can be used to implement proper management techniques. This knowledge can be further expanded upon to give a more in-depth understanding of all size and age classes in the population and the role that the environment and anthropogenic activities play in the population structure
Complicities
No full textThis Open Access book offers a model of the human subject as complicit in the systems that structure human society and the human psyche which draws together clinical research with theory from both psychology and the humanities to advance a more social just theory and practice. Beginning from the premise that we cannot separate ourselves from the systems that precede and formulate us as subjects, the author argues that, in reckoning with this complicity, a model of subjectivity can be created that moves beyond binaries and identity politics. In doing so, the book examines how we might develop a more socially just psychological theory and practice, which is both systems work and intra-psychological work. In bringing together ways of thinking developed in the humanities with clinical psychotherapeutic practice, this book offers one interdisciplinary take on key questions of social and emotional efficacy in action-oriented psychotherapy work
Recommended from our members
Complicities
No full textThis Open Access book offers a model of the human subject as complicit in the systems that structure human society and the human psyche which draws together clinical research with theory from both psychology and the humanities to advance a more social just theory and practice. Beginning from the premise that we cannot separate ourselves from the systems that precede and formulate us as subjects, the author argues that, in reckoning with this complicity, a model of subjectivity can be created that moves beyond binaries and identity politics. In doing so, the book examines how we might develop a more socially just psychological theory and practice, which is both systems work and intra-psychological work. In bringing together ways of thinking developed in the humanities with clinical psychotherapeutic practice, this book offers one interdisciplinary take on key questions of social and emotional efficacy in action-oriented psychotherapy work
Variation in mammal species richness and relative abundance in the Karoo<sup>§</sup>
No full textUnderstanding how climate change and land transformation may impact the distribution and diversity of wildlife species requires landscape-level foundational biodiversity surveys. The Karoo BioGaps Project aims to provide such data and to support the scientific assessment for shale gas development projects in the Karoo basin. In this paper we present the findings of the BioGaps mammal survey, which recorded medium and large mammals across twenty-five 1 km × 1 km sampling sites within the proposed fracking footprint using camera trapping techniques. We use sample rarefaction curves, non-parametric species richness estimators and non-metric multidimensional scaling plots to explore both species richness and community structure. We also used a generalised linear model to investigate how species diversity varies with both site-specific and landscape-level predictors. A total of 38 species were recorded at the majority of sites. Longitude (z = 4.018, p = 0.0005) emerged as the best predictor of species diversity across the study area, which we suggest is linked to the clear east–west aridity gradient. Together these results reveal the cosmopolitan distribution of the mammal taxa in the Karoo and could be used to inform decision-making linked to mining activities in the area.</p
- …
