Bio-economic optimisation of surveillance to confirm broadscale eradications of invasive pests and diseases

Although pest eradications from islands have been successful and impart biodiversity benefits, eradications at regional/national scales are more challenging. Such broadscale eradications incur high repeated costs (e.g. control and surveillance effort) because the entire area cannot be treated at one time, and a progressive ‘treat-evaluate-move on’ approach must be employed. We describe a two-stage model to analyse surveillance data for assessing progress and declaring success of broadscale eradications, and to identify optimal cost-efficient surveillance strategies. Stage I modelling coincides or follows population control within a subset area or management zone (MZ). Surveillance data are analysed to quantify the probability of freedom for a treated MZ (i.e. local eradication), which is used to inform an operational decision to reallocate resources to other MZs, and progress across the region. Importantly, freedom declared individually in all MZs is not necessarily equivalent to a high probability of eradication over the broadscale area, because each MZ will have a probability of being erroneously declared free. After a MZ has been operationally declared free, Stage II surveillance commences to detect MZ-level failures, and to estimate the broadscale surveillance sensitivity and a corresponding probability of eradication. We developed a computer algorithm to identify cost-optimal Stage I and II surveillance strategies for a hypothetical large area. We assessed the following: (1) the balance between local surveillance intensity and spatial coverage; (2) the number of years to declare success in Stages I and II; (3) the stopping probability of freedom (Stage I); and (4) the optimal strategy given variation in the starting-over cost, should a MZ be erroneously declared free. This two-stage approach provides an objective basis for decision-making in wildlife pest/disease eradication, and guidance for implementing optimal bio-economic surveillance strategies.     

Quantitative evaluation of complex surveillance systems for pest and disease detection

Surveillance for disease detection is used primarily for early detection of incursions and to support assertions of freedom from disease.Analytical techniques used to evaluate the efficacy of such surveillance are equally applicable across the domains of invasive species (both plant and animal), diseases and pests of agriculture crops, livestock, and of wildlife. Scenario tree models of surveillance activities may be used to estimate their diagnostic sensitivities, or the probability that the target organism will be detected given that it is present at a defined level. This paper will outline techniques for estimating the sensitivity of both targeted and general surveillance activities, and for the surveillance system as a whole. Probability of freedom from the target organism may be estimated from the surveillance sensitivity, and this Bayesian approach may be extended to estimate current probability of freedom from appropriate use of historical and ongoing surveillance evidence.     

Eradication of Feral Pigs From Pinnacles National Monument

Abstract: Feral pigs (Sus scrofa) have caused considerable damage where they have been introduced around the world. At Pinnacles National Monument, California, USA, managers were concerned that feral pigs were damaging wetland habitats, reducing oak regeneration, competing with native wildlife, and dispersing nonnative plant species through soil disturbance. To address these threats the National Park Service constructed an exclosure around 57 km² of monument land and through cooperation with the Institute for Wildlife Studies eradicated all feral pigs within the area. Trapping, ground-hunting, hunting dogs, and Judas techniques were used to remove feral pigs. Trapping techniques removed most pigs, but a combination of techniques was required to cause eradication. A series of bait sites and transects across the monument helped focus removal efforts and facilitated detection of the last remaining feral pigs in the exclosure. Consistent funding and cooperation from the National Park Service allowed for a seamless and comprehensive program that provided intensive removal of feral pigs. The successful eradication of feral pigs at Pinnacles National Monument should encourage managers in other areas to implement future control or eradication programs.     

Sampling Considerations for Disease Surveillance in Wildlife Populations

Abstract Disease surveillance in wildlife populations involves detecting the presence of a disease, characterizing its prevalence and spread, and subsequent monitoring. A probability sample of animals selected from the population and corresponding estimators of disease prevalence and detection provide estimates with quantifiable statistical properties, but this approach is rarely used. Although wildlife scientists often assume probability sampling and random disease distributions to calculate sample sizes, convenience samples (i.e., samples of readily available animals) are typically used, and disease distributions are rarely random. We demonstrate how landscape-based simulation can be used to explore properties of estimators from convenience samples in relation to probability samples. We used simulation methods to model what is known about the habitat preferences of the wildlife population, the disease distribution, and the potential biases of the convenience-sample approach. Using chronic wasting disease in free-ranging deer (Odocoileus virginianus) as a simple illustration, we show that using probability sample designs with appropriate estimators provides unbiased surveillance parameter estimates but that the selection bias and coverage errors associated with convenience samples can lead to biased and misleading results. We also suggest practical alternatives to convenience samples that mix probability and convenience sampling. For example, a sample of land areas can be selected using a probability design that oversamples areas with larger animal populations, followed by harvesting of individual animals within sampled areas using a convenience sampling method.     

Application of techniques for the quantitative analysis of complex surveillance systems to disease and wildlife management

Scenario tree modelling and associated methods provide tools to quantify the combined value of multiple complex surveillance activities over time. The outputs of this analysis are an estimate of the sensitivity of a surveillance system, and the cumulative probability of disease freedom that surveillance provides over time. The ability to quantify the performance of complex surveillance systems provides a number of new opportunities in the design and application of search and detection activities. One of these is the ability to objectively compare alternative detection strategies. The sensitivity of a strategy (the probability that the target biota would be detected, given that they are present at a defined level) may be balanced against cost and practicality considerations to determine the most effective strategy for a given situation. The paper provides an example of the comparison of two surveillance strategies (structured surveys and abattoir surveillance) for disease detection in animal health (Classical Swine Fever). These techniques may also play an important role in the certification of success in pest eradication operations. The ability to use multiple sources of evidence to evaluate success means that a higher level of confidence can often be achieved at lower cost. Examples of the application of scenario tree modelling in plant health and invasive pests are provided to illustrate its use within eradication programs.     

Population probability models for insect eradication

One of the greatest challenges in eradicating pest species is determining when no further individuals remain: terminating the control programme too early means failure to eradicate, whereas continuing for too long can add considerable expense. Since monitoring tools are usually only qualitative and invariably imperfect, there may be considerable uncertainty about when and if eradication has been achieved. However, it is possible to quantify the efficacy of monitoring tools and to use this together with knowledge of the basic ecology of the target pest to robustly quantify the probability of successful eradication over time. Here, I describe one such approach and demonstrate its use in the large-scale eradication of painted apple moth (Teia anartoides) from Auckland, New Zealand. A population model for the production of male moths was used in conjunction with spatially-explicit pheromone trap locations and attraction radii to determine the daily probability of detecting a hypothetical wild population at a particular location. Over time, these probabilities compounded to decrease the likelihood of painted apple moth presence given an ongoing lack of detection. In this way, spatio-temporal risk maps were produced to inform managers and to suggest when eradication had been achieved to a predetermined level of certainty. The model suggested that eradication was likely to have been successful in the main infestation areas by mid 2005, with subsequent catches likely to represent further small incursions, as corroborated by evidence from mitochondrial DNA and stable isotope markers. While it was plausible that a wild population was present in the Otahuhu area in 2005, it was very unlikely that it remained by the end of 2006. Population probability models have potential for much wider use in border biosecurity and establishment of area freedom, particularly in combination with future automated trapping systems.     

Network theory as a method for developing biosecurity preparedness: Netweork topology of Paratrechina longicornis in the Pacific and within New Zealand

One of the greatest challenges in eradicating pest species is determining when no further individuals remain: terminating the control programme too early means failure to eradicate, whereas continuing for too long can add considerable expense. Since monitoring tools are usually only qualitative and invariably imperfect, there may be considerable uncertainty about when and if eradication has been achieved. However, it is possible to quantify the efficacy of monitoring tools and to use this together with knowledge of the basic ecology of the target pest to robustly quantify the probability of successful eradication over time. Here, I describe one such approach and demonstrate its use in the large-scale eradication of painted apple moth (Teia anartoides) from Auckland, New Zealand. A population model for the production of male moths was used in conjunction with spatially-explicit pheromone trap locations and attraction radii to determine the daily probability of detecting a hypothetical wild population at a particular location. Over time, these probabilities compounded to decrease the likelihood of painted apple moth presence given an ongoing lack of detection. In this way, spatio-temporal risk maps were produced to inform managers and to suggest when eradication had been achieved to a predetermined level of certainty. The model suggested that eradication was likely to have been successful in the main infestation areas by mid 2005, with subsequent catches likely to represent further small incursions, as corroborated by evidence from mitochondrial DNA and stable isotope markers. While it was plausible that a wild population was present in the Otahuhu area in 2005, it was very unlikely that it remained by the end of 2006. Population probability models have potential for much wider use in border biosecurity and establishment of area freedom, particularly in combination with future automated trapping systems.     

Emerging Infectious Diseases in Free-Ranging Wildlife–Australian Zoo Based Wildlife Hospitals Contribute to National Surveillance

Emerging infectious diseases are increasingly originating from wildlife. Many of these diseases have significant impacts on human health, domestic animal health, and biodiversity. Surveillance is the key to early detection of emerging diseases. A zoo based wildlife disease surveillance program developed in Australia incorporates disease information from free-ranging wildlife into the existing national wildlife health information system. This program uses a collaborative approach and provides a strong model for a disease surveillance program for free-ranging wildlife that enhances the national capacity for early detection of emerging diseases.     

Bio-Economics of Large-Scale Eradication of Feral Goats From Santiago Island,Galápagos

ABSTRACT Invasive mammals are premier drivers of extinction and ecosystem change, particularly on islands. In the 1960s, conservation practitioners started developing techniques to eradicate invasive mammal populations from islands. Larger and more biologically complex islands are being targeted for restoration worldwide. We conducted a feral goat (Capra hircus) eradication campaign on Santiago Island in the Galápagos archipelago, which was an unprecedented advance in the ability to reverse biodiversity impacts by invasive species. We removed >79,000 goats from Santiago Island (58,465 ha) in <4.5 years, at an approximate cost of US$6.1 million. An eradication ethic combined with a suite of techniques and technologies made eradication possible. A field-based Geographic Information System facilitated an adaptive management strategy, including adjustment and integration of hunting methods. Specialized ground hunting techniques with dogs removed most of the goat population. Aerial hunting by helicopter and Judas goat techniques were also critical. Mata Hari goats, sterilized female Judas goats induced into a long-term estrus, removed males from the remnant feral population at an elevated rate, which likely decreased the length and cost of the eradication campaign. The last 1,000 goats cost US$2.0 million to remove; we spent an additional US$467,064 on monitoring to confirm eradication. Aerial hunting is cost-effective even in countries where labor is inexpensive. Local sociopolitical environments and best practices emerging from large-scale, fast-paced eradications should drive future strategies. For nonnative ungulate eradications, island size is arguably no longer the limiting factor. Future challenges will involve removing invasive mammals from large inhabited islands while increasing cost-effectiveness of removing low-density populations and confirming eradication. Those challenges will require leveraging technology and applying theory from other disciplines, along with conservation practitioners working alongside sociologists and educators.     

An open spatial capture–recapture model for estimating density,movement, and population dynamics from line‐transect surveys

The purpose of many wildlife population studies is to estimate density, movement, or demographic parameters. Linking these parameters to covariates, such as habitat features, provides additional ecological insight and can be used to make predictions for management purposes. Line‐transect surveys, combined with distance sampling methods, are often used to estimate density at discrete points in time, whereas capture–recapture methods are used to estimate movement and other demographic parameters. Recently, open population spatial capture–recapture models have been developed, which simultaneously estimate density and demographic parameters, but have been made available only for data collected from a fixed array of detectors and have not incorporated the effects of habitat covariates. We developed a spatial capture–recapture model that can be applied to line‐transect survey data by modeling detection probability in a manner analogous to distance sampling. We extend this model to a) estimate demographic parameters using an open population framework and b) model variation in density and space use as a function of habitat covariates. The model is illustrated using simulated data and aerial line‐transect survey data for North Atlantic right whales in the southeastern United States, which also demonstrates the ability to integrate data from multiple survey platforms and accommodate differences between strata or demographic groups. When individuals detected from line‐transect surveys can be uniquely identified, our model can be used to simultaneously make inference on factors that influence spatial and temporal variation in density, movement, and population dynamics.     

Genetic population assignment reveals a long-distance incursion to an island by a stoat (Mustela erminea)

When new individuals from a pest species are detected after an eradication programme, it is important to determine if these individuals are survivors from the eradication attempt or reinvaders from another population, as this enables managers to adjust and improve the methodologies for future eradications and biosecurity. Rangitoto/Motutapu Islands in the Hauraki Gulf (New Zealand) had a multispecies mammalian pest eradication conducted in 2009. A year after this eradication a single stoat was trapped on the island. Using genetic population assignment we conclude that this individual was a reinvader, which probably swam a minimum distance of 3 km from the adjacent mainland. This swimming distance is greater than any previously known stoat incursions. Our results suggest that the original population on these islands was from natural dispersal rather than anthropogenic introduction and that it had some limited ongoing mixing with the mainland population. These findings highlight the invasion/reinvasion potential of stoats across large stretches of water, and will necessitate ongoing biosecurity indefinitely for these islands. The study also highlights the utility of genetic assignment techniques for assessing reinvasion, and emphasizes the need for pre-eradication genetic sampling of all pest species to enable such analyses to be carried out.     

Quantifying the success of feral cat eradication, San Nicolas Island, California

It is usually uncertain when to declare success and stop control in pest eradication operations that rely on successive reductions of the population. We used the data collected during a project to eradicate feral cats from San Nicolas Island, California to estimate both the number of cats remaining towards the end of the project, and the amount and type of surveillance effort required to declare successful eradication after the last known cat was removed. Fifty seven cats were removed between June 2009 and April 2010 and our model estimated that there was a 95% chance that a further 1 to 4 cats remained, with 1 cat being the most likely number. After this time a further two cats were detected and removed and the model predicted this outcome with a probability of 0.25. If managers wished to confirm eradication success at this point, we estimated that 55 km of effort searching for recent evidence of cats over the whole island without detecting any would provide 99% certainty that no cats remained (stopping rule 1). Alternatively, the optimal amount of search effort for evidence that minimized the joint cost of searching and the cost of wrongly declaring eradication was 75 km (stopping rule 2). The equivalent amount of camera-nights (26 cameras were available) required to declare successful eradication were 416 (stopping rule 1) and 1196 camera nights (stopping rule 2). During the confirmation phase, 270 km of sign search effort and 3294 camera-nights surveillance were used from late June 2010, when the last cat was removed, through August 2010, without detecting signs of survivors. Managers can be very confident that eradication has been successful.     

Managing feral cats on a university's campuses: how many are there and is sterilization having an effect?

Worldwide domestic and feral cat (Felis catus) numbers have increased. Concerns regarding high populations of feral cats in urban areas include wildlife predation, public nuisance, and disease. This study aimed to estimate the size of the feral cat population on 5 campuses of the University of KwaZulu-Natal, South Africa, to determine whether sterilization has an effect and to make management recommendations. The study used both the total count and mark-recapture methods to estimate the feral cat population on each campus. The study chose a noninvasive method of taking photographs to "mark" individuals and record those who were sterilized. The study estimated a total of 186 cats on all campuses and density at 161 cats km(-2). There was a negative relationship between sterilization and numbers. Sites with higher sterilization showed a lower proportion of younger cats. At the average sterilization of 55%, the population, according to predictions, would remain stable at fecundity, survival, and immigration rates reported by cat caretakers. However, caretakers underestimated cat abundance by 7 ± 37 SD%. Caretakers' feral cat sterilization and feeding programs appear to provide a service to the university community. Key management recommendations were to increase sterilization to 90% to reduce the population over the long term and to raise funds to support the costs incurred by voluntary cat caretakers.     

A serosurvey for Brucella suis, classical swine fever virus, porcine circovirus type 2, and pseudorabies virus in feral swine (Sus scrofa) of eastern North Carolina

As feral swine (Sus scrofa) populations expand their range and the opportunity for feral swine hunting increases, there is increased potential for disease transmission that may impact humans, domestic swine, and wildlife. From September 2007 to March 2010, in 13 North Carolina, USA, counties and at Howell Woods Environmental Learning Center, we conducted a serosurvey of feral swine for Brucella suis, pseudorabies virus (PRV), and classical swine fever virus (CSFV); the samples obtained at Howell Woods also were tested for porcine circovirus type 2 (PCV-2). Feral swine serum was collected from trapped and hunter-harvested swine. For the first time since 2004 when screening began, we detected B. suis antibodies in 9% (9/98) of feral swine at Howell Woods and <1% (1/415) in the North Carolina counties. Also, at Howell Woods, we detected PCV-2 antibodies in 59% (53/90) of feral swine. We did not detect antibodies to PRV (n=512) or CSFV (n=307) at Howell Woods or the 13 North Carolina counties, respectively. The detection of feral swine with antibodies to B. suis for the first time in North Carolina warrants increased surveillance of the feral swine population to evaluate speed of disease spread and to establish the potential risk to commercial swine and humans.     

Improved surveillance for early detection of a potential invasive species: the alien Rose-ringed parakeet <Emphasis Type="Italic">Psittacula krameri</Emphasis> in Australia

The Rose-ringed parakeet Psittacula krameri is the most widely introduced parrot in the world, and is an important agricultural pest and competitor with native wildlife. In Australia, it is classified as an ‘extreme threat’, yet captive individuals frequently escape into the wild. The distribution and frequency of incursions are currently unknown, as are the potential impacts of the species in Australia. This lack of critical ecological information greatly limits effective biosecurity surveillance and decision-making efforts. We compiled a unique dataset, which combined passive surveillance sources from government and online resources, for all available information on parakeet detections at-large in Australia. We investigated whether geographic variables successfully predicted parakeet incursions, and used species distribution models to assess the potential distribution and economic impacts on agricultural assets. We recorded 864 incursions for the period 1999–2013; mostly escaped birds reported to missing animal websites. Escapes were reported most frequently within, or around, large cities. Incursions were best predicted by factors related to human presence and activity, such as global human footprint and intensive land uses. We recommend surveillance of high (predicted) establishment areas adjacent to cities where a feral parakeet population could most affect horticultural production. Novel passive surveillance datasets combined with species distribution models can be used to identify the regions where potential invasive species are most likely to establish. Subsequently, active surveillance can be targeted to the areas of highest predicted potential risk. We recommend an integrated approach that includes outreach programs involving local communities, as well as traditional biosecurity surveillance, for detecting new incursions.     

Deciphering Serology to Understand the Ecology of Infectious Diseases in Wildlife

The ecology of infectious disease in wildlife has become a pivotal theme in animal and public health. Studies of infectious disease ecology rely on robust surveillance of pathogens in reservoir hosts, often based on serology, which is the detection of specific antibodies in the blood and is used to infer infection history. However, serological data can be inaccurate for inference to infection history for a variety of reasons. Two major aspects in any serological test can substantially impact results and interpretation of antibody prevalence data: cross-reactivity and cut-off thresholds used to discriminate positive and negative reactions. Given the ubiquitous use of serology as a tool for surveillance and epidemiological modeling of wildlife diseases, it is imperative to consider the strengths and limitations of serological test methodologies and interpretation of results, particularly when using data that may affect management and policy for the prevention and control of infectious diseases in wildlife. Greater consideration of population age structure and cohort representation, serological test suitability and standardized sample collection protocols can ensure that reliable data are obtained for downstream modeling applications to characterize, and evaluate interventions for, wildlife disease systems.     

Improving inference for aerial surveys of bears: The importance of assumptions and the cost of unnecessary complexity

Obtaining useful estimates of wildlife abundance or density requires thoughtful attention to potential sources of bias and precision, and it is widely understood that addressing incomplete detection is critical to appropriate inference. When the underlying assumptions of sampling approaches are violated, both increased bias and reduced precision of the population estimator may result. Bear (Ursus spp.) populations can be difficult to sample and are often monitored using mark‐recapture distance sampling (MRDS) methods, although obtaining adequate sample sizes can be cost prohibitive. With the goal of improving inference, we examined the underlying methodological assumptions and estimator efficiency of three datasets collected under an MRDS protocol designed specifically for bears. We analyzed these data using MRDS, conventional distance sampling (CDS), and open‐distance sampling approaches to evaluate the apparent bias‐precision tradeoff relative to the assumptions inherent under each approach. We also evaluated the incorporation of informative priors on detection parameters within a Bayesian context. We found that the CDS estimator had low apparent bias and was more efficient than the more complex MRDS estimator. When combined with informative priors on the detection process, precision was increased by >50% compared to the MRDS approach with little apparent bias. In addition, open‐distance sampling models revealed a serious violation of the assumption that all bears were available to be sampled. Inference is directly related to the underlying assumptions of the survey design and the analytical tools employed. We show that for aerial surveys of bears, avoidance of unnecessary model complexity, use of prior information, and the application of open population models can be used to greatly improve estimator performance and simplify field protocols. Although we focused on distance sampling‐based aerial surveys for bears, the general concepts we addressed apply to a variety of wildlife survey contexts.     

Targeted surveillance of raccoon rabies in Québec,Canada

Data from wildlife disease surveillance programs are used to inform implementation of disease control (e.g., vaccination, population reduction) in space and time. We developed an approach to increase detection of raccoon rabies in raccoons (Procyon lotor) and skunks (Mephitis mephitis) of Québec, Canada, and we examined the implications of using this approach for targeted surveillance. First we modeled the probability of a rabid animal relative to environmental characteristics of sampling locations. Rabid animals were more likely to be found in low-lying flat landscapes that had higher proportions of corn-forest edge habitat and hay agriculture, and that were within 20 km of one or more known rabies cases. From the model, we created 2 complementary risk maps to identify areas where rabid animals were most likely to be sampled. One map accounted for habitat and known rabies case locations, and can be used to define an infection zone from which surveillance can be targeted along the periphery to determine if disease is continuing to spread. The other map only accounted for habitat and can be used to locate areas most likely to contain rabid animals when the disease is present. In a further analysis we compared the 2 most successful methods for detecting raccoon rabies in Québec, given the disease was present. Government trapping operations (active surveillance) detected more cases in the short-term, but citizen notification (passive and enhanced) was more effective after 12 trapping days from which the initial rabies case was found. Our approach can benefit wildlife and public health agencies wanting to assess the disease status of regions by targeting surveillance to habitats most likely to contain infected animals and by defining the duration over which sampling methods are effective. © 2011 The Wildlife Society.     

Framing messages to support feral dog eradication: Both ecocentric and anthropocentric frames work

Due to their impact on biodiversity, human health, and livestock husbandry, we need to eradicate feral dogs. However, eradication usually encounters public opposition, which hinders its implementation and success. Persuasive messages could help us attain public support for feral dog eradication. We evaluated the effectiveness of two message frames to increase intentions to support eradication of feral dogs via lethal methods. Messages addressed the negative impacts of feral dogs on (1) human health and livestock husbandry (anthropocentric frame) and (2) wildlife (ecocentric frame). These frames were randomly assigned to 506 Chilean citizens and Chilean residents in a before-after survey experiment measuring attitudes, subjective norm, and behavioral intentions to support feral dog eradication via lethal methods. Both frames significantly increased behavioral intentions, attitude, and subjective norm, irrespective of participant demographics. Changes in attitude and subjective norm influenced change in intentions. Both frames can aid managers increase public support to eradicate feral dogs.     

Quantifying the errors in animal contacts recorded by proximity loggers

Automated contact detection by means of proximity loggers permits the measurement of encounters between individuals (animal-animal contacts) and the time spent by individuals in the proximity of a focal resource of interest (animal-fixed logger contacts). The ecological inference derived from contact detection is intrinsically associated with the distance at which the contact occurred. But no proximity loggers currently exist that record this distance and therefore all distance estimations are associated with error. Here we applied a probabilistic approach to model the relationship between contact detection and inter-logger distance, and quantify the associated error, on free-ranging animals in semi-controlled settings. The probability of recording a contact declined with the distance between loggers, and this decline was steeper for weaker radio transmission powers. Even when proximity loggers were adjacent, contact detection was not guaranteed, irrespective of the radio transmission power. Accordingly, the precision and sensitivity of the system varied as a function of inter-logger distance, radio transmission power, and experimental setting (e.g., depending on animal body mass and fine-scale movements). By accounting for these relationships, we were able to estimate the probability that a detected contact occurred at a certain distance, and the probability that contacts were missed (i.e., false negatives). These calibration exercises have the potential to improve the predictability of the study and enhance the applicability of proximity loggers to key wildlife management issues such as disease transmission rates or wildlife use of landscape features and resources.