David Choquenot

Setting thresholds for pest control: how does pest density affect resource viability?

Abstract Conservation in New Zealand is largely focused on reducing the impact introduced mammals have on the abundance of indigenous species. Conservation managers have a range of strategies they can employ to control these pests, but the combination that maximises conservation gains depends on the protection each strategy affords, and the scale at which it can be applied. Given a limited budget, the use of threshold pest densities to initiate pest control can increase control effectiveness by reducing opportunity costs. However, complex trophic relationships between pests and resources mean that thresholds which minimise the costs of controlling pests without reducing the viability of threatened populations to unacceptable levels will often be difficult to identify. Here we review three general consumer–resource models in the context of pest control; (1) the damage function based on the functional response of pests to resource abundance, (2) density dependent predator–prey models, and (3) interactive models. Damage functions can be used to set threshold pest densities that achieve tactical but not strategic conservation outcomes. Density dependent predator–prey models can be used to set threshold pest densities that have strategic consequences for resource conservation, but are limited in their scope where pest or resource abundance is influenced by density independent environmental perturbation. Interactive models can be used to identify thresholds for imposition of pest control that are responsive to pest density, resource abundance and prevailing environmental conditions. We advocate this modelling framework as a basis for setting control thresholds for pests in New Zealand.

Testing the relative influence of instrinsic and extrinsic variation in food availability on feral pig populations in Australia's rangelands

Instrinsic variation in the availability of food to animal populations reflects the influence of foraging by the animals themselves. Instrinsic variation in food availability provides a link between population density, subsequent food availability and variation in the rate of population increase (r), operating through density-dependent food shortage. In contrast, extrinsic variation in food availability is caused by environmental influences on food or animal abundance, which are independent of animal foraging. Extrinsic variation in food availability is random relative to that arising through intrinsic shortage. Intrinsic and extrinsic variation in food availability can influence animal populations simultaneously. Intrinsic variation will impart a tendency towards an equilibrium between animal and food abundance. which will be progressively obscured by density-independent variation as the influence of extrinsic factors increases. This study used a large-scale field experiment, in which the density of food-limited feral pig (Sus Scrofa L.) populations was manipulated on six sites, to assess the relative influence of intrinsic and extrinsic variation in food availability. The experiment evaluated the influence of pig population density on r and the abundance of food resources measured as pasture biomass. It was predicted that if intrinsic shortages dominated variation in food availability, pasture biomass and r would decline with increasing pig density. If extrinsic factors dominated variation in food availability, pig density would have no systematic effect on either pasture biomass or r. If intrinsic and extrinsic sources simultaneously affected variation in food availability, higher pig densities would have no systematic effect on r, but would reduce pasture biomass. The simultaneous model predicts reduced pasture biomass because, in the absence of compensatory changes in other sources of variation, the effects of intrinsic and extrinsic factors will be additive. To examine further the degree of interdependence in pig and pasture abundance, a series of stochastic models of the grazing system were estimate and the feedback loop comprising the functional and numerical responses of feral pigs to variation in pasture biomass was manipulated. In the large-scale experiment, neither pasture biomass nor r declined with increasing pig density, suggesting that food availability was dominated by extrinsic factors. However, limitations of the experiment meant that a minor decline in pasture biomass may have gone undetected. Comparison of simulation models, which included and omitted pasture offtake by pigs, indicated that because they were less efficient grazers and persisted at lower average densities relative to other large herbivores, pigs had little influence on variation in pasture biomass.

Assessing Lamb Predation by Feral Pigs in Australia's Semi-Arid Rangelands

Feral pigs prey upon newborn lambs in the semi-arid rangelands of eastern Australia. In this study. two experiments were used to investigate the relationship between feral pig density and the rate of lamb predation. The differential susceptibility of single and twin-born lambs to pig predation was also examined. In Experiment 1, the density of pig populations on four sites subjected to different levels of pig control was monitored quarterly (range: 0.7-6.4 pigs km -2 ) between 1989 and 1991. The rate of lamb loss was indexed by udder-scoring a sample of lambing ewes to estimate the proportion whose lambs had died. The index of lamb loss varied significantly with pig density, the relationship taking the form of an asymptotic increase in loss at progressively higher pig densities. Inclusion of a variable indicating whether pig density increased or decreased over the lambing season (summarizing the cumulative effects of recent seasonal conditions) explained no additional variation in lamb loss. This suggests that the availability of alternative foods did not influence the propensity of pigs to prey upon lambs. Experiment 2 was conducted on three sites with respective pig densities of 0.4, 2.4 and 5.8 pigs km -2 . On each site, two paddocks were established, one of which was electrically fenced to preclude pigs. Approximately 300 pregnant ewes at each site were ultrasonically scanned to determine whether they carried single or twin lambs, and then placed randomly in either the paddock to which pigs had access or the paddock from which they were excluded. The proportion of lambs born that survived was contrasted between the two paddocks at each site to provide a direct estimate of predation rate. The rate of lamb predation increased with feral pig density, reaching a maximum proportional rate of 0.29 on the site where pig density was highest. Comparison of predation rates indicated that twin lambs were on average 5-6 times more likely to be preyed upon by pigs than were single lambs. The relationship between pig density and the index of predation rate estimated in Experiment 1 was modified to predict actual predation rates by substituting the maximum rate measured in Experiment 2. The resulting relationship, in combination with a measure of stochastic year-to-year variation in predation rates, was used to construct distributions describing the probability of sustaining different rates of lamb predation at given feral pig densities. This information enables sheep farmers to contrast the risk of lamb predation with their lamb production objectives and the costs of holding feral pig density below levels corresponding to acceptable rates of lamb predation. By valuing the reduced risk of sustaining high levels of lamb predation and the costs associated with achieving it, a sheep farmer can incorporate reduction in costs and benefits of pig control directly into economic analyses used to manage their enterprise.

A comparison of population estimates and abundance indices for house mice inhabiting beech forests in New Zealand

Populations of house mice were sampled on nine grids in Fiordland National Park between May and November 1999, using live-capture and footprint-tracking tunnel methods. Trapped mice were removed from three grids (approximately 3.24 ha each), and marked and released on the other six. Density estimates were obtained using recapture data from the grids where mice were released (Mh (jackknife) model from program CAPTURE), and rate-of-capture data from grids where mice were removed (Mbh (removal) CAPTURE model). Density estimates were used to evaluate the performance of 4 indices of mouse abundance by contrasting R2-values of their regression on estimated mouse density. Indices evaluated were: minimum number of individuals known to be alive (MNA) (total number of individual mice caught over the course of a trapping session), one-night trap catch (number of mice caught on first night of each trapping session expressed as captures per 100 trap-nights), three-night trap catch (same index estimated from number of mice caught over the first three nights), and tracking-tunnel index (proportion of nine tunnels that had mouse tracks). While MNA, one-night trap catch, and three-night trap catch were all significantly correlated with estimated density, MNA was most strongly correlated, with R2 varying between 0.67 to 0.87 depending on whether 3, 4 or 5 nights’ capture data were used. Variation in tracking-tunnel index was unrelated to mouse density on our grids.

INITIAL EFFECTS OF RABBIT HAEMORRHAGIC DISEASE ON FREE-LIVING RABBIT (ORYCTOLAGUS CUNICULUS) POPULATIONS IN CENTRAL-WESTERN NEW SOUTH WALES

Quarterly spotlight counts of rabbits were conducted at three sites in central-western New South Wales. These counts commenced two years before the arrival of rabbit haemorrhagic disease (RHD) in the winter of 1996. The existing data on quarterly rates of change in rabbit abundance for the three populations provided a unique opportunity to study the effects of RHD on rabbit demography. Prior to the arrival of RHD, all three populations underwent phases of sequential increase and decrease in each year. On the basis of these patterns, RHD had a variable influence on the demography of the three rabbit populations. In 1996–97, the density of two populations declined over an expected period of increase, while at the third site the density increased as expected from pre-RHD patterns. Twelve months after their failure to generate expected positive rates of increase the two affected populations had returned to the normal sequence of increases and decreases in density although still at comparatively low numbers.

Aerial baiting of feral pigs (Sus scrofa) for the control of exotic disease in the semi-arid rangelands of New South Wales

An experiment that held the density of feral pigs constant while varying the effective density of aerially distributed baits was conducted at three sites in north-western New South Wales. Meat baits, containing one of the biomarkers iophenoxic acid, tetracycline or rhodamine B, were distributed at different intensities over each site, and a sample of pigs was shot from a helicopter at each site to determine bait uptake. Serum and tissue samples taken from each pig were analysed for the occurrence of the biomarkers; the proportions of pigs exhibiting biomarkers repre- sented the proportions of the feral pig populations that had consumed baits at different baiting intensities (expressed as baits per unit of pig density). The maximum percentage of sampled pigs that had eaten baits varied from 31% to 72% across the three sites. Bait uptake was regressed against baiting intensity. For two of the trials, the quantity of bait hypothetically required to eliminate a population of feral pigs was extrapolated to be 1577 baits per unit of pig density, while for the third trial 1874 baits per unit of pig density would have been required. Bait-uptake by non-target animals was substantial, posing potential hazards to birds and reducing the availability of baits to feral pigs. Most likely, seasonal conditions affected bait-uptake by feral pigs. We discuss the implications of these results for exotic disease contingency plan- ning.

Comparing cost-effectiveness of commercial harvesting, state-funded culling, and recreational deer hunting in New Zealand

Abstract The 7 taxa of deer introduced to New Zealand are officially regarded as pests but also are valued by hunters and commercial harvesters, who often debate the need for deer population control. Some hunters argue that it should be more cost-effective to enhance existing private hunting effort than to use state-employed cullers to kill deer. To explore that argument, we combined predator–prey and economic theory to predict how net revenue (carcass value minus cost of harvesting it) for a commercial helicopter-based venison-recovery operation was likely to vary with deer density and to result in stable harvest equilibria when marginal net revenue was zero. We then adapted that model to simulate the cost of state-funded deer control and the net satisfaction obtained by ground-based recreational hunters. Key findings were that 1) payment of incentives to commercial harvesters usually will be more cost-effective than state-funded culling, 2) payment of incentives to recreational hunters usually will not be effective unless time costs can be reduced at little monetary cost, and 3) ground-based cullers will be more effective than helicopter-based hunters at attaining low deer densities in dense forest. Key management implications are that commercial hunting can be cost-effectively manipulated to enhance control of deer populations, but neither commercial nor recreational hunting is likely to be a cost-effective alternative to state-funded control where very low densities are required in inaccessible or difficult-to-hunt areas. Although developed for deer control in New Zealand, the models are applicable to any situation in which harvesting is used as a form of population control.

Observations on the impacts of rabbit haemorrhagic disease on agricultural production values in Australia

Rabbit haemorrhagic disease (RHD) may be the most important rabbit control agent to be made available to graziers in Australia since the advent of myxomatosis. Documenting the benefits of RHD to agricultural production values is an important process in determining best-practice strategies for the use of the disease in controlling rabbit populations. In this paper we review previous studies on the impact of rabbits and present recent Australian case studies that tracked the effects of RHD on agricultural production as the disease first spread across the continent. Indirect consequences of RHD, such as changes in costs of rabbit control as monitored through the use of 1080 (sodium monofluoroacetate), are reported. Potential negative impacts such as adverse effects on the wild rabbit fur and meat trade and in the spread of woody weeds are also discussed.

Trophic Interactions Among Native and Introduced Animal Species

In the absence of exotic mammals, the beech forest system is strongly driven bottom-up. The sporadic heavy seeding of beech trees results in a cascade of population increases in the native fauna, without any known reciprocal effects. In dryland ecosystems, the nutrient pulses occur annually during spring flushes of herbaceous plants. In both the little modified beech forest and the highly human-impacted dryland ecosystems, mammalian introductions have been made at both the herbivore and predator trophic levels. These exotic additions have created strong top-down effects on indigenous fauna because predator abundance (stoat, ferret, and cat) is driven mainly by exotic prey species (mice and rabbits). Predator numbers can reach levels not normally possible without the introduced prey, and this can potentially lead to extinction of the native fauna. The worst scenario for native prey occurs when mice and rabbit numbers fluctuate widely. This leads to acute bouts of predation caused by the increases in predator numbers (in the case of stoats), or as ferrets and cats switch to native species following sudden declines in rabbit abundance. We now know enough about some processes in beech forest and dryland ecosystems to build prototype models that will help to predict the wider effects of controlling introduced species, identify critical knowledge gaps, and ultimately guide management decisions to achieve desired biodiversity outcomes.

Bioeconomic modelling in the development of invasive fish policy

Abstract Invasive species policy is either explicitly or implicitly underpinned by the question “When should investment in managing the invasive species stop?” Bioeconomic modelling provides a quantitative framework for considering the benefits and costs of alternative levels of investment in invasive species management by linking the level of investment to the costs of intervention (control) and value of benefits derived. Control costs are typically the product of the number of individuals that have to be removed to either eradicate the invasive species or constrain it at some specified density, and the cost of removing each individual. Impact functions take a variety of forms, but in general are systematically related to the density of the managed population. Where impacts can be accounted in monetary terms (e.g., where an invasive species affects income), control costs and benefits can be directly compared and an optimal level of investment (usually that which maximises return on investment) can be identified. However, where impacts do not have a directly accessible monetary value (e.g., where an invasive species affects conservation values), benefits and costs cannot be directly contrasted. Under these circumstances, bioeconomic modelling can be used to identify management strategies that maximise the level of benefit that can be achieved for expenditure of a fixed budget (benefit maximisation), or minimise the cost of achieving a given level of benefit (cost minimisation).