5. RABBIT CALICIVIRUS DISEASE AND RABBIT CONTROL

The rabbit calicivirus, which causes rabbit calicivirus disease RCD, also known as Rabbit Calicivirus Disease and previously as Rabbit Haemorrhagic Disease, RHD), is a candidate biocontrol agent for rabbits. The disease, which was first noted in China in 1984, occurs naturally and is currently reported to be effectively controlling rabbits in some parts of Europe - most notably in Spain.

A joint Australia/New Zealand research programme, aimed at evaluating RCD as a biocontrol technology, has been running in Australia since 1990. Initial research was based at the CSIRO Australian Animal Health Laboratory. The research included developing tests for detecting the virus, testing humaneness, and checking on host specificity and spread under laboratory conditions.

In March 1995 trials began on Wardang Island off the Australian mainland to study transmission of the virus under conditions more closely resembling those in the wild. However, in October 1995 the RCD virus escaped to the mainland, which led to the commissioning of several projects not mentioned in "Research in Progress 1995/96". These projects examine the likely impact of RCD in New Zealand, and review potential transmitters of the disease.

Other New Zealand-based research examined the role of ferrets and cats in rabbit control and the impact of rabbit control on these predators.

5.1.1

Programme Title: Predator-prey interactions and impacts of rabbits

Programme Leader:Dr G Norbury

Institution:Landcare Research, Alexandra

Summary

Studies of the impact of controlling rabbit populations on the ecology of wild cats and ferrets have indicated that ferrets, and to some extent cats, shift their diet to other fauna following rabbit control. Ferrets ate lizards, insects and hedgehogs and cats ate birds. Ferret and cat numbers declined slowly after rabbit poisoning: the main effect appeared to be on breeding. There was some starvation of ferrets, but not of cats, and little secondary poisoning or dispersal of either animal. It is estimated that about 60% of the ferret population and 30% of cats die anyway each year, of natural causes.

Protection of trial plots of degraded short-tussock grassland from the effects of rabbit grazing resulted in an extra 714 kg/ha of pasture yield (a six-fold increase compared to unprotected plots) for the four months September to December. An extra 334 kg/ha was recorded for January to April. These rapid gains in pasture yield indicate the potential benefits for pastoral production and land conservation following protection from rabbits.

Objective 1: The impact of rabbit control on predator ecology

Research Leader:Dr G Norbury

Objective 2:The impact of rabbit grazing on vegetation

Research Leaders:Dr O Bosch and Dr G Norbury

Description - Objective 1: The impact of rabbit control on predator ecology

To identify the impact of controlling rabbit populations on the ecology of wild cats and ferrets, by:

  • determining the density and survival of wild cats and ferrets;
  • defining the movement patterns of predators;
  • determining the population structure and breeding status of predators;
  • determining the diet of predators; and
  • conducting this research in the presence and absence of rabbits.

Approach & Outcomes

The results of this work have featured in the debate regarding the wider ecological effects of RCD in New Zealand. A final analysis and report is not yet available.

The experimental design consisted of two 2500-6000 ha sites where rabbit control (by 1080 poisoning) took place, and a third 1000 ha site where no rabbit control occurred. The two rabbit poisonings resulted in a 99% reduction in spotlighted rabbits on one site, and a 77% reduction on the other site.

The number of ferrets known to be alive on the control site naturally cycled with peaks in summer/autumn (due to recruitment of young ferrets and to an increase in trappability), and lows in winter/spring (due to high post-recruitment mortality and to a decline in trappability). Based on survival of radio collared ferrets, about 60% of the ferret population naturally dies off every year, so high ferret mortality is a natural feature of ferret dynamics in semi-arid New Zealand.

Ferret numbers on the treatment sites showed population cycles similar to the control site but have since failed to recover to anywhere near their previous levels. The reasons why ferret numbers failed to recover following rabbit control are varied. About 7% of ferrets died from secondary poisoning by scavenging rabbit carcasses. Sixteen percent of ferrets succumbed to starvation. Nearly all resident ferrets remained within home ranges of about 90 ha in area. With the arrival of the first winter, 20% of the ferrets remaining on the treatment site that received a 99% rabbit kill, moved a few kilometers away into areas that supported higher rabbit numbers. But half of these have since returned to their home range. All ferrets on the other treatment site remained on site. Ferret dispersal was not a major factor in accounting for the observed ferret declines. Rabbit control-induced dispersal of ferrets is unlikely to be a significant contributor to the spread of bovine Tb.

Rabbit control appears to have had most effect on ferret breeding. The number of new ferrets caught during the first breeding season was much lower on the treatment sites. If expressed as per capita recruitment, there was only 37-39% recruitment on the treatment sites, compared with 244% on the control site. The second breeding season coincided with very high rainfall and presumably higher alternative prey. Hence, there was 100-286% per capita recruitment on the treatment sites, but this was still significantly lower than 600% recruitment on the control site. This suggests that the failure of the two treatment populations to recover following rabbit control may be due primarily either to a lack of successful breeding or to very high mortality of juvenile ferrets.

The fact that ferret numbers slowly declined following rabbit control, without major dispersal, is a positive result. However, on the treatment site where a 99% rabbit kill was achieved, lizards, insects and even hedgehogs increased in prevalence in the diet, replacing rabbit, since the rabbit poisoning. A similar, although less pronounced diet switch was observed on the other treatment site where only a 77% rabbit kill was achieved. The impacts on these fauna are unknown.

Cat numbers also declined after rabbit control, but only on the site that had the most successful rabbit poison. Again, lack of breeding appeared to be the main cause. Very few cats died from secondary poisoning, none starved, nor did any disperse. Cats have larger home ranges than ferrets and average about 220 ha in size. It is estimated that about a third of the cat population dies every year due to natural causes. A diet shift by cats was observed only on the site where rabbit control was most successful. Only birds were eaten in significant numbers. It appears that cats are more able to withstand the effects of food deprivation following rabbit control than are ferrets.

Early indications from this work show that ferrets, and to some extent cats, will shift their diet to other fauna in a short-term struggle to survive following rabbit poisons. Depending on the success of the poison operation, a few ferrets will move short distances away from poisoned areas (but often only temporarily) in search of alternative prey. Diet changes certainly raise concerns for the safety of protected fauna but their effects may be offset to some extent by the fact that predator numbers appear to be declining anyway due to a lack of recruitment.

Description - Objective 2: The impact of rabbit grazing on vegetation

To quantify the impact of rabbit grazing on vegetation in the High Country, by:

  • identifying the type and amount of forage removed by rabbits from three ecological regions during different growing seasons;
  • correlating the amount of forage removed with estimates of rabbit density;
  • determining the production capability of individual species in these regions; and
  • determining the long-term effects of rabbit grazing on vegetation composition and productivity by analysing existing data from long-term enclosures.

Approach & Outcomes

During the four most productive plant growing months of 1994 (September to December), a six-fold increase in pasture yield was observed after protection from rabbit grazing (139 kg dry weight/ha with rabbits cf. 853 kg DW/ha without rabbits). Rabbit counts were 30 to 42 rabbits per spotlight km. The following four months (January to April) were characterised by reduced pasture growth (3 kg DW/ha with rabbits cf. 337 kg DW/ha without rabbits) and higher rabbit numbers (42-76 rabbits/spotlight km), and was a critical period of herbage depletion. These rapid gains in pasture yield indicate the potential benefits for pastoral production and land conservation following protection from rabbits.

There were significant grazing, season and ‘grazing x season’ interaction effects (all P<0.0001) on total pasture yield. More biomass accumulated when rabbits were absent, more biomass accumulated in spring, and the increase in biomass when rabbits were absent was greatest in summer (Table 1). Protection from grazing during the most productive plant growing months in spring resulted in a six-fold increase in pasture yield. Although the proportional increase in pasture following protection from grazing in summer was far greater than that in spring, overall summer pasture growth was much lower.

The impact varied for different species. There were significant grazing effects during both seasons for sweet vernal (P=0.0313 in spring; P=0.0078 in summer). For spring only, significant effects were found for chewings fescue (P=0.0355), hard tussock (P=0.002), and tussock hawkweed (P=0.0156). For summer only, there was a significant effect for blue tussock (P=0.0313). All species (apart from hard tussock) flowered profusely in the absence of rabbits. Grazing did not appear to significantly reduce the growth of the other species.

A total absence of rabbit grazing during four months resulted in an extra 714kg/ha of phytomass during spring, and an extra 334 kg/ha during summer. Although this result is confined to one habitat type during one spring/summer growing season, it demonstrates the potential gains in pasture yield in the absence of rabbits. The coincidence of relatively low pasture yields in summer (see also Radcliffe and Cossens 1974), and seasonal peaks in rabbit numbers following recruitment from the breeding season in spring, indicates that summer may be a critical period of herbage depletion.

H. lepidulum, one of a number of hawkweeds currently spreading in the high country, grew prolifically in the absence of rabbit grazing during spring. Flowering was also profuse under these protected conditions, unlike that in the grazed plots where flowers were never observed. This species is therefore likely to spread following the cessation of stock and rabbit grazing. Without rabbit grazing, prolific flowering of grasses was also observed, as reported in Australia, again leading to grass recruitment in the longer term.

It was concluded that the information available on the impact of rabbits on New Zealand vegetation is often confounded by livestock grazing and seasonal influences.

The significant growth of herbage following protection from rabbits especially during spring is likely to generate benefits in soil condition by promoting nutrient recycling, water infiltration and protection from wind erosion. Depending on pasture condition, benefits may further accrue to other recolonising plants species and faunal, particularly invertebrate assemblages. While the study demonstrates some of the potential gains in herbage growth following protection from rabbits, the data are indicative only.

Publications

Norbury, G. and McGlinchy, A. (1996): The impact of rabbit control on predator sightings in the semi-arid high country of the South Island, New Zealand. Wildlife Research, 23(2), pp.93-97.

Norbury, D. and Norbury, G. (in press): Short-term effects of rabbit grazing on a degraded short-tussock grassland in Central Otago. New Zealand Journal of Ecology, 20(2).

Heyward, R. and Norbury, G. (submitted): Secondary poisoning of wild cats and ferrets after 1080 rabbit poisoning. Journal of Wildlife Management.

Conferences

Norbury, G. and Heyward, R. (1996): Predator behaviour and some wider ecological issues surrounding rabbit control. Conference Programme and Abstracts of the 8th Australasian Wildlife Management Society Conference, Christchurch, p.44.

Heyward, R. and Norbury, G. (1996): Diet switching of wild cats and ferrets following rabbit control in the semi-arid lands, South Island, New Zealand. Conference Programme and Abstracts of the 8th Australasian Wildlife Management Society Conference, Christchurch, p.34.

Heyward, R. and Norbury, G. (1996): Secondary poisoning of wild cats and ferrets following 1080 rabbit poisoning in the semi-arid lands, South Island, New Zealand. Conference Programme and Abstracts of the 8th Australasian Wildlife Management Society Conference, Christchurch, p.11.

Norbury, G. and Heyward, R. (in press): The response of ferrets to rabbit control. Proceedings of the National Science Strategy Committee workshop on ferrets as vectors of tuberculosis and threats to conservation, Christchurch.

Moller, H.; Norbury, G. and King, C.M. (in press): Ecological and behavioural constraints to effective control of ferrets (Mustela furo). Proceedings of the National Science Strategy Committee workshop on ferrets as vectors of tuberculosis and threats to conservation, Christchurch.

Reports

Norbury, G and Murphy, E (1996): Understanding the implications of rabbit calicivirus disease for predator-prey interactions in New Zealand. Landcare Research Contract Report LC9596/61, prepared for MAF Policy, Wellington.

5.1.2

Programme Title: RCD Wardang Island pen trials

Research Leader:Dr K Murray

Institution:CSIRO

Summary

In March 1995 trials began on Wardang Island off the Australian mainland to study transmission of the virus under conditions more closely resembling those in the wild. However, in October 1995 the RCD virus escaped to the mainland. The rabbits on the island were destroyed and research to examine the likely impact of RCD in New Zealand was increased.

Description

To study the impact of the RCD virus on rabbits, living in warrens in large enclosures, during the breeding season, through monitoring warrens by clinical and pathological surveillance and subsequently by serological testing.

Approach & Outcomes

In March 1995 trials began on Wardang Island off the Australian mainland to study transmission of the virus under conditions more closely resembling those in the wild. Initial results had suggested that the virus might not spread easily. However, in October 1995 the RCD virus was found to have infected rabbits outside the fenced enclosure on the island and all rabbits on the island were destroyed. Subsequently it was found that the virus had escaped to the mainland, possibly carried by bush flies. The disease is now widespread in South Australia. Research into likely impacts of RCD in New Zealand was stepped up.

5.1.3

Programme Title: RCD host specificity trials

Research Leader:Dr B Buddle

Institution:AgResearch, Wallaceville

Summary

In order to check whether indigenous species would be susceptible to the RCD virus, tests were carried out on North Island brown kiwi and short-tailed bats in New Zealand. These followed on from tests carried out in Australia. The final results of testing confirmed that the kiwi and bats are not susceptible to RCD.

Description

To determine whether or not the North Island brown kiwi and short-tailed bat are susceptible to the rabbit calicivirus using the trial protocol used at the Australian Animal Health Laboratory at Geelong in the Australia and New Zealand RCD Programme. The animals were subjected to massive intramuscular doses of virus and then monitored for effects.

Approach & Outcomes

In previous years, many animal species (28 in total) were tested for susceptibility to RCD at the CSIRO Australian Animal Health Laboratory. These included domestic and feral animals, reptiles, Australian native animals and birds. In 1995/96, further trials were conducted on kiwis and bats within New Zealand.

Neither species showed any clinical signs of the disease and tissue samples from the bats produced clearly negative results. However, blood samples from the kiwi showed low levels of antibodies, so further tests were necessary to eliminate the possibility that they were susceptible to the disease.

The final results of testing confirmed that the kiwi and bats are not susceptible to RCD.

5.1.4

Programme Title: Cost benefit analysis of the introduction of RCD into New Zealand

Research Leader:Dr N Brown

Institution:Brown Copeland & Co Ltd

Summary

This study compared the potential costs and benefits of releasing RCD in New Zealand with the costs and benefits of a continuation of the status quo, in which regional councils monitor and enforce "user pays" rabbit pest management strategies. It was difficult to quantify the impacts, but preliminary results were provided. As those who would benefit are different groups from those who would sustain the costs, there would be large equity impacts.

Description

To determine the potential costs and benefits of using RCD as a rabbit biocontrol in New Zealand by:

  • utilising available data to develop a cost/benefit profile from New Zealand agriculture; inclusive of pastoral and possible disease management benefits;
  • scoping the likely costs and benefits in conservation/environment areas utilising relevant resource accounting procedures if appropriate or even possible; and
  • making a brief review of the strengths and weaknesses of cost/benefit analysis as applied to RCD.

Approach & Outcomes

A report was prepared. Much of the information used was sourced from the Australian and New Zealand Rabbit Calicivirus Programme Proponent Committee Submission 1996. The approach adopted was to compare the probable outcome of two scenarios:

  • Option 1 - the strategic introduction and release of the RCD; and
  • Option 2 - a continuation of the status quo where rabbit control is part of regional pest management strategies, with a trend toward greater emphasis on "User Pays" and with regional councils concentrating on monitoring and enforcement functions.

The analysis was conducted from the national perspective. In the analytical framework it was attempted to identify all likely negative and positive impacts from Option 1 compared with Option 2 and, where possible, quantify these impacts in monetary terms. Many of the impacts, however, can only be described qualitatively, since they are basically intangible (unable to be valued in dollar terms). In addition, equity impacts are large and that those which primarily benefit are different groups than those who sustain the costs.

The results are summarised in the table, but it was emphasised that the information provided is only preliminary and, at best, only indicative of the magnitude of the likely costs and benefits. Not only has the proposed release strategy not yet been finalised, but the relationship between RCD and the dynamics of the rabbit population spatially and over time is very uncertain. Second order effects, for instance predator-prey relationships, are also not well researched.

COSTS AND BENEFITS OF THE STRATEGIC RELEASE OF RCD (COMPARED WITH THE "STATUS QUO")
COSTS BENEFITS
RCD Release costs Probably <$ 150,000 Additional Pastoral Production Unlikely on 616,800 ha of High & Extreme Rabbit prone land. Some increase possible on 802,000 ha of Moderate prone land. Further research needed.
RCD Re-Release Costs Indicative - <$50,000 p.a. Additional Production on other Land Classes Small.
RCD Monitoring Costs Indicative - $250,000 p.a. reducing to $100,000 p.a. after 5 years. Saved Rabbit Research Costs Negligible.
Costs for Rabbit Farmers & Processors Capital cost indicative $0.5 million. Loss in sales indicative <$2 million p.a. Saved Rabbit Control & Compliance Costs Indicative $5 million p.a.
Vaccination Costs Indicative $200,000 p.a. Saved Costs from Reduced Numbers of Rabbit Predators which carry Tb Reduced number of ferrets should reduce spread of bovine Tb, although not conclusively proven.
Research & Administration Costs Indicative <$900,000 p.a. reducing to $200,000 p.a. Improved Water & Soil Values Will mitigate degeneration of approximately 286,000 ha of High & Extreme prone land where rabbit control uneconomic. Value unquantifiable.
Risk & Associated Costs of Host Switching High perceived cost, but very low probability. Improved Bio-diversity Values Improvement in Bio-diversity conditions, except that the coverage of woody weeds will probably increase.
Costs on Recreational Shooters Not available. Likely to be low. Benefits from Reduced Use of Poisons Reduced use of 1080. Unquantifiable value.
Costs Incurred from changes in Predator-Prey Relationships Increased predator control costs in certain localities to protect endangered species. Indicative $0.5 million p.a. for 3 years. Reduced Stress/Social Tension Reduced personal/community stress, principally in High & Extreme Rabbit prone areas.
Negative Public Perception of RCD ?

Reports

Brown Copeland & Co Ltd (1996): Cost Benefit Analysis of the Introduction of Rabbit Calicivirus Disease (RCD) into New Zealand. A preliminary analysis (draft form) prepared for MAF Policy, Wellington.

5.1.5

Programme Title: Potential vectors of rabbit calicivirus disease (RCD) in New Zealand

Research Leader:Dr T Crosby and J McLennan

Institution:Landcare Research

Summary

A review of information was undertaken to identify how invertebrates and vertebrates might contribute to the spread of RCD virus with New Zealand, should the disease be accidentally or deliberately released. The most significant potential vector (transmitter) identified was humans. Ferrets, Dominican gulls and eight species of fly are also potential vectors.

Description

To review the current status and relevant behaviour of potential RCD vectors by:

  • determining the species likely to be vectors of RCD;
  • determining the spatial distribution of such vectors and any seasonal variation in their likely role/efficacy; and
  • determining any long-term research needs.

Approach & Outcomes

Researchers sought to identify how invertebrates and vertebrates might contribute to the spread of RCD virus within New Zealand, should the disease be accidentally or deliberately released. Information came from a large number of sources identified in the report. (Landcare Research Contract Report: LC9596/072)

Eight species of fly were identified as potential vectors of RCD, in two categories: direct-contact vectors (transferring the virus to healthy rabbits), and indirect contact vectors (contaminating rabbit grazing areas with the virus). Three flystrike species of blowflies (Lucilia cuprina, L. sericata, and Calliphora stygia) are likely to be both direct contact vectors and indirect-contact vectors. Four carrion species of blowflies (Calliphora hilli, C. quadrimaculata, C. vicina, and Xenocalliphora hortona) are likely to be indirect contact vectors. One species of fleshfly (Hybopygia aria) is likely to be a vector. These flies are more common in the warmer months, but little is known of their seasonality and relative abundance in the most rabbit-prone areas. Biting flies such as mosquitoes and blackflies (sandflies) are considered unlikely to be potential vectors.

Humans are the most significant potential vector of RCD virus. Of other vertebrates, ferrets are the most likely mammalian vector as they capture rabbits underground and are capable of killing adult rabbits. Dominican gulls are the most probable avian vector.

Reports

Crosby, T. and McLennan, J. (1996): Potential vectors of Rabbit Calicivirus Disease (RCD) in New Zealand. A review prepared for MAF Policy, Wellington.

5.1.6

Programme Title: Rabbits as pests in New Zealand: A summary of the issues and critical information

Research Leader:J Parkes

Institution:Landcare Research

Summary

This project was to summarise information about the present distribution, impact costs and control of rabbits in New Zealand and review key issues involving rabbits. The report contains estimates of the areas of land where rabbits are controlled, and methods and costs of control, and identified issues relating to land management, costs to landowners, ratepayers and taxpayers, use of poisons and commercial harvesting of rabbits.

Description

To summarise information on the current distribution, impacts, control, and costs of rabbits in New Zealand and to review some of the key issues involving rabbits by:

  • summarising land management issues that involve rabbits;
  • summarising rabbit impacts on production systems and conservation values;
  • summarising the potential effects of control of tuberculous predators on rabbit populations, and the risks such control poses to rabbit management;
  • summarising the current distribution of rabbits and relate this to their relative densities, broad land-tenure classes, type and degree of control used, breeding seasonality, and the distribution of their predators known to carry bovine tuberculosis;
  • estimating past and present direct costs of rabbit control, and to discuss what opportunity costs these impose on land managers;
  • estimating the present and future costs of uncontrolled rabbit populations;
  • assessing present/likely contributions of commercial harvesting to rabbit control;
  • summarising the implications of potential changes in climatic stability on the stability of rabbit populations; and
  • reporting on the amount and extent of use of 1080 and pindone toxins.

Approach & Outcomes

Rabbits occur over about 150 000 km2 of New Zealand (about 55% of the total area). They are usually absent from native forests and dense scrublands, mature exotic forests, and from tussock grasslands and scrub over about 900m. The area in which rabbits are a major problem is much smaller, but its exact size depends on how "problem" is defined.

Rabbits contribute to many unresolved land management issues. On the most rabbit-prone land they are a major factor in determining the profitable and sustainable use of the land. Therefore they are a major factor to be considered in the Rabbit & Land Management and Semi-Arid Lands Research programmes and tenure reviews for Crown Lease land in the semi-arid areas of the South Island. Rabbits are also pests on conservation values, both directly by eating native plants and indirectly by sustaining high densities of predators. The way these predators are managed when they become pests on conservation values or as carriers of bovine tuberculosis also has consequences for rabbits and how they are managed. Domestic and foreign perceptions of control technologies, such as poisoning or the use of diseases as biological controls, are also management issues.

In 1983/84, control was conducted over 87 081 km2 by night shooting and 3 057 km2 by poisoning. The amount of night shooting had declined to cover about 64,000 km2 in 1987, the last year of Agricultural Pests Destruction Council funding. Present control by landowners and Regional Governments probably covers a smaller area than in 1987 (Regional Councils' control covers only about 31,000 km2). Control actions should define the extent of the problem, but estimates of the areas in different rabbit-proneness classes (assuming proneness = problem) show only about11,700 km2 (8,075 km2 in the South Island and 3,620 km2 in the North Island) in the "extreme" or "high" proneness classes, with only about 300 km2 in the South Island showing "intractable" problems due largely to neophobia and/or poison shyness.

In 1992/93, a survey of 18 properties in the Rabbit & Land Management Programme showed the costs of rabbit control exceeded the income in 10%, 75%, and 90% of land in the moderate, high, and extreme rabbit-prone classes, respectively.

The national costs of rabbits are at least $22 million per annum for the direct costs of control ($7.8 million from landowners, $5.7 million from ratepayers, and $0.45 million from the Department of Conservation [DOC]), plus theunremitted costs to production $8.0 million). This does not include the incalculable costs to non-market resource and conservation values, the contingent costs to trade from risks associated with diseases or use of unacceptable control practices, or the social costs of changes to land use.

New Zealand currently uses more 1080 than that used in other countries. Use of 1080 for rabbit control (about _ of total usage) is now declining as regional government agencies move towards more regular maintenance control strategies, which require less frequent aerial poisoning, and as more control is carried out by landowners who cannot use 1080. Landowners currently use about 200 tonnes of pindone baits annually.

Regional governments are in the process of moving to the new procedures for rabbit control under the Biosecurity Act. This Act stresses action where benefits exceed costs, and where "beneficiaries" and "exacerbators" pay. There is currently little consistency among regional councils in the way they fund, conduct, or monitor rabbit control.

Estimates of the national scale of the problems posed by rabbits are inadequate because accurate figures are not available for areas in different rabbit-proneness classes, and on the extent of control actions - particularly by landowners.

Rabbits may be commercially harvested and sold to one of the six licensed processing factories. Harvesters are paid between 50 cents and $1 per rabbit. The total harvests and the viability of the factories are unknown. It is also not clear whether this harvesting will contribute to rabbit control, particularly in highly prone areas where rabbit numbers have already been reduced by other control methods.

Reports

Parkes, J. (1996): Rabbits as Pests in New Zealand: A summary of the issues and critical information. Prepared for MAF Policy, Wellington.

5.1.7

Programme Title: Modelling the likely impacts of RCD

Research Leader:N Barlow

Institution:AgResearch

Summary

Two models were used to test the likely outcome of introducing RCD on rabbit populations, using data already available. The results suggested that the occurence and intensity of RCD epidemics depends on rabbit density. RCD should have little or no effect on New Zealand lowland rabbit populations which have low density. The effect of predators and other controls would interact with the effect of RCD to delay recovery after an epidemic. Disease vectors (carriers) could cause spring/summer rather than autumn/winter epidemics and repeated epidemics. It was recommended that more quantitative data be acquired to enable more accurate prediction of the effects of RCD in New Zealand. The work is being carried out in 1996/97.

Description

To use models for:

  • determining the likelihood and significance of RCD impacts on rabbits being density-dependent;
  • assessing the potential impact of seasonal and spatial variation in RCD vectors inclusive of the presence or absence of young rabbits;
  • determining the significance of RCD predator interactions if RCD outbreaks occur during periods of high young rabbit survival; and
  • assess the likely rates of recovery of the New Zealand lowland and semi-arid rabbit populations after an RCD outbreak in the presence and absence of predator and/or other supplementary controls.

Approach & Outcomes

Two models were used: a simple epidemic and a detailed population model. For each model, two alternative transmission modes, direct and indirect, were tested. The researchers already had parameters for the rabbit’s dynamics, and data on RCD were obtained primarily from Xu and Chen 1989, Vizcaino and Cooke 1991 and Lenghaus 1993. There was little difference in the behaviour of the models for each of these four alternatives.

The models suggested that the occurrence and intensity of RCD epidemics depends on rabbit density. Based on the Australian (Gum Creek, SA) data, no epidemic should be possible at densities below 1.5 - 2 rabbits/ha. Hence RCD should have little or no effect on New Zealand lowland rabbit populations, which are already suppressed by 90 - 95% and represent some of the lowest stable rabbit populations in the world. It follows that RCD will have no effect on predators or their alternative prey in such areas, if the above is true. There are two important caveats. One is that the contact rate (number of potentially infectious contacts made per infected rabbit per unit time) changes linearly with density. There is no way of telling whether this relationship is actually linear, except by monitoring epidemics and relating their intensity (reduction in rabbit numbers) or occurrence, to initial rabbit density before the epidemic. The second caveat concerns scale and spatial patchiness. If a rabbit population is spatially patchy and epidemics occur within patches, it mle be that the within-patch density varies much less than the density at a larger scale which includes a variable number of patches. Again, this can only be resolved by properly targeted and quantitative monitoring of RCD epidemics in space and time.

The role of vectors is most likely to involve RCD appearing for the first time in populations in spring/summer rather than autumn/winter. Secondly, vectors, which are yet to be identified, could play a role in spreading the disease and causing repeated epidemics in local populations where the disease would persist. In this case the timing of epidemics would again be biased towards spring/summer. Thirdly, and less likely, vectors may affect the nature of these initial or repeated epidemics, rendering the summer ones more intense and any that do occur in winter less intense, through a possible role of the vectors in local as well as global transmission.

Even in the semi-arid areas, where predators have a limited effect on rabbits (probably suppressing their density by around 10%), they do interact significantly with other controls such as an RCD epidemic, slowing down the population's recovery. The interaction is similar for an RCD epidemic in winter or spring. This is because the predator effect on young rabbits, which is what inhibits recovery of the population, occurs not only at the time of the epidemic but also in the succeeding years, particularly the first year after the epidemic when densities are low. In general, the lower the rabbit density the greater the contribution of predation to their suppression. Thus the timing of an RCD epidemic and the presence or absence of young rabbits is immaterial in terms of the positive interaction between the epidemic and the effect of predators.

Given that RCD epidemics would appear to be isolated events in a local rabbit population, recurring with a frequency as yet unknown, the recovery rate of the population depends on its density, the timing of the epidemic, and the presence of other controls. Low-density populations should recover more rapidly than high-density ones, because the RCD epidemic is less intense. For similar reasons, populations should recover more rapidly after spring epidemics than after winter ones, the former having less effect because of the presence of young, immune rabbits. Finally, populations recover less rapidly in the presence of predators or other controls, even though the density may be lower and the epidemic therefore less intense, because the additional controls reduce the rabbits' rate of increase and therefore their potential for recovery. The fact that there is synergy between RCD and other controls carries obvious implications for management. In lowland areas the issue of recovery does not arise unless transmission rates are considerably greater than data currently suggest, since no epidemic is possible. However, if such higher transmission rates should prove to be realised, then the same rules about population recovery will apply as are relevant to the semi-arid areas.

It was recommended that more data be obtained to enable better understanding and predictions through further modelling. This work is being carried out in 1996/97.

Reports

Barlow, N.D. (1996): Modelling the likely impacts of RCD. Prepared for MAF Policy, Wellington.

5.1.8

Programme Title:A strategy for the release and management of RCD in New Zealand

Research Leader:J Parkes

Institution:Landcare Research

Summary

As a component of the evaluation of RCD as a possible biological control for rabbits, a study was undertaken to identify the steps which would need to be taken to maximise the effects of a controlled release of the virus, if such a release was sanctioned. The factors studied included when to release, where to release and how to release. The issue of monitoring post-release was also examined.

Description

To discuss options on the method, location, timing, and frequency of planned releases of RCD in New Zealand, and to put consequent decisions into a framework for the consideration of the benefits, constraints, and uncertainties associated with the planned release and ongoing management of RCD.

Approach & Outcomes

Current epidemiological models of the on-site effects of RCD on rabbit populations suggest outcomes may vary depending on rabbit densities, the persistence and behaviour of the virus and the influence of predators and conventional control.

In high density rabbit populations RCD may not persist but may recur as severe epidemics, perhaps every 3-4 years with full recovery of rabbits between epidemics, or more frequently before full recovery but with less severe epidemics. If it does persist in such high density populations RCD may act as a traditional biological control agent giving sustained suppression of rabbit populations.

In low density rabbit populations RCD may persist and give some moderate suppression of rabbit numbers, or it may have no effect at all.

Whether RCD will persist within or spread naturally between rabbit populations depends on the presence of vectors. The efficacy of potential insect and vertebrate vectors remains unknown in New Zealand.

Despite these models and overseas data, the behaviour of RCD in New Zealand rabbit populations will only be verified by the events. However, risks associated with its release and behaviour can be minimised, and the national, longer-term benefits maximised by making a deliberate decision to use an adaptive management approach. This requires monitoring and research over a two phase release process - a staged initial release followed by a wide release strategy dependant on results.

Adaptive management during the initial release phases will provide the vital information required for the long-term management of RCD especially if the disease does not act as a traditional, persistent biological control but dies out at particular sites.

When to release RCD: The number of immune young rabbits will be the minimum in most rabbit populations between March and August. If the initial release cannot be made before September 1996, delaying the release until autumn 1997 would minimise any risks posed to the success of the agent by the presence of young rabbits. However, it is possible that the spread of RCD from initial release sites may be at a maximum in summer if vectors are more abundant at that time. In the long-term it may not matter when RCD is initially released.

Where to release RCD: The virus should be released first in a limited number of areas with high rabbit densities to test its rates and mode of spread. It should not be released in areas with no rabbit problems, and particularly not in such areas that also contain native biota at risk to any changes in predator behaviour.

Mode of release: The initial releases should be made by inoculating wild rabbits. Subsequent releases can be made by inoculation or by baiting if this technology is developed. Releases by moving infected carcasses should be discouraged to avoid future problems with potential viral attentuation.

Who should release RCD: The initial release should be made by MAF or Regional Council staff. Subsequent releases, if needed, could be made by anyone, particularly if done using a bait - either to restart epidemics or as a biocide.

The future management of RCD will require research on the epidemiology and virology of the disease. Critical data is required on the seasonality, frequency and severity of epidemics and how these factors relate to rabbit densities, the spatial behaviour of the disease, and on any changes in virulence of the virus or susceptibility of rabbits.

The future management of rabbits will almost certainly require the use of conventional control methods even if RCD does persist. Development of an RCD bait to spread the disease if vectors are ineffective and to act as a biocide if the disease does not persist is a high priority.

Reports

Parkes, J. (1996): A Strategy for the Release and Management of Rabbit Calicivirus Disease in New Zealand. Prepared for MAF Policy, Wellington.

Previous Page TOC Next Page

Contact for Enquiries

Farm Monitoring Programme Manager
Monitoring and Evaluation
MAF Policy
PO Box 2526
Wellington
NEW ZEALAND
Phone: +64 4 894 0623
Fax: +64 4 894 0741
Contact this person

 
Print this page Print this page




Biosecurity New Zealand Web Site