Review of eastern grey kangaroo counts and derivation of sustainable density estimates in the Australian Capital Territory

Introduction

Objectives

• 
  To review the rationale for determining the target population sizes or densities of eastern grey kangaroos desired in the ACT to achieve a grazing regime that conserves native animals that rely on the conditions of the ground-layer vegetation in habitats used by kangaroos.
  
• 
  To review the methodology use to estimate the numbers or densities of eastern grey kangaroos in the ACT to validate the culling regime applied to achieve the desired target population sizes.
  

Why should kangaroos be controlled in the ACT?

If kangaroos are to be controlled in the ACT, then what target density (or densities, or range of densities) should be set?

How should the population size or density of kangaroos in reserves be estimated, and how often?

And, is the formula used to calculate how many kangaroos should be culled when the target density is exceeded valid?

In answering these questions we have the information in the KMP which is based on evidence published or known before 2010. There is also some information published since 2010, and information from projects in progress but not yet published. These latter sources may of course already indicate changes required when the KMP is revised.

Background

There are 34 discrete Nature Reserves (and more likely to be added in future) that comprise the Canberra Nature Park. These reserves support a diverse suite of native plants and animals including 19 species listed as threatened under Commonwealth and ACT legislation (ACT 2010), and include two endangered ecological communities; natural temperate grassland and yellow box – red gum grassy woodland.

Eastern grey kangaroos are the dominant vertebrate herbivore (see below) in most of these reserves and are managed under the Kangaroo Management Plan (2010) (hereafter KMP). The overall objective of the KMP is to maintain the presence of eastern grey kangaroos while reducing any environmental (i.e. particularly to the other native plants and animals), economic and social impacts.

Results

Why should kangaroos be controlled in the ACT?

Eastern grey kangaroos are clearly currently the dominant grazer at sites within Canberra Nature Park at which culling of this species is being considered. However, this has not always been the case and may not be the case in the future. The impacts, and hence management of other mammalian herbivores, are outside the scope of our review, but should be considered in the broader context of these reserves. European rabbit (Oryctolagus cuniculus), brown hare (Lepus europaeus), fallow deer (Dama dama), swamp wallaby (Wallabia bicolor), red-necked wallaby (Macropus rufogriseus), and euro (M. robustus) may be present in some of the reserves. Culling eastern grey kangaroos may increase food availability or otherwise make habitat more suitable for some of these species, or conversely taller grass might disadvantage species such as rabbits.

1. Do kangaroos affect the biomass of vegetation?

The best described kangaroo-vegetation system comes from the series of studies undertaken in the arid rangeland, including Kinchega National Park, in western NSW (Caughley et al. 1987). Here the system of red (Macropus rufus) and western grey (M. fulginosus) kangaroos and ground-layer vegetation biomass is largely driven by erratic rainfall (annual mean = 236 mm but highly variable between years with a CV of 45% and variable over short distances). Vegetation biomass increases soon after rain and kangaroos respond to more food (actually a threshold biomass of over 200 kg/ha) by breeding and increasing in numbers. When the vegetation biomass falls below 200 kg/ha kangaroo numbers decline. Kangaroos have little effect at any density on the flush of vegetation after rain falls but they may (depending on their density at the time since their numbers are sometimes not synchronised with the rain fall event) reduce the peak vegetation biomass that might occur or drive the vegetation to an even lower state in droughts. The feedback loops (more vegetation-more kangaroos, less vegetation-fewer kangaroos, and a loop within the pasture itself that slows growth as biomass increases in this system ensures centripetality – a tendancy towards a stable state but not one that is never reached as a new external event (erratic occurrence of rain) resets the system. Caughley et al. (1987) were generally skeptical about the merits of culling kangaroos in national parks to conserve biodiversity values when these dynamics applied, although they did concede that controlling kangaroos to reduce competition with sheep or simply to harvest meat were legitimate management actions.

So, is the advice given in this seminal book relevant to the kangaroos in the Nature Reserves in Canberra? If the eastern grey kangaroos in Canberra do not affect ground-layer vegetation biomass as for the rangeland species of kangaroos, then there is no justification (on biodiversity grounds alone) for controlling their numbers.

The systems are different. Rainfall in Canberra is higher (650 mm) with a much more even annual distribution (CV = 27%) and more even monthly distribution than that at Kinchega (e.g. McIntyre et al. 2010). Rainfall is most variable in March, and in April 2014 when we visited the sites it had been wet and the vegetation was abundant and green (see frontpiece). Under these circumstances plant growth is still driven by rainfall (and season via a strong temperature effect and a rain-temperature interaction (Fletcher 2006)), but as rain is less erratic pasture biomass is more constant and reaches higher levels (means of 569 ± 44 kg/ha with maximums of 2352 kg/h in 2007 (McIntyre et al. 2010) than in the arid rangelands (less than 10 kg/ha in droughts and over 800 kg/h in wet periods) (Robertson 1987).

The KMP and the research on pastures on which it relies (e.g. Fletcher 2006, 2007 at three sites selected to span the local range of climatic variation in grassland habitats, viz. Tidbinbilla, Gudgenby and Googong; and Neave & Tanton 1989 at Tidbinbilla) show that, unlike the arid rangeland case, kangaroos do have a significant affect on grassland biomass in temperate areas under some circumstances. The modelled outcomes in Fletcher (2006) showed more kangaroos equals less grass¹ and vice versa, although this relationship was at times completely overwhelmed by weather. This is inevitable in any vegetation-herbivore system because the plants can respond (grow or die) much faster than the herbivores breed or die. Conversely, the herbivores can overshoot their food supply and be left with too many mouths to feed when the vegetation is in natural decline, e.g. in a drought. The experimental data reported by Neave & Tanton (1989) show (among other things) that the height of most vegetation was significantly reduced when kangaroos were present. The conclusion is from both empirical data and the model derived from them that the temperate system differs from the semi-arid system studied at Kinchega. Kangaroos in the more stable temperate system do, on average over time, have a large effect on the biomass of grasslands.

Following the publication of the KMP in 2010, the only additional published evidence that eastern grey kangaroos affect grass biomass in the Canberra Nature Park is that of McIntyre et al. (2010). They described an experiment being conducted at Mulligans Flat that aims to evaluate how various (but unstated) eastern grey kangaroo densities affect a variety of variables, including plant biomass. Further studies were conducted in 14 Nature Reserves (Anon undated) in 2012/13 to assess the ground-layer vegetation biomass under three herbivore regimes (eastern grey kangaroos and rabbits excluded, kangaroos only excluded, and kangaroos and rabbits at ambient densities with that for kangaroos being measured). The sites have a range of kangaroo densities between 0 and 3.2 kangaroos/ha (mean = 1.2) so encompass the range of interest for setting target densities (0.6 – 1.5/ha). The results from these studies, when published, will provide empirical data on the relationship between kangaroo density and plant biomass to inform the modelled relationships from Fletcher (2006).

2. Does fewer kangaroos change native plant biodiversity?

The evidence that more or less kangaroo grazing advantages or disadvantages native plants over exotic plants is unclear. For example, Table 3.3 in the KMP lists nine threatened plant species from the ACT of which only two (Thesium australe and Leucochrysum albicans var. tricolor) are speculated to be at risk from herbivory – neither are grasses so the threat from kangaroos is moot. Neave & Tanton (1989) show some are advantaged by grazing and some not. A more detailed analysis to discriminate kangaroo effects from other factors (perhaps by concentrating on grass species rather than all species) would be required to answer this question.

The current project (Anon undated) has measured several indices of species richness at plots on 17 sites with and without kangaroos (and under the herbivore experimental regime noted above) and may help answer the question of how kangaroos managed to a target density change native plant biodiversity.

We conclude that kangaroos do change plant diversity but the details are unclear.

3. Does more ‘grass’ and more diverse flora favour valued native fauna?

As with the flora, there are several threatened animal species in the Canberra area that are thought to be adversely affected by kangaroo grazing (KMP 2010). If we go back to the various recovery plans for these species the ‘usual suspects’ are listed rather than much evidence of a ‘culprit’. For example, the plan for the grassland earless dragon (Tympanocryptis pinguicolla) says “the main factors involved in the decline are thought to be …” and goes on to list a page of possible threats (Robertson & Evans 2010). This lack of specific diagnosis is common in threatened species plans and reflects, in part, the difficulties noted in the first sentence of this section.

Since the KMP was published there have been attempts to improve these threat diagnoses. For the grassland earless dragon, Dimond et al. (2012) suggested that the degree of cover during droughts – and by implication the effect of grazing during droughts – was a key factor in the decline of this species. The project underway (Anon undated) is also measuring reptile abundances and species composition across a gradient of kangaroo densities – results are pending.

A third study (Manning et al. 2013) asked whether ambient vegetation biomass indices or kangaroo grazing affected reptile abundance. They concluded that decreasing grazing effectively increased small skink abundance in grasslands with high biomass, but the addition of cover (woody debris) was required in areas with low vegetation biomass caused by high kangaroo densities. This suggests to us that predation on the reptiles (by cats, foxes or birds) is mediated by the degree of cover which is affected by kangaroo grazing. Managing both herbivory and predation may be required to protect rare reptiles.

Howland et al. (unpublished manuscript in review) studied reptile diversity and abundance in 19 reserves with different kangaroo densities. Some reptile species were favoured by low kangaroo densities, others by moderate kangaroo densities, while high kangaroo densities favoured no reptiles. They recommended managing different reserves for different kangaroo densities (from low to intermediate) to favour different suites of reptiles – see also our recommendations.

Barton et al. (2011) showed that reducing the grazing by kangaroos increased the abundance and diversity of beetles, while adding woody debris to sites further improved the status of beetles. They speculated that this increase in invertebrate biomass would advantage insectivores such as the endangered reptiles.