Contact us

Protecting our precious places

Tawa fragments need fencing and pest control

An update from the Landcare Research Forest Remnant Resilience programme funded by the Foundation for Research, Science and Technology.

John Innes from Landcare Research in Hamilton describes a two-year collaborative research project involving ecologists from Landcare Research, University of Waikato, University of Canterbury, and AgResearch.

Previous research on kahikatea remnants has shown that vegetation can restore itself once stock is excluded, even in long-grazed fragments.

New research is now finding answers to some follow-up questions: Do the effects of fencing flow through vegetation into other parts of the ecosystem, such as leaf litter, and the invertebrates that live there?

What does pest control add to fencing, if anything? Does fencing itself affect pest abundance?

Do fragments isolated in a ‘sea of pasture’ act like real islands, preventing reinvasion by pests like ship rats?

Blocks of tawa-mangeao forest Photo: Hamish Dean

Above: South of Cambridge, five blocks of tawa-mangeao forest In Kairangi were protected in 2007 with a 25ha QEII covenant by Whanake Farms Limited.

To exclude grazing stock, the remnants were fenced with contributions from the landowner, Environment Waikato and QEII.

Photo: Hamish Dean

Tawa-dominated forest fragments

We worked in 53 tawa-dominated forest fragments in central Waikato, with an average size of six hectares.

The sites have been fenced for different lengths of time and have had different levels of control of possums and ship rats, the key pests.

As expected, a burst of native seedlings establishes after fencing, with a dense sapling thicket developing by 15 years later. Opportunities for new seedlings are then limited by the lack of available growing space.

Without pest control, young non-palatable subcanopy trees like mahoe and kawakawa quickly dominate at the expense of canopy trees such as mangeao.

Pest control at the time of initial fencing is therefore needed to enable seedlings of palatable species to establish.

The rapid establishment of seedlings following fencing means that there is a very limited window of opportunity for pest control to influence future canopy composition.

So while possum and rat control at any time will improve leaf production and flowering and fruiting, control at the time of initial fencing can improve the whole restoration trajectory of the fragment.

More vegetation and less soil compaction inside fenced fragments created the moist conditions that encourage leaf litter decay.

Leaf litter mass in summer varied from 8 to 14 tonnes per hectare among sites and tended to be higher in fenced fragments.

The rate of decomposition of experimental bags of tawa litter increased from about 40% over a 200-day period in unfenced fragments, to nearly 70% in fragments from which stock had been excluded for many years.

And what of invertebrates? The litter-dwelling invertebrates we targeted, for example, millipedes, earthworms, beetles, flies, thrips and amphipods, are closely involved in the breakdown of organic matter and the cycling of nutrients essential for forest health.

From over 100,000 invertebrates collected in almost 1,000 samples from 30 forest fragments and three reference forest sites, we learned that invertebrate densities found in fragments fenced 30-40 years ago were up to 100 times those found in unfenced fragments.

Our results suggest that there are strong links between vegetation regeneration, invertebrate abundance and nutrient cycling, and that these links can be restored in managed fragments.

However, fences are not entirely beneficial. Our research discovered that ship rats – key, unwanted predators of seeds, invertebrates, lizards and nesting birds – were seven times more abundant in fenced fragments (7 rats per hectare) than in grazed fragments (1 per hectare).

This is probably due to the extra food from the more dense vegetation and litter and the greater number of invertebrates.

Whether or not this is true for other pests such as possums and hedgehogs is unknown.

Rat eating song thrush chicks Photo: David Mudge

Above: A ship rat eating song thrush chicks.
Photo: David Mudge, Nga Manu Images

Furthermore, all eight fragments in which we eradicated rats were reinvaded within a month, despite the pasture gap of up to 250m to the nearest source forest, and regardless of whether the fragments were linked to the source by some kind of vegetated corridor.

Our results identify a previously unsuspected dilemma for fragment conservation.

Fencing to reduce stock access is a key tool to protect the health of vegetation, litter and associated ecological processes, but it also encourages high numbers of ship rats and perhaps other pests.

Maximising the biodiversity values of forest fragments therefore requires both fencing and pest control.

Covenants protecting forest fragments containing tawa

The research is exploring the effects of fencing and pest control on the resilience of tawa-dominated forest fragments that are common on rolling hill country in the Waikato.

Protecting these fragments with QEII covenants helps to assist the viability of our native species and forms valuable corridors and habitat for birds such as tui and kereru.

In the Waikato, 204 QEII covenants protect 353 forest fragments in which tawa is a dominant species, a total of 8,037 hectares.

The recently registered covenants featured below are similar to the fragments studied in the research.

Tawa remnant

Tawa remnant

Above: This tawa-mangeao remnant was protected with a 4ha covenant in 2007 by Peter O’Brien on his farm west of Maungatautari.

Intensive grazing for years in remnants prevents regeneration and reduces litter depth and invertebrate abundance. The research has shown this reduces the suitability of the habitat for ship rats.

Fenced with contributions from QEII and landowner in 2007, new seedlings are already emerging in this covenant.

Photos: Hamish Dean

Tawa remnant

Tawa remnant

Above: South-west of Pirongia, Andrew and Anna Clark protected three forest fragments with a 2.5ha covenant in 2006.

‘Tawa is the dominant species and there are now young ones coming through,’ says Andrew. ‘There’s also been a tremendous increase in tui and kereru since the possums were controlled.’

Originally fenced in 1986, the healthy understorey includes pigeonwood, mapou, mahoe, wheki, mamaku, pate, nikau, lancewood, heketara, rangiora and wineberry.

Excluding stock by fencing remnants increases vegetation regeneration, litter mass and litter decomposition rates.

Unfortunately, the research shows these positive effects may also increase ship rat abundance.

Photos: Malcolm Mackenzie

Tawa remnant

Tawa remnant

Above: Brett and Michelle Miller protected two tawa-mangeao forest fragments south-east of Cambridge with a 12.5ha covenant in 2007.

For isolated fragments such as these, the research shows that up to 250m of pasture between fragments does not stop ship rat reinvasion after eradication.

‘We are working with Environment Waikato on a rat eradication programme,’ says Michelle.

‘Unfortunately, there are areas of bush near us with no pest control which means that possums and rats do come back.’

Photos: Hamish Dean


For more information on controlling rats, see Rats on the New Zealand mainland from Open SpaceTM Issue 66, March 2006.

Your Regional Council is another source for advice on rat control.

For more information on the research, email John Innes


Open SpaceTM Magazine No. 75, March 2009 © QEII National Trust

MoST Content Management V3.0.6374