The Windy Hill Rosalie Bay Catchment Trust (WHRBCT) has recently completed an analysis of their rodent trap-ping programme1 , which was started in 1999 and has increased in effort every year since then. The analysis also includes comparisons with results from Benthorn Farm (BT), Awana (Awana Catchment Trust) and Okiwi (DoC data). The results are sobering.

The most intensive effort has been at Windy Hill and Benthorn farm, where in 2004 c.1800 traps caught c. 3400 rodents over c. 260 ha. That represents c. 13 rats per ha per year. We hope this apparently low catch rate is because numbers have already been reduced by five years of trapping. This is probably so: Fig 1 shows that the initial trapping at Windy Hill (1999) had higher catch rates, while Fig 2 suggests that current catch rates at Windy Hill are significantly lower than in equivalent vegetation types at Awana. Unfortunately there is uncertainty in both comparisons, because different numbers of traps were used over different periods. All the results show the same seasonal pattern for ship rats, with a peak in March/April and a low in the late winter (Fig 3). It is important to have established the island-wide generality of this pattern; it proves that changes at any one site are mainly a result of births and deaths, not migrat-ion to or from the area. The last few years of intensive trapping at Windy Hill have apparently reduced the April peak, but the overwintering population remains unchanged (Fig 4).

As the vegetation recovers from formerly higher levels of rat predation, there is an abundance of food (eg nikau berries, invertebrates). Consequently any relaxation of trapping effort is likely to result in a rapid rebound in rat numbers. Moreover, recent work with ‘tracking tunnels’ suggests that the rat population at Windy Hill is greater than the trapping results suggest. Up to 49% of 120 tunnels were visited by rats in October 2004, when rat numbers should have been minimal. Perhaps the trap-ping has gradually selected a trap-shy rat population. Whatever the explanation, the ‘tracking tunnel’ results are not good for robins, and show that trapping alone is unlikely to get rat numbers down sufficiently for eco-system restoration. Poisons, proven to work in rodent eradication programmes elsewhere in New Zealand, seem to be the only alternative.

From the perspective of an island-wide eradication campaign in the future the results are very useful. Extrapolating, it would require an army of over 400 people and c. 200,000 traps to achieve the results at Windy Hill overall – results which are probably the best achieved anywhere with traps alone, but which we now know are not good enough. We know too that different vegetation types have very different absolute numbers of rodents, and different proportions of ship rats, kiore and mice. Land Units 8 and 9 (tea-tree on slopes and eroding ridges) which cover wide areas on the island, have relatively low ship rat numbers; streamsides and coastal flax have most. Kiore are commonest at grassland/scrub boundaries. We know in fact where rodents are and when they are at their most vulnerable. We have good indications of ‘interactions’ between the different species, and can anticipate an increase in mice if the rats are removed. The basic research is not complete, but I believe we now know enough about the enemy to start planning the campaign.

John Ogden

Figure 1. Rodent indices at Windy Hill, showing the initial decline in catch per 100 trap-nights (an index of rat abundance).

Figure 2. Comparisons between rat abundance in native forest and manuka scrub at Awana, and a mixture of these vegetation types at Windy Hill. Note that the WH results have been multiplied x10 to get them onto the same scale – ie, there are almost 10x fewer rats in similar vegetation types at Windy Hill.

top