New Zealand Journal of Ecology (1996) 20(1):27-35 - PDF document

27 New Zealand Journal of Ecology (1996) 20(1):27-35 ©New Zealand Ecological SocietyJ.A. McLENNAN1, M.A. POTTER2, H.A. ROBERTSON3, G.C. WAKE4, R. COLBOURNE3, L. DEW1,L. JOYCE5, A.J. McCANN1 28NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 20, NO. 1, 1996 Sources of information used in the analysis of adult survival.LocalityNumber of tracking periodsSourceMalesFemalesNorthern brown kiwiWaitere, Hawkes Bay15McLennan Haliburtons, Hawkes Bay46IbidUrewera, Hawkes Bay168J. McLennan and L. Dew, Okarito, West Coast116J. Reid and J. Lyall, Purua, Northland157H. Robertson, Riponui, Northland113IbidRarewarewa, Northland2526IbidHodges, Northland61IbidTangiteroria, Northland1823Potter, 1989Saxon, Northwest Nelson1416McLennan and McCann, 1991bKahurangi, Northwest Nelson63IbidKaramea, West Coast34J. Lyall, 29 Survival estimates for adult kiwi in mainland forests.Brown KiwiGreat Spotted KiwiMalesFemalesMalesFemalesNo. Tracking periods (individuals)108 (99)84 (72)23 (20)23 (18)SE)0.72 0.050.52 0.050.84 0.140.80 Sum (years)78.144.019.218.3No. Deaths8231Mortality (%/year)10.24.515.65.5Av. Mortality (%/year)8.210.7 30NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 20, NO. 1, 1996 Survival of kiwi eggs in mainland forests.Brown KiwiGreat Spotted(33 pairs)Kiwi (15 pairs)Nesting attempts5819No. eggs8319No. hatched26 (31%)7 (37%)No. eaten2 (2%)0No. smashed83No. disappeared62No. rotten13 (16%)3 (16%)No. deserted244No. buried20No. other20No. chick hatchedper pair/year (x SD)0.85 0.870.39 No. chicks fledgedper pair/year (x SD)0.51 0.720.39 31 The fate of kiwi chicks and juveniles in mainland forests. Alive refers to fledging for chicks. In juveniles, aliverefers to individuals which shed their transmitters.AgeNo. observedDiedNaturalDisappearedKilledAliveUnknown Chick4922 (45%)99427 (55%)0Juvenile2715 (55%)1595 (19%)7 32NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 20, NO. 1, 1996 5.8% per annum. There is, however, considerablevariation between areas, depending on the variousvalues of the input parameters. The Urewerapopulation, for example, with its high rates of adultmortality, is declining about six times faster (at 18%per annum) than populations near Whangarei inIt would take about 65 years for the density of akiwi population to decline from 100 km-2 to 2 km-2 ifthe average loss rate is 5.8% per annum (Fig. 1). Theprocess would take 95 years at a loss rate of 4% perannum. We do not know the actual rate over the lastvalues, given Bullers rough indications of kiwidensity in mainland forests at the turn of the century.Population decline ceases if the proportion ofyoung kiwi surviving till adulthood increases from6% (present levels) to 19% (Fig. 1); and recruitmentpopulation stability, predation rates on young kiwihave to fall by about 34% from present levels (c.50%) to 33%. Lower levels of predation are requiredthan those predicted by the model (33%) to sustaincurrent population levels because some juvenilesbefore they reached adulthood. Clearly, some lossesto predators are tolerable; indeed, halving the presentfacilitate population recovery. Eliminating predationon adults and eggs would not in itself promotepopulation recovery, though it would slow the rateDiscussionThis paper presents the first quantitative data on thein mainland forests, and quantitative predictions onthe declines in predation rates required to reversesuggestions of McLennan and Potter (1992, 1993)and others that predators are the primary culprits inZealand.Adult survivalThe vulnerability of a kiwi to predators changesthroughout its life, and different predators areimportant at different stages. Adult brown and greatspotted kiwi suffer few losses (3% per annum)apparently because they exceed the threshold preysize of cats and stoats, the largest of the commonforest-dwelling predators. This threshold is(2200 g) are generally about 20%-30% lighter thanfemales (2900 g; Reid and Williams, 1975). Themainland forests is thus largely attributable to theirevolutionary history, which happened to favour aat 1100-1400 g, remain vulnerable to predatorsthroughout adulthood, and are now found only onThe impact of ferrets and dogs on adult kiwi inmainland forests is both unpredictable and episodic,reductions in local populations. These reductionsinvolve the sudden loss of adults and are much moreTaborsky (1988) documented an episode in WaitangiState Forest where a single dog killed about 500kiwi in a few months. The population in PaerataWildlife Management Reserve in Northland droppedfrom approximately 90 to 30 kiwi between 1988 andepisodes by ferrets or dogs (H. Robertson and M.Potter,unpubl. data). Other populations inNorthland have also collapsed since the 1970s,particularly in areas recently colonised by ferrets andpossums (Miller and Pierce, 1995). Adult kiwiremnants in agricultural landscapes (McLennan andFigure 1: Density of northern brown kiwi in mainlandforests given adult mortality of 8.2% per annum, aand juvenile mortality of 94% per annum (A). Densityassuming a recruitment rate of 19% per annum (B). 33 dogs and ferrets.In this study, one of eight radio-taggedpopulations suffered episodic predation from ferrets.The event itself lasted just a few weeks, but resultedin the disappearance of at least 17% of 40 knownhowever, whether the frequency of episodicpredation events in the samples used to construct ourmainland populations - and thus whether ourestimate of adult survival is high or low.One periodic cause of adult mortality -accidental trapping and poisoning by possum hunters- is under-represented in our sample, mainly becausediscouraged hunting. Extensive surveys show thatsome 10% of the current crop of adult kiwi inindicating that predationŽ by leghold traps has beensignificant over the previous two or three decades(McLennan and McCann, 1994; Potter, 1989; J.McLennan,unpubl. data). Some birds are injuredfatally by traps, so actual capture rates are probablyof maimed birds. Furthermore, some 8% of kiwimarkedŽ by leghold traps probably die each year,capture. This natural attrition would halve theproportion of birds with marks (from 20% to 10%)this period.We have no measures of annual mortality ratesattributable to traps and cyanide baits. However, inyears of high fur prices, such losses may exceed 5%,and significantly reduce the average longevity andEgg survivalKiwi lose few eggs (c.predators compared with most forest birds (average39%; Moors, 1983). This is remarkable given thatkiwi eggs are potentially at risk for up to 85 days,two to five times longer than those of all other forestbirds (Oliver, 1955). The high survival of eggs isprobably attributable to four factors: nest siteselection and camouflage; the attentiveness ofwould-be egg predators; and the large size andweight of the eggs.The relative contribution of the first two factorsvaries between northern and southern populations.Cryptic nests are a feature of kiwi in the Northunattended for up to 10 hours each night.Conversely, in the South and Stewart Islands, kiwioften nest in large and obvious burrows, but shareincubation and seldom leave eggs unattended.the entrance of their breeding burrow with leavesand twigs, presumably because the near continualpresence of an adult in the nest is an adequatedefence in its own right.In part, then, defensive behaviours which mayhave evolved to protect eggs against avian predatorsalso appear to be partially effective againstintroduced mammals. However, the historical legacythe eggs of kiwi are too large to be punctured orremoved by rats, the most numerous egg predator inmainland forests.Chick and juvenile survivalYoung kiwi suffer exceptionally high mortality,especially in their first 100 days of life. Our figuresadulthood, an estimate which is remarkably close toan earlier one (92%) derived from the age structurepredators (McLennan and Potter, 1993).Predators, chiefly stoats, cause about half ofthese losses. Young kiwi have no behaviours ordefences that protect them against introducedmammals. They forage independently from theirdaytime shelters in the early evening, up to an hourbefore adults. They frequently respond to anapproaching predator by freezingŽ rather than byfleeing. This response may be appropriate forpredators which detect prey by sight and/or sound(birds of prey), but is entirely ineffective againstmammals hunting prey by scent. Young kiwiselectively inhabit dense, regenerating vegetation (J.McLennan, P. Miller, unpubl. data), possiblybecause such habitats afford protection from birds ofprey. The same habitats provide little if anyby introduced mammals on young kiwi is the singlemost important factor contributing to the demise ofmainland populations.Management of mainland populationsNorthern brown kiwi will be saved in mainlandforests only by creating relatively predator-freechicks (at least 7 months old) to compensate for adultmortality. The birds have already disappeared fromDeclines can be halted if predation rates on youngMcLENNANet al.: PREDATION ON KIWIS 34NEW ZEALAND JOURNAL OF ECOLOGY, VOL. 20, NO. 1, 1996 ): recruitment and behaviour of juvenilesRange size and denning behaviour of brown (Apteryxaustralis 35 rodents on the eggs and chicks of native andedition. A.H. and A.W. Reed, Wellington, N.Z.reproductive biology of the North Island brownMonitoring breeding of Okarito brown kiwi inCommon and scientific names of mammalianFelis catusMustela furoSus scrofaTrichosurus vulpecula KMustela ermineaMustela nivalis vulgaris