IN THE MATTER of the Resource Management Act 1991

AND

IN THE MATTER of the hearing of applications by

EVIDENCE OF RHYS P. BUCKINGHAM

ON BEHALF OF THE APPLICANT

RELATING TO INDIGENOUS BIRDS,

BATS AND LIZARDS

WITHIN BEECH WORKING CIRCLES 1, 2 AND 3

December 1999

1 Preliminary

My name is Rhys P. Buckingham. I am a field ecologist specialising in surveys for birds and bats. I run a small consultancy business called Wildlife Surveys.

2 Introduction

This document summarises evidence regarding faunal (bird, bat and lizard) values within the Timberlands West Coast Limited (TWCL) forests proposed for sustainable beech management (beech working circles). The predicted impacts on fauna by sustainable beech management as proposed by TWCL are discussed.

TWCL seeks 35 year resource consents to undertake sustainable management of Crown owned beech and beech-podocarp forests within the Buller and Tasman Districts. Sustainable management will involve the extraction of beech trees and a very small proportion of rimu trees across a range of diameter classes. The rates and methods of extraction are designed to ensure that extraction does not exceed the natural rate of replenishment of the forest. Extraction of beech and rimu trees will be selective, and will almost exclusively be carried out by helicopter.

Detailed and extensive bird and bat surveys were carried out by Wildlife Surveys in TWCL proposed sustainable beech forests in the lower Grey Valley, eastern Paparoa Range, Maruia Valley and Inangahua Junction areas between 1994 and 1999

(Buckingham & Nilsson 1994a&b; Buckingham & Brown 1996; Buckingham 1998; Buckingham 1999a). These surveys involved five-minute bird counts, kiwi counts, standard bird distribution mapping, search for rare species including kiwi, blue duck and mohua (yellowhead), and surveillance for bats using electronic bat detectors. Reports and evidence of South Island kokako were investigated. Tony Whitaker carried out specialised surveys for lizards in the Grey Valley, Maruia Valley and Inangahua Beech Working Circles (BWCs) in 1997 and 1999 (Whitaker 1997, 1999).

3 Objectives

4 The faunal resourcebirds, bats and lizards

4.1 Birds

  1. The relative abundance and diversity of indigenous birds varied throughout TWCL working circles. Overall relative densities were high compared to many other South Island forest areas where comparative count data are available (see Table 1). These comparisons give only broad indications of relative abundance differences, as different observers undertook counts during different years, and different times of the year.
LOCATION/DATE COUNT MEAN

(total individuals)

Larrys Creek/Waitahu SF (1998)

10.4

Granville SF (1995)

9.7

Orikaka SF (1998)

9.6

Paparoa SF (1995)

9.0

Maimai SF (1995)

8.9

Maruia SF (1996)

7.1

Maruia SF (1979)*

16.3

South Karamea SF (1979)*

15.7

Granville SF (1979)*

12.2

Orikaka SF (1979)*

11.1

Paparoa SF (1979)*

10.2

Totara SF (1979)*

8.7

Matiri SF (1979)*

7.9

Otira-Kopara SF (1979)*

7.9

Kaniere Scenic Reserve (1979)*

7.8

Rowallan SF (1976)**

7.1

Longwood SF (1976)**

6.8

  1. It is assumed that relative densities and diversities of birds are similar in protected forests administered by DoC adjacent to the TWCL forests proposed for beech management, though surveys were not carried out in the DoC estate during the time that the TWCL surveys were carried out.

  2. Numbers of indigenous forest bird species recorded (including birds using streams and/or lakes within the forest) ranged from 19 at Granville SF in 1994, 24 at eastern Paparoa forests (Moonlight Valley to Maimai SF) in 1995/1996, 24 at Inangahua forests in 1998/1999, to 25 at Maruia SF in 1996/1997. Highest indigenous species and individual counts were recorded at Waitahu SF, in the Inangahua Working Circle (see Table 2).

Granville 1995

Paparoa

1995

Maimai

1995

Shenandoah

1996

Maruia

1996

Orikaka

1998

Waitahu

1998

Mean indigenous species

4.68

4.69

4.13

5.30

5.80

Mean indigenous

individuals

9.70

9.45

7.39

7.10

7.07

9.70

10.73

  1. Of the indigenous bird species recorded, eight are regarded as of conservation importance (after Molloy & Davis 1994). These are great spotted kiwi, blue duck, New Zealand falcon, western weka, New Zealand pigeon, kea, South Island kaka, and kakariki (yellow-crowned parakeet). Of these species, kiwi and blue duck were restricted in distribution, while the other species were found throughout in variable numbers.

  2. Sign of kokako or reports of their presence in TWCL forests were not confirmed (see item 27, Section 4.1). No sign of mohua (yellowheads), orange-fronted kakariki (item 18, Section 4.1) or other very rare species was found.

  3. Great spotted kiwi were found only at Orikaka SF and areas of TWCL forest in the Paparoa Range. No sign of them was found east of the Inangahua and Grey Rivers, including Maruia SF. This supports current knowledge of their distribution in Buller and Westland (John Lyall, pers. comm., October 1998).

  4. Highest kiwi counts were recorded at Orikaka SF, where call rates were similar to those recorded in the contiguous Mount William Range/Ngakawau Valley/Glasgow Range area (Kingett Mitchell 1997) (see Table 3). This population is regarded to be of regional and national significance. Kiwi counts in the lowland Paparoa Range were overall much lower and indicative of a more patchy distribution.

  5. Overall counts indicated that kiwi numbers were lower in highly modified forest compared to unmodified forest areas (see Table 4)*. However the apparent lower numbers of kiwi in highly modified forest could be due to higher predator numbers, or higher incidence of possum trapping/poisoning in those areas
.* Note: although habitat seemed to impact on kiwi counts (F-prob=0.074 for two-hour count; F-prob =0.036 when only first hour of count considered), our data indicate only broad trends that need to be tested by more rigorous survey design

LOCALITY

NO. OF COUNTS

MEAN CALL RATE

(No. of calls per hour)

Stronghold, north-west Nelson

57

6.4

North-west Nelson

152

3.3

Karamea River to Buller River

85

0.9

Mount William/Orikaka

1997

25

2.5

Orikaka SF 1998

13

2.5

Ohikanui SF 1998

12

0.8

Ohikanui SF (unlogged)

1998

7

1.1

Paparoa range and surrounds

107

1.8

Arthurs Pass/Southern Alps

175

1.1

Entire geographical range

640

1.5

Cutover

Partly logged

Unlogged

Mean

1.63

3.63

4.71

Std error

0.829

0.734

1.193

  1. Only two records of blue duck (one seen on Pike River, eastern Paparoa Range and one on Rappahannock Lake, Maruia SF) were obtained in approximately 80 km of suitable streams and rivers searched. Information from local residents indicates that blue ducks are a rare and declining species in this area, but that they are commonly recorded in Big River and Pike Stream in Paparoa SF, where cutover forest and farmland abuts the stream sides.

    Blue duck records in the survey area were found to be few and inconsistent compared to records in the Owen River catchment of the upper Buller River which is regarded as the stronghold for blue ducks in the region (Planning Tribunal 1996; files SUR 503 and PES 143/1 DoC, Nelson). Similarly, blue ducks are concentrated in other parts of Kahurangi National Park such as Roaring Lion River where averages of up to one adult blue duck per kilometre were recorded (Studholme 1999).

  2. New Zealand falcons were widespread but only occasionally encountered within the survey area. One confirmed breeding record was obtained (a nest with 3 eggs found at Orikaka SF on 26 November 1998).

  3. Western weka were patchily distributed within the survey area. Highest counts were recorded in Te Wharau SF, while lowest numbers were encountered in Maruia SF. Weka populations in the Inangahua and Grey Valley areas are considered to be of national conservation importance, because of their relative density in those areas (Buckingham 1999a). Weka have shown sudden, dramatic declines in some other areas (e.g. at Golden Bay). Counts and casual observations indicated that weka were as common, or more common, in shrubland and cutover forest than unlogged forest.

  4. New Zealand pigeons were present throughout, but notably most common in the more diverse forests in the west of the survey area. Highest numbers were recorded at Granville SF, Orikaka SF and forests along the lower Buller gorge, while lowest numbers were encountered at Maruia SF. New Zealand pigeon densities within the survey area were not regarded as exceptional by national standards (Buckingham 1999a), though important local concentrations were found (Table 5).

LOCATION

DATE

% OCCURRENCE

n

Orikaka SF

1998

27

64

Ohikanui SF

1998

53

36

Larry/Waitahu SF

1998

4

56

Granville SF

1994

41

64

Maruia SF

1996

1.5

132

Bruce Bay

1984

73

89

Mataketake

1985

70

228

Big Bay

1986

67

184

Arawata

1985

54

142

Copland Valley

1984

42

63

Waitutu SF

1984

50

227

Catlins Forest Park

19811986

76

79

  1. South Island kaka are present throughout TWCL forests in variable numbers. Notable areas where kaka were recorded in relatively high numbers included the Rappahannock River/Glenroy River area at Maruia SF, Granville SF, Blackwater (Reefton hill-country), Aynsley Creek/Big River area at Paparoa SF, Larrys Forest and Orikaka SF. New Zealand Wildlife Service, and DSIR Surveys in the 1970s also found comparatively high kaka numbers in the Reefton hill-country area (Crook et al.1977; Dawson et al.1978).

KAKA % KAKARIKI % n (counts )

Granville SF

(1994)

20.7 3.6 140

Granville SF

(1995)

12.1 0 66

Blackwater

(1994)

34.1 2.4 41

Paparoa SF

(1995)

8.1 11.7 111

Maimai SF

(1995)

4.5 18.2 110

Larrys Creek/ Waitahu SF(1998)

12.2 8.3 180

Rough Creek, Inangahua (1998)

0 10.0 30

Orikaka SF

(1998)

13.3 14.8 210

Maruia SF

(1996)

8.7 31.1 427

Waitutu SF

(1984)*

47 20 731

Western Southland

(1977)*

17 21 470

South Westland

(19771978)*

4 0.5 492

North Westland/ Buller (1979)*

6 9 995

Note: Kaka densities in some parts of South Westland (Mataketake, Landsborough, Cascade, and Big Bay) approach those at Waitutu SF (O'Donnell & Dilks 1986).

  1. Relative kaka densities in Buller forests were considered high, with local densities approaching those of current stronghold areas on the mainland (see Table 6). Thus the area is considered to be of national importance for kaka.

  2. In all localities surveyed, kaka were less often encountered in modified forest areas than in unlogged forest. Count differences of kaka between cutover and unlogged forest at Granville SF and Blackwater were highly significant (P=0.001) (Buckingham & Nilsson 1994b). Nocturnal counts as well as casual observations indicated lower numbers of kaka in heavily cutover forest compared to other forest habitats in the eastern Paparoa Range forests (Buckingham & Brown 1996). Notably lower counts of kaka were recorded in partly-logged and cutover forest in TWCL Inangahua working circle forests (see Fig. 1) (Buckingham 1999a). Morse (1981) also noted that kaka were uncommon in heavily cutover and fragmented forest in the Buller/north Westland region.

(16) Kea were found throughout the survey area in low numbers. They were more frequently encountered at Maruia SF than other areas. For example they were found in 6.3% of counts at Maruia SF compared with a maximum of 3.9% of counts in the lower Grey River/Inangahua River area. At Maruia SF they were commonly encountered in the heavily-logged lowland areas at Station Creek, presumably because of the large numbers of dead, standing spars in that area (important food source).

(17) Kakariki were found throughout the survey area, with notably highest densities found at Maruia SF (see Table 6). The kakariki population at Maruia SF is considered to be of regional and national importance, being one of the most significant New Zealand mainland populations (see Tables 6 and 7) (Buckingham 1998).

  1. A concerted effort was made to search for orange-fronted kakariki (Category A, conservation priority species) in Maruia SF, but they were not identified in 80 relatively close observations using binoculars. All kakariki identified (in Maruia SF and elsewhere in the TWCL estate) were yellow-crowned kakariki.

(19) Kakariki were observed to be present in cutover forest as well as unlogged forest, and their preference for either habitat was not as distinct as for kaka (Fig. 1). Kakariki were notably common in some areas of cutover forest in the lower Grey River (Buckingham & Nilsson 1994b). In Orikaka SF, kakariki were recorded in higher numbers in cutover forest than unlogged forest in autumn, but the opposite in spring (Buckingham 1999a). In the Inangahua working circle as a whole, mean counts of kakariki were higher in unlogged forest than cutover or partly-logged forest (Fig. 1). In eastern Paparoa lowland forests, kakariki appeared to prefer unlogged or partly-logged hill-country forest to cutover terrace forests (Buckingham & Brown 1996).

LOCATION/DATE % OCCURRENCE

(IN GRID SQUARES)

n

Maruia SF (1996)

77 132

Orikaka SF (1998)

56 64

Blackwater (1994)

50 20

Ohikanui/TeWharau (98)

43 40

Waitahu SF (1998)

41 27

Granville SF (1994)

31 64

Camp Creek (1998)

31 32

Waitutu SF (1984)**

52 227

Arawata (1985)***

49 142

Wakatipu SF (1982)*

36 237

Landsborough (1985)***

23 163

Windbag Valley (1983)***

20 69

Okuru (1985)***

15 76

  1. Riflemen, robins and brown creepers were patchily distributed within the survey area. There is concern that riflemen and possibly brown creepers may be disappearing from lowland forest areas (<600 m a.s.l) in the Buller/north Westland region. Riflemen were recorded at Granville SF in the 1970s (Morse 1981), but they were not recorded there in 1994 and 1995 (Buckingham & Nilsson 1994a&b; Buckingham & Brown 1996). At Maruia SF there appeared to be a correlation between riflemen densities and wasp densities, with riflemen generally being commonly encountered only in areas where wasp densities were relatively low (Buckingham 1998). The only lowland location where riflemen were notably common (i.e. >0.5/count was at Orikaka SF, on eastern ridges between Pensini Creek and New Creek (Buckingham 1999a)

  2. Riflemen showed a distinct preference for unlogged forests, and were absent in heavily cutover forest. Even in partly-logged forest riflemen were uncommon or not recorded (Fig.1) (Buckingham & Brown 1996; Buckingham 1999a).

(22) At Granville SF, robins were inexplicable absent in podocarp/beech hill-country forest (PB5), but were present in modified kahikatea-rimu/beech forest in some of the lower-lying valleys (Buckingham & Nilsson 1994a&b). Otherwise robins were generally common throughout most locations and habitats. Robin densities were considered to be exceptionally high on a national level at Maruia SF (particularly the northern part of the area), and parts of Inangahua Working Circle ; (Buckingham 1998,1999a).

  1. In the Inangahua and Grey Valley forests, robins appeared to be as plentiful in cutover and partly-logged habitat as in unlogged forest (Fig. 2). However, forest fragmentation may affect robin populations as indicated at Maimai SF where no robins were recorded (using tape playback) in one isolated, and heavily cutover forest outlier (Buckingham & Brown 1996).

  2. Brown creepers were infrequently encountered in lowland forest throughout, the highest counts being recorded at Granville SF (0.24/count in November 1995). In higher altitude parts of the TWCL area (>600 m a.s.l), brown creepers were commonly encountered only in the headwaters of Station Creek (Maruia SF) where they were recorded as >0.4/count (Buckingham 1998). Brown creepers did not show any clear preference for modified or unmodified forest in TWCL Buller forests.

  3. Tui were found throughout the survey areas in generally low to moderate densities. However, concentrations were found (e.g. between Slaty Creek and Pike River, Paparoa SF; Granville SF; Waitahu River and western Orikaka SF, Inangahua Working Circle; Glenroy River, Maruia SF). These forest areas (apart from the latter) have a high component of podocarps and/or rata, which are important foods for tui.

  4. Overall tui numbers appear to be higher in unlogged than cutover or partly-logged forest (Fig. 2), but tui use cutover or other modified forest seasonally (Buckingham 1999a).

  5. Sign of presumed South Island kokako (silhouette sightings, calls, moss grubbing) was found at several localities within the survey area (notably Granville SF, Maimai SF, and Glenroy River). In addition, recent reports of kokako were known from Station Creek (Maruia SF), Ohikanui SF, and the upper Inangahua River (McConnochie Creek block) (Robin Inch, pers. comm., April 1996; Neil Taylor, pers. comm, May 1996; Rob Dalley, in litt., 21 September 1998; Gina Boffelman, pers. comm., September 1998). However, unequivocal evidence of kokako presence was not obtained even after considerable investigative effort (Buckingham 1996b; Buckingham 1997; Buckingham 1999a; van Mierlo 1997). Lack of positive evidence does not exclude the possibility of kokako presence.

  6. Accumulative evidence for the presence of South Island kokako has been established since the early 1980s, with sightings, a feather (from Stewart Island), several tape recordings of calls, and associated moss grubbing sign (Buckingham 1996a). The evidence has been generally accepted as sufficient to warrant urgent investigative effort (Innes et al. 1988; Cheryl Mudford in litt.1996; The Ornithological Society Rare Birds Committee via Chris Robertson, in litt.17 January 1997). However, the bird's rarity, scant vocal activity, and secretive behaviour make the prospect of confirming South Island kokako presence very difficult and time-consuming.

Note: Apart for fantail, no overall significant differences were found in counts between habitats as shown for species in the above figures. Counts of riflemen in Orikaka SF were significantly higher in unlogged hill-country forest than adjacent cutover forest.

4.2 Bats

  1. A search for bats using up to four Batbox IIIÖ electronic bat detectors was made throughout the TWCL Buller forests during summer months. Methods followed that described by O'Donnell & Sedgeley (1994) for automatic surveillance of bats. Detectors were placed overnight at forest edge and forest interior locations, near streams and on ridges or saddles.

  2. The only species of bat detected were presumed to be long-tailed bat which are listed under Category B conservation priority by DoC (Molloy & Davis 1994). Results of 1455.5 hours surveillance on 4045 kHz frequency setting indicated a patchy distribution of long-tailed bats, with concentrations in parts of Maruia SF, and isolated records near Reefton (see below). Short-tailed bats (2728 kHz) were not detected in a total of 671 hours of surveillance, though they may be present, as they are difficult to detect (O'Donnell 1997).

  3. Mean number of bat passes per hour detected on 4045 kHz (optimum range for long-tailed bats) are presented in Table 8. The highest pass rates detected were 7.0 passes per hour at Glengarry Stream, 5.1/hour at Thompsons Flat near Lake Daniels, and 1.2/hour at Conglomerate Road (all sites within Maruia Working Circle). Bats were detected at low pass rates at the Inangahua River and tributaries just north of Reefton, and at one location at Rough River near Ikamatua. They were not detected at Granville SF, Orikaka SF, or in areas near Inangahua and the lower Buller Gorge. Nor were they detected in the central part of Maruia SF from Pea Soup Creek to Rappahannock River.

LOCATION

PASSES/HR

TOTAL PASS

No. SITES

TOTAL HRS

Maruia WC

0.185

70

43

378.5

Inangahua WC

0.013

10

98

750

Paparoa WC

0.011

3

46

267

Granville SF

0.000

0

8

60

  1. Overall bat detection rates were low in TWCL Buller forests compared to those in stronghold long-tailed bat populations such as the Eglinton Valley in Fiordland. In the Eglinton Valley, bats were detected in a January average of 65% of monitored sites (Colin O'Donnell, pers. comm., 1996), whereas bats were detected in only 28% of sites At Maruia SF, 5% of sites at Inangahua Working Circle, and 2% of sites at Paparoa Working Circle during summer surveillance.
4.3 Lizards

An assessment of the lizard fauna in parts of the TWCL Buller/north Westland estate was undertaken by Tony Whitaker in March 1997 and March 1999 (Whitaker 1997; Whitaker 1999). The field surveys involved a limited but targeted search within different habitats and altitudinal ranges within the estate, and adjacent areas. The results of these surveys, and other recent surveys (particularly two extensive surveys undertaken by DoC), combined with anecdotal information, have provided baseline information on the herpetofauna of the West Coast and Buller region.

  1. The taxonomy of West Coast lizard fauna is far less understood than for many other parts of New Zealand. Up to 17 lizard species (9geckos, 8 skinks) are now known from the West Coast, with 910 species (5 geckos, 56 skinks) recognised from the north Westland/Buller region (Whitaker 1999).

  2. Surveys have confirmed that skinks are scarce on the West Coast, and are mainly localised in coastal areas (van Mierlo 1998; Miller et al. 1999; Whitaker 1999). Geckos are probably more evenly spread, at low densities throughout forest habitats.

  3. Lizard species known to be present in north Westland and Buller are 'West Coast' forest gecko (Hoplodactylus aff. granulatus), common gecko (H. maculatus), Canterbury gecko (H. aff. maculatus 'Canterbury'), Southern Alps gecko (H.aff. maculatus 'SouthernAlps'), West Coast green gecko (Naultinus tuberculatus), speckled skink (Oligosoma infrapunctatum), 'Grey Valley' skink (O.aff. nigriplantare 'Grey Valley'), brown skink (O. zelandicum), and two other species of skinks (O. sp. 'Paparoa' and O. sp. 'Denniston'). The common skink (O. nigriplantare polychroma) may also be present as it occurs both north and south of the north Westland region (Whitaker 1999).

  4. The only species confirmed present in the TWCL Working Circle areas surveyed (BWC1, 2 and 3) are West Coast green gecko and 'West Coast' forest gecko. Information derived from interviews indicated that forest geckos are widespread and relatively frequently encountered in forest habitats, whereas green geckos are found far less frequently (Whitaker 1999). Skinks appear to be very rareor elusivein the Buller/north Westland region, and are seldom reported. The only possible sighting of a skink within the Timberlands estate was near the Mackley River (Orikaka SF) in April 1998 (Buckingham 1999a). Reports of skinks (or probable skinks) are also known from farmland near the Timberlands estate, for example in the upper Maruia Valley (Whitaker 1997) and at Waitahu, near Reefton (Whitaker 1999)

  5. Two species found or presumed to occur in TWCL Working Circles, West Coast green gecko and speckled skink, are listed on the IUCN Species Survival Commission 'Red List' as species of 'lower risk' but are not listed for conservation priority by DoC (Molloy & Davis 1994). However, West Coast skinks are listed for conservation priority by DoC. The 'Grey Valley' skink and the 'Denniston' skink, considered to be of high conservation priority, are likely to occur within the TWCL estate (Whitaker 1999).

4.4 Population trends

  1. Kiwi, blue duck, western weka, New Zealand pigeon, South Island kaka, kakariki, mohua and endemic bat populations have declined since European settlement, and are continuing to decline (Oliver 1955; Morse 1981; Molloy & Davis 1994; Heather & Robertson 1996). The main reasons for these declines are habitat removal and modification, forest fragmentation, predation, competition, hunting, and disease (Oliver 1955; Morse 1981; Heather & Robertson 1996). The main current cause of indigenous bird declines is accepted to be predation, and competition for food by introduced mammals and wasps (Innes & Hay 1991; Clout et al. 1995; Elliott 1996; Elliott et al. 1996; McLennan et al.1996; O'Donnell 1996a&b; O'Donnell & Phillipson 1996; Wilson et al. 1998).

  2. Little spotted kiwi and South Island brown kiwi were formerly present in Buller/north Westland forests, but are now gone from this region (Smith 1888; NZWS 1986). Mohua have more recently disappeared from the region, the last confirmed report being from forests near Reefton in the late 1970s (Morse 1981; Gaze 1985; NZWS 1986).

  3. Some indications of recent (past 2030 years) bird population trends are shown when the results of our surveys are compared to those carried out by Ecology Division (DSIR) and New Zealand Wildlife Service between 1974 and 1979 (Best & Harrison 1976; Crook et al. 1977; Dawson et al. 1978; Wilson et al. 1988). The only species for which declines are indicated over the past 2025 years are rifleman and brown creeper (Buckingham 1999a). No obvious indications of kaka or kakariki decline are evident when comparing data of the two survey periods. Local information suggested that numbers of blue ducks had declined in eastern Paparoa Range, and pigeons had declined near Inangahua Junction over the past 2030 years.

  4. General patterns of bird distribution appear to be similar in the 1970s to the present day. Some of the key similaries being:
(a) notably high densities of kakariki in Maruia SF;

(b) relatively high densities of kaka in the Reefton hill-country; Inangahua Valley, and parts of Maruia SF;

  1. great spotted kiwi concentrated in the Paparoa Range and forest north of the Buller River;
  2. higher numbers of pigeons recorded in podocarp/beech forests of the lower Grey and Inangahua valleys than in beech forests at Maruia SF;
  3. notably high densities of robins in Maruia SF, and the lower Buller RiverInangahuaReefton area;
  4. patchy distributions and high local density variations of rifleman and brown creeper;
  5. relatively high numbers of riflemen in the southern part of Maruia SF.
  1. Local information and other sources (e.g. Morse 1981; Don 1993) show a concentration of bat records in the upper Grey Valley, Buller River, and Maruia Valley. Bats were recorded as being commonly seen in the lower Grey Valley/Reefton area 40 or more years ago, whereas they are infrequently reported now (Buckingham & Brown 1996).

  2. Throughout New Zealand many species of lizards (particularly forest-dwelling species) have declined mainly as a result of the combined effects of introduced predators and competitors (vespulid wasps being a particularly serious problem in Buller and north Westland) (Whitaker 1999). These impacts, as well as forest fragmentation and habitat loss, have probably resulted in significantly lowered populations of forest-dwelling geckos in the Buller/north Westland region.

5 Predicted impacts on fauna (birds, bats and lizards)

5.1 Predicted impacts by sustainable beech management

  1. Although there have been several studies describing the effects of 'clear-fell' or 'large coupe' logging on indigenous birds (e.g. Coker 1980; Onley 1983; Taylor 1985; Spurr 1985) the effects of 'selection system' harvesting have not been studied. Thus these studies cannot be used to predict the impacts on birds by the proposed TWCL sustainable beech harvesting methods.

  2. O'Donnell & Dilks (1987), from their South Westland study, developed a model for predicting the impacts of selective logging on indigenous forest birds by determining the components of the forest that were preferred by or critical to them. They concluded that the logging of even low volumes of large-diameter (>60 cm, dbh), mature or old trees (podocarps and silver beech) would have severe impacts on the populations of specialist bird species including kaka, kakariki and mohua.

  3. Removal of large-diameter senescent, or dead trees was considered to have a disproportionately severe effect on the birds (or bats) using them (O'Donnell & Dilks 1987). These trees provide nest or roost holes for various species (e.g. endemic bats, kaka, and riflemen), and an important food source (e.g. epiphytes, fungi, and insects).

  4. However, the TWCL management prescriptions state that only very low volumes (<10%) of rimu (critical species for kaka and kakariki) will be taken, and other podocarps (e.g. kahikatea and miro critical for NZ Pigeon) will not be logged. A high proportion (>90%) of dead trees (critical for kaka and kea) will be retained.

  5. The O'Donnell & Dilks model assumed that logging would be based on selected target trees and preferred stem diameters. However the TWCL 'selection system' management proposes to harvest species in proportion to their natural presence across a range of diameter classes to simulate natural processes (e.g. wind-throws).

  6. Also, their study assumed that the removal of critical or focal trees would adversely affect birds using them, with no allowance for the adaptability of some species (e.g. New Zealand pigeon and tui) which may use less preferred or alternative trees.

  7. Specialist bird and bat species known to be present in TWCL Buller forests are great spotted kiwi, South Island kaka, yellow-crowned kakariki, and long-tailed bat. It is predicted, from survey results and other studies, that South Island kaka will be the most vulnerable species to development. Other species that may potentially be impacted by the proposed development in this area are great spotted kiwi, New Zealand pigeon, rifleman and tui.

  8. The impacts of the proposed development on riflemen are difficult to predict, but they are a species that should be monitored closely. O'Donnell & Dilks (1987) found that silver beech (particularly trees of 6080 dbh) was critical for riflemen. It is suspected that mammalian predators and wasps will be a much more limiting factor for riflemen than the minor habitat changes caused by the proposed 'selection system' management.

  9. Management may have some impact on long-tailed bat populations where they are present, as bats actively select tall trees with large trunk diameters that are usually targeted for development (Sedgeley & O'Donnell 1999). They prefer red beech and dead trees. However, the sparsity or absence of bat records in large areas of unmodified forest (e.g. Orikaka SF and Granville SF, central part of Maruia SF) indicates that habitat is not the only limiting factor for bats.

  10. Overall impacts on fauna, if they occur, are predicted to be slight compared to the impacts of traditional forest harvest methods. Earlier (pre-1980) selective logging practices were not found to be obviously detrimental to birds, as the impacts were to an extent ameliorated by subsequent forest regeneration (O'Donnell 1991). The currently proposed beech management design will result in much less change to forest structure and composition than past selective logging practices, thus further reducing the likelihood of any major faunal impacts. In discussing beech management options for forests in the proposed beech project area of Buller and north Westland in the 1970s, Crook et al. (1997) stated that "careful logging for podocarp and some beech, would clearly best maintain continuity of habitats and bird fauna between Ecological Reserves." Backed by extensive survey results they stated that "The original forest structure is retained, no catastrophic change is involved and the modified habitat is capable of supporting all species of native birds in closely similar proportions to those found in unexploited forests."

  11. Our survey results indicated that there may have been some impacts on birds caused by earlier, selective logging practices that removed merchantable podocarp and low proportions of beech (Figs 1 and 2). However the partly-logged forest areas studied in Inangahua Working Circle were relatively isolated forest blocks surrounded by farmland and/or heavily cutover forest, thus forest fragmentation rather than, or more than, selective logging may have influenced these impacts. The current proposed sustainable harvest method would not result in forest fragmentation.

  12. The impact of harvesting timber on the lizard fauna and its habitat is dependent on the harvest methodology used (e.g. clear-felling, coupe, single tree, etc) and is not easily predicted (Whitaker 1999). The 'Selection System' harvesting method proposed by TWCL is unlikely to significantly affect lizard habitat, as the method should not cause any major change to forest composition or structure. Retaining most of the large old trees is important for lizards as they use loose bark, crevices, dead limbs and epiphytes for retreat sites and food resources.

  13. Indirect effects on lizards resulting from the TWCL harvesting regime are almost impossible to predict (Whitaker 1999). Whitaker stated that "As one component of a complex forest ecosystem the lizard fauna is dependent on the health and stability of that ecosystem for its welfare and survival. If the harvesting regime, through some mechanism, triggers environmental perturbations through that ecosystem, lizards may suffer". The most serious threats to the lizard fauna would be increases in densities of mammalian predators or increases in densities of competitors such as introduced vespulid wasps (Whitaker 1997; 1999).

  14. It is not known to what extent, if any, the proposed development will impact fauna indirectly, by resulting in increases of mammalian predators, or increases in competitors (see Section 5.2). Changes in forest structure and composition (expected to be minor), and increased roading may cause some change to predator/competitor compositions and/or densities. Deleterious effects arising from these changes may be reversed by strategic pest management operations (see Section 6.1).

5.2 Other impacts on fauna

  1. Recent research has shown that introduced mammalian predators are the main cause of current indigenous bird declines (Innes & Hay 1991; O'Donnell 1996a). These findings specifically apply to endangered species such as kiwi (McLennan et al.1996), New Zealand pigeon (Clout et al. 1995), kaka (Wilson et al. 1998), kakariki (Elliott et al. 1996), mohua (Elliott 1996; O'Donnell 1996; O'Donnell & Phillipson 1996; O'Donnell et al. 1996), and North Island kokako (Innes et al. 1999).

  2. Browsing mammals, and vespulid wasps have also impacted indigenous fauna, through the effects of competition. Although research has been carried out to determine these effects on specific fauna (e.g. Leathwick et al. 1983; Beggs & Wilson 1991; Moller et al. 1991; Wilson et al. 1998), the full extent of these impacts is largely unknown.

  3. Habitat removal caused by mining and other industries has also impacted indigenous fauna in the Buller/north Westland regions. Notable damage to mainly valley floor habitats by gold mining is found at Granville SF and other parts of BWC1, in the lower Grey Valley. Significant areas of forest have also been removed or modified by coal mining in Inanangahua Working Circle (BWC2). Establishment of electricity power pylons at Orikaka SF has impacted on considerable areas of forest habitat there.

  4. Trapping and poisoning for possums may have impacted indigenous fauna to some extent. These operations may have reduced populations of ground birds such as kiwi and weka, and also foraging birds such as robins. However, reductions in numbers of mammalian pests from these activities may have had beneficial effects on indigenous fauna.

6 Mitigation and compensation

6.1 General

Any development of New Zealand indigenous forests will have some impacts on indigenous fauna. The key issues should define whether or not these impacts are likely to be severe and irreversible (ie. leading to significant declines or causing the extinction of a species), and whether the deleterious effects can be avoided, mitigated against, or compensated for.

TWCL management prescriptions detail the mitigation processes to be put in place. In summary these are:

In addition to these mitigation measures, TWCL have elected to invest a proportion of the annual net revenue derived from beech management, into active forest pest control. Predator control may be regarded as a compensatory measure against the adverse effects of 'selection system' management on indigenous fauna. There is growing evidence for the recovery of species and/or ecosystems following ecological management at mainland sites (O'Donnell et al. 1996; Sim & Saunders 1997; DoC 1999; Innes et al. 1999). Population viability modelling for northern brown kiwi, mohua and North Island kokako have indicated that predator control is likely to have a substantial positive effect on the population survival of these species (McLennan et al.1996; Elliott 1996; Innes et al.1999). It is possible that the net conservation gains from well-designed and monitored predator control may outweigh the negative effects from harvest management.

6.2 Selection and justification of reserves within the TWCL estate

  1. The selection and function of reserves outside the TWCL estate have already been discussed.

  2. Areas reserved from production within the TWCL estate are streamside management zones, recreational zones, wildlife and special forest associations, and special landscape zones.

  3. Wildlife and special forest association reserves within the TWCL Beech Working Circles 1, 2 and 3 were selected to protect areas considered to be of especially high ecological value, based on:
  4. areas where species of conservation importance (Molloy & Davis 1994) were concentrated;
  5. areas where regionally threatened species (e.g. rifleman and brown creeper) were concentrated;
  6. areas where indigenous birds were recorded in notably high densities or diversities;
  7. areas where bird habitats are locally, regionally, and/or nationally unique or endangered (Imboden 1978; Coker & Imboden 1980).

    (4) The main proposed reserves and justification for these reserves within the TWCL Buller/north Westland estate (BWC1, BWC2 and BWC3) are as follows.

  8. Valley floor areas in Granville SF where there are high proportions of podocarps (particularly kahikatea and rimu), and red beech. High bird species diversity, and notable abundance of pigeons. Presence of robins which are absent in hill-slope PB5 hard beech forest. This lowland, podocarp-rich habitat is rare in north Westand.

  9. Area of forest between Aynsley Creek, Granity Creek and Shellback Creek in Paparoa SF has notable concentrations of great spotted kiwi, kaka and kakariki. Diverse forest associations including stunted podocarp species and mountain beech on ridges, and taller podocarp/beech forest in gullies.

  10. Area between Slaty Creek and Kakapo Creek in Paparoa SF was noted for its high podocarp content, floristic diversity, and abundance of birds (especially tui). Possible presence of South Island kokako (calls and response to tape playback).

  11. Mount Courtney escarpment, Orikaka SF. Notable concentrations of kaka, kakariki and tui, and high indigenous species diversity.

  12. Blue Duck Creek, Orikaka SF. High numbers of endemic birds including great spotted kiwi, kaka, kakariki, pigeon and tui. High proportion of kahikatea and rimu.

  13. Pell Creek, Orikaka SF. In the most remote part of this block, relatively high counts of great spotted kiwi and diurnal birds including riflemen and brown creepers were recorded. High podocarp content on mainly broken limestone terrain.

  14. Rough Creek, Inangahua Working Circle. An area of forest on undulating hills in the lower part of Rough Creek is one of the few remaining largely unmodified lowland podocarp/beech forests in the east Inangahua Valley area. Notably high counts of robin, tui and other birds, and relatively high counts of pigeons in autumn.

  15. Rip and Tear Creek, Larrys Forest (Inangahua Working Circle). Notably high kaka numbers with groups of up to ten seen. High component of podocarps (locally exceptional) in this unlogged forest area.

  16. Waitahu Valley. Podocarp-rich terraces provide important habitat for kaka, kakariki and tui. Highest indigenous bird species and individuals recorded in TWCL survey areas (see Table 2).

  17. The Glenroy Terrace area, Maruia Working Circle was notable for the concentration of kaka, kakariki and tui, and high indigenous species diversity. Intensive searches for South Island kokako between 1997 and 1999 resulted in sign (sighting, calls, tape responses, and moss grubbing) of at least one bird present.

  18. Thistle Creek to Rappahannock Lake, Maruia Working Circle. Seasonally high concentrations of kaka and kakariki, and the highest indigenous bird species diversity recorded in Maruia SF in spring 1996. Spectacular landscape features, and a wetland area.

  19. Ruffe Creek/Pea Soup Creek (saddle area), Maruia Working Circle. Highest concentration of kakariki recorded within TWCL survey areas. Also high robin counts and high indigenous species diversity. The large proportion of dead trees in this area is presumed to contribute to the richness of bird habitat there (offering abundant food and nesting sites to species such as kaka, kakariki and kea).

  20. South Station Creek, Maruia Working Circle. Relatively high numbers of bat passes were recorded in this area. Also notable concentrations of riflemen and brown creepers.

  21. Tilted limestone area, including Lake Caslani and Cliff Creek Lake in Shenandoah Forest, Maruia Working Circle. Significantly high numbers of robins, reports of blue ducks, and presence of scaup and black shag colonies on the earthquake-formed lakes.
  1. As well as protecting specific bird and bat habitats as described above, these reserve areas also increase the overall retention of podocarps, large-diameter trees, senescent trees, and standing dead trees that are important to fauna (for feeding, nesting, roosting, or sheltering).

  2. The relatively even dispersal of reserves within the production area may prevent or reduce any potential disruption of use by mobile species.

6.3 Other mitigation/compensation options

The following mitigation options have been outlined in the Assessment of Environmental Effects, and are incorporated within the framework of the proposed management regime.

  1. Retain a high proportion (>90%) of rimu, as this species is highly prefered and seasonally critical for many species of birds (particularly kaka, kea, kakariki, pigeon and tui). Kahikatea and miro will not be harvested as they provide seasonally critical food for pigeons.

  2. Southern rata will not be harvested as it provides important seasonal food for some species of birds (e.g. kaka, kea and tui).

  3. Retain a high proportion (>90%) of large diameter (>60 cm), senescent, and dead standing trees as these are important for fauna (providing nesting, roosting and/or feeding sites for many birds, bats and lizards). The TWCL prescriptions allow for 100% retention of dead standing trees unless safety reasons apply.

  4. In areas where great spotted kiwi are present, logging should be avoided during the kiwi breeding season (late winterearly summer).

  5. Ban dogs from areas where kiwi are present, as dogs are known predators of kiwi (McLennan et al. 1996).

  6. Improve forest health by reducing numbers of mammalian browsers, particularly possums, goats and pigs that are concentrated in some areas (see TWCL fauna survey reports).

  7. Predator control (discussed in detail elsewhere).
7 Monitoring

7.1 General

Monitoring the potential effects of 'Selection System' sustainable beech management on the environment is a major issue within the planning framework. Long-term monitoring systems are required to detect any seriously deleterious effects should they occur. Proposed monitoring systems are designed to monitor gross changes in:

Only the monitoring of avifauna is discussed here. Monitoring of other components of the ecosystem has been covered elsewhere.

7.2 Monitoring birds

  1. If the objective is to determine the effects of 'selection system' beech management on indigenous bird populations, then monitoring should be carried out over a wide area of the proposed management estate (including at least three working circles), and samples replicated to allow sufficient power of analysis.

  2. The study design should incorporate matching treatment and non-treatment sites within each study location (this may require having sites on DoC land). The same sample sites should be used for each repeated survey (i.e. sites permanently marked).

  3. Ideally, a Before-After/Control-Impact (BACI) monitoring design should be used, to compare bird populations before and after development. BACI designs are useful to detect the effect of an activity (Manly & Brown 1999).

  4. Long-term population trends are best monitored using index or relative abundance methods, as these methods are cost-effective, and less intrusive on study populations than absolute density estimate methods (Verner 1981; Dawson 1981; Buckingham 1999b).

  5. Monitoring should be targeted at species most likely to be impacted by management. These species include great spotted kiwi, kaka, kakariki, and rifleman.

  6. Five-minute counts (Dawson & Bull 1975) and standard kiwi call counts (Butler & McLennan 1991) are the most appropriate methods to use for long-term monitoring of birds. Both methods have been used widely throughout New Zealand.

  7. Sampling should be designed to minimise confounding sources of variation, such as observer bias, time of day/year, weather, background noise, etc. The most appropriate time of year for monitoring birds is springearly summer when birds are generally most conspicuous, and their numbers less influenced by flocking, or juvenile recruitment.

  8. More robust monitoring methods (e.g. radio-tagging, banding, territory mapping) may be required to assess population changes of birds in conjunction with predator control operations. These are localised studies requiring a greater degree of accuracy than measuring overall broad population changes (trends) that may result from sustainable beech management.

  9. The long-term monitoring proposal described by Buckingham (1999b) sought the advice of two statisticians (Caryn Thompson, Department of Mathematics and Statistics, University of Otago; Peter Alspach, Hort+Research, Motueka). Consultation with statisticians at the planning as well as the analysis stage is essential to ensure feasibility and accuracy of the study design.

8 CONCLUSIONS

8.1 Conservation significances

8.2 Impact considerations

9 REFERENCES

Beggs, J.R.; Wilson, P.R. 1991. The kaka Nestor meridionalis, a New Zealand parrot endangered by introduced wasps and mammals. Biological Conservation 56: 2338.

Best, H.A.; Harrison, M. 1976. An assessment of the proposed biological reserves in the West Coast beech project area with respect to the avifauna of the region. Fauna Survey Unit report No. 7, 11 p. plus appendices.

Best, H.A.; Ogle, C.C. 1979. Birdlife of the Western Southland beech project area. Fauna Survey Unit report No. 16, 10 p. plus appendices.

Buckingham, R.P. 1996a. South Island kokako: further evidence of remnant populations in the South Island and on Stewart Island: A discussion document prepared for the Department of Conservation. Unpublished Wildlife Surveys No. 3 report. 24 pp.

Buckingham, R.P. 1996b. A search for South Island Kokako Callaeas cinerea cinerea in Granville State Forest and Beech Working Circle 2, North Westland, during November and December 1995. Wildlife Surveys report No. 4 for Timberlands West Coast Limited. 14 pp.

Buckingham, R.P. 1997. WWF-NZ final report: search for South Island kokako 1997. Unpublished Wildlife Surveys report. 24 pp.

Buckingham, R.P. 1998. Bird and bat survey of Maruia State Forest, Murchison District, New Zealand, 19961997. Wildlife Surveys contract report No. 7 for Timberlands West Coast New Zealand Ltd., Greymouth. 106 pp plus maps.

Buckingham, R.P. 1999a. A survey for birds and bats in Inangahua Working Circle (Beech Working Circle 2). Wildlife Surveys contract report No. 25/2 for Timberlands West Coast New Zealand Limited, Greymouth. 146 pp plus maps.

Buckingham, R.P. 1999b. Long term bird monitoring programme: proposed design and review of techniques. Wildlife Surveys contract report No. 26/1 for Timberlands West Coast Ltd Greymouth. 26 pp.

Buckingham, R.P.; Nilsson, R.J. 1994a. A survey of birds within Beech Working Circle 3, comprising parts of Granville SF, Hukawai SF and Tawhai SF, North Westland. Unpublished Contract Report for Pine Plan NZ Ltd, Dunedin. 96 pp. plus Appendix 7 (maps) (separate folder).

Buckingham, R.P.; Nilsson, R.J. 1994b. Seasonal variations in bird counts and a search for yellowheads and bats in Beech Working Circle 3, mid Grey Valley, North Westland. Unpublished Contract Report for Pine Plan NZ Ltd, Dunedin. 27 p.

Buckingham, R.P.; Brown, K.P. 1996. Bird and bat survey of Beech Working Circle 2, eastern Paparoa Range, North Westland, New Zealand. Wildlife Surveys contract report No. 2 for Pine Plan New Zealand Ltd., Dunedin. 76 p.

Butler, D.; McLennan, J. 1991. Kiwi recovery plan. Threatened species recovery plan series No. 2. Department of Conservation, Wellington. 35 pp.

Clout, M.N.; Karl, B.J.; Pierce, R.J.; Robertson, H.A. 1995. Breeding and survival of New Zealand pigeons Hemiphaga novaeseelandiae. Ibis 137: 264271.

Coker, P.M. 1980. Wildlife values of the forested areas of the Hokonui and Catlins. Fauna Survey Unit report No. 23. 185 p.

Coker, P.M.; Imboden, C. 1980. Wildlife values and wildlife conservation in South Westland. Fauna Survey Unit report No. 21. 84 pp. plus appendices.

Crook, I.G.; Best, H.A.; Harrison, M. 1977. A survey of the native bird fauna of forests in the proposed beech project area of North Westland. Proceedings of the New Zealand Ecological Society 24: 113126.

Dawson, D.G. 1981. Counting birds for a relative measure (index) of density. Pp. 1216 in C.J. Ralph & J.M. Scott (Eds). Estimating numbers of terrestrial birds. Studies in avian biology No. 6: 630 pp.

Dawson, D.G.; Bull, P.C. 1975. Counting birds in New Zealand forests. Notornis 22: 101109.

Dawson, D.G.; Dilks, P.J.; Gaze, P.D.; McBurney, J.G.R.; Wilson, P.R. 1978. Seasonal differences in bird counts in forests near Reefton, South Island New Zealand. Notornis 25: 257278.

DoC 1999. Restoration resumé in Rare Bits. The newsletter about threatened species work. Department of Conservation, Wellington. p. 25.

Don, G.L. 1993. Macraes Mining Company Ltd: the ecology of Globe Hill, Reefton; a baseline survey. Unpublished report, Bioresearches Ltd., Auckland, for Macraes Mining Company Ltd., Reefton. 435 pp.

Elliott, G.P. 1996. Productivity and mortality of mohua (Mohua ochrocephala). New Zealand journal of zoology 23: 229237.

Elliott, G.P.; Dilks, P.D.; O'Donnell, C.F.J. 1996a. The ecology of yellow crowned parakeets (Cyanoramphus auriceps) in Nothofagus forest in Fiordland, New Zealand. New Zealand journal of zoology 23: 249265.

Elliott, G.P.; Ogle, C.C. 1985. Wildlife and wildlife habitat values of Waitutu Forest, Western Southland. Fauna Survey Unit report No. 39. 108 p.

Heather, B.D.; Robertson, H.A. 1996. 'The field guide to the birds of New Zealand' Penguin Books (NZ) Ltd, Auckland. 432 pp.

Imboden, C. 1978. The valuation of wildlife habitats. Wildlife: a review. 9. NZ Wildlife Service, Department of Internal Affairs, Wellington.

Innes, J.G.; Hay, J.R. 1991. The interactions of New Zealand forest birds with introduced fauna. Pp 2523-2533 in J.L. Craig; G.S. Dumbell (Conveners) Integrating New Zealand Conservation. Acta XX Congressus Internationalis Ornithologici. Symposium No. 48.

Innes, J.; Hay, R.; Flux, I.; Bradfield, P.; Speed, H.; Jansen, P. 1999. Successful recovery of North Island kokako Callaeas cinerea wilsoni populations, by adaptive management. Biological conservation 87: 201214.

Innes, J.; Owen, K.; Smale, S. 1988. Proceedings of the kokako workshop; FRI, Rotorua, June 79, 1988. Department of Conservation and Ministry of Forestry. 30 pp.

Kingett Mitchell 1997. Great spotted kiwi survey of the upper Waimangaroa and surrounds. Prepared for Solid Energy International. 14 pp. plus appendices.

Leathwick, J.R.; Hay, J.R.; Fitzgerald, A.E. 1983. The influence of browsing by introduced mammals on the decline of North Island kokako. New Zealand journal of ecology 6: 5570.

McLennan, J.A.; Potter, M.A.; Robertson, H.A.; Wake, G.C.; Colbourne, R.; Dew, L.; Joyce, L.; McCann, A.J.; Miles, J.; Miller, P.J.; Reid, J. 1996. Role of predation in the decline of kiwi, Apteryx spp., in New Zealand. New Zealand journal of ecology 20(1): 27-35.

Manly, B.F.J.; Brown, J. 1999. Workshopstatistical methods for environmental management, Christchurch 1011 June 1999. Unpublished report (pages unnumbered).

van Mierlo, R. 1997. South Island kokako survey Grey Valley: 30/10/96-6/12/96. Unpublished report, Department of Conservation, Hokitika. 6 pp.

van Mierlo, R. 1998. West Coast skink (Oligosoma sp.) survey: September 1997May 1998. Unpublished report, West Coast Conservancy, Department of Conservation, Hokitika.

Miller, C., Affeld, K., and Adams, L. 1999. West Coast lizard survey: taxonomy, distribution and habitat requirements. Unpublished report, West Coast Conservancy, Department of Conservation, Hokitika. 27 pp.

Moller, H.;Tilley, J.A.V.; Thomas, B.W.; Gaze, P.D. 1991. Effects of introduced social wasps on the standing crop of honeydew in New Zealand beech forests. New Zealand journal of zoology 18: 171179.

Molloy, J; Davis, A. 1994. 'Setting priorities for the conservation of New Zealand's threatened plants and animals'. (second edition collated by Tisdall, C.) Department of Conservation, Wellington. 64 p.

Morse, P.M. 1981. Wildlife values and wildlife conservation of Buller and North Westland. Fauna Survey Unit report No. 29. 185 p.

NZWS, 1986. Habitats, reserves and corridors for wildlife in North Westland. Unpublished report by New Zealand Wildlife Service.

O'Donnell, C.F.J. 1991. Application of the wildlife corridors concept to temperate rainforest sites, North Westland, New Zealand. Pp 85-98 inD.A. Saunders & R.J. Hobbs (eds.) 'Nature Conservation 2: the role of corridors'. Surrey Beatty & Sons, Chipping Norton, NSW.

O'Donnell, C.F.J. 1996a. Predators and the decline of New Zealand forest birds: an introduction to the hole-nesting bird and predator programme. New Zealand journal of zoology 23: 213219.

O'Donnell, C.F.J. 1996b. Monitoring mohua (yellowhead) populations in the South Island, New Zealand, 198393. New Zealand journal of zoology 23: 221228.

O'Donnell, C.F.J. 1997. How cryptic can a bat be? Rare bits. No. 25.

O'Donnell, C.F.J.; Dilks, P.J. 1986. Forest birds in South Westland: status, distribution and habitat use. New Zealand Wildlife Service occasional publication No. 10. 179 p.

O'Donnell, C.F.J.; Dilks, P.J. 1987. Preliminary modelling of impacts of logging on forest birds in South Westland. Science and Research Internal Report
No. 1, (unpublished), Department of Conservation, Wellington. 27 p.

O'Donnell, C.F.J.; Dilks, P.J.; Elliott, G.P. 1996. Control of a stoat (Mustela erminea) population irruption to enhance mohua (yellowhead) (Mohoua ochrocephala) breeding success in New Zealand. New Zealand journal of zoology 23: 279286.

O'Donnell, C.F.J.; Phillipson, S.M. 1996. Predicting the incidence of mohua predation from the seedfall, mouse, and predator fluctuations in beech forests. New Zealand journal of zoology 23: 287293.

O'Donnell, C.F.J.; Sedgeley, J. 1994. An automatic monitoring system for recording bat activity. Department of Conservation technical series No. 5. 16 p.

Oliver, W.R.B. 1955. 'New Zealand birds'. (2nd Ed.) A.H. & A.W. Reed, Wellington. 661 p.

Planning Tribunal. 1996. Report and recommendations of the inquiry into the national water conservation order for the Buller River, 1996.

Sedgeley, J; O'Donnell, C.F.J. 1999. Roost selection by the long-tailed bat, Chalinolobus tuberculatus, in temperate New Zealand rainforest and its implications for the conservation of bats in managed forests. Biological conservation 88: 261276.

Sim, J; Saunders, A. (Eds) 1997. 'Predator workshop 1997' Proceedings of a workshop held 2124 April 1997, St. Arnaud, Nelson Lakes. 135 pp.

Smith, W.W. 1888. On the birds of Lake Brunner district. Transactions of the New Zealand Institute 21: 205224.

Spurr, E.B. 1985. Effects of beech forest management on bird populations.
New Zealand journal of ecology 8: 145.

Studholme, B. 1999. Blue duck (whio)survey of the distribution and abundance of blue duck in Kahurangi National Park. Department of Conservation internal report No. 30. 59 pp.

Taylor, G.A. 1985. The effects of logging on forest bird communities on the Mamaku Plateau. Unpublished MSc thesis, University of Canterbury, Christchurch. 314 p.

Verner, J. 1981 Introductory remarks: sampling design. Pg. 391 in C.J. Ralph & J.M. Scott (Eds). Estimating numbers of terrestrial birds. Studies in avian biology No. 6: 630 pp.

Whitaker, A.H. 1997. Timberlands West Coast Limited: The herpetofauna of Beech Working Circles 1 and 3, Grey Valley and Maruia Valley, north Westland. Unpublished report, Timberlands West Coast Limited, Greymouth. iv+31 pp.

Whitaker, A.H. 1999. Timberlands West Coast Limited: The herpetofauna of the Inangahua Working Circle (BWC2), north Westland. Unpublished report (draft), Timberlands West Coast Limited, Greymouth. iv+19 pp.

Wilson, P.R.; Karl, B.J.; Toft, R.J.; Beggs, J.R.; Taylor, R.H. 1998. The role of introduced predators and competitors in the decline of kaka (Nestor meridionalis) populations in New Zealand. Biological conservation 83 (2): 175185.

Wilson, P.R.; Taylor, R.H.; Thomas, B.W. 1988. Effect of topography on seasonal distribution of forest birds in the Ohikanui, lower Buller and Inangahua valleys, North Westland. Notornis 35: 217243.