Marine reserve site selection along the Abel Tasman National Park coast New Zealand Consideration of subtidal rocky communities.код для вставкиСкачать
AQUATIC CONSERVATION: FRESHWATER AND MARINE ECOSYSTEMS, VOL. 4, 153-167 (1994) Marine reserve site selection along the Abel Tasman National Park coast, New Zealand: consideration of subtidal rocky communities ROBERT J. DAVIDSON and W. LINDSAY CHADDERTONt Department of Conservation, Private Bag 5, Nelson, New Zealand ABSTRACT 1. At present, marine reserves do not represent the full range of community types throughout New Zealand. 2. To assist with the placement of a marine reserve along the Abel Tasman National Park coast (northern South Island), dominant subtidal laminarian and fucoid algae, echinoids and herbivorous molluscs were quantitatively investigated. Results from 100 quadrats collected from 19 random transects at six selected sites showed that algae and grazer assemblages varied between granite and limestone substrata. 3. Granite had a high percentage cover of crustose coralline algae (mean 82%-9O%), a sublittoral fringe of brown macroalgae and no Ecklonia radiata or red foliose algae. Limestone sites were distinguished by a relatively low percentage cover of coralline algae (mean 13%) and high cover of foliose red algae and E. radiata (2%-36% cover and 0.2-13.9 stipes m-2, respectively). 4. On limestone, molluscs Turbo smaragdus and Cookia sulcata, and the echinoid Evechinus chloroticus were larger than those on granite. On limestone sites with little macroalgae, herbivore size was intermediate. Grazers were more abundant on granite than limestone (mean 34.6 m-2, and 10.8 m-2 respectively). 5. Differences in herbivore composition were recorded between granite substrata, while both algal and herbivore composition varied between limestone sites. 6. We suggest that a variety of environmental factors including substratum influence algal and herbivore assemblages along the Abel Tasman coast. 7. It is recommended that selection of a marine reserve site or sites along the coast of Abel Tasman National Park recognizes differences in community structure both between and within limestone and granite substrata. INTRODUCTION During a qualitative study to identify an area or areas warranting protection in a marine reserve along the Abel Tasman National Park coastline, Davidson (1 992) described contrasting algal communities between subtidal limestone and granite substrata. The author found that on granite, brown macroalgae were absent or most often restricted to a narrow sublittoral fringe, whereas a more continuous algal cover was recorded from localized areas of limestone. He also noted that many brown macroalgae, characteristic of other ?Present address: Department of Conservation, PO Box 3, Stewart Island, New Zealand CCC 1052-7613/94/0201 53-15 01994 by John Wiley & Sons, Ltd. Received 21 July 1993 Accepted 11 February 1994 154 R. J. DAVIDSON AND W. L. CHADDERTON exposed and semi-exposed parts of New Zealand, were absent or rare from the Abel Tasman coast. Furthermore, herbivores on limestone shores appeared larger and in some instances more common than those on granite (Davidson, 1992). Based on these qualitative observations, the author suggested both limestone and granite substrata be included in a marine reserve. There is little literature about the principles for, and the outcomes of, the design of marine reserves (Fairweather, 1990). Because marine reserves are open systems, the argument to include an area including rich and diverse communities which would provide refuges and growth centres (Cairns and Elliot, 1987) seems appropriate. In practice this may mean the inclusion of as many habitats and communities as practical within any marine reserve. This present quantitative study aimed to test the null hypothesis that macroalgae and dominant herbivore communities were not different on adjacent shallow reefs composed of different substrata along the Abel Tasman coast. Understanding ecology processes affecting populations is not possible until patterns of distribution and abundance are well documented (Andrew and Mapstone 1987; Underwood et al., 1991). Until habitat and associated community patterns are documented throughout New Zealand, a rationalization of where marine reserves should be placed is difficult. Most descriptions of shallow subtidal rocky communities in New Zealand are from northern locations (Gordon and Ballantine, 1976; Ayling, 1978; Ayling et al., 1981; Ballantine et al., 1973; Choat and Schiel, 1982; Grace, 1983; Schiel, 1984; Battershill, 1986; Berben and McCrone, 1988; Coffey and Grace, 1990; Hogan, 1991; Jones and Garrick, 1991; Brook and Carlin, 1992; Grange et al., 1992). It has been recognized, however, that these studies may have limited application throughout New Zealand. For example, Schiel (1990), in a wide geographically ranging study, concluded that patterns and densities of Evechinus chloroticus and associated barrens in subtidal rocky areas could not be applied to southern New Zealand. These findings suggest that existing marine reserves do not represent the range of marine community types present in New Zealand. There have been few quantitative subtidal rocky studies for southern coasts, and consideration of marine community relationships is therefore hampered. A second aim of the study was to compare results from rocky subtidal habitats and communities from northern South Island localities with other northern and southern New Zealand shores and temperate overseas examples. Many biological and physical factors influence community composition on subtidal rocky reefs in New Zealand (see Andrew, 1988; Creese, 1988; Schiel, 1988 for reviews) and overseas temperate reefs (Lawrence, 1975; Steneck, 1982; Dayton, 1985; Himmelman, 1986; Deysher and Dean, 1986a; Andrew and Underwood, 1989; Underwood and Kennelly, 1990; Underwood et al., 1991). Which environmental factors are important is difficult to evaluate because they may never be entirely independent of each other (Dayton, 1985). In addition, the influence of substratum has had little attention in the literature. In intertidal studies, Harlin and Lindberg (1977) reported that surface relief influenced colonization and species assemblages at Rhode Island, while Raimondi (1988) found substrate influenced survival of barnacles in the northern Gulf of California. Nienhuis (1969) and Hartog (1972) reported algal zones extended into higher elevations on surfaces with large grains (e.g. limestone). The present study discusses the possible influence of substratum on subtidal biota from the sheltered Abel Tasman coast. STUDY AREA The Abel Tasman National Park coastline is sheltered from large ocean swells within Tasman and Golden Bays. Wave action quickly subsides with a drop or change in wind direction. Sediment input from catchments outside the Abel Tasman coast, combined with regular sea breezes and large tides (4.7 m extreme high tide), maintain water clarity at consistently low levels (2-8 m horizontal distance). Water temperatures ranged between 10" and 22°C (Dix, 1970a). Rocky reefs may extend to a depth of 14 m and are bordered by gently 155 MARINE RESERVE SITE SELECTION, NEW ZEALAND I New Zealand I 170' A * I Golden 173" ... : : ' . ;.; .. . Abel Tasman 175" . ... ,. ......; ,:. ' ' "' ..:,.. . .. .\" National Tasman Bay Kaiteriteri ~ 6 41' . 0 500 m I Figure 1 . Location of sample sites along the Abel Tasman National Park coastline. 1, Taupo Point north-west (limestone); 2, Taupo Point north (limestone);3, Taupo Stack (limestone);4, Taupo Hill (granite); 5, Separation Point (granite); and 6, Kaiteriteri (granite). sloping soft sediment shores. These are composed primarily of broken shell and coarse sands. Rock substrata along the Abel Tasman coast are dominated by granite with < 1070 of rocky shores comprising limestone (Figure 1) (Davidson, 1992). Five sites were selected to minimize environmental variables other than rock type, while the sixth site (Kaiteriteri) (Figure 1) was selected to compare data on the echinoid E. chloroticus collected by Dix (1970a) in 1968-69. Sample sites included three limestone shores, two with considerable algal cover (Taupo Point north, Taupo Stack) and one largely devoid of macroalgae (Taupo Point northwest), and three granite shores, Taupo Hill, Kaiteriteri, and Separation Point (Figure 1). All limestone sites were located within 500 m of a granite site at Taupo Hill but were not contiguous, being separated by soft bottom shores (< 10 m depth) and beaches. All sites in the Taupo Hill area had an almost identical shore aspect (northerly exposed), topography (80"-90" slope) and depth range (0-6.1 m), thereby minimizing environmental variables. Kaiteriteri was located in a shallow, north facing semi-sheltered situation and represented a moderate shore (30"-50°)slope comprised of large boulders on a bedrock base. 156 R. J. DAVIDSON AND W.L. CHADDERTON METHODS A total of 19 random transects were investigated from six selected sites. The location of each transect was determined as the point directly below the diver’s point of entry. Each site was selected as being representative straight coastline, thereby excluding headlands and embayments. At each site, two divers sampled 2-4 transects extending from 6.1 m deep up to extreme low tide. All samples were collected within this shallow zone and no depth stratified sampling was attempted. Quadrats of 1 m2 were placed sequentially along each transect starting at the base of rocky outcrops. A total of 100 quadrats were collected, with between 6-25 quadrats collected at each site. Numbers of Evechinus chloroticus, dominant herbivorous molluscs (Turbo smaragdus, Cookia sulcata, Cellana spp., Cryptoconchus porosus, Trochus viridus, Eudoxochiton nobilis, Maurea punctulata), Ecklonia radiata stipes and percentage cover of coralline spp., foliose red algae, Carpophyllum maschalocarpum and other brown macroalgae within each quadrat were recorded. Prior to sampling, practice measures were made to ensure that estimates of percentage cover were similar (within a 15% target) for the two divers. The same two divers collected all data, from all six sites between 20-24 November 1991. All E. chloroticus, T. smaragdus and C. sulcata were collected from quadrats and measured using callipers immediately after each dive. Additional T. smaragdus and C. sulcata were collected using the same methodology. Maximum test diameter was measured for E. chloroticus, whereas, maximum length from the anterior shell edge to the most posterior part of the shell was measured for the two molluscs. STATISTICAL ANALYSIS To compare herbivore and algal assemblages among the six subtidal sites, all random transects were regarded separately. No quadrat data were excluded from the analysis. Results from each quadrat were averaged for each transect producing a matrix of 13 species x 19 transects. Average densities were classified using the Bray-Curtis dissimilarity index of group average clustering strategy (Clifford and Stephenson, 1975). This analysis progressively grouped transects with similar species composition and was graphically displayed. Representative species from cluster groups were defined using a pseudo F test (Stephenson et al., 1974; Stephenson and Dredge, 1976; Stephenson and Campbell, 1977). This test uses the mechanics of the F test because the outcomes relate closely to those obtained by visual scanning of the data, however, because the data are classified, the randomness and thus the legitimate use of the test is destroyed (Stephenson and Campbell, 1977). Hence the term pseudo F test and the use of ‘noticeable’ differences not ‘significant’ ones. Transformed values of species recordings were tested using two combinations of site-groups. Species were eliminated if the psuedo F value was less than the F value giving 0.01 probability; the species hence conform more rigorously than the 0.05 probability value assigned to site-groups in terminology in Stephenson et al. (1974). Differences in herbivore size between sites were tested by ANOVA (Statistix 3.1). RESULTS A total of 50 limestone and 50 granite quadrats were sampled from 19 transects at six sites along the Abel Tasman National Park (Figure 1). Classification of data (mean m-2) for each transect confirmed two distinct groups separated at the 0.45 level (Figure 2). These groups showed a positive relationship to substrate type labelled A (limestone) and B (granite) in Figure 2. Species conforming to these substrate types were arranged into groups (Table 1). Limestone transects were characterized by foliose red algae, C. sulcata, M . punctulata and other brown algae. All granite shores were characterized by a high percentage cover of crustose coralline algae, an absence of macroalgae apart from a sublittoral fringe, and an abundance of T. smaragdus, Cellana spp. and T. viridus (Table 2). Differences in species abundances between limestone and granite sites were often dramatic. For example, 894 individual T. srnaragdus (mean = 17.9, SE = 2.5) MARINE RESERVE SITE SELECTION. NEW ZEALAND 157 ___ T5 0. T3 T9 TS 0 J 05 Figure 2. Group average clusteringbased on five algal groups and eight herbivore species recorded from 100 quadrats along 19 random transects from six selected sites along the Abel Tasman coastline. A, limestone; B, granite; I, Taupo Point (north-west); 11, Taupo Stack; 111, Taupo Point (north); IV, other granite sites; and V, Separation Point. were recorded from granite, whereas only 44 individuals (mean = 1.O, SE = 0.3) were found on limestone; Celiuna spp. were common from granite transects but were rarely recorded on limestone shores; and foliose red algae and E. rudiatu were recorded from all limestone sites but were not recorded from granite (Table 2). Further classification of cluster analysis at a 0.2-0.3 level recognized five groups, labelled I-V on the dendrogram (Figure 2). Four of the five groups corresponded to sample sites. Only granite transects from Kaiteriteri and Taupo Hill were classified into a mixed group (Figure 2). This mixed group did not include transect data from the Separation Point granite site. Pseudo F tests showed the Kaiteriteri and Taupo Hill granite site were ‘notable’ for high densities of T. smarugdus and high percentage cover of crustose coralline algae (Table 3). Separation Point was represented by high densities of T. viridus, Cellana spp. and crustose coralline algae but was distinguished by the absence of T.smaragdus (Table 4). The Taupo Point (north-west) limestone site had a paucity of macroalgal cover (barren) (Lawrence, 1975) and was ‘notable’ for high densities of C. sulcata, while the Taupo Point (north) site had E. rudiutu which formed a canopy (Table 4). The Taupo Stack limestone site was distinguished by a high percentage cover of foliose red algae, which was also high at Taupo Point (north) (Tables 3,4). Although not distinguished in the results, all coralline algae below the sublittoral fringe on granite were crustose, while most coralline on limestone was erect (geniculate). Table I . Conforming species characterizing the major sample groups distinguished by cluster analysis. Values represent mean number m-2. Site groups Species group Species Granite Limestone A Cellana spp. Coralline algae Turbo smaragdus Trochus viridus 8.7 82 17.9 2.5 0.3 13 1.o 0.3 B Foliose red algae Cookia sulcata Mauria punctulata Other brown algae 0 0.3 23 3.5 0.06 1.0 0 0.7 158 R. J. DAVIDSON AND W. L. CHADDERTON Table 2. Density of animals and algae on granite and limestone sites along the Abel Tasman coastline. Total area for each group was 5 h 2 . ~~ Limestone (n = 50) Granite (n = 50) Number of individuals Mean number m-2 Number of individuals Mean number m-* 247 894 15 438 3 125 6 3 4.94 4 0.58 17.88k2.44 0.30k0.07 8.76 k 0.70 0.06 k 0.03 2.50k0.58 0.12k0.07 0.06 k 0.03 263 44 173 14 10 12 3 25 5.10 t0.71 1.OOkO.28 3.48 kO.55 0.28k0.17 0.20 f0.08 0.24 2 0.07 0.06 f0.06 1.00 k0.12 1758 34.6 k 2.37 544 10.82 f0.96 Coralline algaea Foliose red algaea Ecklonia radiata NA 0 0 82 5 2.4 NA NA 13t1.3 23 k 3.5 3.62 k 0.99 Carpophyllum maschalocarpuma Other brown algaea NA 4.6k 1.7 NA 3.3 NA 0.4 k 0.4 NA 0.7 k 0.3 Species Evechinus chloroticus Turbo smaragdus Cookia sulcata Cellana spp. Cryptoconchus porosus Trochus viridus Eudoxochiton nobilis Maurea punctulata Total herbivores 0 0 178 t 1.3 aPercentage cover m-’. Table 3. Conforming species characterizing the minor sample groups distinguished in cluster analysis. Values represent means m - 2 . Site groups Sample group Species Separation Point Other granite 0.67 0 0 0 90 10.99 0.25 0 0 ~~ I I1 111 IV V Taupo Stack Taupo Point (north) 2.43 0.17 0.66 6.30 13.82 0.50 2 2.66 16 0.38 0.72 _ _ _ _ _ _ _ ~~ Cookia sulcata Ecklonia radiata Foliose red algae Turbo smaragdus Coralline algae Trochus viridus Cellana spp. 11.49 20.63 82 1.65 7.92 Taupo Point (NW) 38 0.16 10 0.17 0.12 22 0.38 13 0.32 0 Bold characters show the highest values for each species group, Overall species densities varied considerably between and within granite and limestone groups (Tables 2, 4). E. chloroticus, for example, was present at all sites with little difference in densities for combined sites on limestone and granite (mean = 5.1, SE = 0.8; mean = 4.94, SE = 0.58 respectively). The highest densities of E. chloroticus were recorded for transects at Separation Point (mean = 9.5, granite) and Taupo Stack (mean= 8.4, limestone) (Table 4). Variability between sites was also recorded for T. smaragdus and T. viridus on granite and C.sulcata, T. smaragdus, foliose red algae, and E. radiata on limestone (Table 4). Total herbivore numbers recorded from granite were higher than on limestone (Tables 2, 4). Variation in total herbivore numbers was also apparent between limestone sites with the lowest numbers of herbivores associated with the highest percentage cover of E. radiata (Table 4). Size comparison of E. chloroticus between limestone sites with macroalgal cover showed no significant difference (F=1.47, p = 0.23). Individuals from these limestone sites were larger than those on granite (F=30.5,p<O.OOl) and those from the limestone site with no macroalgae (Taupo Point north-west) MARINE RESERVE SITE SELECTION, NEW ZEALAND 153 25Keitertterl 20. - Granite 15. N 89 II = 57.3 mm 10. 5. L Separation Point Granite E 6 - 66.2 mm 4 20 Taupo Hill Granite N = 129 X = 57.6 mm 10 z o s91 Taupo Point (north-west1 Limestone N = 113 I = 56.6 mm 15 10 20. Taupo Point lnonhl 16. Limestone 12 N = 128 X = 69.2 mm 8. 4. 30 Taupo Stack '/& Limestone ii = 67.1 mm 10 5 00 20 30 40 Teat 50diameter 60 70 lmml80 90 100 110 Figure 3. Size-frequency distributionsof Evechinus chlorotim from sample sites along the Abel Tasman coastline. Entire distribution in 5 mm intervals. 160 R . J. DAVIDSON AND W. L. CHADDERTON 30. Kaiteriteri rl Granite 25. N = 239 ii = 39.8mm r-l Taupo Hill 16. 14. 12. 12 10- Granite N = 140 8- x = 29.6 mm - .!7 . . . . . . . . . . . . . . . . . . . . 14. Taupo Point (north-west) 12- g Limestone 10. C J 8. E 6. U Y P = 42Omm ............. - l4 Taupo Point (northr 2. Limestone a. - x = 56.6mm 6. 4- m 30. Taupo Stack 25- Limestone 20. N = 220 15. X 0 10 20 30 40 50 = 44.3 mm 60 70 80 90 100 Length lmml Figure 4. Size-frequency distributions of Turbo smaragdus from sample sites along the Abel Tasman coastline. Entire distribution in 1 mm intervals. 161 MARINE RESERVE SITE SELECTION, NEW ZEALAND Table 4. Density of animals and algae for the minor sample groups along the Abel Tasman coastline. Values represent mean m-2. Granite Limestone Separation Point (n = 6) Other granite sites ( n = 44) Taupo Stack (n = 22) Evechinus chloroticus Turbo smaragdus Cookia sulcata Cellana spp. Cryptoconchus porosus Trochus viridus Eudoxochiton nobilis Maurea punctulata 9.50? 2.93 0 0.67 f0.33 11.50f2.33 0 11.OOk 1.73 0 0.33 f0.21 4.56k0.46 20.65f2.60 0.18 f0.07 8.3920.74 0.09f 0.04 0.8650.21 0.07 f0.04 0.05f0.05 8-39? 1.08 0.17?0.10 2.65 f0.74 0.13?0.09 0.17 k0.12 0.1720.10 0 0.43k0.19 3.00+0.98 0.25 f0.18 0.92f 0.53 0 0 0.17 f0.17 0 0.17 f0.17 2.19k0.67 2.69 f 0.71 6.38 f 0.91 0.6920.51 0.38f0.15 0.38 k0.13 0.19k0.19 0.56 ? 0.20 Total herbivores 33.00f 3.34 35.1922.69 12.13+ 1.47 4.83 f 1.46 13.43f 1.27 90k4 0 0 82f2 0 0 102 1.7 36k 5.4 0.17 2 0.08 14 k 2.8 2655.9 13.92f2.28 16 f 2.3 2 f0.8 0.63 f0.24 2.9+ 1.9 0.8 ? 0.4 1.721.7 1.1 k0.8 5f3 0.6 f 0.2 Species Coralline algaea Foliose red algaea Ecklonia radiata C. maschalocarpuma Other brown algaea 0 3.3k3.3 4f 1.8 0 Taupo Point Taupo Point (north) ( n = 12) (north-west) (n = 6) apercentage cover m-2. (F=40.8,p<O.OOl). Less than 2% of E. chloroticus from Taupo Point (north-west) and granite sites reached sizes >75mm diameter, whereas 30% of urchins from Taupo Point (north) and Taupo Stack measured between 75-110mm diameter (Figure 3, Table 4). Comparison of Dix’s (1970,1972)size and density data with data from the same site in the present study (Kaiteriteri) revealed no significant differences for either his 1967 (F=1.06,p = 0.30)or 1968 data (F=0.74,p = 0.39). The population size structure of T. smurugdus varied between all sites (Figure 4). The largest mean and individual sizes were found from limestone sites (Figure 4,Table 4). The lowest mean sizes of T. smurugdus were collected from Kaiteriteri. As the mean size of T. smarugdus decreased, a corresponding increase in animal density was recorded. The highest densities were recorded from sites with the smallest individuals (Figure 4, Table 4). The highest densities and largest C. sulcutu were recorded from limestone although population structure was extremely variable (Figure 5, Table 4). Insufficient numbers of C. sulcutu were available at Separation Point to provide reliable information. DISCUSSION Differences in herbivore and algal assemblages between limestone and granite shores along the Abel Tasman National Park coast were considerable. Granite shores had a barren appearance, with macroalgae most often restricted to a narrow sublittoral fringe. Granite shores had a high percentage cover of crustose coralline algae, high densities of herbivores and an absence of E. rudiutu. In contrast, limestone sites were characterized by the presence of foliose red algae and E. rudiutu. Differences between substrata were supported by qualitative information collected by Davidson (1992)who recorded more macroinvertebrate species from limestone (mean = 43.5,SE = 0.5)than for granite (mean = 30.8,SE = 1 S ) , while Nelson et al. (1992)listed 1 1 species of algae recorded on granite and not limestone and 14 species from limestone but not granite. These qualitative studies and the quantitative data presented in this study suggest that selection of a marine reserve site with the goal of having a variety of community types should include both limestone and granite substrata. Results suggested that although granite shores appeared uniformly barren, they did in fact differ in herbivore composition. Similarly, differences in algal taxa and cover and herbivore composition between 162 R. J. DAVIDSON AND W. L. CHADDERTON Kaiteriteri Granite N = 147 I = 21.2 mm 1 .................... Taupo Hill Granite N = 135 P = 25.2 mm E . . . r . . . . . ... . . . . . .< 25 > Taupo Point (north-west) 20. Limestone 15. N = 99 U 3 ?! U 1 I = 29.5 mm 10. 5. 14. Taupo Point (north) 2 ' Limestone 8. - x = 34.7 mm Taupo Stack Limestone n n 111 111 Lengrh (mm) Figure 5. Size-frequency distributionsof Cookiu sulcutu from sample sites along the Abel Tasman coastline. Entire distribution in 1 mm intervals. MARINE RESERVE SITE SELECTION, NEW ZEALAND 163 limestone sites were considerable. Limestone at Taupo Point (north) was characterized by high numbers of brown macroalgae (E. radiata) and an understorey of foliose red algae, while limestone at Taupo Point (north-west) was distinguished by a virtual absence of macroalgae. Limestone at Taupo Stack was intermediate, with moderate densities of brown macroalgae and high percentage cover of foliose red algae. These differences were supported by cluster analysis. The variation in community composition within each substrate type should also be considered in marine reserve site selection. Reasons for differences in community structure between substrata is probably due to environmental variables. Heterogeneityof substrate surface is known to affect algae (Nienhuis, 1969; Hartog, 1972; Harlin and Lindberg, 1977; Schiel, 1980; Deysher and Norton, 1982; Dayton, 1985). Deysher and Norton (1982) documented greater attachment success from Sargassum muticum when rock surfaces had small grooves and pits, as opposed to smooth surfaces. Harlin and Lindberg (1977) reported that relief of substrate surface could regulate the development of an intertidal algal community. Further, they concluded that many macroalgae preferred to colonize surfaces with greater relief. Harlin and Lindberg (1977) suggested several reasons for this phenomenon related to larger surface space for settlement, reduced predation due to refugia, trapping of detritus, and surface chemistry. In California, Raimondi (1988) recorded greater postsettlement mortality of barnacles Chthamalusanisopoma on intertidal basalt than on granite near the upper limit of its distribution resulting in a 25 cm difference in the vertical limit up the shore. The mechanism responsible for this difference appeared related to the thermal properties of the two rock types. Unfortunately, there is little published information on the influence of substrate on in situ subtidal algal communities. Granite of the Abel Tasman coast represents a uniform and relatively featureless surface dominated by large boulders and bedrock with few cracks or crevices, whereas limestone reefs exhibit variable relief with many cracks, pits, crevices and fissures. Quantitative data from the present study suggest that surface heterogeneity may be an important environmental factor contributing to differences in herbivore and algal composition. The macroalgal floras of granite shores along the Abel Tasman were depauperate. C. maschalocarpum was restricted to an occasional sublittoral band, while E. radiata and C. frexuosum were extremely patchy or rare (Davidson, 1992). Instead granite was dominated by crustose coralline algae. Subtidal rocky barrens have been reported in many regions (see reviews: Andrew, 1988; Schiel, 1988, Underwood et al., 1991). Underwood et al. (1991) suggested that stability of barrens seemed more assured than habitats dominated by foliose brown algae or geniculate corallines, while Johnson and Mann (1986) suggested that on subtidal granite in Nova Scotia barrens stability contributed significantly to inhibition of algal settlement. Davidson (1992) examined aerial photos from the Abel Tasman coast taken 25 years prior to the present study, and reported no evidence of macroalgal beds where barrens existed in the present study. The results suggest that granite barrens of the Abel Tasman are relatively stable. Barrens stability may be further ensured by environmental factors which are unsuitable for macroalgae. High turbidity, sedimentation, low light, high temperatures, low nutrients and low water motion do not favour most macroalgal species (Charters et al., 1973; Jackson, 1977; Devinney and Volse, 1978; Kain, 1979; Norton et al., 1982; Kennelly, 1983; Reed and Foster, 1984; Dayton, 1985; Deysher and Dean 1986b) and may explain the absence of many brown algae common in southern New Zealand (e.g. Macrocystis pyrifera, Marginariella spp., Landsbergia quercifolia, Lessonia variegata, Durvillaea spp .). Unfavourable environmental conditions for algae are compounded by the granite relief which appears to favour herbivores. Many authors have attributed barrens establishment and maintenance to grazing herbivores, primarily sea urchins, (Lawrence, 1975; Ayling, 1981; Choat and Schiel, 1982; Andrew and Choat, 1985; Himmelman and Lavergne, 1985; Himmelman, 1986; Fletcher, 1987; Andrew, 1988; Schiel, 1988; Andrew and Underwood, 1989; Jones and Andrew, 1990; Keats et al., 1990; Underwood and Kennelly, 1990; Underwood et al., 1991). Subtidal studies on molluscan grazers suggest they may also influence algal community composition (see review Creese, 1988). Fletcher (1987) reported that limpets were second only to urchins in the maintenance of crustose coralline barrens. Other authors have recorded an increase in foliose algal species with removal of limpets (Ayling, 1981; Moreno and Sutherland, 1982; Schiel, 1980) and a decline 164 R. J. DAVIDSON AND W. L. CHADDERTON or smothering of crustose corallines (Paine, 1980; Steneck, 1982; Himmelman et ul., 1983). In the present study, the highest combined densities of grazers were recorded for granite barrens. Densities of T. smurugdus and Cellana spp., similar to values recorded by Ayling (1981), Moreno and Sutherland (1982) and Schiel (1990), were present on barren granite reefs; suggesting these molluscs may also help to maintain crustose coralline algae on the Abel Tasman coastline. The results suggest that for Abel Tasman granite, the presence of large numbers of E. chloroticus and molluscan herbivores living on a low relief substrate, representing an ideal grazing surface, is the most probable explanation for the maintenance of extensive barrens. The selection of a marine reserve area should take account of biogeographic differences throughout New Zealand as well as local community differences. Herbivore communities from the Abel Tasman, for example, contrast with those reported from North Island studies. In northern New Zealand, T. smaragdus is usually considered an intertidal species and is relatively rare below low water (Creese, 1988). T. smurugdus was the dominant gastropod from granite sites on the Abel Tasman with counts up to 64 individuals m-2. Of the common subtidal herbivorous molluscs of northern New Zealand (Cellana stellifera, Cantharidus purpureus, Trochus viridus, Micrelenchus sanguineus, Cookiu sulcuta) only Cellana spp., with the exception of one Abel Tasman site, were recorded in comparable numbers (Choat and Schiel, 1982). This evidence suggests that although overall densities of herbivorous molluscs are comparable to values reported in northern New Zealand studies, different species contribute to their relative abundance for sheltered waters of the Abel Tasman. In northern New Zealand, the highest densities of E. chloroticus are recorded from coralline flats, rather than underneath macroalgal canopies (Ayling, 1978; Choat and Schiel, 1982; Andrew and Choat, 1985). On the Abel Tasman coast, E. chloroticus abundances on crustose coralline dominated granite rock were comparable to reported North Island densities. Similarly, lowest E. chloroticus densities on limestone were recorded underneath algal canopies, however, in contrast with northern New Zealand studies, urchin numbers on limestone with sparse E. rudiata but a high percentage cover of foliose red algae, were high. There is growing circumstantial evidence indicating that herbivorous grazers benefit from the close proximity to macroalgal beds. Himmelman and Nedelec (1990) recorded the largest urchins living adjacent to the algal fringe in intensely grazed areas in eastern Canada. In California, Rowley (1990) found urchins Strongylocentrotus purpuratus of > 1.2 mm diameter fed on foliose algae and as a consequence grew 6-7 times faster than urchins feeding on barrens. Dix (1969), at Kaikoura and McShane and Naylor (1991) in Fiordland (south-west New Zealand), recorded dense aggregations of E. chloroticus in small patches devoid of macroalgae, amongst beds of dense seaweeds. From tagging trials, Dix (1969, 1970b) concluded that movement of urchins within these patches or aggregations at Kaikoura was small and individuals appeared to form and occupy shallow depressions or bare spots. Size data for E. chloroticus, T. smurugdus and C. sulcuta from the Abel Tasman support the suggestion that herbivores benefit from this association. The largest individuals were recorded from limestone sites in the present study with a high percentage cover of algae, while smaller sizes were recorded from limestone sites with a low percentage cover of macroalgae. Further, preliminary results suggest that in some areas urchins move little, maintaining small patches of rock grazed clean of macroalgae. This localized action eventually results in the formation of depressions in the limestone. On granite along the Abel Tasman, the sizes of E. chloroticus suggest that urchin growth is limited, as considerably larger individuals were recorded from nearby limestone. This hypothesis needs testing for the Abel Tasman coast. In a review, Andrew (1988) suggested that E. chloroticus could maintain high densities in areas with low food availability through adjustments to growth and reproductive output. Dix (1970a, 1972) proposed that urchins on granite at Kaiteriteri may be limited by food availability. Dix (1970a, 1972) compared E. chloroticus from granite shores at Kaiteriten with limestone sites of Kaikoura Peninsula (northeastern South Island). He suggested that the environmental conditions, particularly the impoverished macroalgal flora on granite of the Abel Tasman were responsible for the observed differences. Our data, collected from Abel Tasman shores with and without macroalgae, support this contention. MARINE RESERVE SITE SELECTION, NEW ZEALAND 165 CONCLUSION In recent years there has been an increased awareness of community complexities in subtidal rocky reefs. Mechanisms responsible for structuring these communities have been discussed in many papers, however, little attention has been focused on patterns of distribution and abundance. In New Zealand, most community description has been from northern North Island, but Schiel(1990) reported that as collection of quantitative data on algal and herbivore assemblages become more widespread, few generalizations remain appropriate. From limited quantitative data available from South Island subtidal rocky shores, it appears that not all patterns described for northern New Zealand are applicable. In addition, differences between sheltered shores of the Abel Tasman and exposed or semi-exposed locations in southern New Zealand also appear considerable. For marine reserves t o be representative (Ballantine, 1991), there must be greater emphasis placed on studies describing patterns and species distributions and abundance. It is also important that once biogeographic patterns have been identified, local variation in community composition be investigated especially in the identification of sites appropriate for marine reserves. Results in the present study suggest that subtidal communities recorded from the Abel Tasman coast are different from subtidal shores described to date elsewhere in New Zealand. On the grounds of protecting representative marine areas throughout New Zealand, the Abel Tasman coast therefore warrants protection as part of a network of marine reserves. Further, to achieve a variety of community types, selection of a marine reserve location along this coast should recognize the contrasting subtidal communities recorded both between and within granite and limestone substrata. ACKNOWLEDGEMENTS We thank Peter Braggins, Bill Franklin, Helen McAllen, Brendon Clough and Charmayne Devine for assistance with field work which was supported from the Nelson and Motueka Department of Conservation offices by Sheryl File and Wendy King. Thanks to Len Edwards for committing staff time and equipment to the study. Statistical analyses were implemented using a variety of computer programs designed or adapted by Dr J. D. Stark (Cawthron Institute, Nelson). We gratefully acknowledge comments by Chris Battershill, Ken Grange, Islay Marsden, Wendy Nelson, Keith Probert, Bob Rowley, K. Hiscock, Jacqui Davidson and anonymous referees. Maps and figures were prepared by Alan Price and Carry Holz. The manuscript was typed by Charmayne Devine. REFERENCES Andrew, N. L. 1988. ‘Ecological aspects of the common sea urchin, Evechinus chloroticus in northern New Zealand: a review’, New Zealand Journal of Marine and Freshwater Research, 22, 415-426. Andrew, N. L. and Choat, J. H. 1985. ‘Habitat related differences in the survivorship and growth of juvenile sea urchins’, Marine Ecology Progress Series, 21, 155-161. Andrew, N. L. and Mapstone, B. D. 1987. ‘Sampling and the description of spatial pattern in marine ecology’, Oceanography and Marine Biology Annual Reviews, 25, 39-89. Andrew, N. L. and Underwood, A. J. 1989. ‘Patterns of abundance in the sea urchin Centrostephanus rodgersii on the central coast of New South Wales, Australia’, Journal of Experimental Marine Biology and Ecology, 131,61-80. Ayling, A. M . 1978. Cape Rodney to Okakari Point Marine Reserve Survey, University of Auckland Marine Laboratory Report. 98p. Ayling, A. M. 1981. ‘The role of disturbance in temperate subtidal encrusting communities’, Ecology, 62, 830-847. Ayling, A. M., Cumming, A. and Ballantine, W. J. 1981. Map of Shore and Subtidal Habitats of the Cape Rodney to Okakari Point Marine Reserve, North Island, New Zealand, in three sheets, scale I:2000, Department of Lands and Survey, Wellington. Ballantine, W. J. 1991. ‘Marine reserves-the need for networks (personal view)’, New Zealand Journal of Marine and Freshwater Research, 25, 115-116. Ballantine, W. J., Grace, R. V. and Doak, W. T. 1973. Mimiwhangata Marine Report, Turbott and Halstead, for New Zealand Breweries. Battershill, C. N. 1986. The Marine Benthos of Caves, Archways and Vertical Walls of the Poor Knights Islands, Unpublished report to New Zealand Ministry of Agriculture and Fisheries, 179 p. 166 R. J. DAVIDSON AND W. L. CHADDERTON Berben, P. H. and McCrone, A. 1988. ‘The Mokohinau Islands: a marine survey’, University of Auckland, Leigh Laboratory Bulletin, 21, 77 p. Brook, F. J. and Carlin, G. L. F. 1992. Subtidal Benthic Zonation Sequences and Fish Faunas of Rocky Reefs in Bay of Islands, Northern New Zealand, Department of Conservation, Northland Conservancy Report, 70p. Cairns, D. K. and Elliot, R. D. 1987. ‘Oil spill impact assessment for seabirds: the role of refugia and growth centres’, Biological Conservation, 40, 1-9. Charters, A. C., Neushul, M. and Coon, D. 1973. ‘The effect of water motion on algal spore adhesion’, Limnology and Oceanography, 18, 884-896. Choat, J. H. and Schiel, D. R. 1982. ‘Patterns of distribution and abundance of large brown algae and invertebrate herbivores in subtidal regions of northern New Zealand’, Journal of Experimental Marine Ecology, 60, 129-162. Clifford, H.T. and Stephenson, W. 1975. An Introduction to Numerical Classification,Academic Press, New York, 229 p. Coffey, B. T. and Grace, R. V. 1990. ProposedMarine Reserve, Hahei: A Preliminary Assessment and Habitat Inventory, Prepared for Department of Conservation, Hamilton, by Brian T. Coffey and Associates Limited, Hamilton, 54 p. Creese, R. G. 1988. ‘Ecology of molluscan grazers and their interactions with marine algae in northeastern New Zealand: a review’, New Zealand Journal of Marine and Freshwater Research, 22, 427-444. Davidson, R. J. 1992. A Report on the Intertidal and Shallow Subtidal Ecology of the Abel Tasman National Park, Nelson, Occasional publication No. 4, Department of Conservation, Nelson/Marlborough Conservancy, 161 p. Dayton, P. K. 1985. ‘Ecology of kelp communities’, Annual Review of Ecology and Systematics, 16, 215-245. Devinny, J. S. and Volse, L. A. 1978. ‘Effects of sediments on the development of MacrocystispyriYeragametophytes’, Marine Biology, 48, 343-348. Deysher, L. E. and Dean, T. A. 1986a. ‘In situ recruitment of sporophytes of the giant kelp, Macrocystis pyrifera (L.) C. A. Agardh: effects of physical factors’, Journal of Experimental Marine Biology and Ecology, 103,41-63. Deysher, L. E. and Dean, T. A. 1986b. ‘Interactive effects of light and temperature on sporophyte production in the giant kelp Macrocystis pyrifera’, Marine Biology, 93, 17-20. Deysher, L. E. and Norton, T. A. 1982. ‘Dispersal and colonisation in Sargassum muticum (Yendo) Fensholt’, Journal of Experimental Marine Biology and Ecology, 56, 179-195. Dix, T. G. 1969. ‘Aggregating in the Echinoid Evechinus chloroticus (note)’, Pacific Science, 23(1), 123-124. Dix. T. G. 1970a. ‘Biology of Evechinus chloroticus (Echinoidea: Echinometridae) from different localities. 1. General’, New Zealand Journal of Marine and Freshwater Research, 4(2), 91-116. Dix, T. G. 1970b. ‘Biology of Evechinus chloroticus (Echinoidea: Echinometridae) from different localities. 2. Movement’, New Zealand Journal of Marine and Freshwater Research, 4, 267-277. Dix, T. G. 1972. ‘Biology of Evechinus chloroticus (Echinoidea: Echinometridae) from different localities. 4. Age, growth and size’, New Zealand Journal of Marine and Freshwater Research, 6, 48-68. Fairweather, P. G. 1990. ‘Ecological changes due to our use of the coast: research needs versus effort’, Proceedings of the Ecological Society of Australia, 16, 71-77. Fletcher, W. J. 1987. ‘Interactions among subtidal Australian sea urchins, gastropods and algae: effects of experimental removals’, Ecological Monographs, 57, 89- 109. Gordon, D. P. and Ballantine, W. J . 1976. ‘Cape Rodney to Okakari Point Marine Reserve: review of knowledge and bibliography to December 1976’, Tune, 22, supplement. 146 p. Grace, R. V. 1983. ‘Zonation of sublittoral rocky bottom marine life and its changes from the inner Hauraki Gulf, North-eastern New Zealand, Tune, 29, 97-108. Grange, K. R., Murdoch, R. C., Battershill, C. N. and Singleton, R. J. 1992. A Survey of the Marine Habitats and Communities of White Island, Bay of Plenty, Report prepared for Department of Conservation by New Zealand Oceanographic Institute, 43 p. Harlin, M. M. and Lindberg, J. M. 1977. ‘Selection of substrata by seaweeds: optimal surface relief‘, Marine Biology, 40, 33-40. Hartog, den. C. 1972. Substratum: plants-multicellular plants, In Kinne, 0. (Ed.), Marine Ecology, Vol. 1. Environmental Factors, Wiley-Interscience, London, 1277-1289. Himmelman, J . H. 1986. ‘Population biology of green sea urchins on rocky barrens’, Marine Ecology Progress Series, 33, 295-305. Himmelman, J. H., Cardinal, A. and Bourget, E. 1983. ‘Community development following removal of urchins, Strongylocentrotus droebachiensis, from the rocky subtidal zone of the St. Lawrence Estuary, eastern Canada’, Oecologia, 59, 21-39. Himmelman, J. H. and Lavergne, Y.1985. ‘Organisation of rocky subtidal communities in the St. Lawrence Estuary’, Nature Canada, 112, 143-154. Himmelman, J. H. and NEdElec, H. 1990. ‘Urchin foraging and algal survival strategies in intensely grazed communities in eastern Canada’, Canadian Journal of Fisheries and Aquatic Science, 47, 1011-1026. MARINE RESERVE SITE SELECTION, NEW ZEALAND 167 Hogan, K. 1991. Kaoura Marine Reserve Proposal. A Preliminary Intertidal, Subtidal and Site Survey, Department of Conservation, East Coast Conservancy, 56 p. Jackson, G. A. 1977. ‘Nutrients and production of the giant kelp, Macrocystis pyrifera, off southern California’, Limnology and Oceanography, 22, 979-995. Johnson, C. R. and Mann, K. H. 1986. ‘The crustose algae, Phymatolithion Foslie, inhibits the overgrowth of seaweeds without relying on herbivores’, Journal of Experimental Marine Biology and Ecology, 96, 127-146. Jones, A. J. and Garrick, A. 1991. Tuhua (Mayor Island) Marine Habitats: A Preliminary Intertidal and Subtidal Survey, Technical series report No. 4., Department of Conservation, Bay of Plenty Conservancy Report, 129 p. Jones, G. P. and Andrew, N. L. 1990. ‘Herbivory and patch dynamics on rocky reefs in temperate Australasia: the roles of fish and sea urchins’, Australian Journal of Ecology, 15, 505-520. Kain, J. M. 1979. ‘A view of the genus Laminaria’, Oceanography and Marine Biology Annual Review, 17, 101-161. Keats, D. W., South, G. R. and Steele, D. H. 1990. ‘Effect of an experimental reduction in grazing by green sea urchins on a benthic macroalgal community in eastern Newfoundland’, Marine Ecology Progress Series, 68, 181-193. Kennelly, S. J. 1983. ‘An experimental approach to the study of factors affecting colonisation in a sublittoral kelp forest’, Journal of Experimental Marine Biology and Ecology, 68, 251-276. Lawrence, J. M. 1975. ‘Onthe relationship between marine plants and sea urchins’, Oceanography and Marine Biology Annual Review, 13, 213-286. McShane, P. E. and Naylor, J. R. 1991. A Survey of Kina Populations (Evechinus chloroticus) in Dusky Sound and Chalky Inlet, South- Western New Zealand, New Zealand Fisheries Assessment Research Document 91/17. Moreno, C. A. and Sutherland, T. P. 1982. ‘Physical and biological processes in a Macrocystispyrifera community near Valdivia, Chile’, Oecologia, 55, 1-6. Nienhuis, P. H. 1969. ‘The significance of substratum for intertidal algal growth on the artificial rocky shore of the Netherlands’, Znternational Revue ges Hydrobiology, 54, 207-215. Nelson, W.A., Adams, N. M. and Fox, J. M. 1992. ‘Marine algae of the northern South Island’, National Museum of New Zealand Miscellaneous Series, 26, 79 p. Norton, T. A., Mathison, A. and Neushul, M. 1982. ‘A review of some aspects of form and function in seaweeds’, Botany Marine, 25, 501-510. Paine, R. T. 1980. ‘Food webs: linkage, interaction, strength and community infrastructure. The third Tansley lecture’, Journal of Animal Ecology, 49, 667-685. Raimondi, P. T. 1988, ‘Rock type affects settlement, recruitment, and zonation of the barnacle Chthamalusanisopoma Pilsbury’, Journal of Experimental Marine Biology and Ecology, 123, 253-267. Reed, D. and Foster, M. 1984. ‘The effects of canopy shading on algal recruitment and growth in giant kelp (Macrocystis pyrvera) forest’, Ecology, 65, 937-948. Rowley, R. J. 1990. ‘Newly settled sea urchins in a kelp bed and urchin barren ground: a comparison of growth and mortality’, Marine Ecology Progress Series, 62, 229-240. Schiel, D. R. 1980. A Demographic and Experimental Evaluation of Plant and Herbivore Interactions in Subtidal Algal Stands. Ph.D. thesis, University of Auckland. 166 p. Schiel, D. R. 1984. ‘Poor Knights marine reserve survey’, Leigh Laboratory Publication, 17, 93 p., University of Auckland. Schiel, D. R. 1988. ‘Algal interactions in shallow subtidal reefs in northern New Zealand: a review’, New Zealand Journal Qf Marine and Freshwater Research, 22, 281-489. Schiel, D. RI 1990. ‘Macroalgal assemblages in New Zealand: structure, interactions and demography’, Hydrobiologia, 192. 59-76. Stenedk, R. S. 1982. ‘A limpet-coralline alga association: adaptations and defenses between a selective herbivore and its prey’, Ecology, 63, 507-522. Stephenson, W. and Campbell, B. M. 1977. ‘The macrobenthos of Serpentine Creek’, Memoirs of the Queensland Museum, 18(1), 75-93. Stephenson, W. and Dredge M. C. L. 1976. ‘Numerical analysis of fish catches from Serpentine Creek’, Proceedings of the Royal Society of Queensland, 87, 33-43. Stephenson, W., Williams, W. T. and Cook, S. 1974. ‘The macrobenthos of soft bottoms in southern Moreton Bay’, Memoirs of the Queensland Museum, 17, 73-124. Underwood, A. J. and Kennelly, S. J. 1990. ‘Ecology of marine algae on rocky shores and subtidal reefs in temperate Australia’, Hydrobiologia, 192, 3-20. Underwood, A. J., Kingsford, M. J. and Andrew, N. L. 1991. ‘Patterns in shallow subtidal marine assemblages along the coast of New South Wales’, Australian Journal of Ecology, 6, 231-249.