Plant communities, landforms, and soils of a geomorphically active drainage basin, Southern Alps, New Zealand PDF

Preview

Summary

This article was downloaded by: [Lincoln University Library] On: 19 November 2013, At: 14:01 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK New Zealand Journal of Bota...

Description

This article was downloaded by: [Lincoln University Library] On: 19 November 2013, At: 14:01 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

New Zealand Journal of Botany Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tnzb20

Plant communities, landforms, and soils of a geomorphically active drainage basin, Southern Alps, New Zealand a

G. H. Stewart & J. B. J. Harrison

b

a

Forestry Research Centre , Forest Research Institute , P.O. Box 31-011, Christchurch, New Zealand b

Department of Geology , The University of New Mexico , 87131, Albuquerque, New Mexico Published online: 05 Dec 2011.

To cite this article: G. H. Stewart & J. B. J. Harrison (1987) Plant communities, landforms, and soils of a geomorphically active drainage basin, Southern Alps, New Zealand, New Zealand Journal of Botany, 25:3, 385-399, DOI: 10.1080/0028825X.1987.10413356 To link to this article: http://dx.doi.org/10.1080/0028825X.1987.10413356

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

New Zealand Journal of Botany. 1987. Vol. 25: 385-399 0028-825X/87/2503-0385$2.50/0 © Crown copyright 1987

385

Plant communities, landforms, and soils of a geomorphically active drainage basin, Southern Alps, New Zealand

Downloaded by [Lincoln University Library] at 14:01 19 November 2013

G. H. STEWART J. B. J. HARRISON* Forestry Research Centre, Forest Research Institute P.O. Box 31-011, Christchurch, New Zealand

Abstract The plant commumtles in a mountainous drainage basin in the Southern Alps, Westland, New Zealand, were classified from 92 vegetation descriptions. Nine forest and shrubland communities were defined from the presence/absence of 264 species and from differences in canopy cover. Their composition and distribution, inferred from ordination and descriptions oflandforms and soils, were determined primarily by the influe~ces of mass movement disturbance and elevatIOn. Erosional and depositional landform units of varying age and surface stability contained soils of.~if­ ferent stages of development. Seral commumtles occupied well-drained and often frequently disturbed recent soils. Mature forest/shrubland dominated stable landforms characterised by yellowbrown earths or gley podzol soils. Although specific relationships of individual species with factors such as soil nutrient status are consistent with other studies in Westland, landform age and surface stability, soil depth, soil drainage, and other physical disturbances, such as treefalls, appear to be equally critical determinants.

Keywords forest communities; landforms; soils; floristics; ordination; mass movements; soil development

INTRODUCTION Mass movement is an important form of disturbance in the western frontal ranges of the Southern Alps because of high rates of erosion and geological activity (O'Loughlin & Pearce 1982). A variety of landforms result from variations in the intensity of *Present address: Department of Geology. The University of New Mexico. Albuquerque, New Mexico 87131, USA. Received 3 July 1986; accepted 6 October 1986

erosional and depositional processes associated with mass movement. The distribution of soils, their stage of development on various landforms, and their associated plant communities often reflect the geomorphic history of such disturbances. The close relationship between vegetation communities, landforms, and soils in Westland has been noted by several authors (Morris 1959; Kennedy 1961; Chavasse 1962). Soil processes have been studied on several sequences in north and central Westland (Stevens 1968; Mew 1975, 1980; Stevens & Walker 1970; Tan 1971; Mew et al. 1975; Ross et al. 1977; Smith & Lee 1984; Sowden 1986), and Wardle (1977, 1980) gave an account of vegetation change on moraine sequences in Westland National Park. Recently, Basher et al. (1985) described soil development and vegetation change in low forests, shrublands, and grasslands in a subalpine basin in Westland. They found that soil development was rapid. Alluvium and colluvium were transformed into recent soils within 150 years and to gley podzols by 2500 years (Basher 1986). Few soils were more than 1000 years old, which suggested a long history of episodic slope instability. Despite low nutrient status, revegetatlon was rapid, with seral communities covering disturbed sites within 50 years and a species composition similar to that on uneroded areas being reached in 500-1000 years. Most of these studies have shown that the morphological and chemical properties of soils reflect a trend of rapid soil development with time (Stevens & Walker 1970; Smith & Lee 1984; Sowden 1986). Although trends in plant succession and soil development have been reasonably well defined for several forest types of the lowlands, apart from Basher (1986) few studies have examined in detail the interactions of vegetation, landforms, and SOlIs in steepland forested areas of Westland. This study describes (1) the plant communities, and (2) their distribution in relation to landforms and soils in a small drainage basin in the western frontal ranges of the Southern Alps. Particular emphasis is placed on mid-elevation forests (600-900 m) containing rata (Metrosideros umbellata*) and kamahi (Wemmannia racemosa).

*Nomenc1ature follows Allan (1961), Moore & Edgar (1970), and Edgar (1973), unless otherwise indicated.

New Zealand Journal of Botany, 1987, Vol. 25

386 STUDY AREA

(1) Tree tier: Emergent trees, canopy trees, and

Camp Creek is a 6 km drainage basin in the Alexander Range (42°40' S, 171 °38' E) near Lake Brunner, central Westland. It lies on the south-eastern boundary of the Alpine Fault in a zone of predommantly biotite schist. High uplift rates and high rainfall (c. 6500 mm/yr, Forest Research Institute 1982) result in high rates of erosion and sediment yield, which, with the influence of former glacial advances, dominate the geomorphology of the area. Soils in the area have been mapped as upland and high country podzolised yellow-brown earths and podzols (New Zealand SOli Bureau 1968a). The altitude of the catchment extends from 150 m a.s.l. to 1800 m at its highest point. In this part of Westland, where beech (Nothofagus) species are rare or local, mixed conifer-broadleaved hardwood forest and shrubland vegetation extends to about 1400 m a.s.l. and is then replaced at higher elevations by alpine grasslands. Numbers of introduced deer and chamois have been high in the past but are now low. Moderate numbers of brush tail possums have been established in the catchment for at least 30 years.

Downloaded by [Lincoln University Library] at 14:01 19 November 2013

2

METHODS Landform classification and soil description A hierarchial classification was used to group landforms into increasingly smaller landscape units (Petersen 1981; Harrison 1985a; see Table I). Landform boundaries were drawn from air photos and field reconnaissance. Soil pits were located on representative slope positions within each unit. Ninety-two soil profiles were described following the termmology of FAO (1974) with the addition of the horizon subscript j, used to denote weak expression of a qualifying suffix (Canada Soil Survey Committee 1978). On the basis of horizonation and depth to bedrock, the described soils (Harrison 1985a) were classified according to the New Zealand Genetic classification (New Zealand Soil Bureau 1968b).* Vegetation sampling and analysis Vegetation descriptions were made in the area surrounding each soil pit. The area sampled (plot) varied from 100 to 400 m 2 to encompass areas of homogeneous vegetation larger than the minimal area (sensu Mueller-Dombois & Ellenberg 1974). The presence of all vascular species was recorded in four physiognomic tiers: *Detailed soil descriptions are held at the Forestry Research Centre, FRI, Box 31-011. Christchurch.

hanes in the main canopy. Where the upper forest tier was less than 6 m a tree tier was not recorded. (2) Subcanopy tree and shrub tier: Shrub, subcanopy tree, and hanes, usually 1-6 m tall, but sometimes up to 10 m. Where the forest was < 6 m, this tier formed the canopy. (3) Ground tier : Herbaceous species and woody seedlings < 1 m. (4) Epiphytes: Species growing on living and dead trees. Species of epiphytic origin but rooted in the soil were recorded in one of the three other tiers. The mean height and total cover of the tree, subcanopy/shrub, and ground tiers were visually estimated. Individual species in each of these tiers were assigned to cover classes based on abundance and cover (Braun-Blanquet 1964). Altitude, aspect, and slope were recorded at each sampling site. Vegetation plots were classified on their floristic compositlOn (i.e., species presence/absence), using Sorensen's "K" index of similarity (Sorensen 1948) and group average clustering strategy. Community types were identified on the cluster dendrograms at the 0.30 level of similarity. Communities were delineated by examination of the cluster dendrograms and 2-way species X plot tables (Allen & McLennan 1983; Hall & Allen 1985). The names of these communities reflect the species with high cover values in any tier, except for two seral communities (8 and 9) which were defined by elevation. Generally vegetatlOn >6 m tall is referred to as forest and that below 6 m as shrub land. Seral communities 40% frequent in one community and at least 20% more frequent than in any other, or > 20% frequent in one commumty and absent from all others. Species that occurred in >40% of samples in one communityand 40% frequent in several communities and at least 20% more frequent in all of these than in all other communities also indicated floristic relationships. In addition to the classification described, ordination was used to investigate within- and between community patterns. Detrended Correspondence Analysis (DECORANA, DCA) was used in the analysis of the 92 plot X 264 species data matrix

Downloaded by [Lincoln University Library] at 14:01 19 November 2013

Stewart & Harrison-Plant community determinants in a drainage basin

387

Two developmental sequences of soils were recognised: soils formed on debris regolith and soIls formed over bedrock (Harrison 1985a; Table 2). Another group were compound soil profiles which contained a buried soil. Shallow recent and yellowbrown earth soils over bedrock were encountered but pooled with similar soils on debris regolith for most aspects of this study (Table 2). Shallow gley podzol soils over bedrock were differentiated from gley podzols on debris regolith by the absence of a B horizon (cf. Basher 1986; Table 2). Soils were assigned to three groups on the basis of stage of development, drainage characteristics, and mferred nutrient status (Harrison 1985a, b). These groups were recent soils, yellow-brown earth soils, and podzolised and gleyed soils (Table 2). Although the geomorphic history of the drainage basin is reflected in the distnbution of landforms, current geomorphic activity can be inferred from the distribution of soils. Two examples of both erosional and depositional landforms have been chosen to illustrate these processes. Erosional landforms 1. Dissected sides lopes (upper valley) The dissected sides10pes of the upper valley are relatively stable and have a gully/spur topography (Table 1). The spurs make up the largest proportion of the landscape unit and are characterised by shallow yellow-brown earth and gley podzol soils over bedrock. Mass wasting is uncommon although debris avalanches occur and result in slow continual deposition in the gullies, as indicated by the

(Hill 1979; Hill & Gauch 1980). Samples and species are ordinated simultaneously so that the final output matrix places samples with similar composition, and species with similar distributlOn in the samples, adjacent to each other. The axes are rescaled so that the difference between sample scores results in a constant rate of species turnover (Gauch 1982). Differences in species composition along a DCA axis often reflects important environmental gradients that underlie community patterns. These gradients were investigated by using non-parametric Spearman rank correlation analysis for the site factors measured, and the location of the sItes along each of the two axes generated by the ordination. RESULTS Landforms and soils The drainage basin was separated into two zones: an upper valley, which contains a combination of erosional and depositional landforms; and a mid and lower valley zone, where an actively degrading stream channel causing high rates of erosion has led to a predominance of erosional landforms (Fig. la, Table 1). Landforms in the upper valley include a cirque, fans, debris cones, dissected sideslopes (ridges and gullies), interfluves (probably remnant terraces), and a large slump. The mid and lower valley is comprised of dissected sldeslopes, a small number of recent alluvial terraces adjacent to the stream channel, and interfluves.

Table 1 Landforms, landform elements, and soil classes, ranked in order of landform surface age and stability, Camp Creek, Westland. Physiographic Unit Upper Valley

Landform

Landform Element

cirque debris cones slump dissected sideslopes interfluve dissected fan dissected valley fioor remnant dissected sideslopes slump fan

Mid and lower Valley

alluvial terrace dissected sideslopes

Soil Class

r, r/r" r/r r, rlr r, rlr, r/yb r, r/r

head gully gully spur gully spur spur toe edge centre

r r

levee, tread and riser gully spur

~

recent, yb

~

yellow-brown earth. pod

~

podzolised. gp

~

yb yb yb

gp pod yb, shallow gp pod yb, gp gp gp, shallow gp shallow gp shallow gp pod yb, gp gp gp, peat

yb r, r/yb

interlluve • r

yb/gp yb yb

gley podzol.

yb, yb/yb yb

pod yb, gpo shallow gp pod yb, gp

388

New Zealand Journal of Botany, 1987, Vol. 2S

l

Downloaded by [Lincoln University Library] at 14:01 19 November 2013

.'"

",,

L ___

~--'~_l1I1D

Upper Valley

(b)

Chtonachlpa rub,. grassland,7 Chlonochloa Pllllen.

~~~~~~~~!r"

gr.ssland Ol•• rl. co/ansal shrub land

Chlonochloa pallens grassland

389

Stewart & Harrison-Plant community determinants in a drainage basin Table 2 The number of sOli pits examined m dIfferent sOli classes, Camp Creek, Westland. Soli Classification recent soils

recent

Downloaded by [Lincoln University Library] at 14:01 19 November 2013

recent/recent recent/yellow-brown earth yellowbrown earth soils

yellow-brown earth

podzolised and gleyed soils

podzolised yellow-brown earth gley podzol shallow gley podzol peat

Honzon Sequence* Ah, C (on debris regolith) Ah, R (over bedrock) Ah,C, Ahb, Cb Ah,C, Ahb, Bwb, Cb

Ah, Ah, yellow-brown earth/yellow-brown earth Ah, yellow-brown earth/gley podzol Ah,

Bw, Bw, Bw, Bw,

C (on debris regolith) R (over bedrock) (C), Ahb, Bwb, Cb (C), Ahb, Erb, Bsb, Cb

Ah, EJ, Bw, C o or Ah, Er, Bs, C (on debris regolith) o or Ah, Er, R (over bedrock) 0, Er, C

Number of Profiles 11 I

17

3

2 16 I

22

2

3 13 31 8 I

53

• Horizon sequence after Hamson (1985a), horizon nomenclature after FAO (\ 974) except subscript j which follows Canada SoIl Survey Committee (1978).

presence of a range of weakly developed, recent and compound recent soils. 2. Dissected sideslopes (mid and lower valley) The greater stability of spurs and spur summits was reflected by the presence of more strongly developed yellow-brown earths or gley podzols (Table 1). In contrast, recent soils, weakly developed yellowbrown earths, and compound soils were found in the gUllies. Depositional landforms 1. Fans (upper valley) A large area of coalesced fans on the true left of Camp Creek (Fig. la; landform 5) represents one of the oldest landforms in the valley. Several large streams deeply dissect this landform, and channel erosion products through to the main stream. The surface of the fan is thus relatively unaffected by recent erosion. Soils formed a catena in which drainage became increasingly impeded in topographically low areas away from the main stream channel. Soils close to the entrenched stream channel therefore had the least impeded drainage and had thin E horizons (Table 1, 2). Thickness of the o and E horizons gradually increased away from the stream channel, until at the toe of the fan this sequence ended with the development ofa peat with o horizons > 80 cm thick. Gley podzols thus predominated on this depositional landform.

2. Debris cones (upper valley) Frequent, episodic deposition on debris cones (Fig. la, landform 4) has resulted in the formation of recent/recent compound soils (Table 1). Soil depth and frequency of burial varied, but no discernible pattern of soil distribution was evident. Deposition had been sufficiently frequent across the surface to prevent the formation of yellow-brown earths. The plant communities Five community types were recognised from 92 vegetation descriptions (Table 3). Four contained a single community and were separated by many indicator and differential species (Table 3, 4, 5); mid elevation forest contained five distinct communities delineated by indicator and differential species, species groups, and canopy and subcanopy tree cover (Table 6).

A. Low elevation forest Kamahi-rata-Quzntinia forest (Community 1) Twenty indicator species separated this community from all others (Table 4). An additional two species occurred in both this community type and mid elevation forest, and four other species in low elevation, mid elevation forests, and mid elevation seral communities (Table 5). Carpodetus serratus

Fig. 1 opposite (a) Physiographic zones and landforms, Camp Creek. Landforms arc: (1) cirque, (2) dissected valley floor remnant, (3) upper valley sldeslopes, (4) debris cones, (5) fans, (6) dissected fan, (7) upper valley interfluve, (8) slump, (9) midvalley interfluve, (10) mid and lower valley sideslopes. (b) DistnbutlOn of forest, shrubland, and grassland co...