Wendy a thompson and jessica allen

Yüklə 0,49 Mb.
Pdf görüntüsü
ölçüsü0,49 Mb.
  1   2   3   4

Flora and vegetation of greenstone formations of the

Yilgarn Craton: the northern Forrestania Greenstone Belt

(Mount Holland area)


Science Division, Department of Environment and Conservation,

PO Box 51, Wanneroo, Western Australia, 6946.

Email: wendyjo.thompson@gmail.com


A quadrat-based survey recorded the flora and associated environmental parameters in the northern portion of the

Forrestania Greenstone Belt. A total of 312 taxa were identified, representing 47 families and 121 genera. Twelve taxa

of conservation significance were recorded, including two declared rare flora (Banksia sphaerocarpa subsp. dolichostyla

and Eucalyptus steedmanii). Three new taxa were identified, including two shrubs, a Hibbertia sp. and Labichea rossii

(Priority 1), and a single grass species, Austrostipa sp. The two undescribed new taxa are proposed for conservation

listing. The 50 quadrats were classified into eight community types. Topography and edaphic factors were influential

in delineating community groups. Only five weed taxa were identified, despite the history of mining in the area. Only

a small portion of the Forrestania Greenstone Belt belongs to the conservation estate. Based on the survey findings,

expansion of the secure tenure is recommended to incorporate a more comprehensive representation of the vegetation

and landforms of the greenstone belt.

Keywords: classification, Coolgardie, florsitic diversity, mallee, vegetation patterns, ultramafics

Conservation Science W. Aust. 8 (3) : 277–294 (2013)

© The Government of Western Australia, 2013

85 km north east of Hyden (Fig. 1). The Forrestania

Greenstone Belt trends north to south, and is c. 100 km

in length and c. 15 km west to east. The focus of this

study is the northern 40 km of the belt. The latitudinal

and longitudinal boundaries of the study area are roughly

32° 00’ S, 32° 25’ S and 119° 40’ E, 119° 50’ E,

respectively. The land tenure for this greenstone belt,

located in the Yilgarn Shire, is unallocated crown land.

Land use history

The area around Hyden primarily supports agricultural

and pastoral activities (Chin et al. 1984). Mineral interests

have focused on the Forrestania Greenstone Belt since gold

was discovered in 1915 (Chin et al. 1984). Further

exploration of the belt led to the discovery of nickel

deposits, which were subsequently mined. At present,

Western Areas NL operates nickel mines in the central

portion of the belt.


The northern Forrestania Greenstone Belt sits in the south-

western portion of the Coolgardie Bioregion, which is

defined as having a semi-arid climate with hot summers

and mild winters (Thackway & Cresswell 1995). Rainfall

events occur throughout the year, with a distinct increase

in mean monthly rainfall during winter (May–August;

Bureau of Meteorology 2010). Average annual rainfall at


Historically, the greenstone belts of the Yilgarn Craton

have been the focus of pastoral settlement, agriculture and

mining. Recent studies of flora and vegetation have

highlighted their species richness (Gibson 2004a, 2004b;

Gibson & Lyons 2001; Gibson et al. 2010; Markey &

Dillon 2008a, 2008b, 2009; Meissner & Caruso 2008a,

2008b, 2008c; Meissner et al. 2009a, 2009b, 2009c) and

the high beta-diversity of banded ironstone ranges and

allied greenstone belts (Gibson et al. 2007; Gibson et al.

2012). This study is a continuation of the survey effort to

document the flora, vegetation communities and

associated environmental characteristics of the greenstone

belts in the Yilgarn Craton.


The northern portion of the Forrestania Greenstone Belt

is situated in the south-central portion of the Coolgardie

Bioregion, near the boundary of the Mallee Bioregion

(Interim Biogeographic Regionalisation of Australia—

IBRA; Thackway & Cresswell 1995) and is part of the

Roe Botanical District (Beard 1990). The belt is located

approximately 90 km south of Southern Cross and c.


WA Thompson & J Allen

Hyden (c. 85 km south-west of the study area) is 343.3

mm, based on records from 1928 to 2010, with the months

of June and December having the highest (50.6 mm) and

lowest (13.7 mm) mean monthly rainfall, respectively. The

average annual maximum is 24.8 °C and average annual

minimum is 9.8 °C for the Hyden area, based on records

from 1970–2009. The highest temperatures occur between

December and March, with mean maximum temperatures

exceeding 29 °C. The lowest daily minimum temperatures

occur between June and September, where mean minimum

temperatures are below 6 °C.


The geology of the northern portion of the Forrestania

Greenstone Belt has been mapped and described on the

Figure 1. Map showing the location of the northern Forrestania Greenstone Belt survey area, with major landforms and

landmarks indicated. The locations of the 50 permanent quadrats are marked by solid triangles (


Flora and vegetation of northern Forrestania Greenstone Belt


Hyden 1:250,000 geological sheet (Chin et al. 1984).

Locally, greenstone refers to the outcropping of ultramafic

and mafics associated with Archaean metavolcanic and

meta-sedimentary rock sequences, often occurring as

ranges positioned within vast areas of granitoid and gneiss

(Cole 1992; Cassidy et al. 2006). The Forrestania

Greenstone Belt is composed of low undulating hills and

ranges, surrounded by sandplains interspersed with

exposed duricrust (Chin et al. 1984). The northern portion

of the belt sits between 400 and 450 m above sea level,

with Mount Holland (477 m) and North Ironcap (454 m)

as prominent features in the landscape. The southern

boundary of the study area is the salt lake, Lake Cronin, a

designated class ‘A’ nature reserve.

Located in the central Archaean Yilgarn Craton, the

greenstone belt is part of a tectonically stable region that

occupies a substantial portion of Western Australia and

was formed between c. 3000 and 2600 Ma (Myers 1993;

Myers & Swagers 1997). The Forrestania Greenstone Belt

trends northwards for most of its length, with a north-

north-west trend at the southern extent. The belt is a

complex series of folds and faults along a northerly syncline

(Chin et al. 1984; Griffin 1990). Primary rock

composition is fine- and medium-grained mafic

amphibolites and metabasalts (Chin et al. 1984). Other

dominant regolith types within the study area include

banded chert, tremolite and talc schists and Tertiary laterite.

A segment of metamorphosed shale and siltstone occurs

in the western part of the belt. Adjacent to the greenstone

belt are Quaternary colluival deposits derived from the

greenstone and substantial areas of remnant Tertiary

sandplain, derived from the laterite. There are banded iron

formations (BIF) within the greenstone belt, although

not as substantial an element as other parts of the Yilgarn

Craton. Within the northern portion of the belt, Mount

Holland and North Ironcap are the dominant BIF features

in the landscape (Chin et al. 1984). The ultramafic rocks

of the greenstone belt are typically of economic interest,

with nickel deposits associated with these geologies;

however, there is limited surface exposure. Lacustrine and

alluvial/aeolian deposits present in the area are generally

associated with playa lakes (Chin et al. 1984).

The soils of the Forrestania area are primarily sands

over clay (Beard 1990). Beard (1981) described the

Forrestania Tableland, which includes the Forrestania

Greenstone Belt, as principally brown and grey-brown

calcareous earths. Soils from the Bounty gold deposit, in

the northern part of the study area, are predominantly

yellow-red brown clayey sands, with ferruginous gravel

and red to light brown clay loams (Lintern 2004).


The northern Forrestania Greenstone Belt is characterised

by mallee communities, with Eucalyptus eremophila  a

significant element (Beard 1990). Beard (1981) classified

the associated vegetation as the Forrestania System, part

of the Swan 1:1,000,000 map sheet. The Forrestania

System constitutes a mosaic of vegetation communities,

primarily driven by the geology of the landscape (Beard

1981). The primary basaltic greenstones support eucalypt

woodlands, where the communities of mallee, scrub heath

and thicket are generally associated with the granite and

quartzite geologies. Beard (1981) mapped the main

portion of the Forrestania Greenstone Belt as Eucalyptus

salmonophloia and E. longicornis on greenstone. The

northern portion, encompassing the study area, also

included  E. eremophila scrub with pockets of

heterogeneous scrub heath. Mount Holland supports a

dense thicket of Allocasuarina campestris and Calothamnus

quadrifidus (Beard 1981).

An extensive survey describing the flora and vegetation

of the southern portion of the Forrestania Greenstone Belt,

including Middle and South Ironcap, Diggers Rock and

Hatter Hill, recorded 342 native taxa and three weed

species (Gibson 2004a). Four vegetation community types

were identified, which were strongly influenced by

associated geology and edaphic factors. Vegetation

communities included species-rich shrublands and mallee

shrublands associated with BIF. Mallee shrublands and

Allocasuarina thickets on skeletal soils were found in

association with massive laterites. Colluvial deposits

supported Eucalyptus urna – E. salubris woodlands with

Melaleuca sp. understorey, with colluvial flats having

species-depauperate mallee communities with E.

calycogona and emergent E. salmonophloia (Gibson


With known mineral deposits in the area, a history of

mining and the high level of beta-diversity recorded

between greenstone belts (Gibson et al. 2012), this study

is a timely assessment of an area that has otherwise

remained undocumented for plant species richness. This

study aimed to record the floristic diversity, describe

vegetation patterns and examine environmental variables

associated with the northern section of the Forrestania

Greenstone Belt.


Between the 22 September and 7 October 2009, fifty

20 × 20 m permanent quadrats were established

throughout the northern portion of the Forrestania

Greenstone Belt. The quadrat locations were chosen using

an environmentally stratified but non-random method

because of the extensive mineral exploration and mining-

related disturbance in the area. The quadrats represented

the topographical, geological and geomorphological

variation across the length and breadth of the belt, and

captured the associated vegetation communities that

characterise the various geologies. The landscape positions

of the sites encompassed a broad topological sequence

from gentle hill crests to the colluvial deposits. Methods

used followed those of previous surveys on greenstone

belts in the Yilgarn Craton (e.g. Markey & Dillon 2008a,

2008b; Meissner & Caruso 2008a, 2008b, 2008c).

Quadrat locations were selected to represent areas subject

to minimal disturbance or modification, although the

whole region has been and still is the focus of mineral

exploration. Thus sites were avoided where evidence of


WA Thompson & J Allen

disturbance (e.g. clearing, exploration, mining) was


The quadrats were marked by four steel fence droppers,

and their locations recorded by a Garmin Map76 GPS.

Photographs were taken at a set distance of 5 m from

each corner. Site physical characteristics (landform, slope,

aspect, litter and bare ground cover, size of coarse

fragments, cover of surface rock fragments and bedrock,

soil colour and texture) were recorded as a series of

descriptive attributes and semi-quantitative scales as

defined by McDonald et al. (1998). Landform description

was based on topographical position (crest, upper slope,

mid-slope, lower slope or flat) and landform element type

(e.g. hillcrest, hillslope, breakaway; McDonald et al. 1998).

Coarse fragments and rock outcrop data were recorded as

specific geologies present and as part of a seven-point class

scale representing percent (%) cover. The seven cover

classes were: zero % cover (0); <2% cover (1); 2–10 %

(2); 10–20 % (3); 20–50 % (4); 50–90 % (5); >90%

(6). Site disturbance was ranked between zero and three,

with zero (0) representing no effective disturbance and

three (3) being extensively cleared. Runoff was assigned

to a scale of six classes (0 = no runoff, 1 = very slow, 2 =

slow, 3 = moderately rapid, 4 = rapid, 5 = very rapid;

McDonald et al. 1998).

Vegetation structure was determined by assigning

dominant taxa to each stratum in the landscape, noting

emergent taxa where appropriate, based on McDonald et

al. (1998). All vascular plants were recorded from within

the plot and assigned a cover class (D >70%; M 31–

70%; S 10–30%; V <10%; I = isolated plants and L =

isolated clumps). Material was collected for verification

and vouchering at the Western Australian Herbarium (WA

Herbarium). Additional specimens were collected adjacent

to the plots, which contributed to the overall species list

for the survey area. Where sufficient representative plant

material was available, it was lodged at the WA Herbarium.

Nomenclature follows Florabase (Western Australian

Herbarium 2010). For this study, the reference to weed

taxa refers to invasive species categorised as alien or

introduced to the area and corresponds with the Florabase


Soil chemical attributes were analysed for each quadrat.

Soil was collected from 20 regularly-spaced intervals across

the quadrat, then bulked and sieved. The <2 mm fraction

was analysed by a Inductively Coupled Plasma – Atomic

Emission Spectrometer (ICP–AES) for B, Ca, Cd, Co,

Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, Pb, S and Zn using

the Mehlich No. 3 procedure (Mehlich 1984). Soil pH

was measured on 1:5 soil-water extracts in 0.01 M CaCl


(method S3; Rayment & Higginson 1992). Organic

carbon content was determined using a modified Walkley–

Black method (method 6A1) and the calculation of total

soil nitrogen (N) was based on a modified Kjeldahl digest

(method S10; Rayment & Higginson 1992).

The classification and ordination analyses were

undertaken on a presence/absence data matrix of 176

perennial taxa occurring at more than a single site, which

was consistent with previous greenstone belt studies

(Gibson 2004a, 2004b; Gibson et al. 2012). The

dissimilarity between sites was determined using the Bray–

Curtis measure and the Resemblance routine in PRIMER

v6 (Clarke & Gorley 2006). The Bray–Curtis measure is

a widely-used assessment of ecological distance, which

reflects differences in relative abundance and compositional

change (Legendre & Legendre 1998; Anderson &

Robinson 2003), and provides quantitative output for

similarity between samples (Faith et al. 1987). Using the

Bray–Curtis similarity matrix, the sites were classified based

on the flexible unweighted pair-group mean average

method (UPGMA, 

β = –0.1) using PATN v3.11 (Belbin

1989). The results provided the basis for grouping the

taxa into ecological groups. A two-way table was derived

from the classification for species grouping and the

dendrogram of site grouping. The species–site similarity

matrix was then subjected to Non-metric Multi-

dimensional Scaling (NMS). An environmental data

matrix that included soil chemical properties and site

physical characteristics was created, which was then fitted

to the NMS ordination using Spearman correlation values

in Primer v6. The resulting figure displayed lines of best-

fit in the NMS ordination. The continuous environmental

variables were normalised prior to fitting the

environmental vectors.

The similarity percentages (SIMPER) analyses

provided information on those species typically found

within each community. The SIMPER routine in

PRIMER determines those taxa contributing the greatest

similarity within a community and dissimilarity amongst

communities (Clarke & Warwick 2001). Those taxa

contributing 10% or more to the similarity within each

community type were reported. Where no individual

species contribution reached the 10% threshold, taxa

constituting 50% cumulative contribution were included.

When ties occurred at the 50% level, all taxa in the tie

were reported.

Relationships between environmental variables were

examined using the non-parametric Spearman rank

correlation routine in Statistix 7.1 (Analytical Software,

Tallahassee, Florida). The environmental variables were

analysed using Kruskal–Wallis non-parametric one-way

analysis of variance and post-hoc significance testing of

means at 

α = 0.05 (Sokal & Rolf 1995) for differences

between community groupings.


Summary information

A total of 305 taxa from 47 families and 121 genera were

identified from the permanent quadrats in the northern

area of the Forrestania Greenstone Belt (Appendix 1). An

additional seven taxa were collected adjacent to the plots.

The dominant families represented were Myrtaceae (63

taxa), Fabaceae (55 taxa), Proteaceae (25 taxa), Asteraceae

(19 taxa) and Poaceae (14 taxa). The genera with the greatest

number of species were Acacia (31 species), Eucalyptus

(23 species), Melaleuca (23 species), Grevillea (9 species)

and Austrostipa (8 species). There were five weed species

Flora and vegetation of northern Forrestania Greenstone Belt


identified from the survey. Three new taxa were identified

during the survey; two were the first collections of the taxa.

Species richness within the quadrats varied from 12

to 38 taxa, with an average richness of 24.1 ± 6.4 (SD)

taxa per quadrat. There was a high proportion of singleton

specimens collected (94 taxa, including 11 annuals) and

26 annuals were identified during the survey. Fifteen taxa

were amalgamated into seven species complexes for

analysis. The resulting matrix used in the classification

and ordination analyses comprised 176 species × 50 sites.

All annuals, singletons and indeterminate specimens were

excluded prior to analysis. The resulting dendrogram (Fig.

2) illustrated a clear separation of survey sites, grouped as

floristic communities, and is discussed below.

Rare and priority taxa

Two taxa classified as Declared Rare Flora (DRF) were

collected during the survey (Table 1). Both populations

of  Banksia sphaerocarpa subsp. dolichostyla and

Eucalyptus steedmanii were known in the WA Herbarium

census. An additional 22 populations of ten priority taxa

were recorded during the survey. Single collections of the

annual Gnephosis intonsa (P1) and the perennial shrubs

Labichea rossii (P1), Acacia kerryana (P2), Baeckea sp.

Parker Range (M Hislop & F Hort MH 2968; P3),

Eutaxia nanophylla (P3) and Grevillea dissecta (P4) were


Other priority taxa records included two new

populations of both Acacia asepala (P2) and Microcorys

sp. Forrestania (P4), and five populations of Eutaxia

acanthoclada (P3). Seven populations of the perennial herb

Stylidium sejunctum (P2) were recorded. Prior to this

survey, there were 21 records on the WA Herbarium

Florabase. We suggest that the priority status of S.

sejunctum be changed to P3.

New taxa

Three new taxa were identified during the survey. Single

specimens of each taxon were collected. One taxon,

Hibbertia croninensis ms, had previously been collected

but was identified as H. aff. oligantha (PERTH


Austrostipa sp. Mount Holland (WA Thompson & J

Allen 948; PERTH 07702833) is a perennial tussock grass

with narrow, erect culms. The exserted nodes are distinct

with dense retrorse hairs and ligules that have ciliate

margins. The lemmas have margins inrolled onto the palea

and short hairs near the apex (A Williams, pers. comm.).

The characteristics of this taxon are unique to Austrostipa

in Western Australia. Further investigation is required to

Figure 2. Summary dendrogram of community types for the

northern Forrestania Greenstone Belt. The eight community

types displayed in the dendrogram were derived from the

classification analyses of the 176 taxa from 46 sites; the four

sites excluded from the analyses are not displayed.

Table 1

Priority taxa recorded from the northern Forrestania Greenstone Belt. Bioregion abbreviations: COO = Coolgardie,

MUR = Murchison, AW = Avon Wheatbelt, ESP = Esperance Plains, MAL = Mallee, JF = Jarrah Forest.







Banksia sphaerocarpa var. dolichostyla




Eucalyptus steedmanii




Gnephosis intonsa




Labichea rossii




Acacia asepala




Acacia kerryana




Stylidium sejunctum




Baeckea sp. Parker Range (M Hislop & F Hort MH 2968)




Eutaxia acanthoclada




Eutaxia nanophylla




Grevillea dissecta



Yüklə 0,49 Mb.

Dostları ilə paylaş:
  1   2   3   4

Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azkurs.org 2020
rəhbərliyinə müraciət

    Ana səhifə