Ok,,L Peripheral Vegetation of Peel Inlet and Harvey Estuary, Western
Total Page:16
File Type:pdf, Size:1020Kb
I Journal of the Royal Society of Western Australia, Vol. 4, Part 1, 1981, pp. 5-ll. ---- 1) -------- rProlok,,l ggj''li rake~ rga/1. ( A.1oi.1 peripheral vegetation of Peel Inlet and Harvey Estuary, Western Australia Dbt~ by D. J. Backshall and P. B. Bridgewater IistD nurn, School of Environmental and Life Sciences, Murdoch University, Murdoch, W.A. 6153 I, I, Manuscript received 25 July 1979; accepted 27 November 1979 SOllle I. Abstract Fifteen vegetation units, distinguished by differences in the floristic composition of vascular plant species, have been recognised from the Peel Inlet-Harvey Estuary system, Western Australia. Analysis of the epontic diatom flora within units confirms their integrity. The spatial distribution of these vascular-plant communities around the estuarine system appears to be controlled by the substrate summer salinity values, together with the vertical distance Ieath from the high water mark. Evidence from aerial photographs extending back 23 years The suggests that dynamic processes are less important than these spatial controls in determining community distribution. Exceptions to this occur only when sudden major geomorphic changes force concomitant vegetation changes. Introduction considered to represent the least disturbed examples During 1977-78 an investigation was undertaken of a variety of soil and vegetational types. Soils of the peripheral vegetation and associated diatom of the estuary fringe are referrable to three soil flora of the Peel Inlet-Harvey Estuary System, West systems-Bassendean and Spearwood Dune systems ern Australia. This estuarine system comprises two and Pinjarra Plain soil system (McArthur and large shallow basins of 65 km2 (Peel Inlet) and Bettenay 1960). Site selection was made by field 50 km' (Harvey Estuary), joined to the sea by a reconnaissance, coupled with use of colour aerial 5km inlet channel. The basins are connected to photography taken in 1976 by the Western Austra each other by a narrow channel, which cut through lian Lands Department. The transects comprised a broad, barely submerged marginal shelf. The sedi regularly spaced releves (samples) one metre square mentology of the area has been described by Logan and spaced 5 or 10m apart, depending on the and Brown (197 5). This survey was confined to the abruptness of vegetation change. A total of 747 tidal marsh environment of both Peel Inlet and releves were collected from the estuaries. Species Harvey Estuary. cover-abundance were recorded for each releve using Ecological studies of Western Australian vegetation the Braun-Blanquet scale (Bridgewater 1971). A have shown considerable methodological variation, set of computer programs was used to print tables from empirical analysis of transects (e.g. Sauer of species and releves for each transect. These 1964), classifications based on dominant strata using tables were then examined, and 'potential differential a range of criteria (Gardner 1942, Speck 1952, species' <PDS) noted. The initial choice of PDS is Beard and Webb 1974) to recent implementation of made from species having an apparently clumped numerical methods utilising floristic information distribution, with usually < 60% presence in the mavel 1975). This paper outlines the application group of releves forming the transect. Selection of ~~the Zurich-Montpellier (Z-M) method as a prac PDS was facilitated using a computer program, tical technique utilising floristic attributes. Addition which incorporates numerical methods as outlined a!ly, the paper evaluates the results from this tech by Ceska and Roemer 0971). Groups of releves mque with independently gathered environmental with similar PDS were then extracted from the data and data obtained from diatom sampling. transect raw tables and entered into a presence table (Table ll. Theoretical implications of Z-M phytosociology have been discussed by Bridgewater 0971), Westoff Each column in Table 1 represents thus the sum and van der Maarel 0973) and Muller-Dombois and of a number of releves from the estuary, with the Elle~berg 0974). These references contain the most frequency of species presence in these releves re ~etai!ed accounts of the techniques available in the corded as a percentage class <I-V). Table 1 allows thnghsh language, and it is not proposed to discuss the structuring of a classification for the tidal marsh em further in this paper. vegetation (Table 2). As only one locality was surveyed, it is inappropriate to utilise the standard Z-M system of nomenclature. Accordingly, the Methods system of naming the vegetation units follows that of At each of eighteen sites (Fig. 1) two or three Bridgewater (1974). For ease of reference all transects were laid from the high water mark to the specific names are derived from Blackall and Grieve edge of the sandy beach ridge. These sites were 0974) and Grieve & Blackall 0975). s Jo\lrnal of the Royal Society of Western Australia, Vol. '4, Part 1; 1981. I -· obtained from four replicate sites for each veget t" i group. a ton ~ The Y axis of the ordination appears to r I strongly to salinity, with hypersaline releves ate ~e l positive end and slightly brackish rei eves at the ( - ne~at!ve. The ~elati~mship of releves along the XI axis IS less eastly dtscernable, but approximates t 0 Artllrocn_en a littoral-beach ridge sequence. Salicorr:Ja ~ 32' 30'.-------, suaeda am 6 Km Juncus ntt}l Vegetation distribution and dynamics Maclwerrn Mela/euca Peripheral vegetation of an estuary is influenced by Schoen us J Apium pro marked variations in substrate conditions in both Sporobo/u. space and tim~. Ho~izontal distance from the littoral Atrip/ex P zone and vertical dtstance from a fluctuating water Parapltoli' Co/tlla em table are two important factors superimposed on other Artllrocne influences. Environmental factors may occur either Trig/oclzin Gallnia lri as a mosaic, or as an attenuating gradient, depending Frankenia upon local conditions. An understanding of the Atrip/ex I spatial relationships of vegetation with these environ· Polypogoz Angianthl mental factors may help in elucidating temporal pr(} Casuarim cesses operating within estuarine vegetation. Figure 3 Hordew11 shows the distribution of physical features, vegetation Mela/euc. Samolus. N units and salinity regimes along a hypothetical tran sect from water edge to established dunes. Distribution of the Arthrocnemum complex com· ponent communities reflect both spatial and temporal sequences which are readily observed in numerous salt marshes of the area. The Arthrocnemum bidens community develops on the less saline rims of sail N.l marsh concavities, but progrades into the Trig/ochin mucronata community at lower levels. In the more pro tee saline centres of the concavities, A. ha/ocnemoides to be occurs as a single species, or in extreme environ· with~ ments, a bare pan develops. Where the concavities As th are less saline, perhaps as a result of more effective silt a1 ground water or tidal flushing the Sa/icornia placec quinqueflora community occurs. Sedimentation of the A tid concavity and subsequent reduction of flushing im· north plies a maturation of the system through time. Where marsl the concavity is linked to the estuary, the Arthroc· by CC nemum complex merges with the Salicomia complex. 1976. Beeftink 0962) found the salinity units of the M1 Venice system useful in environmentally classifyin% more saltmarsh communities. This system provides a classt· certai fication of saline waters into three broad categories the 1 euhaline, with a range of 1.65-2.2% chlorinity, like!) polyhaline, with a range of 1.0-1.65% chlorinity,, ~nd a st< Figure I.-Location map of Peel Inlet-Harvey Estuary clim: showing sites of transect analysis. mesohaline, with a range of 0.3-1.0% chlonmty. Vegetation units 1-3 (Arthrocnemum complex) a~e fied euhaline, using this terminology, while the Trig/o~/!111 flu vi: Vegetation ecology mucronata and Sa/icornia quinqueflora cornmumues Exce Mac The fifteen vegetation units derived from the (units 4 and 5) are polyhaline, with all other com· munities being mesohaline. whic classification (Table 2), augmented by all diatom SpeE species (Table 3), were subjected to Principal Axes Many communities clearly respond to sp~tial Ordination (van der Maarel 1969). This technique features, such as the salinity changes descnbed allows a choice of reference stands which will establish above, rather than temporal factors. Neverthelehs~, 15 the extremes of a gradient instead of emphasizing due to the dynamic nature of shorelines in. t L differences within the groups. region, it is possible to infer some successwnal diat The ordination (Fig. 2) is accompanied by selected relationships between communities. Inference ll1~) Wo' substrate information. Summer salinity values were be given a factual basis by comparison of old aer~\ Dia derived from soil samples obtained from the top photographs with recent runs. In this study aenba estu 10 em of the soil profile. These were collected during photographs from 1957 allowed a time scale to e ffieJ March from all transects, and analysed as a 1:5 soil placed on some successional processes. d con water extract using a P66 digital chlorodometer. Initial development of tidal marshes can be hn~e mu Total ions were measured as a 1:5 soil water ex to emergence of spits or sandbars. SedimentattO 0 not tract using a Sproule electronic conductivity meter. often continues until the bar is joined to the ~~or~ rna This analysis was performed in August. Data were by one extremity (Chapman 1938). Once sufticten 6 rI I Journal of the Royal.Society of Western Australia, Vol. 4, Part 1, 1981. ------..1 Table 1. o; '"' "'">~; •• I Presence table for important species in the 15 vegetation units appears to relate Vegetation units (see Table 2) ine releves at the --~~~~~~--~--~- h releves at the ---· 2 3 1 4 5 6 7 8 9 10 11 12 I i I 1~~~~ eves along the X .----· I ·-'----'--'- t approximates to ha/ocnemoides I V I V I V I V I + II II Arthroc~emumSalicorma qut•tquejlora ...