JOURNAL OF THE ROYAL SOCIETY OF WESTERN AUSTRALIA Vol«g.e 70 ^ Part 3 . 1988 iEoyal Society of Australia To promote and foster science in Western Australia and counteract the effects of specialization President PATRON Her Majesty the Queen VICE-PATRON His Excellency Professor Gordon Reid Governor of Western Australia COUNCIL 1987-1988 J T Tippett B Sc, Ph D Vice-Presidents J S Pate Ph D, D Sc, FAA, FRS M Candy M Sc, FRAS Past President J S Beard M A, B Sc, D Phil Joint Hon Secretaries K W Dixon B Sc (Hons), Ph D L Thomas M Sc Hon Treasurer J Dodd B A. M Sc, Ph D Hon Librarian M A Triffitt B A, ALAA Hon Editor I Abbott B Sc (Hons), Ph D Journal Manager J Backhouse M Sc, Ph D Members W A Cowling B Agric Sc (Hons), Ph D D Bell B Sc (Hons), Ph D S J Hallam M A, FAHA E R Hopkins B Sc, Dip For, Ph D L E Koch M Sc, Ph D K McNamara B Sc (Hons), Ph D J Majer B Sc, DIC, Cert Ed, Ph D Journal of the Ro\al Socict\ of Western Australia 70 (3). 1988. 57-67 Floristic reconnaissance of the northern portion of the Gregory National Park, Northern Territory, Australia D M J S Bowman, B A Wilson & P L Wilson Conservation Commission of the Northern Territory PO Box 38496. Winnellie, N T 5789 Manuscript received March 1987; accepted August 1987 Abstract A tlorisiic reconnaissance of two areas in the northern portion of the Gregory National Park revealed a range of dis- tinct vegetation tvpos. Numerical classification was used lodeierminc 1 3 planicommunities from an area near Victoria River Crossing and 10 communities from Bullita homestead area. Phese communites were shown to form significant associations with landtorm and geologv. The latter were found to be closei\ related, as the topography of the area is sirongl) controlled b\ the erosion of Adclaidcan sediments, some of which have been capped by Lower Cambrian basalts. The plant communities arc best grouped within landform complexes, which include: riverine, plain, undulating tenain. mesa plateau and slope, and plaieau/hill rims. These complexes can be subdivided by substrate material and surface soil texture. Vegetation mapsai a .scale of 1 : 100 000 were unable to diffcremiale all the communities due to the complex micro-pattern o( some vegetation types. The complete list of species recorded at the Victoria River and Bullita areas was found to be closely related to the nearby Keep River National Park, but to be poorly related to the Bungle Bungle National Park area in WA. Other areas of sandstone in the northern coastal region of the NT were found to be poorly related to Gregory. It is suggested that these similarities are associated with rainfall and the relative development of sandstone canyons, which act as refugia for mesic plant species. A major management problem of Gregory National Park relates to feral animals and their sus- pected association with soil erosion and the spread of exotic plants. Introduction A basic requirement of nature conservation is an ap- preciation of the uniqueness and representativeness of nature reserves. Vast areas of the Northern Territory are botanically poorly explored. This lack of basic data ham- pers conclusions as to the conservation value of reserves in protecting rare or threatened vegetation. One approach to overcome this problem is to produce small scale veg- etation maps. This method has the advantage of provid- ing an overview of major vegetation formations, but suf- fers the disadvantage of treating the flora superficially. Another approach is to analyse biogeographic patterns of Herbarium records (Dunlop & Bowman 1986). The util- ity of this method however, is dependent upon the thoroughness of past plant collecting. This paper reports the results of a botanical reconnais- sance of Gregory National Park, in which two small areas asse.sscd to represem major land i\pcs wiihm the park, were .studied in detail. Fieldwork was undertaken to sim- ultaneously ground truth large scale vegetation maps, re- cord data for phvlosociological analysis and collect speci- mens ol all vascular plant species for preservation in the Nonhern Territory Herbarium. Phvtosociological analv- scs were conducted to provide land managers w'iih basic ecological data, and enable comparison with vegetation surveys of other areas in northern Australia, enabling the flora of the park to be placed into a regional context. The Appendix is a Supplemeniary Publicaiion and is not printed with the paper. Copies arc lodged with the Society’s Library (c/- Western Australian Museum, Perth WA 6000) and with the National Library of Australia (Manu- script Section, Parkes Place. Barton ACT 2600). Photocopies may be ob- tained from either institution upon payment of a fee. Climatic data for Victoria River Downs and Timber Creek. Table 1 A - mean precipitation (mm); B = mean number of rain days; C = mean daily maximum Jan Feb Mar Apr May Jun Jul Victoria River Downs A 145 142 106 19 6 2 3 B 1 1 10 8 2 1 0 0 C 37.0 35.5 34.7 34.7 31.7 29.6 29.0 D 25.0 23.9 23.2 19.8 16.3 12.2 10.8 Timber Creek A 197 201 157 25 5 2 1 B 12 12 9 2 1 0 0 C&D N/A temperature CO; D = mean daily minimum temperature (°C). Aug Sep Oct Nov Dec Year 1 4 17 0 1 2 32.3 35.5 37.7 14.1 18.0 22.0 0.4 4 27 0 I 3 61 1 12 618 6 9 50 38.3 38.1 34.5 23.5 24.2 19.8 66 128 813 6 9 55 57 A59038-1 Journal of the Royal Society of Western Australia 70 (3), 1988 Regional environment Climate The climate is monsoonal with five months of summer rains {Table 1). The total annual rainfall (618 — 813 mm) is between one half to one third of that received at Dar- win. and the number of raindays is one half of those re- corded for the northern capital. Summer mean maximum daily temperature and winter minimum daily air tem- perature are more extreme than the north coast of the N T. Soils Soils of the park are generally related to lithology and topographic position. Much of the area consists of steep tablelands and hills with shallow immature skeletal soils (Stewart 1970). Deep soils are confined to gentle lower slopes, where red and yellow earths are common with poorly drained lower slopes possessing cracking clays. Al- luvial soils, invariably cracking clays, have developed on river flats. 10 Figure 1 Map of major land types within the environs of the proposed Gregory National Park. (For location of this area see Fig. 7). The map is based on 1:250 000 satelite imagery. Ground iruihing is restricted to the two study areas in the northern portion of the park. Mapping units correspond to the follow- ing environments; 1 Lateritic plateaux; 2 Sandstone plateaux; 3 Sandstone hills and dissected plateaux; 4 Limestone plains; 5 Limestone hills; 6 Dissected limestone; 7 Basalt plains and plateaux; 8 Floodplains: 9 Survey areas; 10 Park boundary. Geology Gregor>' National Park is situated in the Victoria River Basin (Sweet 1977). During the early Adelaidean, cycles of marine deposition and subsequent uplift and erosion produced sequences of sandstone, siltslonc and carbonate rocks (Sweet 1972). Frequent faulting fractured these sediments. Lower Cambrian basalt was extruded across parts of the landscape, with deep laieritization occuring during the Tertiary (Sweet 1 972). Uplift and erosion since the late Tertiary have exposed older rocks and produced a landscape of wide plains with resistant sediments form- ing plateaux and mesas, some with a remnant lalerilizcd cap. Methods Landsat imagery at a scale of 1 :250 000 was visually in- terpreted to produce a map of gross land types (Fig. 1). From this interpretation, areas of approximately 25 by 25 km. near the Victoria River Crossing and the Bullita homestead, which w'crc assessed as being representative of enviroments within the park and important for public access, were selected for intensive study. Photopatterns of the two areas were delineated on 1:80 000 scale. 1968. black and white aerial photographs and 1:100 000 maps compiled. 58 Journal of the Royal Society of Western Australia 70 (3). 1988 A total 01327. 10 h\ 1 0 m quadrats were placed (181 at Victoria Ri\cr and 146 at Bullita) within identified phoiopatterns. using a combination of systematic sam- pling of tracks. Held ira\ erse and helicopter landings dur- ing late Februar\ and earU March 1986. The presence of all \ascLilar plant species was noted in each quadrat. The structure of surrounding vegetation was classified accord- ing to the scheme of Walker &. Hopkins (1986) and SLibsequcntlv con\ cried to the scheme of .Spechl (1981). The lopogiaphic position off quadrats was classified as cither: mesa top. mesa/hill nm. mesa sidcslopc. mesa gulK. hilltop, hill sidcslopc. plain, permanent water- course. ephemeral watercourse or drainage basin. Rock outcrop was noted as either sandstone, limestone, sandslone/limestone mix. basalt, latcnlc or covered with alluvial deposits. The percent cover of rock, gravel and bare ground was noted. .Surface soil te.Mure was classified as either sand, loam or silt/cla\. Data analysis Florislic and environmental data arc stored on the eco- logical data base system ECOPAK (Minchin 1986). Vic- toria River and Bullita data sets w'crc classified separ- ately. Before analysis, species that occurred in less than three quadrats were deleted from the matrix. The presence/absence floristic data for each site were sub- jected to an agglomerative classification using the UPGMA sorting strategy after calculating a Bray-Curlis similarity matrix using the Numerical Taxonomy Pack- age NTP"(Belbin ct a!. 1985). The classification was im- posed on the relatively continuous variation across the communities. As there are no defined stopping rules in classification, the level of truncation, which was not con- sistent across the dendrograms, w'as determined subjectively after careful inspection with lists of species memberships. The association between the non- parametric environmental variables and the florislic groups was tested for departure from randomness by Chi- square analysis. The centroids of the resulting 23 floristic groups were ordinated by Detrended Correspondence Analysis (Hill & Gauch 198()). Because of limitations in computer space, only those species that ocurred in more than 5 groups (245) were used in the ordination. Community definitions Thirteen floristic groups were recognized at Victoria River (Fig. 2) and 10 florislic groups were recognized at Bullita (Fig. 3). Structural classification (following Spechl 1981) and dominant species for the upper, mid and lower strata for each group, arc shown in Tables 2 & 3. A total of 517 species was encountered in the survey. Their percent frequency occurrence by classification group is indicated in the Appendix. Table 2 Siruclural classification (Specht 1981) and dominant species for upper, mid and lower strata for florislic groups at Victoria River Group No Upper stratum Middle stratum Lower stratum I Low open-woodland of Eucal\pius dichromophloia. E. f'crruginca. E. miniaia. Shrubland of .Acac/a laccaia and Grcvilica spp. Grassland of Plcclrachnc pungens and Eriachne ciliaia. 2 Low woodland of £. dichromophloia and Ery ihrophlcum chlorosiachys. Shrubland of Pctalostigma quadrilocularc. Cochlospermum frascri and Calytrix cxsiipulaia. Grassland of Plcclrachnc pungens and Eriachne ciliaia. 3 Woodland of E miniaia with subdominani Tcrminalia laiipcs and Owenia vernicosa. Shrubland of Buchanania obovaia. Tcmpicionia hookcri and Acacia spp. Open-grassland of Plcclrachnc pungens and Fimbhstylis pauciflora. 4 Forest of Lhistona sp. nova Fouicna scncca. Ficus spp. and Vitcx glabrata. Sparse ferns and C ypcraccac spp. 5 Woodland of E. (cciinca and L\siph\Hum cunninghamii. Shrubland of Ampc/oc;ssus acciosa and Hakca arborescens. Grassland of Hctcropogon coniortus Sorghum plumosum and Schima nervosum. 6 Low open-woodland of E- teclUlca with co-dominant Erythrophleum chlorostachys. Grassland of Sehima nervosum. Themeda avenacea and Eriachne ciliaia. 7 Open-woodland of E. lectifica and E. lerminalis Open-shrubland of Aialaya hemiglauca and Grewia rciusifolia. Grassland of Sehima nervosum Dichanihium fecundum and Chrysopogon fallax. 8 Low woodland of Eryihrophlcum chlorostachys. E. lectifica and E. confertifiora. Open-shrubland of Aialaya hemiglauca and Grewia retusifolia. Open-grass/hcrbland of Plcctrachne pungens. Tacca leoniopeialoides and Fabaceac spp. 9 Open-forest of Ziziphus quadrilocularis. Strychnos Jucida and Cellis philippinensis with emergent E. confertifiora. Sparse Commelina ensifoUa and Passiflora foctida. 10 Low woodland of Melaleuca argeniea. Lophostemon granditlorus. Tcrminalia piaiyptera and Pandanus aquaticus. H Woodland of E. camalduknsis, Naucica orientalis and Ficus coronulaia. Open-grassland of Echinochloa colona and Cynodon dactylon. 12 Closed-forest of Sy/yigium angophoroides. Livistona sp nova and Ficus spp. Open-sedge/herbland. 13 Low closed-forest of Melaleuca symphyocarpa with emergent Melaleuca leucadendra. Sparse cover of Cyperaceae spp. 59 Journal of the Royal Society of Western Australia 70 (3), 1988 95 90 85 8 9 10 11 12 13 FLORISTIC GROUPS Figure 2 Dendrogram of floristic similarities of quadrats placed at Victoria River. The classification was truncated at the thirteen group level. 1-95 -90 > K tr < u 2 CO w Q a? L 14 15 16 17 18 19 20 21 22 23 FLORISTIC GROUPS Figure 3 Dendrogram of the floristic similarities of the quadrats placed at Bullita. The classification was truncated at the ten group level. Table 3 Structural classification (Specht 1981) and dominant species for upper, mid and lower strata of floristic groups at Bullita. Group No Upper stratum Middle stratum Lower stratum 14 Woodland of Lysiphyllum cunninghamii E. teclifica, E. ierminalis, £. pruinosa and Adansonia gregorii. Shrubland of Ampelocissus acetosa and Hakea arborescens. Grassland of Heteropogon contortus and Sorghum plumosum. IS Open-woodland of Lysiphyllum cunninghamii. E. leciifica. Termmalia canescens. Open-shrubland of Carissa lanceolaia A mpelocissus aceiosa and Ftueggea vtrosa. Grassland of Sorghym plumosum Themeda avenacea and Heteropogon contortus. 16 Low open-woodland of E. brevijblia. E. dichromophloia and Terminalia canescens Open-shrubland of Ampelocissus acetosa. Ftueggea virosa and Cochlospermum fraseri. Grassland of Ptecirachne pungens Themeda avenacea and Sehima nervosum. 17 Low open-woodland of E. dichromophloia and E. ferruginea. Open-shrubland of Grevillea pyramidalis and Ampelocissus acetosa. Grassland of Sorghum plumosum and Plectrachne pungens. 18 Tall Shrubland of Acacia lepiocarpus and Acacia lysiphloia with emergent E. teclifica Grassland of Heteropogon contortus Aristida bromnana and Sehima nervosum. 19 Open-woodland of Adansonia gregorii. E. teclifica and £. pruinosa. Shrubland of Dodonaea physocarpa and Ampelocissus acetosa Grassland of Plectrachne pungens and Aristida browniana. 20 Open-forest of Terminalia platyphylla Lophosiemon grandiflorus. Melaleuca leucadendra and Ficus coronuiaia. Shrubland of Fiueggea virosa and Acacia holosericea. Grassland of Heteropogon contortus and Echinochloa colona. 21 Low closed-forest of Cetiis philippinensis. Ficus spp. and Strychnos lucida. Open herb/vineland of Passiflora foetida, Jasminum didymum 22 Open-shrubland o( Acacia laccata and Cochlospermum fraseri Grassland of Plectrachne pungens 23 Low open-woodland of E. ferruginea and E. brevifolia. Open-shrubland of Grevillea angulata. Grevillea refracta and Acacia spp. Open-grassland of Plectrachne pungens and Eriachne spp. 60 Journal of the Royal Society of Western Australia 70 (3), 1988 Relationship of mapping units and floristic communities Tables 4 & 5 show there is a significant relationship be- tween floristic communities and mapping units. However this relationship is not perfect; one floristic group may be significantly associated with more than one mapping unit, as some communities are not possible to differentiate on the maps due to scale and their diffuse boundaries. Two mapping units ( 1 0 & 11) are significantly associated with the same florisitic community (Group 1 4). However map- ping unit ! 1 contains riverine communities (Groups 1 9 & 20) in contrast to unit 10. Table 4 Frequency of quadrat occurrence by floristic group and mapping unit at Victoria River. Asterisks denote values showing significant associations following one sample Chi-square analysis (* P<0.05, ** P<0.01, ***P<0.001). Mapping unit 1 2 3 4 5 Floristic 6 7 group 8 9 10 n 12 13 1 1 0 0 0 18*** 3* 4* 0 3 2 14 **» 0 0 2 18** 0 13* 5 8 0 1 2 I I 0 1 0 3 2 10*** 11* 6** I 0 0 1 0 0 0 2 1 4 2 4 0 0 0 0 2 0 0 0 0 0 5 1 0 0 0 3 2 1 5 6*** 0 0 0 0 6 g*** 0 0 0 1 0 0 0 0 0 0 0 0 7 0 0 0 0 2 4*** 1 0 0 0 0 0 0 Total 33 19 28 II 33 9 7 10 10 3 14 3 1 Table 5 Frequency of quadrat occurrence by floristic group and mapping unit at Bullita. Asterisks denote values showing significant associations following one sample Chi-square analysis (* P<0.05. ** P<0.0l. ***P<0.001). Mapping Unit 14 15 16 17 Floristic Group 18 19 20 21 22 23 8 0 0 0 0 0 0 1 1 0 0 9 5 8 3 0 1 0 0 0 1 0 10 19* 5 0 0 2 0 0 0 I 0 11 41* 2 1 0 1 5* 9* 3 0 0 12 0 0 0 0 0 0 0 0 0 0 13 1 1 1*** 8*** 0 I 0 0 1 0 0 14 0 1 5* 2*** 0 0 0 0 1 6* Total 66 27 17 2 5 6 10 6 3 6 Environmental relationships Tables 6.7.8 Sc 9 show that the vegetation types defined by the numerical classification are strongly related to topographic position and substrate material. Table 10 shows that substrate material is significantly associated with landform. Therefore the the communities can be pri- marily grouped by landform with secondary differen- tiation by substrate material. Table 6 Frequency of quadrat occurrence by floristic group and topographic position at Victoria River. Asterisks denote values showing significant associations following one sample Chi-square analysis (* P<0.05, ** P<0.01. ***P<0.001). Floristic group Topographic position 1 2 3 4 5 6 7 8 9 10 11 12 13 Mesa lop 10* 15** 2 0 0 0 0 0 0 0 0 0 0 Mesa rim 1 1 11*** 2 1 0 0 2 5 0 0 0 0 Mesa side-slope 11** 0 4 0 5 0 1 4* 0 0 0 0 0 Mesa gully 0 0 8*** 7* 0 0 0 0 1 0 0 2** 0 Hill top 4 0 0 0 3 1 0 1 0 0 0 0 0 Hill side-slope 2 1 0 0 0 0 0 1 0 0 0 0 0 Plain 2 0 0 0 9* 8*** 4** 0 0 0 1 0 0 1 Permanent water course 0 1 I 2 10 0 1 2 0 2 13*** 1 Epemeral water course 2 1 0 0 5 0 0 0 4*** 1 0 0 0 Drainage basin Missing I 0 0 2 0 0 0 1 0 0 0 0 0 0 Total 33 19 26 1 1 33 9 7 10 10 3 14 3 1 61 Journal of the Royal Society of Western Australia 70 (3), 1988 Table 7 Frequency of quadrat occurrence by florisiic group and substrate type at Victoria River. Asterisks denote values showing significant associations following one sample Chi-square analysis {* P<0.05. ** P<0.0 1 , ***P<0.001 ). Substrate Type 1 2 3 4 5 6 Floristic 7 group 8 9 10 n 12 13 Basalt 2 0 0 0 1 1 0 1 4** 0 0 0 0 Lateriie ■j*** 1 0 0 1 0 0 1 2 0 0 0 0 Sandstone 22 17 28*** 1 1 9 5 3 8 0 1 0 2 0 Alluvium 0 0 0 0 22*** 3 4 0 4 2 14*** 0 1 Limestone Missing 0 2 1 0 0 0 0 0 0 0 0 0 0 1 0 Total 31 19 28 1 1 33 9 7 10 10 3 14 2 1 Table 8 Frequency of quadrat occurrence by florisiic group and topographic position at Bullila. Asterisks denote values significantly associated, following one sample Chi-square analysis (* P<0.05. ** P<0.01. ***P<0.001). Topographic position 14 15 16 17 Floristic 18 group 19 20 21 22 23 Mesa lop 0 0 2 2** 0 0 0 0 1 6** Mesa rim 0 0 4* 0 0 0 0 2** 0 0 Mesa side-slope 3 21*** 8 0 1 0 0 1 2 0 Mesa gully 1 2 2 0 1 0 1 0 0 0 Hill top 8 2 0 0 1 0 1 1 0 0 Hill side-slope 3 0 0 0 1 0 0 0 0 0 Plain 33*** 2 1 0 0 5** 0 0 0 0 Permanent water course 5 0 0 0 0 0 6*** I 0 0 Ephemeral water course 9 0 0 0 0 0 2 0 0 0 Drainage basin Missing 4 0 0 0 1 0 1 0 0 1 0 0 Total 66 27 17 2 5 5 10 5 3 6 Table 9 Frequency of quadrat occurrence by florisiic group and substrate at Bullila. Asterisks denote values showing significantly associations, following one sample Chi-square analysis (* P<0.05. ** P<0.01. ***P<0.00l). Substrate Type Floristic group 14 15 16 17 18 19 20 21 22 23 Limestone 34 13 1 0 4 0 3 4 2 0 Sandstone 14 1 1 12 2 1 0 0 I 1 6*** Lime/Sandstone 1 3 3** 0 0 0 0 0 0 0 Alluvium Missing 17 0 1 0 0 ^*** 1 7** 0 1 0 0 Total 66 27 17 2 5 5 10 5 3 6 Table 10 Frequency of quadrat occurence by topographic position and geology for Victoria River and Bullila. Asterisks denote values significantly associated following one sample Chi-square analysis (* P<0.05. ** P<0,0l, ***P<0.001). 1 opographic position Lalcrite Sandstone Substrate Type Lime/Sandstone Limestone Basalt Alluvium Mesa top 6*** 30** 0 1 1 0 Mesa rim 3 20 0 3 3* 0 Mesa side-slope 2 41 4* 12 2 0 Mesa gully 0 17 3*** 3 1 1 Hill top 1 5 0 2 0 0 Hill side-slope 0 7 0 12*** 2 1 Plain 0 18 0 14 0 33*** Pcnnancni-watcrcourse 0 8 0 5 0 33*** Ephcmcral-waicrcoursc 0 6 0 8 0 10 Drainage Basin 0 3 0 -) 0 2 lotal 12 155 7 62 9 80 62 SOIL TEXTURE Journal of the Royal Society of Western Australia 70 (3), 1988 SAND CLAY DRY DCA 1 MOISTURE STATUS WET Figure 4 Plot of the centroids of ihe flonslic classifications of the Victoria River and Bullita floristic groups derived from the classification in the first two axes of a Detrended Correspondence Analysis ordination (DCA). The dashed line indicates the division between Victoria River and Bullita. Envelopes are placed around centroids with like geologies and landforms. Mesa plateau communities All plateau and mesa lops support eucalypl low open- woodlands. Floristic groups 1 and 2 dominate the sand- stone and latcritc plateaux at Victoria River. At Bullita Group 17 and 23 occur on sandstone substrate plateaux. Mesa plateau and mesa/hill rim communities The sandstone plateau rims at Victoria River are domi- nated by Eucalyptus miniaia woodlands of floristic Group 3. At Bullita limestone and/or sandstone rims support E. brevifolia low woodlands (Group 16) while limestone rims carry low Celtis philippinensis closed-forest (Group 21 ). Mesa sideslopes Mesa sideslopes at Victoria River support E. dichromophloia woodlands (Group 1). These sand- stone slopes support low woodlands dominated by Erythrophleum chlorostachys (Group 8). At Bullita sand- stone slopes support Lysiphyllum cunninghamii domi- nated woodlands (Group 15) while limestone slopes are either covered in this community or open-shrublands dominated by Acacia laccata (Group 22). Mesa gullies Three communities occur in sandstone gullies at Vic- toria River. Eucalyptus miniata woodlands (Group 3). which also occur on plateau rims, occur in the driest gul- lies. In moist gullies LIvistonia ‘Victoria River’ forests occur (Group 4). In deep protected gullies closed-forests dominated by Sycygium angophoroides shade fern understories (Group 12). The gullies on mesas at Bullita arc shallow and do not support characteristic vegetation types. Hills Hill communities are only significant at Bullita, associ- ated with rounded limestone outcrops. They support tall shrubland dominated by Acacia leptocarpa (Group 18) 63 Journal of the Royal Society of Western Australia 70 (3). 1988 1 5 6 7 0 2 6 8 lOKm Scale ♦ N I Figure 5 Map of repeatable photo-patterns at Victoria River. Mapping units correspond to the following environments: I Plains and Rivers: 2 Sandstone pla- teau sideslopes: 3 Sandstone plateau rims and upper slopes: 4 Sandstone plateau tops: 5 Basalt sideslopes: 6 Laleritised basalt plateaux: 7 Basalt plains. Plains Alluvial plains with residual rock outcrops support three interrelated types of £. leciifica woodlands (Groups 5. 6 & 7). At BuUiia the alluvial deposits on the plains sup- port E. icctifica open-woodlands with emergent Adansonia gregorii trees (Group 19). Lysiphyllum cunninghamii woodlands are found on sites with either limestone or sandstone rock, or alluvial deposits (Group 14). Pennancni water conwwnitics These communities are more diverse and abundant at Victoria River than Bullita. In the Victoria River area. camaldulensjs woodlands are found on levees with deep alluvial soils (Group II). Low Melaleuca argcniea woodlands (Group 10) occur on river banks. In poorly drained depressions low Melaleuca symplnvcarpa dosed- forcst occur (Group 13). At Bullita Lophostenion grandiflora open-forest characterizes the riverine com- munities (Group 20). Ephemeral water communities The dry creeks on basalt plateaux support mixed species open-'monsoon' forest dominated by species such as Ziziphus quadrilocularus and Strvehnos lucida (Group 9). Species diversity of the landform-vcgetation complexes The ephemeral water community is the most species rich community (26.9 species per 100 m-) while the per- manent water communities have the greatest range of species richness (7 — 19.9 species per 100 m-). The plains communities are generally richer than the elevated com- munities with the exception of the low Erythrophlcum woodland (Group 8) which is the second most diverse community (Appendix). 64 Journal of the Royal Society of Western Australia 70 (3). 1988 Figure 6 Map of rcpeaiable phoio-palteins at Bullita. Mapping units correspond to the following environments: 8 Limestone plateaux- 9 Undulating limestone counlrv: 10 Plains and rivers: 11 Plains; 12 Eroded devegelaied areas; 13 Sandslone/Limcslone mesa sideslopes; 14 Sandslone/Limestone mesa tons- Noi ground iruthed. ^ ' Comparison between V ictoria River and Bullita areas Ordinaiion of ihc ceniroids of the 23 commumiies shows that the main differences between Bullita and Vic- toria River is moisture status and the occurence of basalt and limestone (Figures 4 . 5 & 6). The permanent water community at Bullita is tlorisiicalK distinct from the four at Victoria River, the former sharing more species with the limestone hill complex. The limestone mesa complex is distinct from the sandstone and laicriic plateau com- plex possibly due to the formers cla>-rich basic soils which gives it more affinity with the plains communities. The 65 plain communities arc similarat both Bullita and Victoria River even though the geologies of the two areas are dif- ferent. The sandslone/latcntc plateau complex is shared both Bullita and Victoria River. The driest conimuni- ucs in this complex occur at Bulhia while at Victoria River these communities span the moisture gradient (as dclincd by DCA 1) from plateaux with skeletal soils through to deep sheltered canvons. Variation in the pl^alcaux communities at Bullita is associated with ch^^ges in surface soil texture and substrate material A59038-2 Journal of the Royal Society of Western Australia 70 (3). 1988 Comparison of Gregory with other areas in northern Australia The vegetation communities described at Victoria River and Bullita approximate the general descriptions provided for the Ord-Vicioria area by Perry (1970). the Bungle Bungle Ranges by Forbes & Kenneally ( 1 986) and Keep River National Park by Henshall & Mitchell (1979) and Sivertsen & Van-Cuylenburg (1986). More detailed floristic comparisons with the complete Oregon' list (Ap- pendix) were made with the above authors’ species lists, the list for Uluru National Park, (or Ayers Rock. Hooper et a/. 1 973). Katherine Gorge N P (Sivertsen & Day 1 986). Alligator Rivers Region (or Kakadu. Taylor & Dunlop 1985) and Litchfield Park (or Tabletop Range, Kirkpatrick cf a! 1988 and Lynch & Manning 1988) (Fig. 7 and Tabic 1 1 ). Because the above surveys were conduc- ted in different seasons and at different levels of intensity, a conservative measure of floristic similarity was used: number of species in common / the lowest number of species m cither species list. This analysis shows that Keep River has the highest similarity with Gregory (53%) while Ayers Rock has the least number of species in com- mon (10%). Katherine Gorge, the Bungle Bungles. Alli- gator Rivers and Litchfield have lower similarities (42. 37. 33 Si 34% respectively). The similarities of the areas generally rellcct north- south changes in rainfall. The Bungle Bungles, however, arc situated further south and inland than Gregory (Fig. 3) but have a comparable rainfall (Forbes Si Kenneally 1986). The relatively low similarity of this area with Gregory appears to be largely due to the number of rainforest species (eg Fkm virens and Euodia elleryana) that are found in the deep gullies at the Bungle Bungles that are not found at Gregory'. Such refugia appear to be smaller and less common at Victoria River and rare at Bullita. These results concur with the findings of Kirkpatrick el al {\ 988) that sandstone communities arc more spatially variable than the more uniformly distrib- uted savanna communities in the northern coastal regions of the Northern Territory'. Table 11 Floristic similarity of various areas in the Northern Territory to Gregory National Park Location tluru N P (.\yers Rock) Keep River N P Bungle Bungle Range Katherine Gorge N P Litchfield Park (Tabletop Range) Alligator Region (Kakadu) Total Species Recorded 320 301 657 165 423 657 Number of Species in common with 32 160 152 69 144 171 Gregory Similarity to Gregory (%) iO 53 37 42 34 33 East-west differences are reflected in a comparison of the number of species held in common with the .Alligator Rivers Region between Victoria and Bullita (130 vs 92). The sandstone escarpments and associated canyons are larger and more frequent at Victoria River and carry many species in common with Kakadu (eg. Euvalypiua miniata. Ervlhrophleum chlorostachys and Ficus spp.) which do not occur in the drier Bullita environment (Ap- pendix). Further east-west gradients are expressed in the common occurence oi Adansonia gregorii at Keep River and Bullita but not Victoria River. Also the presence of Lirislona ‘Victoria River' at Victoria River and the Bungle Bungles may reflect an association of sandstone geologies or a relict distribution of the species. Perry ( 1 970). Forbes & Kenneally ( 1 986). and Sivertsen & Van-Cuylcnburg ( 1 986) assume that there is a strong rc- lationship'betwccn landform. soils and vegetation. This study has supported this assumption. Perry (1970) notes that "after climate, species distributions arc most strongly controlled by some faclor(s) associated with lithology, particular acidic and basic rocks. This study suggests that local microclimate as determined by landform is very im- portant in controlling vegetation distribution with surface soil texture being a secondary interrelated factor. Implications for park management No obviously fire damaged vegetation was encountered in the course of this sur\'ey. Burnt sandstone plateau veg- etation was obscr\'ed to be floristically richer than adjac- ent unburnt patches. The major management problems appear to be associated with introduced herbivores. The bare areas al Bullita (Mapping Unit 12). apparent in the 1968 aerial photography, were still clearly visible at the lime of the survey. They may have developed in response to high densities of cattle associated with stock yards and animal camps. Feral donkey s will continue to have an im- pact on soil erosion following the exclusion of cattle from the park. Clearly there is need to assess the impact and control of these animals. The second major management problem is associated with the spread of exotic plants throughout the park. The 17 exotic plants encountered in this survey were mainly found on the plain and riverine communities which are also the focus of herbivory. It is likely that a reduction of feral animal populations and subsequent control of erosion will help in controlling the spread of exotics. 66 Journal of the Royal Society of Western Australia 70 (3). 1988 Given the variabilily in plant communities found across the northern part of the Northern Territory, par- ticularly with respect to sandstone areas, there is a need to reserve areas across major environmental gradients to en- sure adequate reservation of plant species and habitats. The Gregory National Park is an important addition to the Northern Territory National Park estate as it en- compasses previously unreserved plains communites and an important area of sandstone vegetation in an arid ex- treme of the Australian monsoon tropics. Acknowledgcmenn Wc ihank Bill Prceland. Clyde Dunlop. Biair Wood and Mike Reed for commems on an earlier version ofthis paper. John Dc Koning and Sonia Tidcmann are thanked for the help and camaraderie during ihe Held work. References Belbin L, Faiih D P & Minchin P R 1984 Some algorithms contained in the numerical taxonomy package NTP. CSIRO Water and Land Resources Technical Memorandum 84/23. Dunlop C R & Bowman D M J S 1 986 Allas of the vascular plant genera of the Nonhem Territory. Australian Flora and Fauna Senes No 6. Forbes S J & Kcnneally K V 1986 A botanical survev of Bungle Bungle and Osmond Range, southeastern Kimberley. Western Australia W Aust Naturalist 16 94-169. Henshall T S & Mitchell A S 1979 Vegetation survev oft he Keep River study area. N T Botanical Bull No 2 CCNT, Darwin. Hill M O & Gauch H G 1980 Detrended correspondence analysis, an im- proved ordination technique. Vegetauo 42: 47-58. Hooper P T. Sallaway M M, l&u P K. Maconochic J R. Hyde K W & Corbett L K 1973 Ayers Rock - Mt Olga National Park environmental study. 1972. And Zone Res Inst Land Conservation Ser No 2. Kirkpatrick J B. Bowman D M J S. Wilson B A & Dickinson K J M 1988 A transect study of the Eucalyptus forests and woodlands of a dissected sandstone and latcrite plateau near Darwin. Northern Tcrrilorv Aust J Ecol 12:339-359. ^ Lynch B T & Manning KM 1988 Land Resources of Litchfield Park. Conser- vation Commission of the N T Technical Rep in press. Mmchm PR 1986 How to use ECOPAK: An ecological data base system. CSIRO Water & Land Resources Technical Memorandum 86/6. Perry R A 1970 Vegetation of the Ord-Victoria area. In: Lands of the Ord- Viciona area. W A and N T. CSIRO Land Research Ser No 28. 1 04- 1 1 9. Sivertsen D P & Van-Cuylcnburg H R M 1986 Land resources of the Keep River National Park. Conservation Commission ofthc N TTech Rep No 22 . Sivertsen D P & Day K 1 986 Land resources of the Katherine Ciorge national park. Conservation Commission of the N T Tech Rep No 20. Spcchi R L 1981 Foliage projective cover and standing biomass. In: Veg- etation classification in Australia (eds A N Gillison & D J Anderson) CSIRO & ANU. Canberra. 10-21. Stewart G A 1970 .Soils of the Ord-Vicioria Area. In: Lands of the Ord- Victoria area. W .A and N T. CSIRO Land Research Scr No 28. 92-103. Sweet I P 1972 Dclamer’c N T 1:250 000 Geological Series — Explanatory Notes. Bureau of Mineral Resources. Aust Oovi PubI Serv. Canberra', Sweet L P 1977 The Precambnan geology of the Victoria Rivers region Northern Territory. Bureau of Mineral Resources. Geology and Geo- physics Bull 168. Taylor J A & Dunlop C R 1 985 Plant communities of the wet- dry tropics of Australia: the Alligator Rivers region. Northern Territorv. Proc Ecol Soc Aust 13: 83-128. Walker J & Hopkins M S 1984. Vegetation. In; Australian Soil and Land Sur- vey Field Handbook, (eds R C McDonald. R F Isbell. J G Speight. J Walker & M S Hopkins) Inkata Press. Mclbournc. 67 i I s H'. ,v I •i & Journal of ihc Ro>al Socict\ of Western Australia 70 (3). 1988. 69-87 Consanguineous wetlands and their distribution in the Darling System, Southwestern Australia C A Semeniuk 21 GIcnmerc Road Wan'Mck WA 6024 Manuscript received May /‘AS’7. acicpied Oc'oiyer 1987 Abstract In the Darling System of Southwestern A\ustralia. similarity in physical setting and causative factors of wetland development produces suites of wetlands with common or inter-related features. These genetically related wetlands are termed consanguineous and form assemblages termed consanguin- eous suites. Consanguineous suites arc identified on cniena of wetland type, wetland geometrv, stra- tigraphy . inferred origin.and water characteristics. In total some 42 consanguineous wetland suites are recognized throughout the Darling System. Consanguineous closely occurring w'ellands can be grouped into discrete areas referred to herein as domains. These domains occur throughout the Dar- ling System cither in recurring patterns (eg such as the basin wetlands within the Bassendcan Dune sys- tem) or in unique localities leg such as Bengci Swamp or Lake Pinjar). Domains can most readily'be related to Ihc large scale gcomorphic units Wetlands within each geomorphic system exhibit charac- teristic and distinguishing shapes. Wetlands of the Bassendcan Dunes are usually round or irregular, isolated to coalesced, basins. Wetlands within the Spearwood Dunes arc irregular to elongate or linear basins occurring in linear chains. Most wetlands within the Quindalup Dunes are very small basins in comparison to those in other gcomorphic systems WT^ilands of the Piniarra Plain and Darling Plateau arc channels and associated flais. Introduction The Darling System, comprising the Swan Coastal Plain. Dandaragan Plateau and Darling Plateau, of southwestern Australia (Fig. 1) contains a wide range of wetland lypes. which vary in size, shape, water character istics. stratigraphy and vcgctaiion. These attributes arc determined by regional fcauires such as geology, gcomorphology. soils, climate and hydrology and local physical/chemical processes such as fluvial processes, acolian processes, groundwater flow and karsiification Each wetland is the culmination of these ancestral and modern processes, inherent developmental stratigraphy, and vegetation influences When the factors of gcomorphic setting, origin and water maintenance are common to a group of wetlands, a marked similarity is evident and wetland types can be seen to be related or con- sanguineous. The wetlands of the Darling System can be compartmentalized into localities, or domains of occur- rence. that reflect the distribution of these related w'ctlands. To dale, w'hilc there have been studies of Individual wetlands and wetland systems in the study area (eg Riggcrt 1966, McComb & McComb 1967. Passmore 1 970. Tingay & Tingay 1976, Congdon & McComb 1976. Wetlands Advisoiw Committee 1977. Watson & Bell 1981), (cw studies have placed wetlands into a regional perspective in terms of their categories and the distri- bution of these categories. Servcniy ct a!. (1971). Arnold & Sanders (1981) and Allen (1981) arc van exception to this in that they attempted to categorize wetlands according lo origin (Servcniy er a/ 1971). or into lake lypes (.Allen 1981). or aiiempied lo locate categories of wetlands geo- graphically in the Perth metropolitan region (Arnold & Sanders 1981) However, their studies did not encompass the full variety of wetlands in the Darling system, and did not extend beyond the Penh region. This paper attempts to provide information on ihe numerous and varied cat- egories of wetlands throughout the region of the Darling System (ic identifying related types or. in the terminology of this pai)cr. consanguineous suites), and also attempts to show their distribution in discrete occurrences (or domains). The objcctrves of this paper therefore arc lo: I define Ihe criteria for recognising consanguineous wetlands, 2 identify and describe consanguineous wetlands, and 3 de- lineate domains which contain consanguineous wetland SLiiics within the Darling System between Moore River and Collie River (Fig I). Regional Setting There are a range of regional physical features which arc important to understanding the development of wetland lypes and ihctr distribution m the Darling System. The physical features arc: geology, gcomorphology, soils and gcomorphic processes; climate; and hydrology'. These fea- tures can directly control the development of wetlands, and their variation cither regionally or locally can pro- duce variahiliiy of wetland ivpes. Geology. Gcomorphology. Soils and Gcomorphic Processes The geology and soils of the Darling Svstem have been described by Northcote ct al (1967). McArthur & Bcitenay (I960). Playford et al. (1976). Wilde Sc Low (1978. 1980). Biggs cl al. (1980). and Wilde Si Walker ( ! 982). The Darling Sy stem comprises two distinct geo- logical provinces separated by the Darling Fault. East of the fault arc Prccambrian crystalline rocks of the Yilgarn Block, with local outliers of Phanerozoic sediments (eg Collie Basin) and a variable regolith cover. To the west of 69 Journal of the Royal Society of Western Australia 70 (3). 1988 (I960). the fault is the Perth Basin, a deep trough filled with Phanerozoic sedimentary rocks, extant up to the Quatern- ary. For this paper two regions of the Perth Basin arc dis- tinguished (Biggs e; a!. 1980); 1 Quaternary surficial de- posits. and 2 Mesozoic rocks. Regional geology has a major influence on the pattern of landforms of the Darling System and consequently Churchward & McArthur ( 1 980) used a geological frame- work as the basis of primary classification of landform- soil units. These units, which occur within the study area, are (Fig. I): the Darling Plateau of Precambrian crystal- line rocks and regolilh; the Collie Basin of Permian and vounger sediments; the Dandaragan Plateau of Mesozoic rocks and regolith; and the Swan Coastal Plain of Quat- ernary surficial deposits. Each of these units has a distinctive suite of large, me- dium and small scale landforms and soils as a result of gcomorphic and pedologic processes. In addition, because of their setting and distinctive stratigraphy, the units may influence development of varsing types of small scale hydrologic patterns. These gcomorphic and hydrologic features have a bearing on determining the type and dis- tribution of wetlands within the various gcomorphic set- tings of the Darling System. Since most wetland types are determined by the large and medium scale gcomorphic structure of’ an area a brief description of the gcomorphology at these scales is presented below. The Darling Plateau is a broadly undulating surface with lateritc overlying Precambrian crystalline rocks. It is separated from the Swan Coastal Plain by the Darling Scarp. The Plateau reaches an average height of 400m above sea level, and is dissected by steep sided valleys with incised channels and by sleep sided valleys with broad, flat, ribbon shaped floodplains and small channels. Both the general character of the rock types and the struc- tural trends infiuence to a marked extent the nature and disposition of wetland types on the terrain of the Plateau. Fluvial gcomorphic processes are dominant and conse- quently channel and flat (floodplain) wetland categories predominate. The Collie Basin forms a large topographic depression within the Darling Plateau. It is underlain by laterite- capped Permian and younger rocks. Landscapes have verx low relief ranging from 200m to 250m above sealevci. As a result, although fluvial processes arc domi- nant. channels tend to be broad, shallow and flat-floored, with wide accompanying ribbon floodplains. Mesozoic rocks underlie the Dandaragan Plateau which extends in a splinter block north from Perth, bound to the east bv the Darling Fault and to the west by the Gingin Scarp.'The Dandaragan Plateau also is a lateritc capped surface, but is less dissected than the Darling Plateau, and its surface, some 200m above sealcvel. is gently undulat- ing. Again the character of the rocks and their weathering/ erosion patterns has a major influence on the develop- ment of wetland types. Fluvial processes predominate but because of the relatively low infernal relief, rivers and creeks are not deeply incised and lend to be broad-based with wide floodplains, gently grading upward into valley slopes. Journal of the Royal Society of Western Australia 70 (3), 1988 The Swan Coastal Plain in its enlirciy extends from Dongara to Bussclton (Gentilli & Fairbridgc 1951) but only the southern-central portion is relevant here. The plain generally is of low relief and is some 20-30km wide. The Qualcrnarv' surficial deposits, of Pleistocene to Holocene age and sedimentary and pcdogcnic origin, blanket most of the plain (Playford ct ai 1976), and the major formations therein correspond to the location of the geomorphic elements of McArthur Sc Bettcnay { 1 960) and McArthur & Bartle ( 1 980a.b). There is a marked zo- nation of distinct large scale landforms either arranged parallel to the coast or associated with major rivers. Within each zone there is an array of distinctive medium and small scale landforms. geomorphic processes and hydrologic patterns that are important to the develop- ment of distinct suites of wetlands. The zones, as docu- mented b\ Woolnough (1920). McArthur & Betlenay (1960) and McArthur ^ Bartle (1980a.b), together with their stratigraphic units, from east to west are (Fig. 1 ): • The Ridge Hill Shelf, underlain by lateritc, clay and sand of the Voganup Formation (Low 1971). occurring along the foothills of the Darling scarp. It is dissected by many microscale channels and contains occasional lakes and sumplands. • The Pinjarra Plain, a flat to gently undulating system of alluvial fans, floodplains and various sized channels; the underlying sediments arc the Guildford Formation (Low 1971). The medium and small scale geomorphology is dominated by channels, flats and plains. • The Bassendean Dunes, an undulating plain of low de- graded quartz sand hills and associated hollows varying in relief from 20m to almost flat: the sands arc Pleistocene and arc termed Bassendean Sand (Playford & Low 1972). The medium and small scale geomorphology is alternating hills and basins, and drainage channels generally arc absent. • The Spearwood Dunes and Yoongarillup Plain (McArthur & Bartle 1 980b). comprising large-scale, lin- ear. continuous parallel ridges (c 20m rclicO and inter- vening narrow and steep-sided depressions. The under- lying materials are predominantly Pleistocene aeolianitcs (Tamala Limestone) blanketed by yellow quartz sand. and. to the south underlying the Yoongarillup Plain, yellow quartz sand. Pleistocene aeolianiles and marine limestone. Large scale to me- dium scale landforms arc depressions and gently undu- lating hills. Drainage channels are absent and the pro- cesses of sheet w'ash. basin sedimentation, karslification and subterranean solution are important geomorphic processes m the development of wetlands. • The Quindalup Dunes encompass Holocene dune ridges, beach ridge plains, tombolos and cuspate forelands along the modern coast; the underlying sedi- ments are Safety Bay Sand. Medium and small scale landforms include parabolic dunes (20m high) with as- sociated deflated areas, linear low ridges p-6m high) and associated depressions, and isolated hills and hol- lows. Locally there arc large lakes originally formed by- marine influences. Semcniuk (1983) described additional surficial forma- tions in the Bunbury area; the Leschenault Formation composed of estuarine sediments, and the Eaton Sand that comprises sand ridges co-linear with the Spearwood Dunes. In addition to the above units the contacts between the various geomorphic units constitute important settings for the development of distinct wetland zones or the de- velopment of transition zones. For instance, the junction between Spearwood Dunes and Quindalup Dunes, and the junction between Spearwood Dunes and Bassendean Dunes contain distinct chains of wetlands (McArthur & Bettcnay 1960: Allen 1980). So too the contact of Pinjarra Plain (alluvial fans) and Bassendean Dunes, and the Dar- ling Scarp itself can develop distinct chains of wetlands. r///;/£7/z' The climate of the Darling System is typically Mediterranean (Gentilli 1 972) with north-south and east- west gradients in precipitation, evaporation, temperature and w'ind. The north of the Darling System is semiarid to subhumid. the central part is subhumid to humid, and the south is humid (Gentilli 1972). Rainfall exceeds lOOOmm/yr in southern areas, and along the margin of the Darling Plateau/Darling Scarp. It decreases to c 600mm/yr both in northern areas and east- wards toward the wheatbcll (Gentilli 1972; Bureau of Meteorology 1975). Rainfall is markedly seasonal occur- ring mostly during May to October (Bureau of Meteor- ology 1 973). In response, wetlands of the Darling System exhibit a seasonal variation in water depth. Bow and water quality. The period of Icnvcsl rainfall coincides with the period of maximum evaporation. Evaporation ranges from 2000mm/yr in the north of the Darling System to c 1200mm/yr in the south. Temperature variations also occur throughout the Darling System, increasing slightly to the north and east. Wind IS important in development of sediments, w'eiland margins, and some wetland types. Wind gener- ates waves on standing water of lakes, sumplands and estuaries, and these w-aves effect sediment winnowing, transport and the development of peripheral beachridges. Wind in the coastal zone is important in developing mar- ine and coastal landforms and their accompanying dis- tinctive wetlands. For instance, dune blowouts developed by wind can form into wetlands; swales in beachridge plains may also develop distinctive wetlands; and, at the large scale, coastal landforms such as barrier dunes (Semcniuk 1 985 ) develop and protect large scale wetlands and estuaries. Winds of the Darling System are controlled by eastward migrating anticyclonic pressure cells (Gentilli 1972) and landbreeze - scabreeze systems. Inland areas experience winds from the southeast, east and northeast in summer and. during the winter w-hen water levels of wetlands are elevated, they receive light and variable wind mainly from the eastern and western sectors, with storms from west and northwest. Coastal areas have relatively caim winds during winter, interrupted by storms emanating mainly from northwest and west; during summer the landbreeze - seabreeze system controls the wind, with landbreezcs emanating from southeast, east and northeast and seabreczes emanating from southwest and south (Scarle & Semcniuk 1985). Hydrology The aspects of hydrology important to understanding the development and maintenance of wetlands are re- charge mechanisms, storage systems, discharge mechan- isms, longevity of water retention and water quality. These differ between wetlands located in the various geomorphic settings, and even between wetlands within 71 Journal of ihc Royal SocietN’ of Western Australia 70 (3). 1988 the same gcomorphic selling, and this can influence the development of different types of wetlands and their bio- logical response. For instance, variability of rainfall can effect volume of surface water, its quality, and type of discharge. Recharge of water into wetlands may be the result of di- rect precipitation, water table rise, groundwater discharge from adjoining areas, or surface runoff, all of which arc seasonally variable. 'I'hc type of recharge into a wetland may be dependent not only on the general hydrological setting but also on the local gcomorphic selling and the wetland stratigraphy. Basins with clay Hoors. forexample. can pond meteoric w'atcr or tcllcuric water discharged from adjoining groundwater mounds, whereas basins with sandy floors in the same gcomorphic and hydrologic selling may develop into different types of weilands. A single hydrological process can produce a range of differ- ent weiiand types, or can produce a spectrum of inter- related wxnland types, because of the variability of landform. stratigraphy and suhsiraies upon which the hydrologic process interacts; eg groundwater seepage re- sults in development of basins in one area and in the de- velopment of creeks in anoihcr. Discharge mechanisms may include vegetation- induced evapo-transpiraiion. direct evaporation, run off. overHow discharge or gravitational pcrcolaiion/infil- iraiion. The relative importance of each of these is related to climatic setting, type of vegetation, gcomorphic setting and stratigraphy. A summary of potential recharge and discharge patterns of wetlands is presented in Fig. 2. Figure 2 Idealized diagram illustrating a range of recharge and discharge mechanisms that maintain wetlands. 72 Journal of the Royal Society of Western Australia 70 (3), 1988 The most obvious hydrological pattern of the Darling System is the seasonality of dynamics and variability of water salinity* due to seasonal rainfall. Volume of surface water in wetlands increases rapidly with the onset of win- ter rainfall and is maintained by a nsc in groundwater levels through to Oclober-November. Thereafter the amount of surface water is reduced by drainage and evap oration, which draws upon groundwater from lower levels and concentrates salts in reduced volumes of waler. caus- ing salinity to increase (Allen 1976. 1981). Biggs ct ill. { 1980) summarized the hydrogeology of the Darling System, and Allen (1976. 1 98 1) described much of the relevant hydrology of the Swan (Coastal Plain as it relates to wetlands. These authors identified the major groundwater storage systems in the region. Biggs ct a! ( 1 980) recognized within the Darling System a number of groundwater zones; a Darling Plateau zone, a Collie Basin zone, and a variable range of zones within the Swan Coastal Plain, The hydrological functions within an area that determine recharge, maintenance, or discharge of water from wetlands can be broadly related to large scale gcomorphic setting and local stratigraphy and iheir re- lationship to the regional hydrologic pattern In the Darling Plateau, for instance, sleep surface gradi- ents and impermeable surface materials result in rapid runoff and channelling, with devclopmcnl of short-lived creek systems or wetland valleys sustained by seepage and/or base llow' On the Dandaragan Plateau, flatter slopes and lower runoffraies result in an extended period of water storage or accumulation in w-ctland valley s. On the Swan Coastal Plain there is a predominance ofbasins and consequently fluvial discharge is not important ex- cept on the Pmjarra Plain. Meteoric input, discharge from groundwater mounds, or water table rise arc the main mechanisms of providing water into these basin wetlands. Surface run off. meteoric input and groundwater dis- charge (seepage) contribute to channel wetlands on the Pinjarra Plain, \llcn (1976. 1981) noted, for instance, that many wetland basins of ihcSwan Coastal Plain are in hydraulic connection with the water table, but also noted that rainwater can be ponded in the wetlands within the Bassendean Dunes. The results of Allen ( 1 976. 1981) can be extended to the basin wetlands of the Qumdaiup Dunes and the Pinjarra Plain where they arc underlain by hardpans oflaicritc. iron-cemented sand, clay and cal- careous mud. Allen (1076. 1981) recognized some wetlands as dis- charge basins for localized springs and broad areas of seepage. These emissions and Bow lines occur where sleep groundwater gradients exist, or where a juxtaposition of tw'o facics with diffcrcni transmissivity occurs (eg Six Mile Swamp. Nine Mile Swamp. Lake Pinjar. Bibra chain). The hydrologic functions and behaviour along junctions of the various gcomorphic units thus can result in distinct wetland belts or chains. These junctions, such as those between all the main gcomorphic elements of Mc.Arihur & Belicnay ( I960), may be zones either of dis- charge (eg contact between Pinjarra Plain and Bassendean Quindalup Dunes) or ponding (e.g. between Spearwood Dunes and Bassendean Dunes). Analytical Methods and Terminology Fie/dnvrk ciaia ha.se The results of this paper are based on fieldwork and in- terpretation of aerial photographs. Fieldwork included re- connaissance surveys of numerous wetlands throughout the region (Fig. 3). Some 20 east-west transects, numerous road traverses and over 300 sites were included in the field documentation of geomorphology, stratigraphy and water quality. 80 of which were monitored seasonally for 3 years (Fig. 3). Ai these sites geomorphology, strati- graphic history and water maintenance were studied in deiail by topographical surveying, shailo\^ angering and trenching (up to 3.5m), drilling (up to 30m). seasonal water sampling, seasonal water depth measurements, and surface fiow observaTions. The information from fieldwork was supplemented by desk studies of aerial photographs, and aerial photograph mosaics at scales of I ’60 000, I ;20 000 and topographic maps at scales of 1; 100 000. and 1:5 000. covering the en- tire Darling System. Each domain identified in this study was examined and described in the field. Additional in- formation on water quality and water depth of numerous wetlands was obtained from the literature (Riggcrt 1966. Tingay & Tingay 1976. Wetlands Advisory Committee 1977. McComb & McComb 1967, Congdon & McComb 1976. Moore ct al. 1984. Passmore 1970, Allen 1976. 1980. 1981. Hall 1985) Classification To identify consanguineous wetlands it is necessary to apply a standard classification scheme and the classifi- cation of C. A. Semeniuk (1987) is adopted here. This classification utilizes the two primary components of wetlands, the “landform*’ and “wetness" components (Tabic 1). Using subdivisions of cross-sectional wetland geometry there are recognized; basins, channels, and flats. The maps of wetland suites in this paper differentiate wetlands only to this level. Combining wetness and landform attributes results in 7 categories of common wetlands: I permanently inundated basin = lake: 2 seasonally inundated basin = sumpland: 3 seasonally waterlogged basin - dampland: 4 permanently inundated channel - river. 5 seasonally inundated channel = creek: 6 seasonally inundated flat = floodplain: and /seasonally waterlogged fiat - palusplain. The detailed information in Tabic 2 of this paper differentiates wetlands to the level of one of these categories. Waler and landform descriptors (Table I ) are used to further augment the no- menclature of the primary categories and discriminate in- dividual wetlands. The classification as used in this paper is applied to varying degrees of detail. All wetlands can be readily classified as to basin, channel or flat, but the extent of waler permanence is not known in some cases so that classification into one of the 7 primary categories was not always possible. In addition, water salinity and its seasonal variability were not known for every' wetland produced in the maps of this paper. Lake Joon'dalup. for instance, could be classified using the full nomenclature with the full range of descriptors. White Lake. Bcermullah Lake and Bidaminna Lake could be classified as lakes with a partial listing of descriptors, because in these cases the consistency of water quality is not known. On the other hand some of the wetland’ basins in the Bunbury- Binningup area could not be classified further than "basin” and the descriptors of size and shape only could be applied. The scale terms applied to gcomorphic units in this paper tollow' Semeniuk (1986). Salinilv terms are after Hammer ( 1 986). The term estuary is used as defined by Day (1981). The environment of the estuary encompasses lim'netic and lit- toral landforms that, in southwestern Australia, include bays, headlands, reaches, shoals, shelves, spits, tidal flats, deltas, tidal deltas and exchange channels. 73 Journal of the Royal Society of Western Australia 70 (3), 1988 Journal of the Royal Society of Western Australia 70 (3), 1988 Table 1 WETLAND COMPONENTS FOR USE IN CLASSIFICATION Permanently inundated Seasonally inundated WET LAND Water Permanence Cross-Sectional Shape Basin Channel Seasonally waterlogged Fresh Subsaline Hyposaline Mesosaline Hypersaline Brine Water Salinity Poikilohaline Stasohaline Consistency of Water Salinity Size Plan Shape < Megascale Macroscale Mesoscale Microscale Leptoscale < Linear Elongate Irregular Ovoid Round Straight Sinuous Anastomising Irregular 75 Basins Channels Description of wetland suites Journal of the Royal Society of Western Australia 70 (3), 1988 c .2^ V- ta a.^ ■o x; — tj re 2 i— >■ «= c o . tJ ooZ-oaa i£T 3 V? H ij i-* ^ S es 5* c V re > i_i'— c O o CO olos 3 o > — c/l s •o c QO re ■ ^ ii TO <=^j TO o , t. O. i! ^ c c re O.C — u-r. re c o a §I = c c-o. Cl"® t: 3 C ” oo •§ ? “ re ^ u TO TO TO r- 32 C , TO ^ 3i aO TO t/i c_i'“ ■7^ re'-5 2 0 — 2 -c re iA 32 00 o o o2 P'S o ^ TO re e ^ 2 u “2 ta re e|)§f^re8S i_ ;« V V5 < tJ 5^ O ^ t- •TO Si C.O O C o C • O C ao-; eg C UJ i C« C*c - I TO-e 0.2? X ^5 5 c re - O U ‘~ — fli o re gTO®t ^ C TO o. o O W TO c C/JOOTD -.53 8 ^ c 00 O (rt -g 5 o 1- /TO-Ot- j= G - o O a2^ c o = TO TO-^ •oc .5 o ^ re 3 O TO b ^ Ci o re trt uo O. >. c c .•i; C/) ^ TO 23 TO — re c 2 O — TO O C o C.I O “ O ^ ^ re S TO — O o Si o 3ix: > *“ re 3 ^ T3 ^ > p 2 o ci o ^•5 o-= o ii- >. n •o ox; o E iJ TO TO "O >> u. C O ^ o o re 32 >XJ -- o b >• T3 C 3 TO TO >' r- C/)C/2X1 C O *o > re lu O t" TO 2i O. i/I , P P re^ ;« •Oi. a 3 '0 ; o — ' O 3 • 7 O O 3 S O b O 0.0^ /I i/t OO C c ” ^ ul 3 E o re o'^-g a ^c.O-CQ.3g 1 ■« — TO OO _ ^ O A" ?s:E^ ^ = 0*0 O^TO '^i2 re o ■E^'o ■V 60 - X c J TO O re O o .TO 3 C c o D X; 73 — o E i_ o — TO 00 re p li TO P'C 5 >r TO O CD TO E 3-§ SI ,Si=^i E X> X-- f/3 X5 t?i TO C re TO 2 re S £ c 3 - S- ■2=1 I 2-§§-l” &.E — re c ^aaS-^r TO oO . O c 6 P,~J 5“^ ~n £ re.e^ 0-— c. ■g J-| o O ■J: C S s I s s; = Z re> 2 3 ^ ' 2 4> >-> u. :^:2 pO®C > -O c /5 TO ■ '*' . C ID TO TO u -. iitssre^ j'; ^ c/5 — p CO — v5 b e ^ b 0/ "D '!■ Si -3 /) & c c TO ^ t ^ c re 5 j ^ TOTOTO $TOTO re o . o >V— E 2' re ly 3 re ' c. o« o i o.E , C TO rS"^ V! i-g^-2 S€.b a.2 |-•:5« Qf o£ o ^TOO — re- re ° ^ S E ai§a”“ CO TO O 3=?3 So re o c o - 3 oTO a oto _ i. TO E I l|i C c ir“- o b ^ = c j c 00 _ TO 5 ^ 2 J re m-Ei*' E:;^ S OTO X re i re TO 5 _ o re TO — . o o re iL — 1 c ~ Wh XJ i c > re > o o TO re -^TO — 00 CO re cj re t lo 5§S •TO 5 ^ O li Co = 0^0 3 O0 -; O- C TO CO 3 X re iicO re P 6o5 CO O ./ O CO TO V -7 -TO y 00 '-0 =:2 fP 3 c- 3 TO . ■^S C O, re “ o 2 "=TO 8 B OX' re § c 5to £ CO re<_ re O Is CO ^ E = O O 'S a re c 9Z o O-TO- 2P re - 60 ^ c.E g X— b 5-* C re > fc O O TO re 2 00 ^ -3 g ^ Ex ii: ^ re P -.'2 ® c>— F o-TO " 5j, 5 S S.b ^■S 2 “• Si’lgcc' O TO O c— — •?r CJ— .5 TO TO •& «/) o ■ ■ ^ 2 -b c re 3 D ^ CO — it .J^ C CJ H 3^0 £ ^ Ij TO - g g-2 g ^•TO 2 re b ^ o a o c. c Q.TO Cj to C03-TOTO 'p "TO c— — x c o D. >- « - c re TO CO _ CJ S re b c _£ re O- o re ^ ■£ 2 E re S£ - 3 3)i ^ TO E re5 SS^EocS. 8£fe|c3 o o TO re C- 3 ,-b f £ o o^-= o CO X3 C re CO "3 CJ TO •»§ «a 5°. “p CO *o C Otb o t C O. (/) u. a ? E O O TO b "p 3 ^.E Ob; >• ° b 3 CO Cj O CO TO _ cu o r* TO c E <= O • E o re — _ > 60 o CO ■§2 Cj TO a X « £ re _ TO Cj E C E 0^2 76 Journal of the Royal Society of Western Australia 70 (3), 1988 o oa M c c -o 0-- V ra ^ Scg cot u ^ S 2^ o -l-o 3*2 -i' C ii-S t Is « > O i- ~ “c 5-S )£.E 0,5. *“ ™ ^ fc. O -C o ** i: o iS > ' XSE-o oa c m 1- o »3Sl <* O “T -O U ;*5 w c -E c/) u P — o-^ o t- o> t, w "5 'P lE •o «« « ' .3 > C o « :£ o g.^ C4?°‘o ra Si 5 t ;> W Q O u > aail^ ■2 c w .'S 5 >-En u< a 3 O 0 Cl ^ ^ r: cs 1 C 3 S o“ «=i: o ^ c'2 1 *-S «3 5^3 P o o O E i> §£i >^o S u s O- o 3 i/> •2 5 c iS c a o CDc/5 ■fg =!■ OCL 1/1 a. £ oa 4/ IS' o ii ^ S 0> «/) •> £*o S£ Sc ipil iJ « 3 E'S -i a "O-B "2 S ’^’=•5 c c E 3 ii M 3 C O t -3 J . c o .— C5 fc- 1— O P X Oi- 0.0 o.-t « C ”, n. 'C = C3 _ " O 50 '!■' C 3^ i- ^57. o — 4/ > — J a.. S“! S 42 •3rn cX.E-J u O Xi.C p; 3 ■o o T3 C l| 1 ”^ S )- 50 3 S a ■u -2 u. I* 4> C c“ Cj £ n 3'P C P “• O 3 u obp: — •o oocoD c « P OB ft> S.E-g — ■ U 1/5 ” 50 4) c > ,5 3 0. O CQ c/3 to — t./ 3 C/) oO O 0/) ss? s « E J3 p a— c £«>-;o c/3 £ O ■p ;■§ fl ! o -n c/3 , , aj CO - c i o S ,c 5 c ^ ‘'J o 50 c £ OxtCl,cCD Z £2 r-PJ ,5 >- n) c* 5 J. 50 E H JS-OS CD £ CO O 5,. ^ ^'g •p ^2 § 3S OQ E "O 4/ T3 C p o § o. *- p 2,0 £?«t‘ o CS Cj 5rt C ^ •oOoi ■p i£si p "p N o J= C t: — JZ OB 50 O O.JZ U o <- -p Oj — •— 50 c 50 o. C Is-II^'Ss E"o £=« = CTTJ OB ffl *S^' |t3' .5 c C j= o 2 i J= o S S 2: o iX 2 re 5 / f ^ r- P "O S E = o o S'! > 2 re c c O _re M O- C S £ •o c re t/i °il £.?P OB ^ ® £S g JO -o £ ou CJ re ^ c,i'iS •=■5 V c n'O - «)'P re - 03 3 3: o'. o C I- O O re S ' re 00 , c/3 list il - w. 0/ 03 5 «CD ‘P 2^'- JZ (U ^ C-P P.- S "P p o Z re re cn O j ^ 0ic/3rs > < 1 ^ re ^ > U P re o ?. -5 -5< 50 — •“ “o E 3 re o.* C: . — ■ I 3 p s 5 1/5 O ■u V. Ho a P re 5^ o 5/1 3 ^ 2 o P £Pa>v- ,P U 0 4/ 50 U O’OX) JP re § >'"0 a S” c P.,P oj re 1- 1- o P O tM — -p-o c c re n re re p. cu 5/5 5/1 cr £ ««• I- 4^ 1/ 4J O a a E Q<^ S 4/ 4/ ■O P C 3 4/ "O 50 S ffl S ' 2 « 2 2 ^ 4>,C 2-E-' ^ re s 0“ ' S S^aio.E-i re OB reoS c"^ S'S re •S StJ.E p Z^BaO - r-j 5^^ *T U. © u-TO 3 O 5« p 2 re 51 re O P- 2 V5 c -S >.2 r- i_ w re ^ 4/ Sb P. ^ 2 -S re i/) •£ '0 U (U '-' 2 4* i- 2 25 p. .£ 5« re n. > £ p re'^TS P D. 4; c ' 2 ? = =3^ 2 re O 2 P o -g & c "O O 2 0.-P ^re X> Cl •o ^ 4> 1- c o 're >> s-S 50 E ” c p o 2 oo o-E ^ > o o N -•PC 50 c re 4/ 50 3 Q, 1/5 Q. >; 2 ' 2 ?-o re re.c c U a 2 i '2 q; O 4J 3 O V. TJ C E 'So 3B c P 60 l— 5— O O - J= — 4/ ,T3C^ «^Sfflr>jrn 50 77 Journal of the Royal Society of Western Australia 70 (3). 1988 .2 u rS ra ra o o o ca -5 oc > n 3?:§1 ooCi. a ~ is C <«- = -5 " fl S “"2 — ^ V) C C ri . = 'E155 “ C < -o b S a ” ^ O 1° re u ro V5 o o !2 u re c u ofr- • — 4> re S E a C “ — -- re -o '-. re j: c ^ "o. re 2 2? ^ ■o ?a§ c = U. < ^ "D ^ C lA t w- O ^ re oO O > «5 — "-S-'g £ ^ o S “ ~ ‘'5 «/5 — Q. ej ‘A ^ ol E o o = c re a '« ri « 51 S'^xi'5 3 ■?: re c O o pj=aS a: oaa I -i re o >-JC0 1 « c i. I -5 CD J3'-5 - o « N 00 S’ , u <= 2 u N ra 'SL o o - Q re 0-0 o XJ >< ct 2 o ii > 2^ SiS . = == a , 3 is re S rec: o 5 -E-S-E orarerecre — _ «r± j=:i: P o o E o >- ^ a«^ C.U. •0.== c it; c u re Ej2^ JS Q. ('i re-a .2 w C 2?5 «<— C 2 ® t,_re X S — ooji ^ '— '- -c ■E 2 • o '^,2 ^ <^ *U " re 2 iM q.:2 o “■•E ^ _ o 2? ^ — £ 13 re — o re "O = E o t» c c 2 O c x: u o o re X; -J oO X S re r^. - "O e-c C il-s , re -5 T-5 ^ o « ■oiO, c.— 2 «< -- 3 T3 XI re ^ -J ^X'X C ‘reSPre^E— pc® uCDSD " So: pi ^ 4^ -J 3: -•>■ 1^21^ luEujQ-fc- ^ re S3 5j S : 2 tUX = 1 vAi LU sO 3 CD li “2 ‘re ” ^1 h 22 '"''^<2|Sbfe;= re O >2 o ooc c 3 c O'” ^ w ^ j. OO a, ;3 Q sc(£=?i o ^ cJK ” T ^ 2 o 4> ^ 2 aS £ N ,_o c,« 5 u ^ re ” u. C re ^ o O re S3 3 re O > X O' o=c •0.3 c re 2 a a-o P o M c/2 U, 0.1 i- 3 o c ■X re ._ 0.^ re o> 2 ^ uca P 4> O > 0 2 00 . 4/5 , ^ ST OJ ^ O .£ o2 = ® 2 ^ "re — ^ u ^ a 0. '^“2^ re-c^gS h o o X sj . v» c 2 re ‘ ■: V/ w CC O U o a<5 o-c o re ^ ■ — a o E c “* '5b C *-?4 $ _ re Oy, re = X c c t £ is.ai -< Km ^€»6^ = -JsnUJ o X rtX re > re o 3 a o .— tn ^ ■o-^2 OT3 a. 2 o > o c «•- c'3; >j.iS 4; o o .re 3 5 a^-C Or® re i« 'j 5 2 ^ ».— >-. . -o 25 £ . ^ 15 3 £2? _ o f- ^ S a,' II sill i rex — -2 oo^ re S S5S.S g.S o • ^ w ^ o V) O ^ i,E I-' 2 re a X 2 2 S 00 OO oo -D C C re 2 2 • 7 :; ^ a a E 2 2 o 3 2 re o c/5 o ac 3 o Q 2 _ o XI O 4= S P P X P gS oj O CO ^ re It 4 > us -CDc/5rNiC^-OrnCiSc/5 0.5 ii re m ■3 c re P g“ 2-=>= c o C 3 0 = — ® '® 2 E ® c *- 00 re ” S'.® >, re^2 — o 1 : C re a C.J re 2 -V > ® 0- s« o O' P $ X o u a c o a, re _ TO ^ c F2 u'”- C « ” S >--S re Pc^J-^2® ^-'1-2^8 ^2 0^0 3 > O r- T~ X 3 ;i?5P= c = .s-- „ 3 re X > re 3 X.® 5 I - 7' (/) c : re ” .ca u a OB c c re. 2 ^ X c/5 — V! O 1> C 0/ «<>2 S re ! X 4/ *re re a,^ =*S 'fc 3 = S c 2f^§ 1^1 u t/ £; W t/ c o c « u c /5 X aB«« c: a a 02<^i o £i.2 ”£ cx ^ 10-3 2 ^e-re < 3^0 ®2 O *re 5 re a re O-^ X c ? 2 :« a. c 3 X u O 1/i.— — 3 O a >/) X p — 2 a Q. c/i Sx-^ 2 o ^ re O 2 XX o OA ^ I =■“ ,.S22 : 2 3 c!3 ■X2 55 C C S3 X 00 a’ u 3 ?P«’3 ®-^S< c V) re ' o OJ o QT— VJ S re^ c-o « S u-5 =5'ix H « i 2 o .E ^ O u 'O re Cu. 2 o S p 0 2 <-> v-c = 3 re r\ 4 j C/ a ■E=« ^ 5 o ® c c o 3 =
3.^ — •— VO -1 78 Journal of the Royal Society of Western Australia 70 (3). 1988 C 60 > c: ^ OO = •5 1 “^ > O TO I O ^ c/} 2 -n s C o — : ^ = ■O > 3 O o = TO.D-- O ^ E aSi -Ho-2 C V) C u. •“ >. E !U c ^S'E 5-“ 5^1 ^ o S c = eg 2 ^ C ^ ‘^ssi aj TO — > 3 ^ O C'^5 ■ ^ _ trt ;/) W 7=; C O'u = _ ^ tl Q. a « -50 sti S 2 a 6 o 20 ^. ■ •='j .n -o ^ §XJ 50 ^ ? -a r* •= o *=■ ^ o u 2 ■O Q-ti. oai m c/5 = 1 g’&SsI ^cis£ TO — [A T i-acQ^ jj TO C P TO TO £ O aC^ = C TO O 1/5 _ t> TO 1-1 O VI 4> . SO V) C-O 3 O f- c ? no -. 28 g C <- TO JS c8 a o CO J= U Cl . . o "O c/5 c ^2 S 0,2 c-oxiCj c/5 oc UJ TO 3 C 00 c P > TO >» • r — U c □■ o o . CL o ' >'C /] 2 «1-.= o- |£.£ H !£.« = 2 W ^ ^ CJ ^ £|" 3 o >. ^o^a. 3 S TO - c «. •p ciE‘3.2 .2|S«.= TO'5 VI, = ®?^ ?? w ^ eg ^ y H O' CD .52 'u U] 2; >. ^ o t > 5 o o 3 i — O 2 V5 3 0 .XLU ^ i/i ->± ^ o o5 SfO ^ !r O 30 , 3 ,^ 2'-c Q. ^ ,S<^ -O ^ ^ Wiil O’D'O i- Jr — ___ajTO~ TO = = > 3 u O C C O C7 O ■“ TO . , C c’-i eg w r- I £ S-2 Q0T3 2 o TO P-'o >> o O 0 X a c®^ . -s ^ '” C Cl73 VJ TO •^-2^ 8^-= = O c ^ H a-q o 2 o 2 a o TO O’oOc/5 0 c . , 2 >=b V> O Q 2 '^ § 2 c o >.22 3 "O »- rr o 3 U. VI 8 o- ■D 3 O TO So C/5.E o . . ,3 C HU’ o ^ o o o > . S5 O-g P TO 3 t/5 rr TO < 2 _TO OOs^j c O.E ^ O V) ><^.tC ^ C S o2 S3 {j V5 O -Dci’g- ca. TO^u*^ i- V> — *T* 2 > V5 S p2c-i^l8^ TO g toCTS ^-2 HUo S-S H a 8.^ o c o ;ocqq TO c £-42 ■•3 -2 ^ g t !/5 c/5 O 0 ^ > QC 3 0 0 >.>•- C^oc ^ .2 cfc V) O Q 'u— c c 2 3 ■- c « 8 o >.Ev2 3 - 01 - flJ 3 Um (/^ 00 I ^ O U 3 -— •— c ; s ac o O § '-■Hu.'g2;;ri2 1 TO O CL_0 TO c C c 3 E_2 ,.£i j=^ t c TO CL ■$13 U. Sop , 1 - o > OC'u J= O 3 2 O CQ 0-0. o ^ t: o CD O O' coo ^ o .3 rr; 4c 3 1- 73 '•^2 > 73 o c 2 ■35 o ^ S ;g|-- iC "O ! 5 P £ o p S-£-o2,S ^cEsSoolj O O TO 2 ua VI o — 3 : O O TO ,8ixEEe _TO _TO , • D. O. V5 o 73 52 -?^ > o £ p o n 2 CiCcc a u c "2 3 2P0^ .2 o o o o VI > T3 *3 3 vi>3TO.— 5 . E C I — ^ 3 ; P &a|i K c S-= C.2 £ o r 3 3 x: o OcXQ S E O ^ TO > O ^ = ££ o o -* c)5 5^a! o'i = “^£■50 -2c: ^ 52 o ^ O c .hI|I| si & 3 X CL o TO U- 5 73 TO £ o - V5 Vi o c St TO O Sc 73 C ..TO |i c ^ 3 O -lit o 3 O 2?-T} TO . = 3 : TO > ,3 Q. 60 o O 3 ■oxTS 52- £ £ C — O C p Q lie O ^ TO -op ^•-i2 ^oca- c o o S.2 - TO SPf TO Vi • ■O'C >' V- c . > ^ 5 ; 3 = e ^ p rr ? P !r. T3 S o N C C O TO 3 3 cr o a TO o TO 5 n ^ o ad c £ SSSaS Vi VI 73 C 3 3 O oo £ 3 3 •— eg "C toS 2 TO E S ; a Vi (■73,:^ ; O P I o P TO "O £■§£•§-2 I2||a 3 o ^ 73 ^ 4^ 3 > c c ^ ^ " ^g cj eg eg = o*-^£ V! -U Vi TO o e TO TO :t' ©a cc: c O « ■S£ oiQ o .::i o-= O V) 79 Journal of the Royal Society of Wcsiern Australia 70 ( 3 ). 1988 c .2 u pa « ca C3 ,0 j=‘“ o o T3 ■" O T 3 ”D W C 7 ^ c aa « CoS M O c J= = .5 j- ca « c to a cxi e ^ D. .5 o ^ i/i o — rt ;/! X i> ■CO o 5 -^ « ^ tj -J oU C.'*“ "O E o, = c, aoja “^- 2 -oS. 5 H o 3 c O re P3 00 re t TT -r; o o rer c D. P, ^ S. 5 c Q. c re > -— "rj ^ c o TO X - r\ja Ci. -a — o 3 O > - o 3 '^ ^ u re 1/1 C r® tJ O O 3 ? Cl, "O t:‘0 3 re is re ■ x: E' C 2 ® 2' 2 i. S iJ _o .c O 5 S 5 '? E oO i- c re re C 3 O i :t P o Q- ■□ -O 00 re c *a _, — Q 2 ifi re re _ — uj.Eft. OO O o.r: S 5 --« = o 3 re re o ^ S E ■a $ liefi- £ L- 3 0J3^ 13 X £ 5 « U 60 -^ i •i£ u p3 o re 0.0-^ a re -3 > re > «/> X P tf! £i 3 O re fj O. re— 3 C 3 CC /5 g >^- S,^C c SOE-a^ re E « re c/ e|e ■ - S Sj o »_ si s S X o <2 iw •— •o p = re «e- -P re c ^ o o re J/5 ;.i, P'0 ^ S c-g ' X re -— — ■O o 00 . _ ' L_> ^ c !^x oo-O C ■£ ■ LJ wi > o u u o p.i--£ s isei-f X > 3 5 p * o IP — W ;/} = 4^ re 2^C! • r4 rn 'T sD c ^ Efc ,/> o U c c P a-r re---g — Oj • aC 5 .n Ct 3 £ yore CLO o Cl c H ^ 2 y. — a — ttj c*a O - so $ C C 3 W« wn ^ t- re c o -r re > _Q Pc N — E c £ 2 g re E P L- ^ cr ooO- o- — O 2 p o re 3 O'Cc: ay re o D.E g c L. C o U'uc: X >> > ^- u 7: > o Q o "o '"rej2 7 - 3 > o: y O-d"® Da'yX CQ 1^ reOi ■P != ^ CQ a:Ci “ J 3 s|i ct: a: ^ 00 00 c OOC c .E.E E 3 re y S-2 ccre^o?“ c c oo~ = re re re 3 re TO . 3 OO^^i^SXa; etc • rsi r«S re" y^ <0 r^' 06 O' 5 II L. C X 2y o r- re re c fc $x ^ X X «t£-g a :g '-> ox re lPt- ■" H-c X op o ^ Ep2cux o E c — c 3 o y •“ s e.e «■- 2 - Ml X O c t> 3 ™ vj.E Kx 2 u o ■gl ”3 o E _y X E ^ X C ^ re-P o o Qici; oO- — /S c/1 X Q re X o£) -"s g*® c E o re ^ o p OC 2 J 3 Z> E - u • -X — C rs ro .j- zx i 1 ^ ll«£^ y re •* o ”0 c X £ -x S o = ■5 re;^ ■" 00 > >X 3.0 3 x 2 o o U. P5 00 ^ E ooci: 3 ,.. o •Cii c o re 'X 2 s^l o 2 E 00 — 2o ^ ft. g y X ® g V- re re re o j- 2«« c c £-3 X re o re -3 "o F - 2-po re a.E o.ft- I a: - - "O '^ ■‘ E £*0 '1/ X I — ■ 33 .^ S-i • > o ‘"73 „ O o ■O y X y •j i> ^ ‘ * X -3 E 2 y-3 1/ 3 ^ y eJJ - ■£ P-Jf S re 4/ t i y 2 2-- LU^ efti X re y y oo"— o re o l-S O o ea 2 E S - sx-ig 3 ' 5_2 re > re o y y — y re 3 ^ '•'X O a. X o 3 re 5 O ''' O y C fti > CC^ ?c a P O > -3 X CtJ y •= E S'? S E o ’3 y T 3 j~ O c2 c re -r o c/i ? > 0>;X 3 ^ O V) ? S2 O £ I- 2 £ S .C^C 5 re re 7: y a a 2 --^ -3 £ £ o 0.2 3 E o re 00 yXCL y re ye f 2.2 E £ c re — CQ ^ >.0 E^ re a ox - O -P > 2 I .r, O O o o E S-= O O y y ^ ^2 re X y O ft 3 80 Journal of the Royal Society of Western Australia 70 (3)> 1988 Consanguinity The concept of consanguinity intends to convey the no- tion of relationship between wetlands. In the geological literature the term consanguineous or consanguinity (Bates & Jackson 1980) is applied to materials, such as ig- neous or sedimentary' rocks, where a genetic relationship exists: these materials may occur closely associated in space and time and commonly have a similar (geologic) occurrence and similar characteristics. In this paper the term is applied gcomorphically with the same intention /e to denote relationship. Relationship in type and origin of wetlands, if it exists in a given area, is due to a similarity of causative factors and physical setting. Thus, if there is a similarity of climate, hydrology, geomorphology, geomorphic processes and developmental history, it may- be expected that a suite of similar wetlands, or con- sanguineous wetlands, will result. The criteria for assessing consanguinity are used on the assumption that causative factors are inter-related. For instance, geometty^ of wetlands is dependent upon geomorphic setting. Wetland size is related to ancestral geomorphology, wetland origin, evolutionary stage and amount of water. Water recharge and water maintenance mechanisms depend on wetland stratigraphy, geomorphic setting and present hydrologic regime. Wetland water sal- inity also depends on stratigraphy, geomorphic setting and present hydrologic regime. Wetland stratigraphy is related to wetland origin and vegetation, and vegetation depends on water depth, water permanence and soils. The criteria for identifying consanguineous wetlands are: 1 occurrence of wetlands in reasonable proximity to each other, although proximity alone may be no indication of wetland relationship as other factors such as geomorphic processes and hydrologic regime may be- come significant (Fig. 4A. B. D): 2 a similarity in wetland size and shape (Fig. 4A): 3A recurring pattern of similar wetland forms, i.e. a single wetland type predominates, or an assemblage of wetland types predominate (Fig. 4A. B. C): or 3B heterogeneous pattern representing a spectral range of inter-related wetland forms, or an association of dis- similar but genetically related wetlands: these could result where there arc similar underlying causative factors eg fluvial or hydrological processes (Fig. 4C. E, F): 4 similar stratigraphy and hence similar developmental history: 5 similarity of water salinity and its dynamics: 6 similarity of hydrological dynamics (eg whether wetlands arc recharged and maintained by ponding, seepage, surface runoff, groundwater rise: Fig. 4F): and 7 similar origin eg karslification (Fig. 4D). The criteria are applied in sequence as in a dichot- omous key. Each criterion is applied in turn to progress- ively discriminate between wetland types to determine whether suites of wetlands are related. Criteria 1 to 3 can be applied from information obtained from aerial photo- graphs and short field surveys; criteria 4 to 7 require in- creasing amounts of field information. Ideally all the cri- teria should be applied. In some instances the criteria relating to hydrology and salinity dynamics can only be fully applied after at least one year of seasonal sampling. However, much information can also be obtained for cri- terion 5 from short survey water sampling, and for cri- terion 6 from the analysis of stratigraphy and geomorphology (eg water table depth, clay beds, peat beds, drainage lines, seepage lines, paperbaVk vegetated flats) within a context of local hydrologic and regional hydrological patterns. .At one extreme, a suite of consanguineous wetlands may incorporate a system of very closely related wetlands of similar size, shape, water characteristics, soils and stra- tigraphy. At the other extreme another suite of con- sanguineous wetlands may incorporate wetlands that dif- fer in shape, stratigraphy or some other features, but represent a range of inter-rclatcd forms. These forms may be related only genetically, or may represent a spectral range of types (Fig.4). In other words, there may be local scale heterogeneity but the component wetlands of the consanguineous suite are inter-related and linked because of underlying causative factors. Riverine wetlands are an e.xample of this. Riverine wetlands, that is, those wetlands associated with fluvial areas, may consist of channels, bordering floodplains, extensive palusplains and occasional basins (sumplands and damplands), which alternate along the length of the system, all developed/evolved in conjunc- tion, and superficially may be viewed as a group of hetero- geneous wetlands. The whole wetland system, however, has developed as an internally heterogeneous but inte- grated unit. 81 Journal of the Royal Society of Western Australia 70 (3). 1988 Another example of internal heterogeneity within a consanguineous suite is afforded by groups of basin wetlands in geomorphologic settings in which a seasonal water table rise is the principal mechanism of water re- charge. There may be a similarity of size, shape, soils and water quality between the w'ctlands. but because the undu- lating landsurface is situated at \ arious heights above the water table there is developed a variable and random oc- currence of lakes, sumplands and damplands. In this set- ting the wetlands differ only with respect to the longevity of their water, and so develop into an inter-related suite of consanguineous wetlands representing a spectral range from lakes through to damplands. W'here inundation of a broad area of basins has occurred there may be coalescing of the smaller basins into a single large lake. Thus the spec- tral range may incorporate small damplands. sumplands and lakes with the occasional larger lake. In some ex- amples. because the more inundated basins (lakes) have had a consistently different water history', their sediment margins and perhaps shape may have evolved differently and consequently the suite may consist of an assemblage of 2 wetland forms c.g. round medium to large scale lakes, and irregular medium scale sumplands and damplands. An association of wetland types within a consanguin- eous suite may also occur in response to complex geomorphology and hydrology. For example an area with small scale variations from basins to flats and with small scale lenses of clay. sand, muddy sand, or calcrete, may produce a range of wetland types’despite there being only one hydrological mechanism. .Alternatively, a single geomon^hic structure such as a flat may produce several wetland types in response to hydrological variations. Con- sanguinity thus is established on the basis that wetlands occur in the same vicinity with common or inter- related key features. It should bo noted that vegetation is not used as a cri- terion to identify consanguineous suites. Vegetation is considered to respond to underlying physical and chemi- cal factors of a wetland and consequently it is not a pri- mary causative factor of many wetland features. The in- fluence of vegetation, however, is taken into account in that vegetation formations may produce peats and peaty soils w hich arc considered in the analysis of stratigraphy of wetlands. Markedly dissimilar wetlands of course are not con- sanguineous. For instance wetlands within the Bibra Lake chain arc not consanguineous with those that form the V'anchep to Joondalup chain, in that they do not share similarity of size, shape, stratigraphy or mechanisms of water maintenance. In this case they do not even share similar origins. Domains The concept of domain in this paper intends to convey the notion of the occurrence, in discrete areas, of sets of consanguineous w'ctlands. The term is non-genetic but there is the inference that wetlands that occur in these dis- crete areas are influenced by similar major causative fac- tors acting on the areas to produce consanguineous wetlands. Recognition of domains rests on identifying lo- calities of consanguineous wetlands. The first step in this procedure ts to identify wetlands in the same geomorphic selling. Thereafter it is necessary to isolate those tracts of landform that have wetlands with similar geometry, size, spacing and disposition and phototones on aerial pho- tography. A domain boundary then is drawn around a set of consanguineous wetlands. Results* this study Types of consanguineous wetlands Based on the criteria described above, some 42 types of consanguineous wetland suites are recognized in the Dar- ling System. These suites are named according to a geo- graphic locality where the given suite is best developed. The consanguineous suites have been identified by their aerial photograph patterns within a context of geomorphic (landform) setting. Thereafter further differ- entiation of suites was based on field surveys To determine stratigraph). hydrologic patterns, w'ater quality, medium and small scale landform patterns and geomorphic pro- cesses. Examples of consanguineous wetland suites arc il- lustrated in Fig. 5. These examples are drawn from 1 0km x 10km areas and illustrate the range of w'ctland types, their size and shape and disposition within each of the type examples of a nominated wetland suite. As such, they provide pic- torial information on the geometric features of each suite to enable ready discrimination between them. Table 2 presents detailed information on the characteristics of the various consanguineous wetland suites. As such, it pro- vides more specific information of the features of each suite to enable further discrimination between them, if used in conjunction with the criteria. Figure 5 Examples of the 42 consanguineous suites of wetlands (see Table 2 tor description of the characteristics of each suite). Each of these maps are drawn from 10km x 10km areas’ on 1:60 000 aerial photographs. 82 Journal of the Royal Society of Western Australia 70 (3), 1988 Many of the suites correlate strongly with the gcomorphologic systems described by McArthur & Bettenay (I960), which is not surprising since the ge- ometry and water characteristics of wetlands in general re- flect gcomorphic selling, geomorphic processes, hy- drology and geomorphic hislor>'. Therefore the descriptions of the suites that follow arc presented within the broader scale categories (or framework) of the large scale geomorphic units of the Darling System { = gcomorphic elements of McArthur Sc Bettenay 1 960). The wetland suites are described in groups representative of the gcomorphic elements and the interfaces between the elements. In all, there arc 14 broad categories of geomorphic elements and their interfaces that provide the framework for description of the consanguineous wetland suites: from west to east these are: 1 Quindalup Dunes 2 Quindalup Dunes-Spearwood Dunes, or Quindalup Dunes-Yoongariliup Plain interface. 3 Spearwood Dunes. 4 Yoongarillup Plain. 5 Spearwood Dunes-Bassendean Dunes interface. 6 Bassendean Dunes. 7 Bassendean Dunes/Pinjarra Plain transition zone, or Bassendean with fluvial features. 8 Pinjarra Plain. 9 Estuaries. 10 Coastal plain rivers U Darling Plateau. 12 Darling Plateau/Dandaragan Plateau interface. 13 Dandaragan Plateau. 14 Collie Basin The wetland suites that are within these categories of gcomorphic setting are listed in Table 3. The distribution of consanguineous wetlands in domains throughout the Darling System is shown in Fig. 6. An idealized illus- tration of the distribution of the consanguineous wetland suites m relationship to the regional-large scale geomorphic framework is shown in Fig. 7. 83 Journal of the Royal Society of Western Australia 70 (3). 1988 84 Journal of the Royal Society of Western Australia 70 (3). 1988 Table 3: List of suites & symbols correlated with main geomorphic units of the Darling System Geomorphic Abbreviation Consanguineous Abbreviation Map No. setting of geomorphic wetland suites used in paper in setting used (Fig. 6) Fig. 5 in paper Quindalup Qu Cooloongup Qu.l 1 Dunes Becher Qu.2 2 Peelshurst Qu.3 3 Quindalup- Yoongarillup Interface Q/Y Preston Q/Y.l 4 Spearwood S Yanchep S.l 5 Dunes Balcatta S.2 6 Coogee S.3 7 Siakehill S.4 8 Yoongarillup Y Clifton Y.l 9 Plain Kooallup Y.2 10 Spearwood/ S/B Bibra S/Bl 11 Bassendean Interface Hamden S/B2 12 Bassendean B Pmjar B1 13 Dunes Gnangara B2 14 Jandakot B3 15 Riverdale B4 16 Bassendean/ Pinjarra B/P Beermullah B/Pl 17 Transition Mungala B/P3 18 or Muchea B/P2 19 Bassendean with Bennett Brook B/P4 20 Fluvial Features Bengcr B/P5 21 Pinjarra Plain P Keysbrook PI 22 Estuaries E Moore Estuary El 23 Swan Estuary E2 24 Peel-Harvey Estuary E3 25 Leschenaull Estuary E4 26 Coastal Plain R Moore River R1 27 Rivers* Swan River R2 28 Ellen Brook R3 29 Goegrup R4 30 Dandaragan Dp Red Gullv Dpi 31 Plateau Coorang Dp2 32 Clewley Dp3 33 Mogumber Dp4 34 Dandaragan Plateau Dp/D Wannamal Dp/D 35 Darling Plateau Interface Darling Plateau D Walgunga D1 36 Little Dardanup D2 37 Harris River D3 38 Nalycrin D4 39 Holham D5 40 Brockman D6 41 Collie Basin C Schotts Cl 42 * Coastal plain rivers are equivalent, in pan, to the Pinjarra Plain of McAnhur and Betienay (1960). From Figure 7 it is evident that basin wetlands domi- nate the Quindalup, Spcai*wood, Yoongarillup and Bassendean units. Basins arc replaced by channels and flats in the Pinjarra Plain unit and its transition with the Bassendean unit. The flats are often associated w ith chan- nels and extend for some distance from them, but flats may also occur where channels are absent. Estuaries form discordant water bodies across the large scale geomorphic units. The scarp of the Darling Plateau is marked by in- cised microscale channels, which derive from one of five channel associations that dominate the Darling Plateau. This pattern is interrupted to the north of the system by the Dandaragan Plateau, which is dominated by basin wetlands that are very shallow and grade into flats associ- ated with microscale creeks. The Collie Basin occurring in the south Darling Plateau provides another contrast with linear bifurcated shallow floodplains and sumplands. The consanguineous wetland suites that arc common and recur in domains throughout the Darling System in- clude the Gnangara Suite. Jandakol Suite, Keysbrook Suite and wetlands of the Darling Plateau. Others are less common but nonetheless may still recur throughout the region {eg Nalycnn Suite), while others arc regionally unique features (eg each of the estuaries. Kooallup Suite, and Yanchep Suite). The consanguineous wetland suites of the Spearwood/ Yoongarillup Plain system. Quindalup Dunes, Quindalup Dunes- Spearwood Dunes interface and Collie Basin sys- tems also tend to be unique and restricted to single do- mains. In the Spcarw'ood Dune system north of Mandurah for instance, the system of wetlands differen- tiates into 4 separate suites, each occurring in its own single domain: the Yanchep Suite, the Balcatta Suite, the Coogee Suite and the Stakchill Suite, indicating that these wetlands although superficially similar, in that three of the four groups tend to be linear or chain systems, strictly arc incomparable (Table 2). Each of the estuarine systems also qualifies to be recognised as separate suites and consequently each domain of the estuarine wetlands in the Darling System is regionally unique. The consanguin- eous wetland suites of the Bassendean Dunes. Bassendean Dunes/ Pinjarra Plain transition, the Pinjarra Plains and the Darling Plateau on the other hand are most common and recur throughout the Darling System in several separ- ated domains. Corrdaiion with soil/landforrn units The relationship (or correlation) between broad cat- egories of wetlands of this study w ith the subdivisions of the Darling System into geomorphic elements is well pro- nounced. This relationship underscores the strong influ- ence of large scale and medium scale landform associ- ations and their geomorphic processes in determining type and distribution of consanguineous wetland suites. However, it is apparent that a number of separate do- mains can occur w'ithin a given geomorphic clement. In the Quindalup Dunes there are 3 suites; the Spearwood Dunes and Yoongarillup Plain have 6. the Bassendean dunes have 4. and the Pinjarra Plain has only one. The Darling Plateau contains 5 suites. The occurrence of con- sanguineous wetland suites within each of the geomorphic clcmcnis and ihcir interlaces is shown in Table 3. The correlation of domains of consanguineous wetland suites with all of the landform soil units of Churchward & McAnhur (1980) is not so well marked. Certainly in the Darling Plateau there is a strong correlation between the wetland suites D 1. D2. D3. D4 and D5 with Ihelandform- soil units of Churchward & McArthur ( 1 980), but on the Swan Coastal Plain the finer scale subdivisions of the geomorphic elements such as Bassendean Dunes by Churchward & McArthur ( 1980) do not generally corre- late with the distribution of the domains of this study. Some specific geomorphic units such as Quindalup Dunes have not been further subdivided into finer scale landform-soil units by Churchward & McArthur (1980) but this same unit differentiates readily into a number of different wetland domains. The subdivision of Karrakatla soil and Coilesloe soil associations within the Spearwood Dunes also does not correlate with any of the divisions of wetland domains identified in this study. 85 Journal of the Royal Society of Western Australia 70 (3), 1988 diagram. Discussion The results of ihis paper may have direct application to regional studies and regional assessments of wetlands. The recognition of consanguineous suites of wetlands in discrete domains can provide a perspective of groups of wetlands as part of the Darling System. The conclusion that wetlands can be grouped as similar, related types may be used in comparative studies. The assessment of the representativeness of wetlands within a geomorphic unit or throughout the region, and the regional and local sig- nificance of wetlands, are important in comparative en- vironmental studies, particularly in management of wetlands. This assessment can be based on domain infor- mation to determine whether a wetland type is wide- spread and common, or unique. Thus the approach using consanguineous wetlands and their occurrence in do- mains provides a primary basis for that comparison. The approach using domains can also lorm the basis for com- parative studies of specific wetland features such as veg- etation cover, faunal use of wetlands, and similarity of geomorphic and hydrologic processes that have formed and maintain wetlands, in that it may be assumed that the studies would be intentionally based specifically either on similar or dissimilar wetlands. Acknowledgenmits: 1 thank Dr V. Semcniuk for construc- tive discussion and critical review through various stages of manuscript production, and Dr D. Glassford. I. LeProvost. Dr A. Tingay and P.A.S. Wurm for construc- tive comments on the final manuscript. Research for this paper was supported by VCSRG Ply Ltd. Research and Educational Consultants: assistance in manuscript prep- aration was provided by LeProvost, Semeniuk and Chalmer. Environmental Consultants. Journal of the Royal Society of Western Australia 70 (3), 1988 References Allen A D 1976 Outline of the hydrogeology of the superficial formations of the Swan Coastal Plain. W Aust Geol Surv Ann Repi 1975, 31-42. Allen A D 1980 The hydrogeology of Lake Jandabup. W Aust Geol Surv Ann Kept 1980. 32-40, Allen AD 1981 Goundwater resources of the Swan Coastal Plain near Perth Western Australia In; Groundwater Resources of the Swan Coastal Plain (ed B R Whelan) Proceed Symp CSIRO Dtv Land Res Manag/WA State Comm Water Res Found Aust. Perth 1981, 29-74. Arnold J & Sanders C 1981 Wetlands of the Swan Coastal Plain In: Groundw'aier Resources of the Swan Coastal Plain (cd B R Whelan) Pro- ceed Symp CSIRO Div Land Res Manag/WA Stale Comm Water Res Found Aust. Perth 1981. 81-95. Bales R L & Jackson J \ ! 980 Glossary of Geology. American Geological In- stitute. Virginia. Biggs E R Leech R £ J & Wilde S A 1980 Geology, mineral resources and hydrogeology of the Darling System. Western Australia In; Atlas of Natu- ral Resources Darling System Western Australia. Dept Conserv & En- viron. Univ W Aust Press, 3-20. Bureau of Meteorology 1 973 The climate and meteorology of West Australia In; West Ausc Year Book 12. Melbourne. 36-54. Bureau of Meteorology 1975 Climatic Averages, Western Australia. Aust Govt Pub! Serv. Canberra. Churchward H M & McArthur W M 1 980 Landformsand soils of the Darling System. Western Australia In: Atlas of Natural Resources Darling Sys- tem Western Australia. Dept Conserv & Environ. Univ W Aust Press. 25-33. Clarke E de C Pnder R T & Teicherl C 197 1 Elements of Geology for Aus- tralian Students. Univ W Aust Press. Congdon R A &, McComb A J 1976 The nulnents and plants of Lake Joondalup. a mildly eulrophic lake experiencing large seasonal changes in volume. J R Soc W Aust 59: 14-23. Day J H (ed) 1 98 1 Estuarine ecology with particular reference to South Af- rica. Balkema. Rotterdam. 1-6. Gentilli J 1972 Australian Climate Patterns. Nelson Academic Press, Melbourne. Gentilli J & Fairbridge R W 1951 Physiographic diagram of Australia. Geo- graphic Press. Columbia University. N Y. Hall J 1 985 The hydrogeology of Lake Mariginiup. Penh. Western Australia. Geol Surv W Aust Rept 14. Prof Papers for 1983, 1-13. Hammer U T 1986 Saline Lake Ecosystems of the World. Junk. Low G H 1971 Definition of two new Quaternary formations in the Penh Basin. W Aust Geol Surv Ann Rep 1970. 33-34. McAnhur W M & BanlcG A 1980a Landformsand soils as a basis for Urban Planning in the Perth Metropolitan Nonh-W'^esi Corridor. Western Aus- tralia. CSIRO Aust Div Land Resour Manag Ser. McAnhur W M & Bartle G A 1980b Soils and land-use planning in the Mandurah-Bunbury coastal zone. Western Australia. CSIRO Aust Div Land Resour Manag Ser No 6. McAnhur W M & Beiienay E 1 960 The development and distribution of soils of the Swan Coastal Plain. Western Australia. CSIRO Soil PubI No 16. McComb J A & McComb A J 1967 A preliminary account of the vegetation of Loch McNess. a swamp and fen formation m Western Australia. J R Soc W Aust 50: 105-112. Moore L Knott B & Stanley N 1984 Stromatolites of Lake Clifton. Western Australia: Living structures representing the ongin of life. Search, 14: 309-314. Nonhcoie K H Beiienay E Churchward H M & McAnhur W M 1967 Allas of Australian Soils. Sheet 5. Perth-Albany-Espcrance Area, with explana- tory data. CSIRO Aust Melbourne Univ Press. Passmore J R 1970 Shallow coastal aquifers in the Rockingham district. Western Australia. Water Res Found Aust Bull 18 Playford P E Cockbain A E & Low G H 1976 Geology of the Perth Basin. Western Australia. Geol Surv W Aust Bull 124. Playford P E & Low G H 1 972 Definitions of some new and revised rock units in the Penh Basin. W Aust Geol Surv Ann Rep 1971, 44-46. Riggen T L 1966 Wetlands of Western Australia. Dep Fisheries & Fauna W Aust. Scarle DJ &Semcniuk V 1985 The natural sectors of the inner Roiinest Shelf coast adjoining the Swan Coastal Plain. J R Soc W Ausi 67:1 16-136. Semeniuk C A 1987 Wetlands of the Darling System - a geomorphic approach to habitat classification. J R Soc W Aust 69; 95-111. Semeniuk V 1983 The Quaternary history and geological history of the Australind-Leschcnaull Inlet area. J R Soc W Aust 66: 7I-8T Semeniuk V 1985 The age structure of a Holocene barrier dune system and its implication for sea level history reconstructions in southwestern Aus- tralia. Mar Geol 67: 197-212. Semeniuk V 1986 Terminology for geomorphic units and habitats along the tropical coast of Western Australia. J R Soc W .Aust 68: 53-79. Servenly Owen & Pirrot 1971, cued in Clarke Edc C Pnder RT Teichen C, Elements of Geology For .•\ustralian Students, Univ WA Press. Tingay ,A &Tingay S R 1976 The Wetland of System 6. Dept Conserv & En- viron Bull 28. Watson L E & Bell D T 1981 The ecology of Star Swamp and surrounding bushlands. North Beach. Western Australia. J R Soc W Aust 63: 103-1 1 7. Wetlands Advisory Committee 1977 The Status of Wetland Reserves in Sys- tem 6. Report of Wetlands Advisory Committee to the Environmental Protection Auihonty. Dept Conserx' &. Environ. Wilde S A & Low G H 1978 Perth, Western Australia. W Aust Geol Surv 1:250 000 geological map senes and explanatory notes. Wilde S A & Low G H 1 980 Pmjarra. Western Australia. W Aust Geol Surv 1:250 000 geological map series and explanatory notes. Wilde S A & Walker I W 1982 Collie. Western Australia. W Aust Geol Surv 1:250 000 geological map series and explanatory notes. Woolnough W G 1920 The physiographic elements of the Swan Coastal Plain. J R Soc W Aust 5; 15-19. 87 INSTRUCTIONS TO AUTHORS The Journal publishes {after refereeing) • papers dealing with original research done in Western Australia into any branch of the natural sciences; • papers concerning some biological or geological aspect of Western Australia; • authoritative overviews of any subject in the natural sciences, integrating research already largely published in the more specialized national or international journals, and interpreting such studies with the general membership of the Society in mind; • analyses of controversial issues of great scientific moment in Western Australia. Prospective authors of papers in the last two categories should consult the Hon. Editor for further advice. Contributions should be sent to The Honorary Editor, Royal Society of Western Australia, Western Australian Mu- seum, Francis Street, Perth, Western Australia, 6000. Publication in the Society’s Journal is available to all categories of members and to non-members residing outside Western Australia. Where all authors of a paper live in Western Australia at least one author must be a member of the Society. Papers by non-members living outside the State must be communicated through an Ordinary or an Honorary Member. Submission of a paper is taken to mean that the results have not been published or are not being considered for publication elsewhere. Free reprints are not provided. Reprints may be ordered at cost, provided that orders are submitted with the return galley proofs. Authors are solely responsible for the accuracy of all information in their papers, and for any opinion they express. Manuscripts. The original and two copies must be submitted. They should be typed on opaque white paper with double- spacing throughout and a 3 cm margin on the left-hand side. All pages should be numbered consecutively, including those carrying tables and captions to illustrations, which appear after the text. Illustrations, both line drawings and photographs, are to be numbered as figures in a common sequence, and each must be referred to in the text. In composite figures, made up of several photographs or diagrams, each of these should be designated by a letter (e.g. Figure 2B), To avoid risk of damage to original figures, authors may retain these until after the paper is accepted. The copies of the figures accompanying the manu- script must be of good quality. Authors are advised to use the most recent issue of the Journo! as a guide to the general format of their papers. Words to be placed in italics should be underlined. To facilitate editing, papers must be accompanied by a table of contents, on a separate sheet, showing the status of all headings. References must be set out as follows; Paper Jackson A 1931 The Oligochaeta of South-Western Australia. J R Soc W Aust 17:17-136. Twigg L. Major J D & Kotula R 1983 the influence of fluoroacetate producing plants upon seed selection by seed harvesting ants. Mulga Res Centre W Aust Inst Technol, Bentley, Ann Rep 6:75-80. Book Jacobs M R 1955 Growth Habits of the Eucalypts. For Timb, Bur, Canberra. Chapter in book Deli J 1983. The Importance of the Darling Scarp to Fauna. In: Scarp Symposium (ed J D Majer) W Aust Inst Technol, Bentley, 17-27. The Title should begin with a keyword. The Abstract should not be an expanded title, but should include the main substance of the paper in a condensed form, The metric system {SI units) must be used. Taxonomic papers must follow the appropriate international Code of Nomenclature, and geological papers must adhere to the International Stratigraphic Guide. Spelling should follow the Concise Oxford Dictionary. Authors should maintain a proper balance between length and substance, and papers longer than 10 000 words would need to be of exceptional importance to be considered for publication. Authors will be charged page costs (currently $30 per page) if papers exceed 8 printed pages. Short papers (2-4 printed pages) are particularly sought as these often ensure full use of the 32 pages available in each part. Illustrations. These should be prepared to fit single or double column widths. Illustrations must include all necessary letter- ing, and be suitable for direct photographic reduction. No lettering should be smaller than 1 mm on reduction. To avoid un- necessary handling of the original illustrations, which are best prepared between 1.5 and 2 times the required size, authors are advised to supply extra prints already reduced. Additional printing costs, such as those for folding maps or colour blocks, will be charged to authors. Supplementary Publications. Extensive sets of data, such as large tables or long appendices, may be classed as Sup- plementary Publications and not printed with the paper. Supplementary Publications will be lodged with the Society’s Library (C/- Western Australian Museum, Perth, WA 6000) and with the National Library of Australia (Manuscript Section, Parkes Place, Barton, ACT 2600) and photocopies may be obtained from either institution upon payment of a fee. JOURNAL OF THE ROYAL SOCIETY OF WESTERN AUSTRALIA CONTENTS VOLUME 70 PART 3 1988 Page Floristic reconnaissance of the northern portion of the Gregory National Park, Northern Territory, Australia D M J S Bowman, B A Wilson & P L Wilson 57 Consanguineous wetlands and their distribution in the Darling System, Southwestern Australia C A Semeniuk 69 Edited by I Abbott Registered by Australia Post — Publication No WBG 0351 No claim for non-receipt of the Journal will be entertained unless it is received within 12 months after publication of Part 4 of each Volume The Royal Society of Western Australia, Western Australian Museum, Perth A59038/8/88— 700 GARRY L. DUFFIELD, Government Printer, Western Australia VOLUME 70 PAFiT 3