classification of australian clastic …...calculated for seven different clastic coastal...

JOURNAL OF SEDIMENTARY RESEARCH, VOL. 72, NO. 6, NOVEMBER, 2002, P. 858–870Copyright q 2002, SEPM (Society for Sedimentary Geology) 1527-1404/02/072-858/$03.00

CLASSIFICATION OF AUSTRALIAN CLASTIC COASTAL DEPOSITIONAL ENVIRONMENTS BASED UPONA QUANTITATIVE ANALYSIS OF WAVE, TIDAL, AND RIVER POWER

P.T. HARRIS1, A.D. HEAP2, S.M. BRYCE3, R. PORTER-SMITH1, D.A. RYAN3, AND D.T. HEGGIE3

1 Geoscience Australia and Antarctic CRC, University of Tasmania, GPO Box 252-80, Hobart TAS 7001, Australia2 Geoscience Australia, School of Geography and Environmental Studies, University of Tasmania, GPO Box 252-78, Hobart TAS 7001, Australia

3 Geoscience Australia, GPO Box 378, Canberra ACT 2601, Australiae-mail: [email protected]

ABSTRACT: A statistical assessment of wave, tide, and river power wascarried out using a database of 721 Australian clastic coastal deposi-tional environments to test whether their geomorphology could be pre-dicted from numerical values. The geomorphic classification of eachenvironment (wave- and tide-dominated deltas, wave- and tide-domi-nated estuaries, lagoons, strand plains, and tidal flats) was establishedindependently from remotely sensed imagery. To our knowledge, sucha systematic numerical analysis has not been previously attempted forany region on earth.

The results of our analysis indicate that a relationship exists betweenthe ratio of annual mean wave power to mean tidal power and thegeomorphic development of clastic coastal depositional environments.Deltas and estuaries are associated with statistically significant differ-ences in mean wave and tidal power. Statistically significant distinc-tions between populations of deltas, estuaries, strand plains, and tidalflats are also associated with river discharge and river flow rate (de-fined as discharge divided by open water area). Our results supportthe hypotheses of previous workers that wave, tide, and river powerexert the principal control over the gross geomorphology and faciesdistribution patterns in clastic coastal depositional environments. Meanvalues and confidence limits of wave power, tide power, and river flowfor the coastal depositional environments targeted in this study mayprovide a basis for the comparison of modern environments as well asconstraints for paleo-reconstructions.

INTRODUCTION

The geomorphic evolution of different clastic coastal depositional environ-ments (including deltas, estuaries, lagoons, strand plains, and tidal flats) iscontrolled by the relative importance of three main factors: sediment supply,physical processes, and sea level. In their reviews of this subject, Boyd et al.(1992) and Dalrymple et al. (1992) provided a synthesis and a ternary classi-fication, emphasizing the geomorphic relationships that exist between differentclastic coastal depositional environments over space and time (Fig. 1A–C). Theconceptual wave–river–tide ternary classification, originally proposed by Co-leman and Wright (1975) and Galloway (1975) for deltas, was adopted byBoyd et al. (1992) and applied to clastic coastal depositional environments(Fig. 1B, C). This new classification uses relative wave, tide, and river powerto position each environment within the ternary diagram. Boyd et al. (1992)and Dalrymple et al. (1992) expanded this simple two-dimensional ternarydiagram into a three-dimensional prism to incorporate the evolutionary rela-tionships between the spectrum of clastic coastal depositional environments,where the z axis represents relative time, with reference to changes in relativesea level and sediment supply (Fig. 1A).

Boyd et al. (1992) and Dalrymple et al. (1992) proposed the ternaryprism (Fig. 1A, B) mainly as a means to portray the relative importanceof the three processes. These workers plotted individual systems within thediagram on the basis of their perceived wave, tide, and river power, ratherthan based on measured values. To improve our understanding of the rel-ative importance of these three key physical processes in governing thegeomorphology of coastal depositional environments, the next logical stepis to quantify the relative power (i.e., sediment transporting capacity) of

waves, tides, and rivers for a statistically significant number of cases.Therefore, the aim of this study was to test the hypothesis of Boyd et al.(1992) and Dalrymple et al. (1992) that wave, tide, and river power arethe major controls on the geomorphology of modern clastic coastal depo-sitional environments. This test was conducted by plotting a population ofcoastal depositional environments having a known geomorphology withina ternary diagram, with axes based on numerical estimates of wave, tide,and river power. The results provide not only the first calibration of theBoyd et al. (1992) and Dalrymple et al. (1992) model but also the firstpublished continental-scale approach to the quantitative classification ofclastic coastal depositional environments. In undertaking this work, we an-ticipate determining boundary conditions characteristic to each environmentin the context of the existing ternary classification (Fig. 1A, B) and estab-lishing the first quantitative approach for comparing coastal depositionalenvironments of the same classification. As such, our approach providesthe basis for identifying clastic coastal depositional environments most sus-ceptible to changes in global wave climates as a result of global warming,and a mechanism for constraining paleo-environmental conditions for an-cient coastal systems discovered in the rock record.

Quantification of Wave, Tide, and River Power

In the coastal zone, sediment erosion, transport, and deposition are physicalprocesses that are closely related to the power of waves and tides (e.g., Swiftand Thorne 1991). The geomorphology of clastic coastal depositional environ-ments is strongly influenced by the relative contribution of waves and tides ingoverning the transport and deposition of available sediment (e.g., Davis andHayes 1984). The ratio of wave power to tide power is thus used to classifyeach environment along the base of the ternary diagram (Fig. 1A).

The energy of a gravity wave (E, J m22) in the ocean is:

E 5 1/8(rgH2) (1)

where r is the density (kg m23) of the sea water, g is acceleration due togravity (m s22), and H is the wave height (m) (Soulsby 1997). The wavepower (units J m21 s21) is the energy times the wave group speed (Cg 5L/T), where L is the wavelength and T is wave period. We have estimatedwave ‘‘power’’ as the energy per wave period (E/T, units J m22 s21) andignored wavelength, which is complex to calculate for shallow-water waves(i.e., ocean tides). However, our ‘‘power’’ estimate accounts for the in-crease in near bed orbital velocity that occurs for decreasing period inwaves of a given height (e.g., Soulsby 1997). Because ocean gravity swellwaves and tides are both wave phenomena, the ratio of wave power to tidepower can be estimated by:

2 2Power ratio 5 wave power/tidal power 5 K[H /T] /[H /T] (2)wave tide

where K is a dimensionless coefficient that corrects for the greater powerof tidal waves over ocean gravity waves.

Calculation of the vertical axis (river power) of the ternary diagram (Fig.1B) is a function of the fluid power (P, N m21 s22) available to transportsediment (Swift and Thorne 1991):

P 5 Ut (3)

859CLASSIFICATION OF AUSTRALIAN CLASTIC COASTAL DEPOSITIONAL ENVIRONMENTS

FIG. 1.—A) Evolutionary classification ofclastic coastal depositional environments (afterDalrymple et al. 1992 and Boyd et al. 1992).The long axis of the three-dimensional prismrepresents relative time, with reference torelative sea level (transgression) and sedimentsupply (progradation). B) The triangular sectionsof the prism reflect relative wave/tide power (thex axis) versus the rate at which the river deliverssediments to the coast (y axis). Deltas occupythe uppermost area of the ternary diagram.Estuaries are located in the intermediatetrapezoidal area and coastal strand plains andtidal flats occupy the base of the ternarydiagram. C) Plan-view maps for idealized coastaldepositional environments, showing therelationships between wave and tidal power,prograding and transgressive environments, anddifferent geomorphic types.

where U is the mean river current (m s21) and t is the bed shear stress (Nm22). Although U is proportional to the river discharge through the cross-sectional area of the river channel, the estimation of t requires informationon the shape of the velocity profile and bottom roughness characteristics(e.g., Sternberg 1972).

Australian Clastic Coastal Depositional Environments

Our study encompasses the clastic coastal depositional environments ofthe entire Australian continent to ensure that a wide range of environmentsis included in the analysis. Because the impetus for this study came aboutas an investigation into the ‘‘health’’ of Australia’s coastal waterways(Heap et al. 2001), the clastic coastal depositional environments consideredin this paper are restricted to those associated primarily with Holoceneterrigenous sediment sources. As such, extensive regions of the coastline,including most of southern and parts of western Australia, where trans-gressive wave–dune systems predominate, have not been not considered.

Information on the wave, tide, and river regimes around the Australiancoast have been discussed in relation to the evolution of depositional en-vironments in numerous previous case studies. Examples include studiesof macrotidal estuaries (e.g., Wright et al. 1973; Cook and Mayo 1978;Harris 1988; Woodroffe 1996); wave-dominated estuaries (e.g., Roy 1984);

strand plains and cheniers (e.g., Roy et al. 1980; Rhodes 1982; Murray-Wallace et al. 1999); tidal flats (e.g., Semeniuk 1981); and deltas (e.g.,Johnson 1982; Lees 1992; Jones et al. 1993; Woodroffe and Chappell1993). However, to ensure that only consistently collected and comparabledata are used in the analysis, we have recalculated the wave, tide, and riverpower for all Australian clastic coastal depositional environments consid-ered in this study.

Previous systematic collations of information have been used to clas-sify Australian clastic coastal depositional environments on the basismainly of climate, freshwater discharge, and tidal range (e.g., Bucherand Saenger 1991, 1994; Eyre 1998; Digby et al. 1998). However, noneof these studies included waves as a classification criterion. The distri-bution of continental shelf zones dominated by swell waves, tides, cy-clone-induced currents, and intruding ocean currents was described byHarris (1995) on the basis of available published data, but this classi-fication was not extended to the coastal zone. From these studies it isclear that Australia’s coastal environment spans a range of climaticzones (tropical to temperate) and contains examples from the completespectrum of clastic coastal depositional environments of varying wave,tide, and river influence, and is therefore an appropriate location tocarry out this study. We acknowledge, however, that by restricting our

860 HARRIS ET AL.

TABLE 1.—Mean and standard deviation of the tide power, wave power, (dimensionless) wave/tide power ratio, mean annual river discharge, and river flow ratescalculated for seven different clastic coastal sedimentary environments around Australia.

Type ofSystem

Tide Power(J m22 s21)

Wave Power(J m22 s21)

Wave/TidePower Ratio

River Discharge(m3 s21)

River Flow(3 1026 m s21) n

Wave-DeltaTide-DeltaWave-EstuaryTide-Estuary

398 6 437828 6 1190177 6 280

2244 6 2220

181 6 19242.8 6 63.4318 6 24741.7 6 55.6

4.10 6 10.50.406 6 0.794

27.3 6 52.90.179 6 0.338

19.2 6 26.933.4 6 67.626.66 17035.5 6 71.0

11.4 6 18.68.45 6 16.23.46 6 10.51.38 6 5.11

8169

14599

StrandplainTidal FlatLagoonOther

696 6 12601730 6 2120416 6 1000

—

127 6 18160.6 6 74.1266 6 202

—

2.686 5.810.450 6 0.81914.1 6 23.2

—

1.68 6 2.331.49 6 2.25

0.248 6 0.259—

6.09 6 11.82.84 6 8.26

0.503 6 1.63—

43273

1159

study to the Australian continent some clastic coastal depositional en-vironments, such as those influenced by glacial processes and thoseassociated with major river systems, are not represented, and our resultsmay not be directly applicable in these situations.

METHODS

Australian Database of Clastic Coastal Depositional Environments

The database used in this study is the Australian Estuarine Database(AED; http://www.ozestuaries.org), which contains physical informationfor 780 Australian coastal waterways (Bucher and Saenger 1991; Digby etal. 1998). All of these waterways were identified as ‘‘estuaries’’ based ona definition similar to that proposed by Pritchard (1967), as: ‘‘semi-en-closed bodies of water and adjacent wetlands which have input both frommarine inundation and terrestrial runoff’’ (Digby et al. 1998, p. 15). For awaterway to be contained in the AED, it also must be large enough to beshown on a 1:100,000 scale topographic map and have a catchment areaof at least 15 km2 (Digby et al. 1998).

These 780 coastal waterways span a range of climatic and oceanographicsettings and display a range of geomorphic types. They include a range ofclastic coastal depositional environments, together with other nondeposi-tional environments such as rocky embayments and channels between is-lands and the coast that are not included in the Boyd et al. (1992) classi-fication (Fig. 1C). Because in the present study we are concerned with thegeomorphology of clastic coastal depositional environments, we haveadopted the definitions of deltas, strand plains, lagoons, and tidal flats assummarized by Boyd et al. (1992) and for estuaries we have adopted thedefinition of Dalrymple et al. (1992).

An independent assessment of the geomorphology of the 780 waterwayscontained in the AED (Bucher and Saenger 1991; Digby et al. 1998) wasundertaken in this study by a visual inspection of maps, nautical charts,LANDSAT-TM images, and aerial photographs (see acknowledgments).Each waterway was classified, using the principles of Boyd et al. (1992)and Dalrymple et al. (1992), as wave- or tide-dominated estuaries, wave-or tide-dominated deltas, lagoons, strand plains, or tidal flats (Fig. 1C).Only the geomorphology visible in the images was used to determine theclassification. In cases where the waterway was judged to be an embay-ment, or a channel between an island and the coast, or where the diagnosticgeomorphic features were ambiguous or unclear, a ‘‘mixed/other’’ categorywas used. This resulted in a total of 721 clastic coastal depositional envi-ronments targeted for further analysis.

Quantitative Wave, Tide, and River Data

To quantify the wave power around the Australian coast, heights andperiods of surface gravity waves were estimated using the Australian Bu-reau of Meteorology’s regional atmospheric model. Surface wind speedestimates generated by this model provided input to the Wave Model,WAM (Hasselmann et al. 1988; Komen et al. 1994) to yield estimates ofmean wave height and period. These data are 6-hourly predictions of sig-

nificant wave height (Hsig.) and mean wave period, gridded at a spatialresolution of 0.18 (11 km), for the period March 1997 to February 1998,inclusive. The gridding procedure employed a weighted inverse distance(weighting 5 4) algorithm, which produces representative interpolationsin situations where the original data are infrequent and irregularly distrib-uted. The mean annual Hsig. and periods were then calculated for eachcoastal waterway by assigning the model grid value closest to the locationof the waterway mouth.

To quantify the tidal power along the Australian coast, the maximumspring tidal range for 423 tide gauges located around Australia (AustralianNational Tide Tables 2000) was gridded at a spatial resolution of 0.18 (11km) using an inverse distance algorithm with a weighting of 4. The max-imum spring tidal range for each coastal waterway was then determinedby assigning the model grid value closest to the mouth. Two tidal periodswere used for the calculation of tidal power: diurnal and semidiurnal. Thedifference between diurnal and semi-diurnal tides is defined in the Austra-lian National Tide Tables (2000) on the basis of the ratio of four majortidal constituents as

semidiurnal 5 (K1 1 O1)/(M2 1 S2) , 0.5, anddiurnal 5 (K1 1 O1)/(M2 1 S2) . 0.5,

where K1 is the lunisolar diurnal, O1 is the principal lunar diurnal, M2 isthe principal lunar, and S2 the principal solar tidal component. The waveand tide power, derived here, are considered to best represent the conditionsat the coast (i.e., at the mouth), and in most cases they probably do notreflect conditions inside the waterway.

We were not able to derive a reliable estimate of river power becausein situ measurements of river currents and boundary shear stress are notavailable for most Australian waterways. However, Digby et al. (1998)estimated the mean annual river discharge for the 780 Australian waterwayscontained in the AED as

3 21mean annual discharge (m s )25 [catchment area (m ) 3 mean annual rainfall (m)

3 runoff coefficient]. (4)

The mean annual discharge was then divided by the open water area ofeach waterway, which was measured from LANDSAT TM images and,where available, from aerial photographs, to calculate river flow (m s21).River flow is proportional (but not equivalent) to river power (Equation 3).

Although we have calculated mean annual values for wave and tidepower and river flow, it is important to note that the discharge of manyAustralian river systems is event-driven (Neil and Yu 1995; Wasson et al.1996). The mean annual river flow does not reflect extreme river powervalues associated with flood events, when most, if not all, sediment trans-port occurs. Because of this, it is probable that the rate and degree ofinfilling of Australian coastal depositional systems by river-derived sedi-ment are more closely associated with the frequency and duration of stormsrather than with average annual sediment yield from the catchments. Al-


FIG. 2.—Distribution of 721 different clastic coastal depositional environments around Australia based on the interpretation of aerial photographs and LANDSAT TMimagery. The 59 ‘‘mixed/other’’ coastal environments are not shown. Five regions are identified, on the basis of the combination and distribution of coastal depositionalenvironments (see also Table 2).

though sediment discharge data are available for about 20 waterways (e.g.,Harris 1995; Wasson et al. 1996) they have not been considered in ouranalysis.

RESULTS

Spatial Distribution of Coastal Depositional Environments

The visual inspection of the waterways resulted in the geomorphic clas-sification of 721 clastic coastal depositional environments using the criteriaof Boyd et al. (1992) and Dalrymple et al. (1992), and 59 cases of ‘‘mixed/other’’ types (Table 1). The data indicate that tidal flats are the most com-mon coastal depositional environment in Australia (n 5 273), followed bywave-dominated estuaries (n 5 145), tide-dominated estuaries (n 5 99),

wave-dominated deltas (n 5 81), tide-dominated deltas (n 5 69), strandplains (n 5 43), and lagoons (n 5 11).

The spatial distribution of these environments around the coast exhibitsa distinct zonation, such that five major coastal regions can be identified:southeast coast; southwest coast; northwest coast; Gulf of Carpentariacoast; and northeast coast (Figs. 2, 3; Table 2). The southeast and southwestcoasts are wave-dominated environments, whereas the northern coastal ar-eas (northwest, Carpentaria, and northeast) are mainly tide-dominated(Figs. 2, 3). Although wave-dominated estuaries occur almost exclusivelyin the southeast and southwest, wave-dominated deltas and strand plainsare widely scattered across the southeast, northeast, Gulf of Carpentaria,and to a lesser extent in the northwest (Fig. 3). This pattern conforms tothe general distribution of wave- and tide-dominated shelf environmentsdescribed by Harris (1995), such that tide-dominated estuaries, deltas, and

862 HARRIS ET AL.

FIG. 3.—Maps showing the distribution of theseven different clastic coastal depositionalenvironments around Australia based on theinterpretation of aerial photographs andLANDSAT TM imagery. Note that thedepositional environments targeted in this studyare restricted to those influenced by Holoceneterrigenous sediment; these environments areabsent across most of southern and parts ofwestern Australia.


TABLE 2.—Number (and percentage) of Australian clastic coastal depositional en-vironments listed by geographical area.

nSoutheast

CoastSouthwest

CoastNorthwest

Coast

Gulf ofCarpentaria

CoastNortheast

Coast

Wave-DeltaTide-DeltaWave-EstuaryTide-EstuaryStrandplainTidal FlatsLagoon

8169

1459943

27311

27 (33%)0

114 (79%)2 (2%)9 (21%)

12 (4%)8 (73%)

1 (1%)0

18 (12%)000

1 (9%)

4 (5%)11 (16%)

051 (52%)

4 (9%)95 (35%)

1 (9%)

14 (17%)24 (35%)

7 (5%)17 (17%)21 (49%)80 (29%)

1 (9%)

35 (43%)34 (49%)

6 (4%)29 (29%)

9 (21%)86 (32%)

0

FIG. 4.—Stability diagram for clastic coastaldepositional environments around Australia.Tide-dominated systems plot on the upper left-hand side of the diagram whilst wave-dominatedsystems plot on the lower right-hand side. The xand y axes are log of wave and tidal powerestimated for 721 coastal depositionalenvironments. The line dividing the two groupswas drawn by eye and has a slope of 3.2.

tidal flats occur mostly adjacent to tide-dominated shelves, wave-dominatedestuaries, lagoons, and deltas, and strand plains occur mostly adjacent towave-dominated shelves (Fig. 2). However, each of the five coastal regionscontains a distinct mixture of coastal depositional environments.

The southeast coast contains 79% of Australia’s wave-dominated estu-aries but also significant numbers of wave-dominated deltas, strand plains,and lagoons plus some tide-dominated environments (Table 2). This mix-ture of environments distinguishes this region from the southwest coast,where there are also mainly wave-dominated estuaries, but only one wave-dominated delta and no strand plains; the southwest coast lacks prograda-tional coastal depositional environments. There are no tide-dominatedcoastal depositional environments on the southwest coast.

The northwest coast is characterized by the largest number of tide-dom-inated estuaries (n 5 51). These are supplemented with significant numbersof tidal flats and some tide-dominated deltas. Wave-dominated environ-ments are rare, with only 9 out of 166 environments being wave-dominated(Table 2).

The Gulf of Carpentaria coast is characterized by the largest number ofstrandplains (n 5 21). This region also has significant numbers of tidalflats, tide-dominated estuaries, and tide-dominated deltas. However, thegreater influence of waves in this region relative to the northwest coast ismanifest by the occurrence of wave-dominated estuaries and wave-domi-nated deltas (Fig. 2; Table 2).

The northeast coast is differentiated from the Gulf of Carpentaria coastby its abundance of both wave- and tide-dominated deltas (n 5 69) butfewer strand plains (Fig. 3; Table 2). The northeast and Gulf of Carpentariacoasts are similar in that the total number of tide-dominated environments

(tide-dominated estuaries, deltas, and tidal flats) is about three times aslarge as the total number of wave-dominated environments (wave-domi-nated estuaries, deltas, strand plains, and lagoons). Hence, although thesecoastal regions are classed as being mainly tide-dominated, there is a great-er mixture of wave- and tide-dominated environments on the northeast andGulf of Carpentaria coasts than in any of the other three regions (northwest,southwest, and southeast; Figs. 2, 3; Table 2).

Wave, Tide, and River Power

A plot of wave and tidal power shows that Australia’s clastic coastaldepositional environments are partitioned into two populations (Fig. 4) thatare distinguished by environments that display wave-dominated geomor-phology (wave-dominated deltas and estuaries, strand plains, and lagoons)and environments that display a tide-dominated geomorphology (tide-dom-inated deltas and estuaries, tidal flats). A line of best fit separating thesetwo populations can be drawn with a slope of 3.2. This slope was used asthe coefficient K in Equation 2 to center the distribution of coastal depo-sitional environments at a wave/tide power ratio of 1 (i.e., Log (W/T ) 50).

Coastal depositional environments that display wave-dominated geo-morphology number 280, with a mean annual wave power of between 1.563 1023 and 1,190 J m22 s21 (average 5 246 6 235 J m22 s21). Tidalpower for these environments ranges from 7.9 to 6000 J m22 s21 with anaverage of 327 6 621 J m22 s21. Coastal depositional environments thatdisplay tide-dominated geomorphology number 441, with a mean annualwave power of between 2.47 3 1023 and 497 J m22 s21 (average 5 53.66 69.2 J m22 s21). Tidal power for these environments ranges from 31.6to 10,900 J m22 s21 with an average of 1,700 6 2,070 J m22 s21.

A statistical t-test analysis of the data demonstrates that the average waveand tide power of wave-dominated environments are significantly different(at 95% confidence limits) from the average wave and tide power of tide-dominated environments (Fig. 4). The differences between the averagewave/tide power ratios for wave-dominated environments (5.23 6 12.9, n5 280) are also statistically significant compared with tide-dominated en-vironments (0.119 6 0.231, n 5 441).

In order to investigate whether there are statistically significant differ-ences in the mean values of tidal power, wave power, river flow, and riverdischarge between the seven clastic coastal depositional environments (Fig.

864 HARRIS ET AL.

FIG. 5.—Histograms arranged in order of decreasing values (from top to bottom) for tide power, wave power, river discharge, and river flow in relation to the occurrenceof difference Australian clastic coastal depositional environments. The occurrence of statistically significant differences between populations of environments in each columnis illustrated in Fig. 6. The range in values specific to each type of environment provides constraints that may be applied to paleoenvironmental reconstructions andinterpretations (see text for discussion).

1B), a one-way analysis of variance (ANOVA) was calculated using thestatistical software package STATISTICAy (Fig. 5; Table 1). The ANO-VA indicated that there are significant differences between category meansfor tidal power, wave power and river flow. The Scheffe’s test was thenapplied as a post-hoc comparison to determine which combinations ofmeans in the ANOVA were significantly different from each other at the95% confidence limit (highlighted cells in Fig. 6). These tests provide astatistical check on the significance of the relationships between the geo-morphology of each of the seven clastic coastal depositional environmentsand wave power, tide power, river discharge, and river flow. The significantdifferences also confirm situations where a positive correlation exists be-tween the independently derived geomorphology and wave, tide, and rivervalues. Interestingly, no significant differences were identified between themean river discharges of the seven coastal depositional environments.

Plotting the data on ternary diagrams (Figs. 7A, B) illustrates a strongseparation of coastal depositional environments along the x axis (wave/tidepower ratio). Tide-dominated environments plot almost without exceptionon the right side of the diagram and wave-dominated estuaries and lagoonsclearly plot on the left side. However, strand plains and wave-dominateddeltas exhibit much scatter around the center of the ternary diagram, in-dicating that they occur where the wave/tide power ratio is close to 1.

There is generally wide scatter for all environments along the vertical(river discharge/river flow) axes (Figs. 7A, B). The separation of the dif-ferent populations varies, depending upon which parameter (river dischargeor river flow) is used. For river discharge, deltas and estuaries plot asoverlapping and widely scattered groups, but strand plains, lagoons, andtidal flats all have a mean annual discharge , 10 m3 s21 (Fig. 7A). Asnoted above, no significant differences were identified between the meanannual river discharges of the seven coastal depositional environmentsbased on Scheffe’s test (Fig. 6).

Nearly all deltas have a mean annual river flow of . 1 3 1027 m s21

(Fig. 7A), which corresponds to deltas being significantly different fromestuaries and tidal flats based on Scheffe’s test (Fig. 6). Estuaries, strandplains, and tidal flats exhibit a wide scatter against mean annual river flow

(Fig. 7B), whilst lagoons have a mean annual river flow of , 1 3 1027

m s21 (Fig. 7B).

DISCUSSION

The hypothesis of Boyd et al. (1992) and Dalrymple et al. (1992), thatwave, tide, and river power controls the geomorphology of clastic coastaldepositional environments is generally supported by the results of ourstudy. A clear relationship exists between the ratio of mean annual wavepower to mean annual tidal power (i.e., the wave/tide power ratio) andthe geomorphology of coastal depositional environments (Fig. 4). Thereare also statistically significant differences in the mean annual wave pow-er, mean tidal power, and mean annual river flow between the seven typesof clastic coastal depositional environments targeted in our study (Figs.5, 6).

Our results show that most clastic coastal depositional environments ploton either the wave- or tide-dominated side of the ternary diagrams (Fig.7). An exception to this are the strand plains, which have the widest scatterin terms of wave/tide power ratio, and they tend to plot towards the centerrather than on the wave-dominated half of the ternary diagram (Fig. 7), asexpected (Fig. 1B). We have not separated chenier-ridge complexes fromstrand plains, because of the difficulty in discriminating between the twoin the remotely sensed images used. Hence these depositional environmentsare plotted together, even though they may be associated with differentwave/tide power regimes. Wave-formed chenier ridges are separated byfine-grained, tidal-flat deposits and are generally considered to be more‘‘mixed’’ wave/tide environments (e.g., Dalrymple 1992), which probablyexplains their scattered distribution in the ternary diagram (Fig. 7).

An interesting observation is that the matrix of statistically significantdifferences between clastic coastal depositional environments (Fig. 6)shows that not all environments are differentiated by the same independentvariables, either separately or in combination, and that some environmentsare not differentiated by any of the variables. For example, the differencesbetween wave-dominated deltas, tide-dominated estuaries, and tidal flats


FIG. 6.—Matrices of significant differencesbetween populations of clastic coastaldepositional environments around Australia.Highlighted cells indicate statistically significantcombinations based on the Scheffe’s test for thedifference in means at the 95% confidence level,applied using the statistical software packageSTATISTCAy. Blank cells indicate that nosignificant difference was detected.

were significant for all three of the independent variables (wave power,tide power, and river flow; Figs. 5, 6). On the other hand, there was nostatistically significant difference between strand plains and tide-dominateddeltas, wave-dominated deltas, or lagoons identified by any of the inde-pendent variables (Fig. 6). For the other three environments, however, itappears that there may be some other factor (or factors) that controls thegeomorphology.

One variable that is not accounted for in the Boyd et al. (1992) andDalrymple et al. (1992) classification (Fig. 1) is the initial volume of theincised valley. It is implied in the diagram (Fig. 1A) that all valley sys-tems start out having a comparable volume (accommodation space),which is then filled with river and marine sediment under a steady sealevel. However, some valleys are larger than others, depending on a widerange of geological parameters, and hence they do not infill at the samerate. Valleys having identical wave, tide, and river power may have al-ready infilled and become deltas (where the accommodation space wasrelatively small), or they may still be estuaries (where the accommodationspace was relatively large). Hence variations in the initial volume ofincised valleys may explain some of the scatter exhibited by the data(along the vertical axes) in Figs. 7A and 6B. Nevertheless, the results ofthe present study suggest that there are statistically significant differencesbetween groups of clastic coastal depositional environments based onriver discharge and river flow.

River Discharge and Flow

In Australia, river discharge is larger for deltas and estuaries than forstrand plains and tidal flats but does not vary significantly between deltasand estuaries (Fig. 5; Table 1). The mean annual river discharge is notsignificantly different between the seven clastic coastal depositional en-vironments (Fig. 6). However, a statistical t-test demonstrates that themean annual river discharge for deltas and estuaries combined (28.5 6113 m3 s21, n 5 394) is significantly different (at the 95% confidencelevel) from the mean annual river discharge of strand plains and tidalflats combined (1.51 6 2.26 m3 s21, n 5 316; Fig. 6A). Furthermore,lagoons have a significantly lower river discharge rate than any otherenvironment (Fig. 5; Table 1). This result is consistent with the view ofBoyd et al. (1992), that strand plains, lagoons, and tidal flats occur alongcoastlines that do not have any significant river input (Fig. 1C). In ourstudy, these environments were found not to have incised valleys, butcontained small drainage areas that extended only as far as the immedi-ately adjacent hinterland.

Despite significant overlap in the populations, statistically significantdifferences do occur between mean annual river flow values for some ofthe different environments (Figs. 5, 6). Most significantly, wave- and tide-dominated deltas can be discriminated from wave- and tide-dominatedestuaries (Figs. 6, 7B). Although the difference in mean annual river flowrate between the seven coastal depositional environments is scattered

866 HARRIS ET AL.

FIG. 7.—Triangular classification diagrams ofAustralian clastic coastal depositionalenvironments plotting log wave/tide power ratioversus: A) log mean annual river discharge, andB) log mean annual river flow. The slope of theline in Fig. 4 was used as a correctioncoefficient (K ) in Equation 2 for plotting thedata so that at the data in the ternary plot wascentered on a wave/tide power ratio 5 1. Notethe Murray–Darling River (Fig. 2) has a meanannual discharge of 2,032 m3 s21 and plotsbeyond the limit of the ternary diagrams.


FIG. 7.—Continued.

(Figs. 5, 7B), a t-test indicates that the combined mean annual river flowfor wave- and tide-dominated deltas (1.0 3 1025 6 1.8 3 1025 m s21,n 5 150) is significantly different from the combined mean annual riverflow rate of estuaries, strand plans and tidal flats (2.9 3 1026 6 8.8 31026 m s21, n 5 571) at the 90% confidence level. The infilling ofestuaries to form deltas results in a smaller open water area, and it follows

that deltas should have a larger river flow rate than estuaries (given asimilar river discharge). However, tidal creeks and lagoons associatedwith tidal flats and strand plains also have small open water areas andthey exhibit a wide range of river flow rates (i.e., large standard devia-tions), depending on the size of their catchment areas (Table 1). Somestrand plains, with small open water areas, have river flow rates similar

868 HARRIS ET AL.

to some deltaic systems (Fig. 6). Hence, the type of depositional envi-ronment and their stage of development may explain some of the scatterin the river-flow data.

Dalrymple et al. (1992) suggested that the river sediment load (and notnecessarily the water discharge) exerts the most control over the geomor-phology of clastic coastal depositional environments. If this is true, thenriver power plays only a secondary role in controlling the geomorphologyof coastal depositional environments. Testing of this hypothesis must awaitthe development of a reliable database containing sediment yields for Aus-tralia’s river catchments, which is currently not available (Harris 1995;Wasson et al. 1996).

Flux of River Sediment to the Coast

The distribution of different geomorphic classes of clastic coastal de-positional environments provides insight into the flux of river sediment tothe coast during the Holocene. In Australia estuaries are more commonthan deltas, which is also true globally and is attributed to the relativelybrief time available in the Holocene for estuaries to infill under a steadysea level (Nichols and Biggs 1985). The abundance of estuaries over deltasin Australia also probably reflects the continent’s relatively arid climate,low relief, and hence relatively low sediment yield (Milliman and Syvitski1992).

Two-thirds of all of Australia’s deltas and nearly half of the progradingstrand plains and tidal flats are located on the northeast and Gulf of Car-pentaria coasts. The abundance of these progradational coastal depositionalenvironments (Fig. 1C) indicates that the hinterland of this region has arelatively high sediment yield in comparison with other regions of the con-tinent. Geological investigations of the postglacial evolution of the north-east margin of Australia (e.g., Harris et al. 1990; Woolfe et al. 1999; Dun-bar et al. 2000) have revealed that this margin has been characterized bysignificant input of fluvioclastic sediment during most of the Holoceneassociated with the dominant quasi-monsoonal climate regime in the region(e.g., Suppiah 1992).

Wave-dominated estuaries that occur on the southeast coast are onlypartially filled with postglacial sediment, most of which is marine-derivedsiliciclastic sediment (e.g., Roy 1984; Nichol and Murray-Wallace 1992;Nichol et al. 1994; Nichol et al. 1997). The part of the southeast regionthat borders the Great Australian Bight (Fig. 2) is drought-dominated andreceives little or no runoff (e.g., Erskine and Warner 1988). Clastic coastaldepositional environments in the southeast regions of Australia are thuscharacterized mostly by partially filled lagoons, also indicating the domi-nance of marine sediment deposition in these systems. In contrast, coastalsystems throughout southern and western Australia are dominated by ma-rine-derived sediment from carbonate sources, including the adjacent shelfand coral reefs (e.g., Short in press). Similarly, published case studies fromclastic coastal depositional environments of the northwest coast (e.g.,Woodroffe et al. 1989; Woodroffe and Chappell 1993; Woodroffe 1996)show that the (mainly) tide-dominated estuaries found here are also onlypartially filled with marine-derived Holocene sediment. The abundance ofestuaries and lagoons containing ample accommodation space and marine-derived sediment along the southwest, southeast, and northwest coasts ofAustralia implies that these margins have been characterized by moderateto low flux of terrigenous sediment during the Holocene.

The Relative Influence of Wave and Tide Power onCoastal Geomorphology

An important factor that plays a role in determining the geomorphologyof clastic coastal depositional environments is the relative power of wavesand tides, acting separately and in combination. As noted by Davis andHayes (1984), it is possible, for example, for a coastal environment to bewave-dominated even along macrotidal coasts, if the wave power is great

enough. An interesting result of our analysis is that both wave-dominatedand tide-dominated deltas have significantly lower mean wave and tidalpower than wave-dominated and tide-dominated estuaries (Figs. 5, 6). Inother words, our results suggest that deltas are formed in lower-energyenvironments than estuaries. This may be because the higher-energy sys-tems tend to lose more sediment to the adjacent shelf and coastal areas,thus inhibiting delta development. Delta sediment may be smeared outalong the coast where the receiving basin is characterized by relatively highwave or tide power.

The average tidal power for the 780 coastal environments (1,200 J m22

s21) is greater than the average wave power (130 J m22 s21; see Fig. 4)because power is a function of the wave height squared (Equation 2). Hencethe absolute value of fluid power expended around Australia is highest onthe macrotidal northwest coast (Fig. 2), which is characterized by trans-gressive tide-dominated estuaries and tidal flats, with strong tidal currentsand continuously turbid water (e.g., Wright et al. 1973; Semeniuk 1981).Tide-dominated deltas are most abundant and best developed in central andnorth Queensland (Fig. 2), where the coast is characterized by upper-microto lower-meso ranges and a broad, shallow-gradient, progradational coastalplain (Hopley 1982; Belperio 1983). Wave power along this coast is atten-uated by the Great Barrier Reef, which acts to block out long-period swellwaves generated in the Coral Sea (Hopley 1982; Hopley 1984). The dis-tribution of tide-dominated deltas suggests that they are confined to coastswhere tidal power is relatively weak, allowing sediment supplied by riversto be deposited close to the river mouth, and thus permitting the delta toprograde seaward (e.g., Galloway 1975).

Wave-dominated deltas are also generally found along relatively low-energy coasts. The formation of deltas on the northeast Queensland coastis probably facilitated by the sheltering effect of the Great Barrier Reef asmentioned above (Hopley 1982; Orpin et al. 1999). In contrast, the high-energy exposed southeast coast has relatively few deltas.

The mean values and bounding limits of tide power, wave power, riverdischarge, and river flow predicted by our study (Table 1; Fig. 5) mayprove useful for the characterization of paleoenvironments. If an ancientdeposit is interpreted as a tide-dominated estuary, for example, then itmight be inferred (from Fig. 5) that the mean annual river discharge intothe estuary was at least ; 1 m3 s21, that its tidal power was at least; 100 watts m22, whilst its wave power was probably less than ; 100watts m22. Taken together with information from the fossil record, thesevalues might provide a means of testing the validity or otherwise of basicenvironmental interpretations.

Climate Change and the Stability of CoastalDepositional Environments

Our results show that many Australian coastal depositional environmentsare poised along the wave–tide power boundary (Figs. 4, 7) and may belikely to change from one geomorphic state to another with changes to thewave/tide power ratio. Geomorphic changes resulting from an increase inwave power are likely to be most obvious for tide-dominated environments.The most significant change would be the replacement of existing shore-normal features (e.g., tidal sand banks) with shore-parallel features (e.g.,barriers and beach ridges). Coastal depositional environments that presentlycontain permanently open tidal inlets may become subject to intermittentor complete closure, and prograding environments may begin to recedebecause of increased alongshore sediment transport.

Significant attention has focused on the rise in global sea level inducedby anthropogenic greenhouse warming and the effects this may have onthe coastal zone (e.g., Daniels 1992; MacDonald and O’Connor 1996).Another apparent response to the warming of the Earth’s atmosphere is achange in the ocean wave climate, manifested as increased wave heightsassociated with more intense storm events (e.g., Carter and Draper 1988;WASA 1998; Gulev and Hasse 1999; Allan and Komar 2000; Grevemeyer


et al. 2000). Our approach of quantifying wave, tide, and river power pro-vides a useful method for identifying coastal depositional environmentsmost susceptible to these changes.

Changes to the geomorphology of many clastic coastal depositional en-vironments on the northeast coast of Australia would have occurred duringthe Holocene because of the presence of a ‘‘high-energy’’ wave window(Hopley 1984) that existed because reef growth lagged behind sea-levelrise. Clastic coastal depositional environments along the northeast coastwould have originally evolved as wave-dominated systems during the earlyto mid-Holocene and then as lower-energy wave- and tide-dominated sys-tems as the tops of the reefs eventually caught up with sea level. Similargeomorphic changes have been described by Pendon et al. (1998), whodocumented the transition from tide domination to wave domination ofestuaries in southwestern Spain, where tidal prisms have been reduced bysiltation at the mouth during the late Holocene. Although these estuariesdid not evolve because of a change in wave/tide power ratio along thecoast, this study demonstrates that coastal depositional environments canevolve from one (wave- or tide-dominated) state to another under a constantrelative sea level. The process-based classification derived here could beuseful for identifying environments most susceptible to altering their geo-morphology in response to changes in the coastal wave/tide power regime.

CONCLUSIONS

Our study has confirmed that the general geomorphology of clastic coast-al depositional environments can be predicted from numerical estimates ofwave, tide, and river power. The conceptual ternary diagram of Boyd etal. (1992) and Dalrymple et al (1992), showing the distribution of envi-ronments as being controlled by wave, tide, and river power, is supportedby our results. In particular, we conclude that there is a statistically sig-nificant difference between 721 wave- and tide-dominated clastic coastaldepositional environments in Australia. The large database used here toderive mean wave, tide, and river power values (Table 1) enables quanti-tative comparisons to be made between modern coastal depositional sys-tems, and provides useful constraints for paleo-reconstructions of ancientcoastal systems.

Although we have found differences between combined groups of en-vironments and river discharge and river flow, there is significant scatterin the data related to the seven categories of clastic coastal depositionalenvironments targeted in our study (Figs. 7A, B). Possible explanations forthis scatter include: (1) that not all incised valleys had a comparable ac-commodation space at the onset of Holocene highstand; (2) variable sea-level conditions throughout the Holocene; and (3) that river power is notrepresented by river discharge or river flow. Further, sediment flux ratherthan river power may exert a primary control on the overall geomorphologyof clastic coastal depositional environments, as hypothesized by Dalrympleet al. (1990). Reliable estimates of pre-European river sediment loads inAustralia are required to test this hypothesis.

Our results indicate that deltas are generally found along coasts havingconditions of lower wave and tide power than coasts where estuaries arepredominant. A possible explanation is that higher wave and tide powerresults in estuaries losing a greater percentage of sediment to the adjacentshelf and coastal areas, thus inhibiting delta development.

The large database of clastic coastal depositional environments compiledhere provides insights into continental-scale sediment loads to the coast.Two-thirds of deltas and nearly half of the prograding strand plains andtidal flats are located on the northeast Queensland and Gulf of Carpentariacoasts, indicating that the hinterland of this region has a relatively highsediment yield in comparison with other regions of the continent.

ACKNOWLEDGMENTS

We thank the reviewers R. Boyd, A. Short, S. Nichol, R. Dalrymple, and Dr. V.Passlow for their comments on an earlier version of this manuscript. Dr. D. Green-

slade of the Bureau of Meteorology Research Centre, Melbourne, is thanked forproviding the wave data used in this study. The authors are grateful for the financialsupport of the National Land and Water Resources Audit in undertaking this study.This manuscript is a contribution of the Cooperative Research Centre for CoastalZone Management and is published with permission of the Executive Director, Geo-science Australia. Line drawings of all 780 Australian waterways contained in theAED (Bucher and Saenger 1991) can be viewed at: www.ea.gov.au/coasts/information/reports/estuaries/index.html.

REFERENCES

ALLAN, J., AND KOMAR, P., 2000, Are ocean wave heights increasing in the eastern NorthPacific?: EOS, Transactions of the American Geophysical Union, v. 81, p. 561.

AUSTRALIAN NATIONAL TIDE TABLES, 2000, Canberra, Australian Government Printing Service.BELPERIO, A.P., 1983, Terrigenous sedimentation in the central GBR lagoon: a model from the

Burdekin region: Bureau of Mineral Resources, Journal of Australian Geology and Geo-physics, v. 8, p. 179–190.

BOYD, R., DALRYMPLE, R., AND ZAITLIN, B.A., 1992, Classification of clastic coastal depositionalenvironments: Sedimentary Geology, v. 80, p. 139–150.

BUCHER, D., AND SAENGER, P., 1991, An inventory of Australian estuaries and enclosed marinewaters: an overview of results: Australian Geographical Studies, v. 29, p. 370–381.

BUCHER, D., AND SAENGER, P., 1994, A classification of tropical and subtropical Australianestuaries: Aquatic Conservation: Marine and Freshwater Ecosystems, v. 4, p. 1–19.

CARTER, D.J.T., AND DRAPER, L., 1988, Has the northeast Atlantic become rougher?: Nature, v.332, p. 494.

COLEMAN, J.M., AND WRIGHT, L.D., 1975, Modern river deltas: variability of processes and sandbodies, in Broussard, M.L., ed., Deltas—Models for Exploration: Houston, Houston Geo-logical Society, p. 99–149.

COOK, P.J., AND MAYO, W., 1978, Sedimentology and Holocene history of a tropical es-tuary, Broad Sound, Queensland: Australian Bureau of Mineral Resources, Bulletin,v. 170, 206 p.

DALRYMPLE, R.W., 1992, Tidal depositional systems, in Walker, R.G., and James, N.P., eds.,Facies Models: Response to Sea Level Change: St Johns, Newfoundland, Geological As-sociation of Canada, p. 195–218.

DALRYMPLE, R.W., ZAITLIN, B.A., AND BOYD, R., 1992, Estuarine facies models: conceptual basisand stratigraphic implications: Journal of Sedimentary Petrology, v. 62, p. 1130–1146.

DANIELS, R.C., 1992, Sea-level rise on the South Carolina coast: two case studies for 2100:Journal of Coastal Research, v. 8, p. 56–70.

DAVIS R.A., JR., AND HAYES, M.O., 1984, What is a wave dominated coast? Marine Geology,v. 60, p. 313–329.

DIGBY, M.J., SAENGER, P., WHELAN, M.B., MCCONCHIE, D., EYRE, B., HOLMES, N., AND BUCHER,D., 1998, A physical classification of Australian estuaries: Report prepared for the UrbanWater Research Association of Australia, No. 4178, Southern Cross University, Centre forCoastal Management, Lismore NSW, 46 p.

DUNBAR, G.B., DICKENS, G.R., AND CARTER, R.M., 2000, Sediment flux across the Great BarrierReef Shelf to the Queensland Trough over the last 300 ky: Sedimentary Geology, v. 133,p. 49–92.

ERSKINE, W.D., AND WARNER, R.F., 1988, Geomorphic effects of alternating flood- and drought-dominated regimes on NSW coastal rivers, in Warner, R.F., ed., Fluvial Geomorphology ofAustralia: Sydney, Academic Press, p. 223–244.

EYRE, B., 1998, Transport, retention and transformation of material in Australian estuaries:Estuaries, v. 21, p. 540–551.

GALLOWAY, W.E., 1975, Process framework for describing the morphologic and stratigraphicevolution of deltaic depositional systems, in Broussard M.L., ed., Deltas—Models for Ex-ploration: Houston, Houston Geological Society, p. 87–98.

GREVEMEYER, I., HERBER, R., AND ESSEN, H.H., 2000, Microseismological evidence for a chang-ing wave climate in the northeast Atlantic Ocean: Nature, v. 408, p. 349–352.

GULEV, S.K., AND HASSE, L., 1999, Changes of wind-waves in the North Atlantic over the last30 years: International Journal of Climatology, v. 19, p. 1091–1117.

HARRIS, P.T., 1988, Large scale bedforms as indicators of mutually evasive sand transport andthe sequential infilling of wide-mouthed estuaries: Sedimentary Geology, v. 57, p. 273–298.

HARRIS, P.T., 1995, Marine geology and sedimentology of the Australian continental shelf, inZann, L.P., and Kailola, P., eds., The State of the Marine Environment Report for Australia,Technical Annex 1: The Marine Environment: Canberra, Department of the Environment,Sport and Territories, p. 11–23.

HARRIS, P.T., DAVIES, P.J., AND MARSHALL, J.F., 1990, Late Quaternary sedimentation on theGBR continental shelf and slope off Townsville, Australia: Marine Geology, v. 94, p. 55–77.

HASSELMANN, K., AND WAMDI GROUP, 1988, The WAM model—A third generation oceanwave prediction model: Journal of Physical Oceanography, v. 18, p. 1775–1810.

HEAP, A.D., BRYCE, S.M., RYAN, D., RADKE, L., SMITH, C., SMITH, R., HARRIS, P., AND HEGGIE,D., 2001, Australian Estuaries and Coastal Waterways: A Geoscience Perspective for Im-proved and Integrated Resource Management, AGSO Record 2001/07: Canberra, AGSO—Geoscience Australia, 118 p.

HOPLEY, D., 1982, The Geomorphology of the Great Barrier Reef: Brisbane, Wiley & Sons,453 p.

HOPLEY, D., 1984, The Holocene ‘high-energy’ window on the central Great Barrier Reef, inThom, B.G., ed., Coastal Geomorphology in Australia: Sydney, Academic Press, p. 135–150.

JOHNSON, D.P., 1982, Sedimentary facies of an arid-zone delta: Gascoyne Delta, Western Aus-tralia: Journal of Sedimentary Petrology, v. 52, p. 547–563.

870 HARRIS ET AL.

JONES, B.G., MARTIN, G.R., AND SENAPATI, N., 1993, Riverine–tidal interactions in the monsoonalGilbert River fandelta, northern Australia: Sedimentary Geology, v. 83, p. 319–337.

KOMEN, G., CAVALERI, L., DONELAN, M., HASSELMANN, K., HASSELMANN, S., AND JANSSEN,P.A.E.M., 1994, Dynamics and Modelling of Ocean Waves: Cambridge, U.K., CambridgeUniversity Press, 532 p.

LEES, B.G., 1992, The development of a chenier sequence on the Victoria Delta, Joseph Bon-aparte Gulf, northern Australia: Marine Geology, v. 103, p. 215–224.

MACDONALD, N.J., AND O’CONNOR, B.A., 1996, Changes in wave impact on the Flemish coastdue to increased mean sea level: Journal of Marine Systems, v. 7, p. 133–144.

MILLIMAN, J.D., AND SYVITSKI, J.P.M., 1992, Geomorphic/tectonic control of sediment dischargeto the ocean: the importance of small mountainous rivers: Journal of Geology, v. 100, p.525–544.

MURRAY-WALLACE, C.V., BELPERIO, A.P., BOURMAN, R.P., CANN, J.H., AND PRICE, D.M., 1999,Facies architecture of a last interglacial barrier: a model for Quaternary barrier developmentfrom the Coorong to Mount Gambier coastal plain, southeastern Australia: Marine Geology,v. 158, p. 177–195.

NEIL, D., AND YU, B., 1995, Fluvial sediment yield to the Great Barrier Reef lagoon: spatialpatterns and the effect of land use, in Hunter, H.M., Eyles, A.G., and Rayment, G.E., eds.,Proceedings of the Conference on Downstream Effects of Land Use: Queensland, Depart-ment of Natural Resources.

NICHOL, S.L., AND MURRAY-WALLACE, C.V., 1992, A partially preserved last interglacial estua-rine fill: Narrawallee Inlet, New South Wales: Australian Journal of Earth Sciences, v. 39,p. 545–553.

NICHOL, S.L., BOYD, R., AND PENLAND, S., 1994, Stratigraphic response of wave-dominatedestuaries to different relative sea-level and sediment supply histories: Quaternary case studiesfrom Nova Scotia, Louisiana and Eastern Australia, in Dalrymple, R.W., Boyd, R., andZaitlin, B.A., eds., Incised-Valley Systems: Origin and Sedimentary Sequences: SEPM, Spe-cial Publication 51, p. 265–283.

NICHOL, S.L., ZAITLIN, B.A., AND THOM, B.G., 1997, The upper Hawkesbury River, New SouthWales, Australia: a Holocene example of an estuarine bayhead delta: Sedimentology, v. 44,p. 263–286.

NICHOLS, M.M., AND BIGGS, R.B., 1985, Estuaries, in Davis, R.A., ed., Coastal SedimentaryEnvironments: New York, Springer Verlag, p. 77–186.

ORPIN, A.R., RIDD, P.V., AND STEWART, L.K., 1999, Assessment of the relative importance ofmajor sediment-transport mechanisms to the Great Barrier Reef lagoon: Australian Journalof Earth Sciences, v. 46, p. 883–896.

PENDON, J.G., JIMENEZ, I., LOPEZ, M., MORALES, J.A., AND BORREGO, J., 1998, Evolution of es-tuarine facies in a tidal channel environment, SW Spain: evidence for a change from tide-to wave-domination: Marine Geology, v. 147, p. 43–62.

PRITCHARD, D.W., 1967, Observation of circulation in coastal plain estuaries, in Lauff, G.H.,ed., Estuaries: Washington D.C., American Association for the Advancement of Science, p.3–5.

RHODES, E.G., 1982, Depositional model for a chenier plain, Gulf of Carpentaria: Sedimentol-ogy, v. 29, p. 201–221.

ROY, P.S., 1984, New South Wales estuaries: their origin and evolution, in Thom, B.G., ed.,Coastal Geomorphology in Australia: Sydney, Academic Press, p. 99–121.

ROY, P.S., THOM, B.G., AND WRIGHT, L.D., 1980, Holocene sequences on an embayed high-energy coast: an evolutionary model: Sedimentary Geology, v. 26, p. 1–19.

SEMENIUK, V., 1981, Sedimentology and the stratigraphic sequence of a tropical tidal flat, NorthWestern Australia: Sedimentary Geology, v. 29, p. 195–221.

SHORT, A.D., in press, The distribution and impact of carbonate sands on southern Australianbeach–dune systems, in Carbonate Beaches 2000: Key Largo, Florida, American Society ofCivil Engineers and U.S. Geological Survey.

SOULSBY, R., 1997, Dynamics of Marine Sands: London, Thomas Telford, 249 p.STERNBERG, R.W., 1972, Predicting initial motion and bedload transport of sediment particles

in the shallow marine environment, in Swift, D.J.P., Duane, D.B., & Pilkey, O.H., eds.,Shelf Sediment Transport: Stroudsburg, Pennsylvania, Dowden, Hutchinson and Ross Inc.,p. 61–82.

SUPPIAH, R., 1992, The Australian summer monsoon: a review: Progress in Physical Geography,v. 6, p. 531–549.

SWIFT, D.J.P., AND THORNE, J.A., 1991, Sedimentation on continental margins, I: a general modelfor shelf sedimentation: International Association of Sedimentologists, Special Publication14, p. 3–31.

WASA, 1998, Changing waves and storms in the northeast Atlantic? The WASA group:American Meteorological Society, Bulletin, v. 79, p. 741–760.

WASSON, R.J., OLIVE, L.J., AND ROSEWELL, C.J., 1996, Rates of erosion and sediment transportin Australia, in Walling D.E., and Webb B.W., eds., Erosion and Sediment Yield: Globaland Regional Perspectives: International Association of Hydrological Sciences, Special Pub-lication, p. 1–11.

WOODROFFE, C.D., 1996, Late Quaternary infill of macrotidal estuaries in northern Australia,in Nordstrom, K.F., and Roman, C.J., eds., Estuarine Shores: Evolution, Environments andHuman Alterations: Chichester, U.K., John Wiley & Sons, p. 89–114.

WOODROFFE, C.D., AND CHAPPELL, J., 1993, Holocene emergence and evolution of the McArthurRiver delta, southwestern Gulf of Carpentaria, Australia: Sedimentary Geology, v. 83, p.303–317.

WOODROFFE, C.D., CHAPPELL, J., THOM, B.G., AND WALLENSKY, E., 1989, Depositional model ofa macrotidal estuary and floodplain, South Alligator River, Northern Australia: Sedimentol-ogy, v. 36, p. 737–756.

WRIGHT, L.D., COLEMAN, J.M., AND THOM, B.G., 1973, Processes of channel development in ahigh tide range environment: Cambridge Gulf–Ord River delta, Western Australia: Journalof Geology, v. 81, p. 15–41.

WOOLFE, K.J., NAISH, T.R., AND PURDON, R.G., 1999, Lowstand rivers need not incise the shelf:an example from the Great Barrier Reef, Australia, with implications for sequence strati-graphic models: Geology, v. 26, p. 75–78.

Received 2 October 2001; accepted 9 April 2002.

classification of australian clastic …...calculated for seven different clastic coastal...

Documents