Z O O B E N T H I C V A R I A B I L I T Y

A S S O C I A T E D W I T H A F L O O D A N D D R O U G H T

I N T H E H A W K E S B U R Y E S T U A R Y , N E W S O U T H W A L E S :

S O M E C O N S E Q U E N C E S F O R

E N V I R O N M E N T A L M O N I T O R I N G

A. R. JONES

Division o f Invertebrates, The Australian Museum, P.O. Box A285, Sydney, N.S.W. 2000

(Received February 1989)

Abstract. Changes in the Hawkesbury zoobenthos associated with a major flood and drought are described and comparisons made in the patterns of change among estuarine reaches and among sites within reaches. The consequences for environmental monitoring and management are discussed. Replicate grab samples were taken from four sites in each of the lower, middle and upper reaches. Reductions in the mean number of species per grab (S) following the flood were significant only in the lower reaches and at one site in the middle reaches. Increases in S accompanied the drought in all reaches but intra-reach variation in temporal patterns occurred in both S and in the most abundant species found. Thus, major weather events are associated with temporal changes whose patterns differ on both small and large spatial scales. Consequently, the results from fixed-factor sampling designs, which are widely used, may be unrepresenta- tive of other areas. Unfortunately, the alternative approach of stratified random sampling will probably be both prohibitively expensive and difficult to implement in the complex estuarine benthic habitat. Further, short-term studies will probably be grossly unrepresentative of natural temporal variation. Attempts to reduce expenses by using only one or two abundant species as characterising communities or as indicators of physicochemical conditions may be unreliable because of variation in both time and space in dominant species and the lack of pollution-response knowledge for local species.

Introduction

Eco log ica l m o n i t o r i n g p r o g r a m s measu re progress t o w a r d res tor ing and ma in t a in -

ing ecosys tem in tegr i ty (Hirsch , 1980). This s t a t ement impl ic i t ly assumes tha t the

base l ine s ta te to which the ecosys tem shou ld be re s to red is known . Such know-

ledge is re la t ive ly easi ly o b t a i n e d i f the ecosys tem in ques t ion exists na tu ra l ly in

a s ta te o f s tab le equ i l ib r ium, bu t this seems to occur ra re ly (Connel l and Sousa ,

1983). Ins tead , b io log ica l c o m m u n i t i e s d i sp l ay unp red i c t ab l e t e m p o r a l va r i ab i l i t y at

va r ious spa t i a l scales (Lewis and P la t t , 1982; U n d e r w o o d e t a l . , 1983; D a y t o n and

Tegner , 1984) and , in add i t i on , communi t i e s a re spa t ia l ly he te rogeneous (Sousa ,

1984).

Consequen t ly , base l ine desc r ip t ions shou ld seek to e n c o m p a s s ranges o f na tu r a l

va r i a t i on in bo th space and t ime. In fact , the i m p o r t a n c e o f t e m p o r a l va r i ab i l i ty in

aqua t i c s tudies is such tha t ent i re issues o f j o u r n a l s have been devo ted to the top ic

e.g. Es tuar ies 8 (2A) (1985), H y d r o b i o l o g i a 129 (1) (1985) and H y d r o b i o l o g i a 142

(1986). M a n y o f these s tudies have dea l t wi th so f t - sed imen t zooben th i c communi t i e s

whose a t t r ibu tes m a k e t h e m a p p r o p r i a t e for m o n i t o r i n g (Bi lyard, 1987).

Environmental Monitoring and Assessment 14: 185-195, 1990. �9 1990 Kluwer Academic Publishers. Printed in the Netherlands.

186 A.R . JONES

A principal factor causing variability is natural disturbance (Sousa, 1984). Es- tuaries experience both sudden disturbances such as floods and more protracted changes during droughts. Although some accounts of flood-associated changes in the benthos exist (McLachlan and Grindley, 1974; Boesch et al., 1976; Stephenson et aL, 1977; Hodgkin, 1978; Saenger et al., 1980; Jones, 1987), little is known about the effects of drought. This is unfortunate because such knowledge is important for management purposes. For example, the increasing impoundment of rivers mimics droughts by reducing freshwater input to estuaries with a variety of possible conse- quences (Armstrong, 1982; Harris, 1984).

A second management-related goal is simply the quantification of the extent of natural change i.e. baseline variation. Short-term baseline studies are unlikely to encompass major weather events and hence may be grossly unrepresentative of natural variation if these events cause ecological change. Further, short-term studies may be unable to identify regular cycles in abundance of particular species. Unpred- ictable flooding, such as occurs in S.E. Australia (Rochford, 1959), may interfere with these cycles (Jones et al., 1988). Knowledge of such natural variation and cycles and their influencing factors are useful in that it enables an assessment to be made of the significance of 'unnatural' (anthropogenic) changes (Green, 1979).

During the course of a long-term (seven years) project on the zoobenthos of the Hawkesbury Estuary, a major flood and drought both occurred. Associated changes in the number of species (S) and in which species were most abundant ( = dominant) are described in this paper. In particular, the pattern of change is compared among estuarine reaches and also among sites within a reach and the implications of such variability for baseline monitoring and sampling design are discussed.

Methods

F I E L D AND LABORATORY

All sampling sites were located between the mouth of the Hawkesbury and its junction with the Colo River (Figure 1). These sites were grouped into three blocks on the basis of mean salinity.Block A (lower reaches, sites 1--4), block B (middle reaches, sites 5-8) and block C (upper reaches, sites 9-12) experienced high, inter- mediate and low salinities, respectively (see Jones, 1987).

All sites were sampled at three-month intervals from February (summer), 1978 until August (winter), 1979 inclusive. Subsequent sampling intensity varied among sites. Sites 1-6 were sampled at three-month intervals until August, 1981 with the exception of May (autumn), 1980, for which no samples were taken at sites 1-4. Sites 7 and 8 were sampled every three months until May, 1980 and thereafter in the summers of 1981 and 1982. Sites 9-11 were sampled at six-month intervals until February, 1982. Site 12 was not sampled subsequent to August, 1979.

Four replicate 0.05 m 2 Smith-McIntyre grabs were taken at each site and time. Material retained on a 1 mm sieve was preserved in buffered 10% (V/V) formalin.

B E N T H I C V A R I A B I L I T Y I N F L O O D A N D D R O U G H T 187

~'~.1 PACIFIC OCEAN

EN BAY

MOOt

o , I

kms

MANGRC

HAWKESBURY RIVER ==~

Sydney 3 34"$ 15'E

11f

Fig. 1. Map of the Hawkesbury Estuary showing sampling sites. Sites 1-4, 5-8, and 9-12 are located in the lower, middle and upper reaches, respectively.

188 A . R . JONES

Polychaete, crustacean and mollusc specimens were sorted and identified under stereomicroscopes in the laboratory.

River-discharge data were obtained from the Metropolitan Water Sewerage and Drainage Board's gauging station at Penrith approximately 80 km upstream of site 12.

D A T A ANALYSIS

Data for the mean number of species per grab (S) were analysed as follows. Pre- and post-flood samples were compared by a three-factor analysis of variance (ANOVA): 3 blocks (fixed factor) x 4 sites (nested within block) x 2 times (fixed factor). Samples taken during the drought could not be analysed similarly because missing data led to unbalanced designs. Instead, four separate two-way (site x time), fixed-factor ANOVAs were done on data from sites 1-4, sites 5 and 6, sites 7 and 8, and sites 9-11, respectively. In addition to samples taken during the drought, data from the two sampling periods preceding (winter and spring 1978) and succeeding (post-summer 1981) the drought were included in analyses to provide reference points. Student-Newman-Keuls (SNK) a posteriori multiple comparisons (Sokal and Rohlf, 1981) were used to locate significant differences in S among sampling times. Data were analysed untransformed because significant heterogeneity of variance was absent (Cochran's test, P>0.05).

Some of the following results appear in summary form only, because the presence of significant interactions among ANOVA factors and variation in those species most abundant precluded concise presentation of results. Detailed results are available from the author.

Results

P H Y S I C O C H E M I C A L M E A S U R E M E N T S

Peak discharges caused by the flood occurred in late March 1978 (Figure 2a). These were followed by high discharges in June 1978, very low discharges throughout 1979 and 1980 caused by the drought, and moderate discharges in 1981.

Sites 1, 3, 4, 9-12 were all 4-5 m deep. Sites 2, 5, 6, 7, and 8 were 10, 12, 8, 20, and 6 m deep, respectively. The sediments were as follows. Sites 1, 3, 4 and 7 were sandy mud; sites 2 and 5 muddy sand; sites 6, 8-12 medium to coarse sand. More detailed information appears in Jones (1987).

F L O O D - A S S O C I A T E D C H A N G E S IN NUM B E R OF SPECIES (S)

Significant differences in S occurres between the pre-flood (February 1978) and post-flood (May 1978) sampling times (three-way ANOVA; F = 128.7, P<0.001; Table I). However, the size of these differences varied both among blocks (F t ime • block interaction ---- 24.9, P < 0.001) and among sites within at least one block (F t ime • site interaction = 2 . | , P ~ 0 . 0 5 ) . Significant declines after the flood were found

B E N T H I C V A R I A B I L I T Y IN F L O O D A N D D R O U G H T 189

7 - (a) t r

-,/- 6 - 0

~ 5 -

~ a:o 3-

0 2 - I I ( I I I I I I J I t 1 I t I I

3o- (b) >-- ,, X

2 0 - /

,o_ v

O - ) 1 I I I I ~ I I I I i I I i I I

1 4 -

1 2 -

,,_ r ~ \ h , ~'~ ~ '~ '~-I "~ / 1 V \ 9 - /

o-'- \ ..M ",:,\ I �9 ii �9 SITE

3 - 2 - 1 - O- I I I I I t I I I t I I I I I I I

12- 1 1 - (d) 1 0 -

9 - 8 - 7 -

$ 6- 5- ~ / / xQ~__. , _ ~ , : ~ , i r S I T E 1 0 4 -

21_- ~ ~ . ~ J S I T E 9

0

I I I I I I I I I I I I I I I I I F M A N F M A N F M A N F M A N F

1 9 7 8 1 9 7 9 1 9 8 0 1 9 8 1 1 9 8 2

TIME

Fig. 2. Temporal distribution of river discharge and mean number of species per grab (S) during the period encompassing the flood (March 1978) and drought (January 1979 - February 1981). F, M, A, N = February, May, August, November, respectively. Standard errors have been omitted for clarity but ranged from 0.40-2.08 (lower reach sites), 0.25-1.71 (middle reach sites) and 0.00-0.85 (upper reach sites).

190 A . R . JONES

at all sites in the lower reaches and at one of the four sites in the middle reaches (SNK tests, Table I). No significant differences occurred in the upper reaches or at three of the four sites in the middle reaches.

DROUGHT-ASSOCIATED CHANGES IN NUMBER OF SPECIES (S)

During the period encompassing the drought, significant temporal changes in S occurred at all site groupings analysed (sites 1-4: F=53.9, P<0.001; sites 5-6: F=8.3, P<0.001; sites 7-8: F=8.4, P<0.001; sites 9-11: F=65.5, P<0.001; Figure 2). However, inter-site variation in temporal patterns was found for all site groupings (two-way ANOVAs, site x time interaction terms significant).

In the lower reaches, the two pre-drought times (winter and spring 1978) produced few species at all sites (Figure 2). Subsequent differences were rarely significant although the highest values were usually found in spring 1979 or the summer or winter of 1980. The two post-drought samplings were not poorer in species than during most of the drought. The difference between most and fewest species per grab was greater at sites 3 (27.75) and 4 (25.5) than at sites 1 (18.0) and 2 (15.0) nearer the mouth.

In the middle reaches, fewest species occurred in the first pre-drought sampling (Figure 2), but this value was not significantly different from those in at least four drought samplings (SNK tests). In fact, the sites with sandy sediments (6 and 8) showed virtually no significant temporal differences over the entire sampling period (SNK tests). Inter-site variation in S occurred in the timing of highest values, in post-drought patterns of change and in the range from highest to lowest numbers (Figure 2).

TABLE !

Mean numbers of species per grab for samples before and after the flood (February 1978 and May 1978, respectively). Horizontal lines connect mean values not significantly different (P>0.05) as determined by SNK multiple

comparisons

Site Feb. 1978 May 1978

1 18.75 7.00 Lower 2 13.25 6.50 Reaches 3 11.00 3.75

4 15.00 3.50

5 4.00 1.75 Middle 6 7.00 2.25 Reaches 7 4.50 2.50

8 2.50 1.50

9 3,75 ~,~0 Upper 10 1.75 1.50 Reaches 11 3,00 2.00

12 2.2~ 2.25

BENTHIC VARIABILITY IN FLOOD AND DROUGHT 191

In the upper reaches, the last three drought samplings (summer and winter 1980, summer 1981) yielded the most species (SNK tests, Figure 2). Pre-drought values were low but rarely significantly different from early drought times. Post-drought samples usually yielded intermediate numbers of species (Figure 2).

A B U N D A N T S P E C I E S

Changes in species dominance associated with the flood and drought occurred but varied in extent among the three estuarine reaches (Table II). Variation among sites within a reach also occurred, especially in the lower and middle reaches.

Flood-associated changes in dominance were assessed by comparing the pre-flood and first post-flood sampling times. In the lower reaches, the identity of the two most abundant (= dominant) species changed at all four sites following the flood (Table II). However, while the dominant species at sites 1, 3, and 4 showed taxonomic similarities, those at site 2 were often different (Table II).

Inter-site similarities and differences in flood-related changes also occurred in the middle reaches (Table II) but here, the identity of the two most abundant species did not always change after the flood. In the upper reaches, a single species, the polycheate Ceratonereis limnetica Hutchings and Glasby, maintained its domi- nance at all sites despite the flood (Table II). The polychaete Notomastus estuarius Hutchings and Murray persisted as the second dominant at sites 9 and 10 but not 11.

Drought-associated changes were assessed by comparing the drought sampling times among themselves and with both the two pre- and two post-drought samplings. In the lower reaches, changes in the identity of the two most dominant species were well marked at the onset of the drought but less so after its conclusion (Table II). Changes occurred during the course of the drought at all sites but inter-site differ- ences in both the pattern of this change and at any one time were apparent. Some species (e.g. the polychaete Carazziella victoriensis Blake and Kudenov at sites 1 and 2, and the crustacean Grandidierella gilesi Chilton at site 3) were persistently domi- nant throughout the drought and sometimes remained so subsequently. Other species, particularly the polychaete Nephtys australiensis Fauchald, were dominant before the drought but rarely thereafter (Table II).

In the middle reaches, changes associated with the drought's onset were less marked than in the lower reaches. However, data for sites 5 and 6 indicate that changes following the drought's conclusion were more marked than further downs- tream. Changes in dominance during the course of the drought were again evident as were inter-site differences (Table II).

In the upper reaches, the polychaete Ceratonereis limnetica was usually the most abundant species at all sites, although other species became dominant late in the drought and thereafter. The mollusc Fluviolanatus subtorta (Dunker) became the second dominant after the onset of drought at sites 9 and 11 but not at sites 10 and 12. Notomastus estuarius became the most abundant species late in the drought or subsequently.

192 A.R. JONES

TABLE II

The two most abundant species at each site and time (F =February, M =May , A = August, N = November).

More detail appears in Jones (1987). The flood occurred between Feb. and May 1978; the drought commenced between Nov. 1978 and Feb. 1979 and concluded between Feb. and May 1981. The code numbers for the species occurring frequently below are: 18=Ceratonereis limnetica, 21=Neph ty s australiensis, 32 =Augener i a verdis, 34 = Leitoscoloplos bifurcatus, 35 = Scolopolos (Scoloplos) simplex, 40 = Boccardiella limnicola, 41 = Carazziella victoriensis, 65 = Notomastus estuarius, 73 = Terebellides s troemi (all polychaetes), 77=Gastrosaccus dakini, 101 =Grandidierella gilesi, 121 =Urohaustorius metun gi, 13 5 = Callianassa arenosa, 141 = A marinus paralacustris (all crustaceans), 161 = Fluviolanatus subtorta, 169=Notospisula trigonella (both molluscs), - = n o data. Detailed taxonomic information

appears in Jones et aL (1986) and Jones (1987)

Site 1978 1979 1980 1981

F M A N F M A N F M A N F M A N F

Lower reaches 1 73 135 174 135 101 135 63 41

41 21 11 21 41 41 135 101

2

41 41 41 41 101 32 41 41 41 63 101 73 101 41 41 32 32 101

63 41 31 135 34 135 135 41 41 174 41 34 34 34 32 32 34 80 21 34 34 41 41 21 135 43 34 34 41 32 32 34 82 101

73 21 21 21 41 73 73 73 73 - 75 101 101 41 41 73 101

101 135 135 174 101 21 101 101 101 46 41 41 49 101 41 73

73 135 21 169 101 73 73 73 75 73 41 104 101 32 169 73 169

49 21 169 21 174 49 64 169 73 101 101 75 75 169 73 41 73

Middle reaches 5 21 35 77 35 169 169 21 169

6

7

8

71 71 75 73 75 169 49 169 169

80 21 21 21 109 80 169 112 169 28 73 75 73 21 21 49 49

21 35 77 169 169 169 121 169 121 77 169 169 169 21 121 169 121 17 77 35 17 21 21 21 77 77 116 121 121 121 90 77 77 41

40 40 40 40 40 103 40 40 21 101 101 40 109 74 141 161 100 100 21 103 28 28 75 100

21 18 18 77 109 21 35 35 169 169 169 90

35 109 120 36 35 77 77 21 21 77 37

Upper reaches 9 18 18 18 18 18 18 18 - 18 - 74 - 65 - 65 - 18

65 65 120 161 161 161 161 65 18 18 65

10 18 18 18 18 18 18 18 - 18 - 74 - 18 - 65 - 18

65 65 141 65 120 120 35 18 161 18 65

11 18 18 18 18 18 18 18 - 18 - 40 - 18 - 40 - 40 65 140 141 141 161 161 161 40 18 40 18 65

12 18 18 18 18 18 18 18 - - 161 - 18 - 18 120 141 141 141 141 141 141 18 88 65

B E N T H I C V A R I A B I L I T Y IN F L O O D A N D D R O U G H T 193

Discussion

Major weather events such as floods and droughts affect the physicochemical composition of estuaries. In the Hawkesbury Estuary, these events were associated with changes in the number of benthic species and in which species were most abundant. Variability in the patterns of temporal change occurred both among estuarine reaches and among sites within a reach (i.e. at two spatial scales). Factors potentially causing ecological changes include salinity, erosion and/or deposition of sediments, dissolved oxygen, flushing and availability of colonists. Discussion of these factors is beyond the scope of this paper but appears elsewhere with respect to floods (see Introduction) and drought (Jones, in prep.).

The finding of variability in pattern of change at different spatial scales raises questions concerning the sampling design of baseline programmes. If such pro- grammes seek to provide unbiased estimates, then stratified, random sampling designs are appropriate (Green, 1979). However, in large estuaries, this approach requires numerous samples for reasonable precision of estimates (Tenore, 1972). Because of the highly labour-intensive nature of processing soft-sediment samples, the expense may be unacceptably high. Furthermore, the practice of stratification and randomly locating samples becomes difficult and complex in remotely-sampled, invisible environments characterised by sometimes correlated and sometimes inde- pendent gradients of potentially important factors such as salinity, sediments and depth (Jones et al., 1986). Consequently, most estuarine benthic ecologists have opted for a fixed-factor, systematic design, especially with respect to the location of transects. While such a design may discern changes more powerfully than stratifi- cation (Cox, 1958), the high ecological variability observed in the Hawkesbury, even among sites in close proximity and with similar physicochemical attributes, implies that extrapolation from a particular site to other areas will be unreliable.

The choice of sampling design can be optimised through a pilot programme; a recommended practice (Green, 1979; Andrew and Mapstone, 1987). Sample sizes necessary for acceptable precision of estimates can be determined and design de- cisions then made in light of the resources available and the feasibility of stratifying the habitat and sampling at random.

Spatial and temporal changes also occurres in the dominant species found. This variability, combined with scant knowledge of the general biology, life history and response to pollution of Australian species makes the use of the 'indicator' species approach highly suspect, as found elsewhere (Botton, 1979). However, in the species-poor upper reaches of the Hawkesbury, the polychaete Ceratonereis limneti-

ca proved to be dominant at all sites and almost all times sampled. Because of this persistent high abundance, and because life-history knowledge is available (Glasby, 1984), C. l imnetica deserves further study as a potential biomonitoring and manage- ment tool, especially as the upper reaches of estuaries are often subject to much human impact.

In conclusion, major weather events were associated with significant ecological

194 A . R . JONES

changes. Fur the rmore , the pa t te rn of t empora l change varied a m o n g estuar ine

reaches and a m o n g sites wi th in a reach. Whi le such knowledge assists env i ronmen ta l

m a n a g e m e n t , the var iabi l i ty found has the fol lowing impl icat ions for managemen t :

shor t - te rm projects will no t represent na tu ra l var ia t ion adequately; ex t rapola t ion

f rom localised studies to other areas will be unrel iable; and the use of the indicator

species approach will of ten be risky because of na tu ra l var ia t ion in abundance .

A c k n o w l e d g e m e n t s

J. K. Lowry, R. T. Spr ingthorpe , P. A. Hutchings , H. Pax ton , W. F. Ponder ,

W. B. R u d m a n , P. Co lman , G. C. B. Poore , H. Lew Ton , M. D r u m m o n d , J. Young

and N. S. Jones provided t axonomic assistance. F. Byers and R. Ba teman analysed

sediments and m a n y volunteers sorted samples. Keith McGuinness provided statisti-

cal advice and Peggy O ' D o n n e l l and an a n o n y m o u s referee improved the manuscr ip t .

T h a n k s are due to all the above and especially to J o h n Reed, the boat skipper, and

Char lo t te Watson-Ruse l l , A n n a M u r r a y and Rob in Marsh who made great contr i-

bu t ions to the entire project .

References

Andrew, N. L. and Mapstone, B. D.: 1987, 'Sampling and the Description of Spatial Pattern in Marine Ecology', Oceanog. Mar. Biol. Ann. Rev. 25, 39-90.

Armstrong, N.: 1982, 'Responses of Texas Estuaries to Freshwater Inflows', in V. S. Kennedy (ed.), Estuarine Comparisons, Academic Press, New York, pp. 103-120.

Bilyard, G. R.: 1987, 'The Value of Benthic Infauna in Marine Pollution Monitoring Studies', Mar. Pollut. Bull. 18, 581-585.

Boesch, D. F., Diaz, R. J., and Virnstein, R. W.: 1976, 'Effects of Tropical Storm Agnes on Soft-Bottom Macrobenthic Communities of the James and York Estuaries and the Lower Chesapeake Bay', Ches. Sci. 17, 246-259.

Botton, L. L.: 1979, 'Effects of Sewage Sludge on the Benthic Invertebrate Community of the Inshore New York Bight', Estuarine Coastal Mar. Sci. 8, 169-180.

Connell, J. H. and Sousa, W. P.: 1983, 'On the Evidence Needed to Judge Ecological Stability or Persistence', Amer. Nat. 121, 789-824.

Cox, D. R.: 1958, Planning o f Experiments, Wiley, New York. Dayton, P. K. and Tegner, M. J.: 1984, 'The Importance of Scale in Community Ecology: A Kelp Forest

Example with Terrestrial Analogs', in P. W. Price, C. N. Slobodchikoff and W. S. Gaud (eds.), A New Ecology, Wiley, New York, pp. 457-482.

Green, R. H.: 1979, Sampling Design and Statistical Methods for Environmental Biologists, Wiley, New York.

Glasby, C. J.: 1984, 'The Population Structure and Reproductive Biology of Ceratonereis limnetica (Polychaeta: Nereididae) at Lower Portland, Hawkesbury River', M.Sc. Thesis, University of Sydney.

Harris, J. H.: 1984, 'Impoundment of Coastal Drainages of Southeastern Australia, and a Review of its Relevance to Fish Migrations', Aust. Zool. 21, 235-50.

Hirsch, A.: 1980, 'Monitoring cause and effects - ecosystem changes', in D. L. Worf (ed.), Biological Monitoring for Environmental Effects, Lexington Books, Lexington, pp. 137-142.

Hodgkin, E. P.: 1978, 'An Environmental Study of the Blackwood River Estuary, Western Australia, 1974-1975', W.A. Dept Cons. and Envir., Report No. 1, p. 78.

Jones, A. R., Watson-Russell, C. J., and Murray, A.: 1986, 'Spatial Patterns in the Macrobenthic Communities of the Hawkesbury Estuary, N.S.W.', Aust. J. Mar. Freshw. Res. 37, 521-43.

BENTHIC VARIABILITY IN FLOOD AND DROUGHT 195

Jones, A. R.: 1987, 'Temporal Patterns in the Macrobenthic Communities of the Hawkesbury Estuary, N.S.W.', Aust. J. Mar. Freshw. Res. 38, 607-624.

Jones, A. R., Murray, A., and Skilleter, G. A.: 1988, 'Aspects of the Life History and Population Biology of Notospisuta trigonella (Bivalvia: Mactridae) from the Hawkesbury Estuary, Southeastern Australia', The Veliger 30, 267-277.

Kennedy, V. S. (ed.): 1982, Estuarine Comparisons, Academic Press, New York. Lewis, M. R. and Platt, T.: 1982, 'Scales of Variability in Estuarine Ecosystems', in V. S. Kennedy (ed.),

Estuarine Comparisons, Academic Press, New York, pp. 3-20. McLachlan, A. and Grindley, J. R.: 1974, 'Distribution of Macrobenthic Fauna on Soft Substrata in the

Swartkops Estuary, with Observations on the Effects of Floods', Zool. Afr. 9, 211-233. Rochford, D. J.: 1959, 'Classification of Australian Estuarine Systems', Arch. Oceanografia Limnolgica

11, 171-177. Saenger, P., Stephenson, W., and Moverley, J.: 1980, 'The Estuarine Macrobenthos of the Calliope River

and Auckland Creek, Queensland', Mere. Queensl. Mus. 20, 143-161. Sousa, W. P.: 1984, 'The Role of Disturbance in Natural Communities', Ann. Rev. Ecol. Syst. 15, 353-91. Sokal, R. R. and Rohlf, F. J.: 1981, Biometry: The Principles and Practice of Statistics and Biological

Research, 2nd ed., W. H. Freeman and Co: San Francisco. Stephenson, W., Cook, S. D., and Raphael, Y. I.: 1977, 'The Effect of a Major Flood on the Macroben-

thos of Bramble Bay, Queensland', Mem. Queensl. Mus. 18, 95-119. Tenore, K. R.: 1972, 'Macrobenthos of the Pamlico River Estuary, North Carolina', Ecol. Monogr. 42,

51-69. Underwood, A. J., Denley, E. J. and Moran, M. J.: 1983, 'Experimental Analyses of the Structure and

Dynamics of Mid-shore Rocky Intertidal Communities in New South Wales', Oecologia 56, 202-219.