Macro-invertebrate species diversity as a potential

universal measure of wetland ecosystem integrity in

constructed wetlands in South East Melbourne
Sadiqul Awal∗ and Daniel Svozil
Applied Aquaculture, Dept of Agriculture and Animal Science, Northern Melbourne Institute of TAFE (NMIT)
Corner of Dalton Road and Cooper Street, Epping Victoria 3076 Australia
∗
Corresponding author: sadiqulawal@nmit.edu.au

The use of macroinvertebrates in the assessment of wetland ecosystem integrity is an increasingly

common tool used for management and conservation. The sensitivity of macroinvertebrates to ecological
fluctuation makes them reliable and appealing indicators of ecosystem integrity. However, there is little or
no published data available for assessment of wetland ecosystem integrity on the basis of macroinvertebrate
species diversity in constructed wetlands of metropolitan Melbourne. The aim of the following study was
to assess significant differences in macroinvertebrate diversity in three constructed wetlands in South
East metropolitan Melbourne and consequently, suitability as a universal measure of wetland ecosystem
integrity.
Three wetlands were chosen randomly, with the requirement that they were entirely man-made, from a
list of constructed wetlands in South East metropolitan Melbourne. Between 481 and 629 organisms were
found in each wetland comprising 16 different taxa. The one-way ANOVA for species richness (P > 0.05,
F = 0.19) and Shannon-Weiner diversity index (P > 0.05, F = 2.54) indicate no significant differences
between the wetlands in both of these measures. The data collected in the present study compared with
published species richness and Shannon-Weiner diversity index data suggests macroinvertebrate species
diversity can be used as a universal measure of wetland ecosystem integrity in constructed wetlands in
South East metropolitan Melbourne. This is important where there is need for a rapid and streamlined tool
for assessment of ecosystem integrity and consequently, the management and conservation of constructed
wetlands.

Keywords: Shannon-Weiner index, richness, assessment, temporal, indicators

Introduction structed for the purpose of receiving and treating

storm-water and stream inputs (Melbourne Water,
The importance of wetlands lies in their envi- 2003).
ronmental, economic and social values (Bennett In view of Melbourne Water’s treatment aims of
and Whitten, 2001) and is equally applicable reducing pollution from nitrogen and other contam-
to constructed wetlands in developed and urban- inants in local waterways and the receiving waters
ized areas (Azous and Horner, 2001). Of the of Port Phillip Bay, appropriate strategies for man-
117 wetlands in Melbourne, Australia, managed agement and conservation of wetlands must be em-
by Melbourne Water, a large number are con- ployed. One of these strategies involves assessment

472
Aquatic Ecosystem Health & Management, 13(4):472–479, 2010. Copyright 
C 2010 AEHMS. ISSN: 1463-4988 print / 1539-4077 online

DOI: 10.1080/14634988.2010.526439
 Awal and Svozil / Aquatic Ecosystem Health and Management 13 (2010) 472–479 473

of wetland ecosystem integrity, which is defined by quently, little is known of the macro-invertebrate
De Leo and Levin, (1997) in the context of con- community assemblages, diversity and richness in
servation and economic value, as desirable stability these constructed wetlands and their suitability as a
domains that guarantee optimal exploitation rates. It measure of wetland ecosystem integrity, for the pur-
is often referred to as bioassessment, biomonitoring pose of conservation or bio-monitoring programs. In
or biological monitoring (USEPA, 1997). response to this, the following study set out to assess
The assessment of ecosystem integrity involves the relative dynamics of macroinvertebrate species
sampling, identification and analysis of some biotic diversity in three constructed wetlands. The spe-
elements in a water body, such as phytoplankton, cific aims of the study were to collect baseline data
bacteria, larval and adult aquatic insects, and other for species richness and Shannon-Weiner species
macroinvertebrates. Vertebrates such as fishes, am- diversity index and establish any significant dif-
phibians, bird and mammal species may also be ferences in these data between the three wetlands.
used in assessment of ecosystem integrity, but in Further, this data would be used to determine if
some cases are more laborious and time consuming macro-invertebrate species diversity could be ap-
to sample (Adamus and Brandt, 1990). Macroinver- plied to these constructed wetlands for the assess-
tebrate community centrality and interrelationship ment of ecosystem integrity, and as a rapid, universal
with other organisms in food webs (Anamaet et al., measure for comparison of constructed wetlands in
2005) as well as their sensitivity to changes in the other, similarly urbanized and developed geograph-
trophic state of a water body, makes them highly re- ical regions.
liable and appealing for use as indicators of ecosys-
tem integrity and in bio-monitoring programs.
However, there is very little published data avail- Materials and Methods
able for assessment of wetland ecosystem integrity
Description of study sites
on the basis of macroinvertebrate species diversity
in constructed wetlands of South East metropoli- The three representative sites chosen for the
tan Melbourne (Bryant and Papas, 2007). Conse- study shown in Figure 1 were selected randomly

Figure 1. Relative locations of the three wetlands. (A) Troups Ck wetland, (B) Police Road Retarding Basin, and (c) Dandenong
Retarding Basin. (Image source: Google Maps, 2010.)
474 Awal and Svozil / Aquatic Ecosystem Health and Management 13 (2010) 472–479

Table 1. Three constructed wetlands selected as sites for macroinvertebrate sampling.

Name Location Macrophyte coverage

Dandenong Stud Rd, Dandenong Emergent ≈ 15%
Retarding Basin Submerged ≈ 30%
Open water ≈ 60%

Troups Ck wetland Belgrave-Hallam Emergent ≈ 10%

Rd, Narre Warren Submerged ≈ 5%
Open water ≈ 85%

Police Rd Cnr Wellington and Emergent ≈ 40%

Retarding Basin Police Rd, Mulgrave Submerged ≈ 25%
Open water ≈ 35%

from a list of locations in the South East metropoli- of the whole wetland (Sharma and Rawat, 2009).
tan Melbourne region as described in Melbourne Davis et al. (1999) similarly suggests water bodies
Water’s Water Sensitive Urban Design web page be divided into 4 sectors labeled, north, east, south
(Melbourne Water, 2009). The three wetlands all and west, or however is deemed appropriate, allow-
constructed between 2003 and 2004 treat similar ing one site to be randomly selected by the numbered
discharge types from a combination of residential, grid method as per Sutherland (2006) in each sector.
commercial, industrial and roadside areas in the re- Random selection eliminates influence of confound-
gion. Macrophyte coverage is described in Table 1. ing variables such as substrate, macrophyte growth
and detritus (Barber and Kevern, 1973), on macro-
Sampling point selection invertebrate distribution through the wetland.
Sampling at fortnightly intervals is recom-
Random selection of sampling points in each mended (NRM, 2008) to overcome confounding
wetland, as per Spieles and Mitsch (2000), was used, effects of seasonal changes in hydrology, adult
with 4 sites being selected so as to be representative macroinvertebrate migrations and local climate
 Awal and Svozil / Aquatic Ecosystem Health and Management 13 (2010) 472–479 475

(USEPA, 2002; Balla and Davis, 1995). Samples gree of normality present was sufficient for the use
from each sector were taken in each wetland and of one-way ANOVA.
pooled once a fortnight for 3 months, between
29/4/2009 and 30/9/2009. Results
Barlow et al. (1982) describe sweep and dredge
nets as being most satisfactory in capturing motile A total of 16 taxa across the 3 wetlands were
macroinvertebrates, with dredge nets allowing identified with 481 organisms collected from Police
greater areas to be sampled. Sweep netting is also Road Retarding Basin (PRRB), 629 from Troups
described by Sutherland (2006) as the most suit- Creek Wetland (TCW) and 530 from the Dandenong
able sampling method as it captures the lower water Retarding Basin (DRB) wetlands. The taxa found in
column and upper sediment layers and was used in the samples and the 5 dominant taxa in each wetland
conjunction with benthic grabs to allow sampling are shown in Table 2.
of deeper sediment layers (Davis et al., 1993). Ben- The Shannon-Weiner diversity index (SWDI)
thic grabs were performed with a hand trowel and was calculated for PRRB, TCW and DRB wetlands
sweeps with a 500µm mesh sweep net as per Sharma every fortnight after sampling and identification.
and Rawat (2009). Samples were preserved in 70 per One way ANOVA at the 0.05 significance level was
cent ethanol. Samples were spread on a Petri dish for performed on species richness, or number of species
viewing and identification under a dissecting micro- per sample, recorded across the study period. The
scope, using identification keys from Ingram et al. range and means were 3–9 and 6.17; 3–9 and 6.33;
(1997), Miller (1983), and Gooderham and Tsyrlin 4–8 and 5.92 for PRRB, TCW and DRB, respec-
(2002). Samples were identified to the furthest level tively. The three wetlands did not significantly differ
of classification possible. (P > 0.05, F = 0.19) in terms of species richness.
The same one way ANOVA at the 0.05 signif-
Data analysis icance level was performed on SWDI calculated
across the study period. The range and means were
The Shannon-Weiner diversity index, a function 0.84–1.48 and 1.19; 0.47–1.50, and 1.03; 0.74–1.62,
of species richness and abundance (Wilhm and Dor- and 1.31 for PRRB, TCW and DRB wetlands, re-
ris, 1968; Nzengy’a and Wishitemi, 2000) was used spectively. The three wetlands, though not signif-
to determine species diversity. The equation for icantly different (P > 0.05, F = 2.54) were more
Shannon-Weiner diversity index, where H is diver- likely to show significant differences in species di-
sity, p is proportion or number of individuals of each versity than species richness.
taxon or species in sample and ln is log1 10, is shown Temporal variation in SWDI was found to have
below. an increasing trend in all wetlands. Temporal vari-
ation in SWDI and species richness can be seen
H  = −
(P × ln( p)) in Figures 2 and 3, respectively, between April
and September 2009. The SWDI range documented
The value calculated is negative; however, this for constructed wetlands is between 2.1 (Anamaet
was written as a positive figure in the results for et al., 2005) and 0.4 (Webb and Mitsch, 2001),
practical reasons. Summary statistics for macroin- in the United States. SWDI values of 1.9 to 0.46
vertebrate species richness and Shannon-Weiner observed in the 3 wetlands may be representa-
diversity index were calculated to determine nor- tive of the true range in constructed wetlands in
mality of data as per Boix et al. (2007). Analysis of metropolitan Melbourne, since they fall within this
variance as employed by Muli (2005) and Nzengy’a range.
and Wishitemi (2000) was the most accessible and One way ANOVA for species richness with a P-
affordable data analysis method for assessing signif- value of 0.83, (P > 0.05, F = 0.19), indicates that
icant differences in Shannon-Weiner diversity index the macro-invertebrate species richness in the 3 wet-
and species richness, using MS Office Excel. Tem- lands is similar. This may appear to be contradicted
poral variation of Shannon-Weiner diversity index by the fluctuation in temporal variation of SWDI
and species richness was plotted on a time graph between April and October, seen in Figure 3. How-
(Anamaet et al., 2005). Summary statistics on raw ever, if considered in conjunction with the dominant
data show they were not normal; however the de- taxa listed in Table 2, it could be suggested that the
476 Awal and Svozil / Aquatic Ecosystem Health and Management 13 (2010) 472–479

Table 2. Taxa found in the Police Road retarding basin (PRRB), Troups Creek wetlands (TCW) and Dandenong retarding basin
(DRB) wetlands including 5-dominant taxa in each wetland.

PRRB TCW DRB

General General General 5 dominant

FAMILY taxa 5 dominant taxa taxa 5 dominant taxa taxa taxa
Trichoptera (Order) Chironomidae
Coryxidae Coryxidae
Oligochaeta (Order)
Lymnaedae
Coenagrionidae
Chironomidae
Cordullidae
Coryxidae
Lymnaedae
Oligochaeta
Aeshnidae
Dysticidae
Dysticidae
Lymnaedae
Coe nagrionidae
Ceratopogonidae
Chironomidae
Coenagrionidae
Glossiphonidae


Coryxidae



Notonectidae



Lymnaedae



Stratyomidae


Ceratopogonidae



Chironomidae



Parastacidae

Atyidae

Dysticidae



Amphipoda (Order)

5 dominant taxa in each wetland may be ecologi- Discussion

cal generalists, capable of exploiting a wide range
of conditions (Kratzer and Batzer, 2007). As a re- Of the 16 taxa found in the three wetlands, the
sult, the consistent occurrence and relatively high majority of these may be considered cosmopoli-
frequency of the same 5 taxa in samples from their tan with the exception of Trichoptera, Atyidae and
respective wetland is reflected in the average species Amphipoda in PRRB, Glossiphonidae, Stratyomi-
richness of the 3 wetlands: being 6.17 for PRRB, dae and Parastacidae from TCW and Trichoptera,
6.33 for TCW and 5.92 for DRB. Parastacidae and Atyidae from DRB wetlands. As

Figure 2. Time graph showing temporal variation in Shannon–Weiner diversity index.

 Awal and Svozil / Aquatic Ecosystem Health and Management 13 (2010) 472–479 477

Figure 3. Time graph showing temporal variation in species richness.

Bryant and Papas (2007) suggest, it is likely these ‘weaker’ or less dominant taxa, and may be more
taxa are common to most constructed and natu- sensitive to adverse conditions (USEPA, 2002) such
rally occurring wetlands and other water bodies in as pollution or excessive nutrient concentration.
the Eastern, Northern and Western metropolitan re- They may simply be less competitive within the
gions of Melbourne. The community composition ecosystem and this may explain the fluctuation in
of macroinvertebrates in each wetland was domi- SWDI values over the study period. Their rela-
nated by 5 main taxa listed in Table 2, in order of tively inconsistent occurrence between fortnightly
highest to lowest frequency. samples explains their lack of influence on average
It is known that increased nutrient load and thus species richness and the similarity of species rich-
reduced dissolved oxygen affect a reduction in over- ness between the 3 wetlands.
all macroinvertebrate diversity (Cyr and Downing, Collection of physico-chemical data, including
1988). One way ANOVA for SWDI with P-value of dissolved oxygen, total dissolved solids, salinity, pH
0.09 (P > 0.05, F = 2.54) indicated a comparatively and temperature for each wetland was required to
higher probability for significant difference, than validate any differences in macroinvertebrate diver-
for species richness. This may have resulted from sity between the wetlands (Bryant and Papas, 2007).
higher nutrient loads in one or two of the three wet- However, the volume of work required for the lim-
lands due to long or short term enrichment through ited fortnightly time frame for sampling meant this
run-off and other nutrient inputs; however, the part of the study could not be carried out and only
source and magnitude of the variation in enrichment macroinvertebrate diversity was assessed.
and water quality could not be quantified or con- The application of macroinvertebrate diversity as
firmed, as this could not be completed in the course a measure of wetland ecosystem integrity was con-
of the study. The variation in macrophyte coverage firmed as being a reliable tool for the purpose of
of the three wetlands may have also been a factor in conservation and management of these constructed
the increased probability for significant difference wetlands. The similarities in the species richness
in SWDI; however, the percentages shown in and Shannon Weiner diversity index suggested such
Table 1 are only estimates and cannot be seen as an results may be common, with some regional or lo-
influencing factor without further investigation. cal variation, and so macro-invertebrate diversity
The remaining, less frequent and abundant taxa may be considered a universal measure of wetland
found in each wetland may be what are termed ecosystem integrity.
478 Awal and Svozil / Aquatic Ecosystem Health and Management 13 (2010) 472–479

Conclusions Boix, D., Sala, J., Gascon, S., Martinoy, M., Gifre, Brucet, J.S.,
Badosa, A., Lopez-Flores, R., Quintana, X.D., 2007. Com-
The similarity of collected data for species rich- parative biodiversity of crustaceans and aquatic insects from
ness and Shannon Weiner diversity index to pub- various water body types in coastal Mediterranean wetlands.
Hydrobiologia 584, 347–359.
lished data (Anamaet et al., 2005; Webb and Mitsch, Bryant, D., Papas, P., 2007. Macro-invertebrate communities in
2001) indicated that these measures of macroinver- artificial wetlands: the influence of macrophytes, Arthur Ry-
tebrate diversity are universally suitable for use in lah Institute for Environmental Research, Melbourne.
constructed wetlands. The accuracy and validity of Cyr, H., Downing, J.A., 1988. The abundance of phytophilous
this data would be strengthened with further simi- invertebrates on different species of submerged macrophytes.
lar studies in alternative climates, other constructed Freshwater Biology 20, 365–374.
Davis, J.A., Rosich, R.S., Bradley, J.S., Growns, J.E., Schmidt,
wetlands and other geographical locations.
L.G., and Cheal, F., 1993. Wetlands of the Swan Coastal Plain
The temporal variation in SWDI and species –Vol. 6: Wetland classification on the basis of water quality
richness showed a definite trend, and while this and invertebrate community data. Western Australia Water
agrees with known climatic and ecosystem varia- Authority, EPA, Leederville, Perth. http://www.environment.
tion (Adamus and Brandt, 1990; Kratzer and Batzer, murdoch.edu.au/groups/aer/reports.html.
2007), further studies may assist in determining if Davis, J.A., Horwitz, P., Norris, R., Chessman, B., Mcguire, M.,
local variation in wetland ecosystems in different B. Somer, B., Taylor, K., 1999. Wetland Bio-assessment Man-
ual (Macroinvertebrates), Aquatic Ecosystems Research –
geographical locations and climates is significantly
Murdoch University, Murdoch, Western Australia.
different, or similar enough to one another so as to De Leo, G. A., Levin, S., 1997. The multifaceted aspects of
be negligible. ecosystem integrity. Conservation Ecology 1(1), 3.
The overall implications from the present study Kratzer, E.B., Batzer, D.P., 2007. Spatial and temporal variation in
were that, there was high potential for its univer- aquatic macroinvertebrates in the Okefenokee Swamp, Geor-
sal application in measuring ecosystem integrity in gia, USA. Wetlands 27(1),127–140.
most if not all constructed wetlands despite location, Melbourne Water, 2003. Wetland Nature’s filter, URL: http://
www.education.melbournewater.com.au/content/rivers and
climate or ecosystem characteristics. drainage/wetlands -naturesfilter/wetlandsnaturesfilter.asp
Further studies to determine if these conclusions Accessed 4/4/09.
apply during spring and summer would be desirable Melbourne Water, 2009. Water Sensitive Urban Design-Case
to strengthen the results of the present study, since studies, Melbourne. Accessed 7/4/09.URL: http://wsud.
the study was completed over autumn and winter. melbournewater.com.au/content/case studies/case studies.
Whether pollution and macrophyte coverage have asp
Muli, J.R., 2005. Spatial variation of benthic macroinvertebrates
any confounding effects on the reliability of these
and the environmental factors influencing their distribution
measures should also be observed. in Lake Victoria, Kenya. Aquatic Ecosystem Health & Man-
agement 8(2), 147–157.
Natural Resource Management (NRM), 2008. Wetland ecosys-
tem condition: macro-invertebrate indicator species. Com-
References monwealth of Australia, Canberra. http://www.nrm.gov.au/
Adamus, P.R., Brandt, K., 1990. Impact on quality of inland publications/factsheets/me-indicators/inland-aquatic/pubs/
wetlands of the United States: A survey of indicators, tech- wetland-condition-species.pdf
niques and applications of community level bio-monitoring Nzengy’a, D.M., Wishitemi, B.E.L., 2000. Dynamics of benthic
data, Office of Research and Development, United States macroinvertebrates in created wetlands receiving wastewater.
Environmental Protection Agency (USEPA), Washington International Journal of Environmental Studies 57(4), 419–
DC. 436.
Anamaet, E.R., Taylor, J., Mitsch, W.J., 2005. Macro-invertebrate Sharma, R.C, Rawat, J.S., 2009. Monitoring of aquatic inverte-
diversity and abundance in two created wetlands in Ohio, brates as bio-indicators for assessing the health of wetlands:
Ohio State University, Ohio. A case study in the Central Himalayas, India. Ecological In-
Azous, A.L., Horner, R.R., 2001. Wetlands and urbanization: dicators 9, 118–128.
Implications for the future. CRC Press LLC, Florida. Spieles, D.J., Mitsch, W.J., 2000. Macro-invertebrate community
Barber, W.E., Kevern, N.R., 1973. Ecological factors influencing structure in high and low nutrient constructed wetlands. Wet-
macroinvertebrate standing crop distribution. Hydrobiologia lands 20(4), 716–729.
43(1–2), 53–75. Sutherland, W.J. (Ed.), 2006. Ecological Census
Bennett, J.W., Whitten, S.M., 2001. Social and private values of Techniques–2nd Ed. Cambridge University Press, New
wetlands. Environment Australia and Land and Water Re- York.
sources Research and Development Corporation, Canberra, USEPA, 1997. Wetlands: Biological Assessment Methods and
Australia. Criteria development workshop. Office of Water, United
 Awal and Svozil / Aquatic Ecosystem Health and Management 13 (2010) 472–479 479

States Environmental Protection Agency, Washington, Webb, K.D., Mitsch, W.J., 2001. Macroinvertebrate diversity
DC. and abundance in two experimental wetlands from top-down
USEPA, 2002. Methods for evaluating wetland condition: Devel- and bottom-up interpretations. Olentangy River Wetland Re-
oping an invertebrate index of biological integrity for wet- search Park, Ohio State University, Ohio.
lands. Office of Water, United States Environmental Protec- Wilhm, J.L., Dorris, T.C., 1968. Biological Parameters for Water
tion Agency, Washington, DC. Quality Criteria. Bio-Science 18(6), 477–481.
Copyright of Aquatic Ecosystem Health & Management is the property of Taylor & Francis Ltd and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.