Liao et al.

Zoological Studies (2015) 54:14

DOI 10.1186/s40555-014-0101-8

RESEARCH Open Access

Meiofaunal communities in a tropical seagrass

bed and adjacent unvegetated sediments with
note on sufficient sample size for determining
local diversity indices
Jian-Xiang Liao1*, Hsin-Ming Yeh2 and Hin-Kiu Mok1

Abstract
Background: Seagrass beds are highly diverse and productive marine habitats for many associated organisms in
nearshore coastal waters. The differences in abundance, diversity, and community structure of benthic invertebrates
between seagrass beds and adjacent unvegetated sediments have been stated, whereas most studies are primarily
focused on macrofauna or based on a comparatively long distance, i.e., more than 10 m. The present study is
designed to test if the community structures of meiofauna, especially the free-living nematodes, differ between
seagrass beds and adjacent unvegetated sediments on a meter scale.
Results: There are 21 meiofaunal taxa and 63 nematode genera that have been identified from a tropical seagrass
bed of Thalassia hemprichii in Ludao, Taiwan. Although the compositions of higher meiofaunal taxa are undistinguished,
according to correspondence analysis, the assemblages of nematode genera differ substantially between the seagrass
bed and unvegetated sediments. Regarding the nematodes, approximately 50% of genera are restricted to the seagrass
bed whereas 6% are restricted to unvegetated sediments, which indicate both habitats possessing distinct infaunas. The
number of replicates for reasonable estimation of the local diversity index is calculated by the randomization technique.
For local seagrass beds, only a single core is sufficient for reliably estimating meiofaunal diversity, but at least three cores
or a sample size of 300 individuals is needed for the nematode community.
Conclusions: Nematode assemblages provide more particular differences between seagrass and unvegetated habitats
than meiofaunal communities on small spatial scales. Both seagrass beds and adjacent unvegetated sediments harbor
specific meiofaunal communities, and hence, the conservation strategy for seagrass should also consider the peripheral
bare area of seagrass beds.
Keywords: Meiobenthos; Thalassia hemprichii; Community composition; Marine nematode; Randomization; Diversity
index

Background addition, various detritus and organic particles in seagrass

Seagrass beds, which distribute widely in tropical and beds enhance the density and heterogeneity of infauna
temperate coastal regions, are one of the most productive (Hall and Bell 1993; Danovaro et al. 2002). Meiofauna,
marine ecosystems (Duarte and Chiscano 1999). Thalli, benthic organisms between 1 mm and 42 μm, is consid-
roots, and rhizomes of seagrasses offer associated animals ered a trophic level between microfauna and macrofauna
microhabitats and shelters (Osenga and Coull 1983; Orth (animals larger than 1 mm). However, most studies deal-
et al. 1984; Castel et al. 1989; Atilla et al. 2005). In ing with seagrass beds merely focus on macrofauna, even
though meiofauna can consume from 10% to more than
50% of the total primary production in a seagrass system
* Correspondence: [email protected]
1
Department of Oceanography, National Sun Yat-sen University, Lienhai Road
(Danovaro et al. 2002). Moreover, most members of meio-
70, Kaohsiung 80424, Taiwan fauna are less influenced by hypoxia and predation than
Full list of author information is available at the end of the article

© 2015 Liao et al.; licensee Springer. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction
in any medium, provided the original work is properly credited.
Liao et al. Zoological Studies (2015) 54:14 Page 2 of 10

those of macrofauna and have distinct responses to envir- are used to examine the difference between seagrass beds
onmental changes (Josefson and Widbom 1988; Castel and adjacent unvegetated sediments on a meter scale.
et al. 1989; Giere 2009). Among meiofaunal communities, Meiofaunal samples were collected quantitatively from sed-
free-living nematodes usually constitute the most abundant iments below the sickle seagrass Thalassia hemprichii and
taxa, albeit seagrass studies of meiofauna have primarily peripheral unvegetated substrates in Ludao, Taiwan. The
focused on epifauna, i.e., harpacticoid copepods. Nematode sampling effort (i.e., the number of replicates) required for
assemblages shift consistently with ambient sediments, and estimating the local diversity index in a seagrass habitat is
they are considered an ecological indicator (Bongers and determined using a randomization test to provide a basis
Ferris 1999). for future experimental design of comparative researches.
The unvegetated area around seagrass patches facilitates
infauna dispersal and can be a matrix habitat, but few sea- Methods
grass studies have looked into this bare region (Boström Study site and sampling methods
et al. 2006). The abundance and community structure of Shihlang is a recreational diving site in Ludao, also known
nematodes show substantial differences between vegetated as Green Island, situated off the southeastern coast of
and peripheral unvegetated sites (Fisher and Sheaves 2003; Taiwan (Figure 1). The shore is composed of a sandy
Hourston et al. 2005; Monthum and Aryuthaka 2006; platform extending 50 m seaward and a T. hemprichii
Fonseca et al. 2011; Leduc and Probert 2011), although no meadow (22°39′08″ N, 121°28′25″ E) in the littoral zone.
difference between both habitats has also been reported Patches of seagrass distribute at a region 150 m in length
(Ndaro and Ólafsson 1999). The variability in nematode and 12 m in width along the coast. The distance between
populations on a small spatial scale (centimeter scale) the fringes of seagrass beds to the mean high water line is
accounts for 30% and 43% in intertidal and shallow about 18 m. Because of the freshwater input, the salinity
subtidal sediments, respectively (Hodda 1990; Li et al. reaches approximately 20 to 25 practical salinity units
1997). However, the distances between samples or quad- (PSU) during low tide. Megafaunal bioturbators are fre-
rats in most seagrass studies are separated by more than quently present, mainly composed of crustaceans and
10 m or unmentioned (except Hourston et al. 2005; 2 m), echinoderms.
whereas the other unanticipated local environmental Regarding the narrow distribution of the seagrass patches
factors (e.g., water depth) exceed the effect of seagrass and avoiding intense environmental changes, four 5 × 5 m
(Mills and Berkenbusch 2009). Comparisons between side-by-side sampling quadrats were selected: three seagrass
seagrass beds and peripheral unvegetated sediments at patches and an adjacent unvegetated control quadrat.
small-scale distances elucidate the seagrass effect on The vegetated quadrats densely grew T. hemprichii and
meiofaunal communities. patches with sparse region, where taking cores was
Diversity indices are widely used in ecological research, avoided. Within each quadrat, five replicate cores for
especially for comparative purposes. The Shannon-Wiener meiofauna were sampled by a lower end cutoff syringe
index (H’) is the most popular one for measuring diversity (diameter = 2 cm), which was pushed into the sediment to
in a community. Because of the sample size dependence, a depth of 5 cm. On each sampling occasion, the replicates
a pertinent sampling effort is required for comparison were at least 1 m apart from each other. All samples
between various communities. The randomization tech- collected from the area were 20 m long and 5 m wide at
nique has been used to estimate the sufficient sample approximately 30-cm water depth during low tide on 17
size for obtaining an approximate value of the diversity March 2012. Samples of meiofauna were anesthetized
index (Soetaert and Heip 1990). However, Hurlbert’s using 7% MgCl2 solution to promote tissue relaxation and
modified rarefaction curve (Hurlbert 1971) or the expected then preserved with formaldehyde to a final concentration
species richness, ES(n), a sample size-independent index of of 4%. The sediment grain size profile was very similar to
species richness (Sanders 1968), is much more often used the sandy platform, and five additional replicate cores
to estimate the diversity of meiofauna (e.g., Shimanaga were taken for sediment granulometry as a reference.
et al. 2004; Leduc et al. 2010). In addition, most relevant
investigations are performed in temperate regions or the Laboratory procedures
deep sea, but the same procedure may not be appropriate Meiofaunal samples were stained with 0.5 g/L Rose
for tropical habitats. To our knowledge, there is an absent Bengal for 1 h and washed through sieves of 1,000- and
or scarce report on the marine meiofaunal communities 42-μm mesh size. All the meiofauna retained on the
in Taiwan, and a relevant method for determining local 42-μm sieve were extracted, counted, and identified to
diversity indices might improve the future comparison the major taxon level under a stereomicroscope. From
with the other regions. each replicate, 100 nematodes (or all individuals if fewer)
The present study aims to inspect the community were picked out, transferred to pure glycerol, mounted
structure of meiofauna, particularly nematodes, which onto permanent slides, and identified to the genus level
Liao et al. Zoological Studies (2015) 54:14 Page 3 of 10

Figure 1 Map of the sampling site in Shihlang (Ludao) with indication of the quadrats. SC, unvegetated quadrat; SEA, SEB, and SEC,
seagrass quadrats.

using the pictorial keys (Platt and Warwick 1988; Warwick cluster analysis and CA, if the replicates from identical
et al. 1998). The trophic types of nematodes were classified quadrats were not clustered an obvious group, the abun-
by the morphology of buccal cavities (Wieser 1953). For dance data of these replicates were pooled. The numbers
granulometric analysis, sediment samples were first washed of individuals among the core replicates were highly
with freshwater to remove salt, dried at 60°C for 48 h, and variable, so the mean abundance of meiofauna was chosen
then separated out on mesh sizes of 1,000, 500, 250, 125, as a representative sample size for calculating efficiency.
and 62 μm. Regarding nematodes, 100 individuals, the most commonly
selected number for taxonomic identification, were used to
Statistical analyses simulate a given sample. All abundance data of seagrass
Multivariate and univariate statistical methods were used beds were pooled for the randomization simulation by Yeh’s
to discriminate meiofaunal communities and nematode computer program (Yeh and Ohta 2002). Pooled data were
assemblages between seagrass beds and unvegetated randomly selected to achieve the predetermined number of
sediments. Community structures were examined using sample sizes. The significant differences of ten replicates of
classification and ordination methods to test the con- the Shannon-Wiener index (H’) among every dissimilar
sistency of the core replicates. Cluster analysis based on sample sizes were examined using the Kruskal-Wallis
the similarity of the Morisita-Horn index (Horn 1966), test. Dunn’s test for multiple comparisons was used to
which was independent of sample size and diversity, was determine which values differed significantly if the
conducted to group the subsamples. The unweighted Kruskal-Wallis test revealed a significant difference.
pair group method with arithmetic mean (UPGMA)
was used to construct the dendrograms. Correspondence Results
analysis (CA) was selected as the ordination method for Environmental and biological variables
appropriating to species abundance data. Scaling type 1 The quadrats covered an area 20 m long and 5 m wide,
(Legendre and Legendre 1998) was chosen for preserving where salinity was approximately 20 PSU during the
the chi-square distances between sites. For meiofauna and sampling occurrence at low tide. Sediment grain sizes
nematode assemblages, the abundance data were square (n = 5) showed peaks around coarse (500 to 1,000 μm) or
root transformed to reduce the skewness before the multi- very coarse (1,000 to 2,000 μm) sand, with the average
variate analysis. The square root transformation was used median grain size 537 ± 51 μm and very few silt (<62 μm)
to balance between retaining quantitative information and contents (<1%).
downweighting the dominant taxa. A total of 21 meiofaunal taxa were identified in the
The randomization technique was used to calculate a present study (Table 1). Meiofaunal density in the top
sufficient number of replicates for achieving a reasonable 5 cm of sediments varied between 188 and 2,978 individ-
estimate of the diversity index in the sampling site uals 10 cm−2, whereas mean densities showed 1,172 and
(Soetaert and Heip 1990). According to the results of 1,345 individuals 10 cm−2 for vegetated and unvegetated
Liao et al. Zoological Studies (2015) 54:14 Page 4 of 10

Table 1 Abundance of higher meiofaunal taxa from Consistency of core replicates

seagrass beds and adjacent unvegetated sediments at The univariate analyses cannot detect any significant
Shihlang, Ludao differences in meiofaunal abundance, nematode abundance,
Higher Seagrass (n = 15) Unvegetated (n = 5) and nematode diversity between seagrass and unvegetated
meiofaunal taxa
Mean SD Mean SD habitats. According to the multivariate analysis, most sub-
Acari 1.5 2.0 1.3 0.6 samples were clustered at a high similarity level (>0.80)
Amphipoda 3.2 4.8 0.6 0.3 for higher meiofaunal taxa, whereas replicates of various
habitats were undetermined and combined randomly
Bivalvia 0.2 0.8 0.6 0.3
(Figure 2a). For nematode genera, all replicates from
Ciliophora 11.0 11.2 3.8 1.0
the seagrass quadrats were clustered to a single group at a
Copepoda 160.7 201.6 607.6 82.3 similarity level of 0.51, whereas one unvegetated replicate
Cnidaria 0.8 1.9 1.3 0.6 was highly dissimilar from the others (Figure 2b).
Foraminifera 0.6 0.3 The results of CA were similar to what cluster analysis
Gastropoda 1.3 0.3 displayed. For higher meiofaunal taxa, the eigenvalues of
the first three axes were 0.207, 0.145, and 0.078 and
Gastrotricha 3.6 4.0 0.6 0.3
accounted for 27.7%, 19.5%, and 10.4% of the variation,
Hirudinea 0.2 0.8
respectively (Figure 3a). Except for one seagrass replicate
Isopoda 7.4 15.8 (SEC3 in Figure 3a, which possessed rare taxa, i.e., the
Nauplius 16.6 45.1 138.9 30.6 only rotifer in this study and a sipunculan) in the second
Nematoda 853.3 731.1 475.8 126.0 quadrant, the other replicates were positioned near the
Nemertea 1.1 2.0 origin and distributed along the first axis. A seagrass
replicate (SEB5 in Figure 3a, which possessed the fewest
Oligochaeta 5.9 8.2
nematodes showing a low nematode/copepod ratio as in
Ostracoda 5.7 7.1 22.3 4.7
the unvegetated samples) and the unvegetated samples
Polychaeta 71.5 97.6 69.4 12.6 were aggregated together. For nematode genera, the
Rotifera 0.2 0.8 eigenvalues of the first three axes were 0.582, 0.287, and
Sipuncula 0.4 1.1 0.6 0.3 0.254 and accounted for 21.5%, 10.6%, and 9.4% of the
Turbellaria 28.7 20.5 15.3 4.3 variation, respectively (Figure 3b). In addition to one
unvegetated replicate (SC4 in Figure 3b, which revealed
Undetermined egg 4.5 1.0
an extremely distinct generic composition from the other
The mean and standard deviation (SD) of each meiofaunal taxon are shown in
individuals per 10 cm2. replicates: 71 individuals of Longicyatholaimus, 10 of Mesa-
canthion, and 2 rare taxa, Oxystomina and Thalassironus,
sites, respectively. Nematoda was the most abundant only appearing in this sample) in the fourth quadrant,
taxon (72.8%) in the seagrass quadrats, followed by the other replicates were distributed along the second
Copepoda (13.7%) and Polychaeta (6.1%). However, axis. All seagrass replicates were located near the origin
Copepoda accounted for 45.2% and dominated the and well separated from the replicates of unvegetated sedi-
unvegetated site, followed by Nematoda (35.4%), nau- ments. The non-metric multidimensional scaling (MDS)
plius (10.3%), and Polychaeta (5.2%). The remaining ordination for meiofauna and nematodes (data not shown)
groups represented less than 6% of total meiofauna. both represented a considerable degree of similarity to the
Regarding nematodes, densities within each sample results of CA but with a high stress value (0.16).
ranged from 51 to 2,682 individuals 10 cm−2. A total of Cluster analysis and CA revealed that the nematodes
63 nematode genera belonging to 20 families were iden- showed dissimilar community structures between seagrass
tified from 1,680 individuals (Table 2). Ten genera (6.3% beds and unvegetated sediments. Therefore, the random-
of total) were restricted to unvegetated sediments, and ization test was constructed afterward only according to
30 genera (47.6%) solely occurred in seagrass beds. the seagrass samples.
Daptonema was the most dominant genus and accounted
for similar percentages in both habitats, whereas most Randomization test
genera (44 genera) made poor contributions (<1% of total All of the 15 cores from seagrass quadrats were combined
abundance). Four genera, i.e., Daptonema, Perspiria, in this analysis. For meiofauna, the mean abundance of
Robbea, and Paralinhomoeus, contributed 50.2% to the the seagrass samples (i.e., 368 individuals) was used to
total nematode abundance at seagrass sites, whereas simulate a sample size for the randomization test. The
three genera, i.e., Daptonema, Longicyatholaimus, and cumulative diversity curve (Figure 4a) flattened from the
Spirinia, made a larger contribution to the abundance first sample size (n = 368) to the total abundance (n = 5,521;
(60.1%) at the adjacent unvegetated site. H’ = 1.0), and the Kruskal-Wallis test showed no significant
Liao et al. Zoological Studies (2015) 54:14 Page 5 of 10

Table 2 Percentages of nematode genera from seagrass beds and adjacent unvegetated sediments at Shihlang, Ludao
Nematode genus Family Trophic type Seagrass Unvegetated
Acanthonchus Cyatholaimidae 2A 0.2 0.0
Alaimella Leptolaimidae 1A 0.1 0.0
Anticoma Anticomidae 1A 1.3 0.3
Bathyeurystomina Enchelidiidae 2B 0.2 0.0
Calyptronema Enchelidiidae 2B 0.2 0.0
Camacolaimus Leptolaimidae 2A 0.5 0.0
Chromadorina Chromadoridae 2A 1.1 0.0
Chromadorita Chromadoridae 2A 6.1 3.4
Daptonema Xyalidae 1B 21.2 23.3
Demonema Selachnematidae 2B 0.3 0.0
Desmodora Desmodoridae 2A 0.1 0.6
Desmodorella Desmodoridae 2A 0.0 0.3
Desmolaimus Linhomoeidae 1B 1.2 0.3
Dichromadora Chromadoridae 2A 0.0 0.6
Diodontolaimus Leptolaimidae 2A 1.1 0.0
Elzalia Xyalidae 1B 0.8 0.3
Epsilonema Epsilonematidae 1A 1.7 2.0
Eubostrichus Desmodoridae 1A 0.0 1.1
Eurystomina Enchelidiidae 2B 0.2 0.3
Filoncholaimus Oncholaimidae 2B 0.1 0.0
Gammanema Selachnematidae 2B 0.2 0.3
Graphonema Chromadoridae 2A 0.0 6.0
Halalaimus Oxystominidae 1A 0.5 0.0
Halaphanolaimus Leptolaimidae 1A 0.2 0.0
Halichoanolaimus Selachnematidae 2B 0.4 0.9
Leptolaimus Leptolaimidae 1A 0.5 0.0
Longicyatholaimus Cyatholaimidae 2A 0.0 22.1
Megadesmolaimus Linhomoeidae 1B 0.5 0.0
Mesacanthion Thoracostomopsidae 2B 0.1 3.2
Metachromadora Desmodoridae 2A 4.9 0.9
Metalinhomoeus Linhomoeidae 1B 3.0 0.0
Meyersia Oncholaimidae 2B 0.5 0.3
Microlaimus Microlaimidae 2A 0.2 0.0
Molgolaimus Desmodoridae 1A 0.9 0.0
Neochromadora Chromadoridae 2A 2.5 0.0
Odontanticoma Anticomidae 2A 0.1 0.0
Oncholaimus Oncholaimidae 2B 2.0 1.4
Oxystomina Oxystominidae 1A 0.0 0.6
Paracyatholaimus Cyatholaimidae 2A 0.9 0.0
Paralinhomoeus Linhomoeidae 1B 7.4 3.4
Perspiria Desmodoridae 2A 13.4 0.6
Phanoderma Phanodermatidae 2A 0.2 0.0
Phanodermella Phanodermatidae 1A 0.2 0.0
Pomponema Cyatholaimidae 2A 0.6 2.0
Liao et al. Zoological Studies (2015) 54:14 Page 6 of 10

Table 2 Percentages of nematode genera from seagrass beds and adjacent unvegetated sediments at Shihlang, Ludao
(Continued)
Prochromadorella Chromadoridae 2A 0.4 0.6
Promonhystera Xyalidae 1B 0.1 0.0
Pseudocella Leptsomatidae 2A 0.3 0.0
Pseudochromadora Desmodoridae 2A 5.9 1.7
Pseudonchus Desmodoridae 2B 0.0 1.7
Rhinema Monoposthiidae 2A 0.9 0.0
Robbea Desmodoridae 1A 8.1 0.6
Spilophorella Chromadoridae 2A 2.6 4.3
Spirinia Desmodoridae 2A 0.5 14.7
Steineria Xyalidae 1B 0.9 0.0
Symplocostoma Enchelidiidae 2B 0.1 0.0
Symplocostomella Enchelidiidae 2B 0.1 0.0
Synonchus Leptsomatidae 2A 0.1 0.0
Thalassironus Ironidae 2B 0.0 0.9
Thalassomonhystera Monhysteridae 1A 0.1 0.0
Theristus Xyalidae 1B 3.2 0.0
Trichotheristus Xyalidae 1B 0.0 0.3
Vasostoma Comesomatidae 2A 0.0 0.3
Viscosia Oncholaimidae 2B 1.7 0.9
The percentages of the ten most dominant nematode genera for both habitats are shown in italics. Trophic type: 1A, selective deposit feeder; 1B, non-selective
deposit feeder; 2A, epistrate feeder; 2B, predator or omnivore.

difference among the cumulative sample sizes. For 100 and 200 individuals revealed significant differences
nematodes, 100 individuals, the most commonly selected from the other cumulative sample sizes. The cumulative
number for taxonomic identification, were used to simu- sample size of 200 nematodes merely shows a significant
late a given sample. After the two replicates (n = 200) difference with 1,300 individuals but is indistinguishable
were pooled, the cumulative diversity curve (Figure 4b) from the ultimate sample size (n = 1,332) by Dunn’s test
flattened to the total number of individuals (n = 1,332; (Figure 4b). For reliably estimating the local diversity
H’ = 2.9). The Kruskal-Wallis test showed a significant index of nematode genera, a cumulative sample size of at
difference among various sample sizes (p < 0.001). Only least 300 individuals is recommended.

Figure 2 Dendrograms based on the similarity of the Morisita-Horn index for (a) higher meiofaunal taxa and (b) nematode genera.
SC, replicates from the unvegetated quadrat; SEA, SEB, and SEC, replicates from each seagrass quadrat. Numbers indicate the number of replicates.
Liao et al. Zoological Studies (2015) 54:14 Page 7 of 10

Figure 3 Scatterplots by correspondence analysis of (a) higher meiofaunal taxa and (b) nematode genera. The abundance data have
been square root transformed.

Discussion beds (e.g., Hicks 1986; Ansari and Parulekar 1994; Ndaro
According to the multivariate analyses, the nematode and Ólafsson 1999; De Troch et al. 2001), but a contrary
abundance and diversity recorded in the seagrass bed being result has also been reported (Leduc and Probert 2011).
higher than those in the adjacent unvegetated sediments Shallow seagrass beds may not substantially influence the
are consistent with previous studies (Castel et al. 1989; abundance of copepods as the deeper ones do (De Troch
Danovaro et al. 2002; Hourston et al. 2005; Monthum and et al. 2001). The abundance of copepods is positively
Aryuthaka 2006; Leduc and Probert 2011). However, an related to the coarser sediments (Coull 1985), whereas
unexpected pattern was observed in this study where sediments underneath seagrass are commonly finer than
copepods display a reverse trend to nematodes, i.e., the those of unvegetated areas. In addition, the dense canopies
abundance of copepods inside the seagrass bed is lower of seagrasses prevent the growth of microphytobenthos,
than that in the unvegetated area. Most surveys have which are the main diet of copepods (Leduc et al. 2009;
mentioned that copepod densities are higher in seagrass Leduc and Probert 2011). The epibenthic and interstitial

Figure 4 Results of the randomization test for (a) higher meiofaunal taxa and (b) nematode genera. Standard deviations are shown as
vertical bars. The sample sizes with an asterisk reveal the significant differences of the Shannon-Wiener index from the other sample sizes.
Liao et al. Zoological Studies (2015) 54:14 Page 8 of 10

copepod taxa have distinct environmental demands and Metalinhomoeus, Paralinhomoeus, Pomponema, Spilophor-
stress endurance (Hicks 1986; Giere 2009), so classifying ella, Spirinia, and Theristus (Heip et al. 1985), to occur as
subgroups of copepods may clarify their preference for the the dominant genera. The salinity stress drives the abun-
microhabitats in seagrass beds. dance of few euryhaline species while rich nutrient supply
The seagrass samples bear a rough resemblance in by seagrass creates a luxuriant microbial food resource,
meiofauna as revealed by multivariate analyses but with which attracts the rare deposit-feeding nematodes from
exceptions. The differences in meiofaunal communities peripheral sediments (Fisher and Sheaves 2003). Definite
between seagrass beds and adjacent unvegetated sediments differences are present in nematode assemblages between
may be hardly detectable at a meter scale, and most higher both habitats, and shifts in nematode generic composition
meiofaunal taxa, e.g., copepods and nauplii, have substan- have occurred on a meter scale. For Chromadoridae,
tially higher dispersal rates than nematodes, which affect Chromadorina and Neochromadora are restricted to
their temporary local distribution (Commito and Tita seagrass beds, whereas Dichromadora and Graphonema
2002). The active seagrass-associated taxa could forage for are restricted to unvegetated sediments. For Desmodoridae,
foods in peripheral unvegetated sediments but keep close Robbea, Metachromadora, Perspiria, and Pseudochroma-
to seagrass beds (Orth et al. 1984). The wave disturbance dora are dominant in seagrass beds, whereas Eubostrichus,
and salinity changes in the coastal region tend to mitigate Pseudonchus, and Spirinia are restricted to or dominant in
small-scale patchiness and local diversity of nematodes unvegetated sediments. Various genera of the same family
(Lambshead and Boucher 2003). However, the composi- may have distinct preferences for vegetated and unvege-
tions of nematode genera have various arrangements in tated habitats, or the distribution pattern could be the
this study, even at a scale within several meters. competitive result in which similar species compete for
Regarding the trophic types of nematodes, epistrate the plentiful food sources in seagrass sediments. In
feeders (2A; trophic type by Wieser 1953) are generally addition to the chromadorids, desmodorids, and xyalids
dominant in sandy sediments and non-selective deposit common to both habitats, linhomoeids (e.g., Paralinho-
feeders (1B) are highly abundant in muddy regions, moeus and Metalinhomoeus) are obviously occurring in
whereas a recent study has pointed out that the trophic seagrass beds, whereas cyatholaimids (e.g., Longicyatholai-
structure is coupling with the potential food sources mus) dominate in unvegetated sediments. Linhomoeids
rather than the other environmental factors (Danovaro are also restricted to the seagrass beds in Australia, and it
and Gambi 2002). Daptonema, which possesses the has been suggested that they are tolerant to high organic
highest percentage (>20%) in both habitats, dominates matter content and hypoxia in the sediments underneath
wide seagrass habitats, especially in sandy sediments seagrasses (Fonseca et al. 2011). Paralongicyatholaimus,
(Ndaro and Ólafsson 1999; Hourston et al. 2005; Fonseca which resembles Longicyatholaimus, is the dominant genus
et al. 2011). Although Daptonema has been treated as a in seagrass beds and adjacent unvegetated area in Thailand
non-selective deposit feeder, many trophic studies have (Monthum and Aryuthaka 2006).
found that they also feed on diatoms at the surface of Regarding local seagrass beds, only a single core or a
sand grains and can be regarded as epistrate feeders cumulative sample size of 368 individuals is sufficient
(Heip et al. 1985; Moens and Vincx 1997). Comparing for estimating the diversity of meiofauna, but at least
with the uniform distribution of Daptonema, most three cores or a cumulative sample size of 300 individuals
epistrate feeders (most members of Chromadoridae and is needed for nematode diversity. The taxonomic categories
Desmodoridae), which have the highest relative dispersal (family, genus, or species) obviously influence the require-
rate of nematodes (Commito and Tita 2002), show a ments of appropriate sample sizes. The Shannon-Wiener
substantial patchy colonization in this study. Many index (H’), used as a criterion for the randomization test, is
studies have also stated the dominance of epistrate less sample size dependent than Hill’s N1 index (Hill 1973;
feeders in seagrass sediments (e.g., Hopper and Meyers Soetaert and Heip 1990), but H’ has still been applied
1967; Ndaro and Ólafsson 1999; Danovaro and Gambi extensively to compare community diversity and provides
2002; Fonseca et al. 2011). The seagrass samples consisted an appropriate estimate for a community structure. The
of 8.1% of Robbea, a selective deposit feeder (1A), diversity of nematodes (H’ = 2.9), albeit only for genera,
which is coated with particular ectosymbiotic bacteria is lower than the species diversity in Thailand (3.4 to
and occurred extremely rare (0.6%) in unvegetated sedi- 3.7; Monthum and Aryuthaka 2006) but resembles the
ments. A similar deep-dwelling genus, Catanema, also Australian values (2.66 to 3.5; Fisher and Sheaves 2003).
dominates Australian seagrass beds (Fisher 2003; Fisher The nematode assemblages dominate by a small number
and Sheaves 2003). of taxa but still show moderate diversity in tropical sea-
Freshwater input in the sampling site (20 PSU during grass beds (Fisher and Sheaves 2003).
sampling occurrence) allows salinity-tolerant nematodes, At least 300 individuals are recommended for steady
including Chromadorita, Daptonema, Metachromadora, estimation of the local diversity of nematode genera.
Liao et al. Zoological Studies (2015) 54:14 Page 9 of 10

However, 6 out of 15 of the seagrass samples and 3 out Author details

1
of 5 of the unvegetated samples possess less than 100 Department of Oceanography, National Sun Yat-sen University, Lienhai Road
70, Kaohsiung 80424, Taiwan. 2Coastal and Offshore Resources Research
individual nematodes in this study. The use of cores with Center, Fisheries Research Institute, Yugang North 3rd Road 6, Kaohsiung
fairly wide diameters (e.g., 3.6 cm) is suggested for achiev- 80672, Taiwan.
ing reliable quantitative data. A similar result showed that
Received: 26 August 2014 Accepted: 30 December 2014
two 10-cm2 subsampling cores are required to monitor
nematode genus diversity in subtidal stations of the North
Sea (Li et al. 1997). Water flows or bioturbated activities
References
improve oxygen penetration in sandy habitats and then in-
Ansari ZA, Parulekar AH (1994) Meiobenthos in the sediment of seagrass
duce vertical migration of nematodes to a deeper (beneath meadows of Lakshadweep atolls, Arabian Sea. Vie Milieu 44:185–190
5 cm) sediment layer (Steyaert et al. 2003; Giere 2009). Atilla N, Fleeger JW, Finelli CM (2005) Effects of habitat complexity and
The seagrass bed in Shihlang possesses a large amount of hydrodynamics on the abundance and diversity of small invertebrates
colonizing artificial substrates. J Mar Res 63:1151–1172
coarse sands, where nematodes tend to distribute more Bongers T, Ferris H (1999) Nematode community structures as a bioindicator in
deeply (McLachlan et al. 1977; Heip et al. 1985), and thus, environmental monitoring. Trends Ecol Evol 14:224–228
the vertical profile of the nematode community can be Boström C, Jackson EL, Simenstad CA (2006) Seagrass landscapes and their
effects on associated fauna: a review. Estuar Coast Shelf Sci 68:383–403
clarified by sampling deeper than 5- or 10-cm depths Castel J, Labourg PJ, Escaravage V, Auby I, Garcia ME (1989) Influence of seagrass
(Steyaert et al. 2003; Leduc and Probert 2011). beds and oyster parks on the abundance and biomass patterns of meio- and
macrobenthos in tidal flats. Estuar Coast Shelf Sci 28:71–85
Commito JA, Tita G (2002) Differential dispersal rates in an intertidal meiofauna
assemblage. J Exp Mar Biol Ecol 268:237–256
Conclusions
Coull BC (1985) Long-term variability of estuarine meiobenthos: an 11 year study.
The community structure of marine nematode genera Mar Ecol Prog Ser 24:205–218
distinguishes the seagrass habitats from adjacent unvege- Danovaro R, Gambi C (2002) Biodiversity and trophic structure of nematode
tated sediments at a scale within several meters, whereas assemblages in seagrass system: evidence for a coupling with changes in
food availability. Mar Biol 141:667–677
higher meiofaunal taxa are undetermined according to Danovaro R, Gambi C, Mirto S (2002) Meiofaunal production and energy transfer
multivariate analysis. The copepod abundance is unex- efficiency in a seagrass Posidonia oceanica bed in the western Mediterranean.
pectedly higher in peripheral unvegetated habitats than in Mar Ecol Prog Ser 234:95–104
De Troch M, Gurdebeke S, Fiers F, Vincx M (2001) Zonation and structuring
seagrass beds in the present study. Classifying copepods factors of meiofauna communities in a tropical seagrass bed (Gazi Bay,
into epibenthic or interstitial subgroups may elucidate Kenya). J Sea Res 45:45–61
their preference for microhabitats. Regarding nematodes, Duarte CM, Chiscano CL (1999) Seagrass biomass and production: a
reassessment. Aquat Bot 65:159–174
Daptonema, a non-selective deposit feeder, dominates both Fisher R (2003) Spatial and temporal variations in nematode assemblages in
habitats. Longicyatholaimus is abundant and restricted to tropical seagrass sediments. Hydrobiologia 493:43–63
unvegetated sediments, whereas some genera, e.g., Perspiria Fisher R, Sheaves MJ (2003) Community structure and spatial variability of marine
nematodes in tropical Australian pioneer seagrass meadows. Hydrobiologia
and Robbea, are restricted to dominating in sediments 495:143–158
under seagrass. According to randomization analysis, only a Fonseca G, Hutchings P, Gallucci F (2011) Meiobenthic communities of seagrass
single core can reliably estimate meiofaunal diversity, beds (Zostera capricorni) and unvegetated sediments along the coast of New
South Wales, Australia. Estuar Coast Shelf Sci 91:69–77
whereas three cores are sufficient to estimate the local Giere O (2009) Meiobenthology: the microscopic motile fauna of aquatic
diversity of nematodes. Albeit the unvegetated areas may sediments, 2nd edn. Springer-Verlag, Berlin
not contain as diverse infauna as vegetated sediments, they Hall MO, Bell SS (1993) Meiofauna on the seagrass Thalassia testudium—population
characteristics of harpacticoid copepods and associations with algal epiphytes.
could be a matrix habitat and provide corridors between Mar Biol 116:137–146
fragmented seagrass patches (Boström et al. 2006). In Heip C, Vincx M, Vranken G (1985) The ecology of marine nematodes. Oceanogr
addition, both habitats compose specific infauna, and Mar Biol Ann Rev 23:399–489
Hicks GRF (1986) Distribution and behaviour of meiofaunal copepods inside and
the unvegetated site should also be considered in the outside seagrass beds. Mar Ecol Prog Ser 31:159–170
conservation strategy for seagrass. Hill MO (1973) Diversity and evenness: a unifying notation and its consequences.
Ecology 54:427–432
Hodda M (1990) Variation in estuarine littoral nematode populations over three
Competing interests spatial scales. Estuar Coast Shelf Sci 30:325–340
The authors declare that they have no competing interests.
Hopper BE, Meyers SP (1967) Population studies on benthic nematodes within a
subtropical seagrass community. Mar Biol 1:85–96
Authors’ contributions Horn HS (1966) Measurement of “overlap” in comparative ecological studies.
JXL designed and carried out the experiment. JXL and HMY analyzed the Am Nat 100:419–424
data. JXL, HMY, and HKM finalized the manuscript. All authors read and Hourston M, Warwick RM, Valesini FJ, Potter IC (2005) To what extent are the
approved the final manuscript. characteristics of nematode assemblages in nearshore sediments on the
west Australian coast related to habitat type, season and zone? Estuar Coast
Shelf Sci 64:601–612
Acknowledgements Hurlbert SH (1971) The non-concept of species diversity: a critique and alternative
We thank Shih-Wei Su and Yen-Wei Chang for their help in the fieldwork. parameters. Ecology 52:577–586
We are grateful to two anonymous reviewers for their constructive Josefson AB, Widbom B (1988) Differential response of benthic macrofauna and
suggestion. meiofauna to hypoxia in the Gullmar Fjord basin. Mar Biol 100:31–40
Liao et al. Zoological Studies (2015) 54:14 Page 10 of 10

Lambshead PJD, Boucher G (2003) Marine nematode deep-sea biodiversity –

hyperdiverse or hype? J Biogeogr 30:475–485
Leduc D, Probert PK (2011) Small-scale effect of intertidal seagrass (Zostera muelleri)
on meiofaunal abundance, biomass, and nematode community structure. J Mar
Biol Ass UK 91:579–591
Leduc D, Probert PK, Duncan A (2009) A multi-method approach for identifying
meiofaunal trophic connections. Mar Ecol Prog Ser 383:95–111
Leduc D, Probert PK, Nodder SD (2010) Influence of mesh size and core
penetration on estimates of deep-sea nematode abundance, biomass, and
diversity. Deep-Sea Res I 57:1354–1362
Legendre P, Legendre L (1998) Numerical ecology, 2nd English edition. Elsevier
Science B. V, Amsterdam
Li J, Vincx M, Herman PMJ, Heip C (1997) Monitoring meiobenthos using cm-, m- and
km-scales in the Southern Bight of the North Sea. Mar Env Res 43:265–278
McLachlan A, Erasmus T, Furstenberg JP (1977) Migrations of sandy beach
meiofauna. Zool Afr 12:257–277
Mills VS, Berkenbusch K (2009) Seagrass (Zostera muelleri) patch size and spatial
location influence infaunal macroinvertebrate assemblages. Estuar Coast
Shelf Sci 81:123–129
Moens T, Vincx M (1997) Observations on the feeding ecology of estuarine
nematodes. J Mar Biol Ass UK 77:211–227
Monthum Y, Aryuthaka C (2006) Spatial distribution of meiobenthic community
in Tha Len seagrass bed, Krabi Province, Thailand. Coast Mar Sci 30:146–153
Ndaro S, Ólafsson E (1999) Soft-bottom fauna with emphasis on nematode
assemblage structure in a tropical intertidal lagoon in Zanzibar, eastern
Africa: I. Spatial variability. Hydrobiologia 405:133–148
Orth RJ, Heck KL Jr, van Montfrans J (1984) Faunal communities in seagrass beds:
a review of the influence of plant structure and prey characteristics on
predator–prey relationships. Estuaries 7:339–350
Osenga GA, Coull BC (1983) Spartina alterniflora Loisel root structure and
meiofaunal abundance. J Exp Mar Biol Ecol 67:221–225
Platt HM, Warwick RM (1988) Free-living marine nematodes. Part II. British
chromadorids, Brill/Backhuys, Leiden
Sanders HL (1968) Marine benthic diversity: a comparative study. Am Nat 102:243–282
Shimanaga M, Kitazato H, Shirayama Y (2004) Temporal patterns in diversity and
species composition of deep-sea benthic copepods in bathyal Sagami Bay,
central Japan. Mar Biol 144:1097–1110
Soetaert K, Heip C (1990) Sample-size dependence of diversity indices and the
determination of sufficient sample size in a high-diversity deep-sea environment.
Mar Ecol Prog Ser 59:305–307
Steyaert M, Vanaverbeke J, Vanreusel A, Barranguet C, Lucas C, Vincx M (2003)
The importance of fine-scale, vertical profiles in characterising nematode
community structure. Estuar Coast Shelf Sci 58:353–366
Warwick RM, Platt HM, Somerfield PJ (1998) Free-living marine nematodes. Part
III. Monhysterids, Field Studies Council, Shrewsbury
Wieser W (1953) Die Beziehung zwischen Mundhöhlengestalt, Ernährungsweise und
Vorkommen bei freilebenden marinen Nematoden. Arkiv Zoologi 4:439–484
Yeh HM, Ohta S (2002) Influence of velocity and types of beam trawl towing on
deep-sea demersal fish and decapod crustacean samples. J Oceanogr
58:505–517

Submit your manuscript to a

journal and benefit from:
7 Convenient online submission
7 Rigorous peer review
7 Immediate publication on acceptance
7 Open access: articles freely available online
7 High visibility within the field
7 Retaining the copyright to your article

Submit your next manuscript at 7 springeropen.com