Journal of Applied Phycology

, Volume 26, Issue 4, pp 1759–1771 | Cite as

Optimal colonization and growth of marine benthic diatoms on artificial substrata: protocol for a routine use in bioindication

  • Catherine Desrosiers
  • Joséphine Leflaive
  • Anne Eulin
  • Loïc Ten-Hage
Article

Abstract

Benthic diatoms growing on hard substrata are used for their bioindication ability in freshwater quality monitoring. Artificial substrata are needed in cases where any natural substrate is present or to achieve similar sampling conditions between sites. Prior to use marine benthic diatoms for monitoring, a standardized protocol for sampling on artificial substrata must be set up. Two major types of information are required: (1) the time needed for a diatom community to be well developed and mature (climax stage); (2) the optimal growth conditions, given that the substrataum nature and texture are important parameters for the initial phase of biofilm development and can influence the future diatom assemblage. Three substrataum types were tested: frosted Plexiglass®, frosted glass, and rough enameled tiles. They were submerged for 8 weeks and sampled weekly. The experiment was conducted at five sites of distinct morphology and water chemistry, along the coastal area of Martinique Island, French West Indies. Development of diatom community was studied through biofilm dry weight, valve density, species richness, and species relative abundances. Globally, substratum type had no significant effect on any parameter. Frosted Plexiglass® was found to be the most interesting substratum because of higher valve densities and practical use. The asymptotic phase of biofilm development was encountered between 5 and 8 weeks depending on site and parameter. A compromise between community development and vandalism or loss through time was fixed to 5 weeks. This period is longer than for stream environments and is valid for tropical oligotrophic marine environments.

Keywords

Artificial substrata Bioindication Caribbean coast Colonization Diatoms 

Introduction

Diatoms represent the greatest percentage of algae initially colonizing a surface submerged in the sea (Hendey 1951; Edyvean et al. 1985). Marine biofilms growing on hard substrata contain high densities of epilithic diatoms, belonging to a wide range of diverse species (Patil and Anil 2005b; Cooksey et al. 1984). Diatoms are primary producers with a short life cycle, abundant, and easy to sample. Since the 1950s up to now—and even earlier with the Saprobic System of Kolkwitz and Marsson (1902)—studies conducted in freshwater environments revealed that species assemblages varied depending on environment trophic level (Prygiel and Coste 1996; Lange-Bertalot 1979; Zelinka and Marvan 1961; Kelly et al. 2008; Potapova and Charles 2007). This sensitivity to environmental parameters has been explored for the use of diatoms as bioindicators of water quality and several indices were developed (Kelly and Whitton 1995; Lenoir and Coste 1996; Lecointe et al. 1993). Thus, sampling methods on natural and artificial substrata have been established for streams and lakes (Brown 1976; Siver 1977; AFNOR 2007; CEN 2003). Artificial substrata can be used to standardize sampling methods on sites of various natural bottom type or depth. In the case of water quality monitoring, the use of diatom community growing on mud—epipelic—or on sand—epipsammic—is excluded since its development is rather under the influence of nutrients from sediment pore waters (Agatz et al. 1999) and the diatom community is thus not representative of the water column conditions. The time needed for optimal growth conditions of biofilm on artificial substrata and the effect of substratum type on community structure are two factors that have already been studied for bioindication use in streams (Lane et al. 2003; Acs and Kiss 1993; Blinn et al. 1980; Kelly 2000). Exposure time of artificial substrata needs to be sufficient for the development of stable diatom communities.

In coastal environments, epilithic diatoms are studied principally for their role in biofouling and little has been done on their bioindication ability (Desrosiers et al. 2013; Hillebrand and Sommer 1997; Vermeulen et al. 2011). Investigations have been conducted on how the nature and roughness of artificial substrata could affect biofilm biomass. It appears that some metallic substrata, like copper, have toxic effects on organisms (Edyvean et al. 1985; Sekar et al. 2004). On the other hand, the mineralogical composition of substrata can be a source of organic and mineral silica and enhance the growth of siliceous organisms like diatoms (Penna et al. 2003). Surface microtexture of substrata should also be considered as a factor influencing primary fixation and colonization on a clean surface. An irregular surface offers more space than a flat one for initial colonization by organisms (Blinn et al. 1980; Sekar et al. 2004; Totti et al. 2007).

The goals of the present study were to test the development of marine diatoms on various artificial substrata and define the colonization time needed to reach a stable community. It was a necessary step for the creation of a standardized sampling protocol, before further research on the bioindication abilities of marine benthic diatoms (Desrosiers et al. 2013). Three different types of substrata were tested for their ability to (1) insure first colonization and support high diatom density and (2) allow the development of a diverse and equilibrate diatom population. The three substrata have similar roughness: frosted glass (g), frosted Plexiglass® (p), and rough enameled tiles (t).

Materials and methods

Study sites and experimental setup

The study was carried out on five sampling sites on the coastal area of Martinique Island (French West Indies), surrounded to the east by the Atlantic Ocean and to the west by the Caribbean Sea (Fig. 1). The sites represented various water chemistry and hydromorphology, in order to integrate the variability of biofilm development according to site conditions (Table 1). One site was a reef environment on the Atlantic coast and was submitted to waves action (Atl1) whereas the other reef environment was on the southern Caribbean coast and had smooth (to exceptionally heavy) swell (Car2). Another site was a rocky shore on the Caribbean coast, largely frequented by divers and fishermen (Car1). The last two sites were located in Fort-de-France Bay, one influenced by river plumes (Bay1) and the second one near the outlet of a water treatment plant (Bay2). The goal of choosing sites with distinct characteristics was to get responses from various communities.
Fig. 1

Location of sampling sites

Table 1

Physical description of sites and water chemistry at 3 m depth and subsurface (WFD), mean values (n = 8–10), (min–max). Kruskal–Wallis test for differences between sites

Parameters

Atl1

Bay1

Bay2

Car1

Car2

Localization

Atlantic coast

Bay

Bay

Caribbean coast

South Caribbean coast

Environment

Reef

Old reef and mud

Sand

Rocky shore

Reef

Depth (m)

17

10

13

17

9

Oxygenation (mg L−1 O2)

7.5 (6.8–7.8)

7.7 (7.2–7.9)

7.8 (7.4–8.6)

7.8 (7.3–8.2)

7.6 (7.0–8.5)

Water temperature (°C)

27.0 (26.5–27.8)

27.0 (26.4–28.1)

27.0 (26.5–28.1)

27.2 (26.5–28.1)

27.2 (26.7–28.3)

pH

8.2 (8.1–8.4)

8.3 (8.1–8.5)

8.3 (8.2–8.5)

8.3 (8.2–8.5)

8.3 (8.1–8.5)

Salinity (PSU)

34.5 (34–35)

34.6 (34.1–35.3)

34.8 (34.3–35.3)

34.8 (34.3–35.4)

34.8 (34.2–35.3)

Phosphorous (PO4, μmol L−1)a

0.53 (0.05–2.34)

0.35 (0.05–1)

ND

0.21 (0.05–1.05)

0.79 (0.02–5.68)

DIN (NO3, NO2, NH4, μmol L−1)a

2.83 (0.23–12.71)

0.88 (0.2–3.65)

ND

0.9 (0.2–2.9)

1.82 (0.2–9.27)

Chla (μg/L−1)a

0.25 (0–0.72)

0.64b (0.1–1.3)

ND

0.16 (0–0.3)

0.21 (0.09–0.5)

Turbidity (FNU)a

0.42 (0.14–1.5)

0.84c (0.19–3.9)

ND

0.19 (0.04–0.36)

0.22 (0.16–0.4)

Waves

Moderate to high

None

None to low

None to low

Low to moderate

Wind force

Moderate

Low to moderate

Low to moderate

Moderate to high

Low to moderate

Current speed

None to moderate

None to low

None

None to high

None to moderate

aDIREN 972, WFD samplings years 2007, 2008

bDifferent from Atl1, Car1, Car2

cDifferent from Car1, Car2

Diatom development on three substratum types was tested weekly up to 8 weeks, during the months of March and April 2011, which corresponds to the dry and windy season. Frosted Plexiglass®, frosted glass, and rough enameled tiles were substrata chosen to be neutral in term of nutrient source, and their rough surface offers good adhesion characteristics (Desrosiers et al. 2013). Panels of 180 cm2 square area were submerged vertically at ±3 m depth in the water column. This position near the surface is recommended to get maximal light penetration while the position close to the bottom contributes to the establishment of epipelic and epipsammic community (Lewis et al. 2002; Munda 2005). Above 3 m depth, eventual damage due to boats would become a problem. A total of 24 panels (three panels × 8 weeks) for each substratum type and for each sampling site were fixed on an aluminum frame and all submerged at the same time.

Biofilm sampling and in situ measurements

Each week and for the five sampling sites, three panels of each substrate were scraped with a razor blade, in strict compliance with the limits of the sampling surface in order to get quantitative results. The biofilm of each panel was maintained in separate tubes filled with a known volume of filtered seawater: the first sample was preserved with a known volume of formalin and used for the determination of diatom valve density, diatom richness, and species relative abundance; the second and third samples were kept on ice until the filtration for the determination of biofilm dry weight in the laboratory.

Conditions at sampling sites such as wave, current, and wind force were described each week. Vertical measurements of temperature, pH, dissolved oxygen, and salinity were taken every 2 meters in the water column with a Hanna HI 9828 multiparameter probe.

Laboratory analysis

In the laboratory, a known volume (5–20 mL) of each formalin-preserved sample was placed in beakers with concentrated hydrogen peroxide (30 %), potassium dichromate, and hydrochloric acid to remove organic matter and dissolve calcium carbonate, without heating to avoid breakage of large and weakly silicified valves (Vermeulen et al. 2011; Kelly 2000). Cleaned samples were rinsed several times with distilled water and a known volume was placed on a coverslip and left to dry. Then coverslips were mounted on slides in a high refractive index medium, Naphrax®. Three slides were prepared for each sample. Valve density was determined by counting the entire valves on each slide, without distinction of species, in 100 fields at ×1,000 magnification using a Leica DMLB light microscope. Species richness was obtained by a presence/absence count of the species seen in 100 fields, on two different slides. This requires species to be distinguished between themselves with no need to name them. Species relative abundances were obtained by counting abundance of each species, up to a total of 300 valves. This type of count requires advanced taxonomical work (Hein et al. 2008; Lobban et al. 2012; Lopez Fuerte et al. 2010; Riaux-Gobin et al. 2011; Witkowski et al. 2000; Krammer and Lange-Bertalot 1991, 1997a, b) to identify each species and then allocate an abundance to each species. This work was done for the “p” only.

For the cold sample, two 100 mL subsamples of each of the two samples were homogenized and filtered on Whatman GF/C (1.2 μm) 47 mm filters. Blank filters were done by filtration of filtered seawater. Filters were dried at 80 °C for 24 h in a laboratory oven, and immediately weighted on an OHAUS MB45 precision balance, sensitivity 0.001 g.

Data analysis

Biofilm biomass was estimated by diatom valve density (× 103 cm−2) and by biofilm dry weight (mg cm−2). The latter represents the entire biofilm growth, including various organisms from autotrophic to heterotrophic and from eukaryotes to prokaryotes. Diatom community structure was described for all substrata by diatom richness. Differences between various sampling conditions (substrate type, sites, weeks) were tested by a Kruskal–Wallis test for nonparametric data.

For p, species relative abundances were used to describe more precisely the diatom community. Differences between weeks were tested by a one-way ANOSIM using Bray–Curtis distance. Simpson diversity index (1 − D) and evenness (J′) were calculated.

Results

An overview of the biofilm and diatom community growth on the three artificial substrata tested—frosted Plexiglass ® (p), frosted glass (g), and rough enameled tiles (t)—over 8 weeks, is presented in Fig. 2. Mean values (five sites) of biofilm dry weight, diatom valve density, and diatom richness are given for each substrate and each week. Valve densities were similar between the three substrata and growth was progressive with the highest value reaching 1,770 × 103 valves cm−2 on “g” at week 5. In week 6, biofilm mean dry weight was globally higher on p, with 11.4 mg cm−2 being the highest value at. Species richness increased exponentially up to week 3, then stabilized around 40–45 species. Results were similar between substrata and the maximal richness was 57 species in week 3 on g.
Fig. 2

Biofilm biomass (valve density, dry weight) and community structure (species richness) on each artificial substrata Plexiglass® (p), glass (g), and enameled tiles (t) at each of the 8 weeks of growth. Boxplot: median values of sites (n = 10, valve density, n = 15), with standard error (box) and min and max values (whiskers)

Diatom species occurring in the biofilm on p, for all sites and weeks, which belonged to 84 genus and 155 species were present with a relative abundance ≥1 % in at least one sample (Table 2). Most of the species on biofilm were pennate forms, with only six centric species. Nitzschia genus had the greater number of species (31), followed by Amphora and Mastogloia.
Table 2

Species occurring on Plexiglass® substrata, with relative abundance ≥1 % in at least one sample

Species

Achnanthidium sp01

Climaconeis riddleae A.K.S.K.Prasad

Licmophora sp1

Achnanthidium sp3

Climaconeis sp2

Licmophora sp3

Actinocyclus subtilis(Gregory) Ralfs in Pritchard (c)

Cocconeis archaeana

LIcmophora sp5

Amphora abludensSimonsen

Cocconeis convexaGiffen

Licmosphenia sp1

Amphora cf. helenensis

Cocconeis diruptaGregory var.dirupta

Lunella bisectaSnoeijs

Amphora cf. obtuse

Cocconeis mascarenica Riaux-Gobin & Compere

Mastogloia acutiuscula Grunow in Cleve var. elliptica Hustedt

Amphora coffeaeformis (Agardh) Kützing var. coffeaeformis

Cocconeis molesta Kützing var. crucifera Grunow in Van Heurck

Mastogloia borneensis Hustedt in A. Shmidt Atlas

Amphora coffeaeformis (Agardh)Kützing var. aponina (Kützing) Archibald & Schoeman

Cocconeis scutellum var. posidonia

Mastogloia corsicana Grunow in Cleve & Möller

Amphora incrassate

Cocconeis sp14

Mastogloia crucicula (Grun.) Cleve var. crucicula

Amphora kolbei Aleem

Craspedostauros sp1

Mastogloia cuneata(Meister) Simonsen

Amphora micrometra Giffen

Cyclophora tenuisCastracane

Mastogloia decipiensHustedt

Amphora pseudohyalinaSimonsen

Cyclotella sp2(c)

Mastogloia decussata Grunow in Cleve

Amphora pseudotenuissimaWachnicka & Gaiser

Diadesmis sp2

Mastogloia delicatissimaHustedt

Amphora sp03

Entomoneis pseudoduplexOsada & Kobayasi

Mastogloia erythreae var. erythreae form 1

Amphora sp09

Entomoneis punctulata (Grunow) Osada & Kobayasi

Mastogloia fallax Cleve

Amphora sp12

Entomoneis sp03

Mastogloia graciloides Hustedt

Amphora sp17

Entomoneis sp4

Mastogloia grunowii A. Schmidt

Amphora sp19

Falcula sp1

Mastogloia inaequalis Cleve

Amphora tenerrimaAleem & Hustedt

Gomphonemopsis obscurum (Krasske) Lange-Bertalot

Mastogloia laterostrata Hustedt

Amphora tumida Hustedt emend Sar, Sala, Hinz & Sunesen

Grammatophora oceanica Ehr.

Mastogloia manokwariensis Cholnoky

Amphora vaughanii

Gyrosigma coellophilum

Mastogloia peragalli Cleve

Ardisonnea sp1(c )

Gyrosigma sp4

Mastogloia sp33

Bacillaria paxillifera (O.F. Müller) T.Marsson

Haslea ostrearia(Gaillon) Simonsen

Navicula duerrenbergiana Hustedt in Schmidt et al.

Bacillaria paxillifera var.tumidula(Grunow) Witkowski, Lange-Bertalot & Metzeltin

Hyalosira delicatulaKützing var.gibbosa(Ostrup) Witkowski, Lange-Bertalot & Metzeltin

Navicula johanrossii Giffen

Bacillaria socialis (Gregory) Ralfs in Pritchard

Hyalosira interrupta (Ehrenberg) Navarro

Navicula salinicola Hustedt

Berkeleya fennica Juhlin-Dannfelt

Hyalosira sp1

Navicula sp01

Berkeleya hyalina (Round & Brooks) Cox

Hyalosynedra laevigata(Grunow) Williams & Round

Navicula sp15

Berkeleya scopulorum(Brebisoon) Cox

Hyalosynedra sp03

Navicula sp27

Bleakeleya notata(Grunow) Round in Round Crawford & Mann

Koernerella spAshworth, Lobban & Theriot

Neosynedra provincialis (Grunow) Williams & Round

Caloneis sp02

Licmophora paradoxa (Lyngbye) Agardh

Neosynedra tortosa(Grunow) Williams & Round

Chamaepinnularia wiktoriae (Witkowski & Lange-Bertalot) Witkowski, Lange-Bertalot & Metzelti

Licmophora remulus Grunow

Nitzschia acicularis (Kützing) W.M.Smith

Species

Nitzschia longissima(Brebisson ex Kützing) Ralfs in Pritchard

Nitzschia sp23

Protokeelia sp02

Nitzschia microcephalaGrunow in Cleve & Moller

Nitzschia sp25

Pteroncola sp02

Nitzschia panduriformis var.1

Nitzschia sp27

Protokeelia sp02

Nitzschia panduriformis var.3

Nitzschia sp31

Pteroncola sp02

Nitzschia perindistincta Cholnoky

Nitzschia sp34

Pteroncola sp03

Nitzschia perminuta(Grunow) M.Peragallo

Nitzschia sp36

Rhopalodia pacificaKrammer

Nitzschia ruda Cholnoky

Nitzschia sp37

Rhopalodia sp01

Nitzschia sp01

Nitzschia sp38

Rhopalodia sp04

Nitzschia sp02

Nitzschia sp55

Rhopalodia sterrenburgiiKrammer in Lange-Bertalot & Krammer

Nitzschia sp04

Nitzschia sp58

Seminavis robusta Danielidis & D.G. Mann

Nitzschia sp06

Nitzschia ventricosa Kitton

Seminavis sp01

Nitzschia sp08

Olifantiella gorandensis Riaux-Gobin & Al-Handal

Seminavis sp02

Nitzschia sp09

Olifantiella mascarenicaRiaux-Gobin & Compere

Seminavis sp03

Nitzschia sp10

Olifantiella sp01

Stauroneis retrostauron

Nitzschia sp13

Olifantiella sp4

Striatella unipunctata(Lyngbye) Agardh

Nitzschia sp14

Perideraion montgomeryi Lobban, Jordan et Ashworth

Surirella scalaris Giffen

Nitzschia sp15

Proschkinia complanata (Grunow) D.G. Mann in Round Crawford & Mann

Thalassiosira sp05(c)

Nitzschia sp17

Protokeelia bassonii Round

Toxarium undulatumJ.W.Bailey (c)

Nitzschia sp21

Protokeelia cholnokyana(Giffen) Round & Basson

 

In bold, species with abundance ≥5 % in at least one sample

c centric diatoms

Comparison of the study sites

Environmental parameters

The five sampling sites were chosen to represent various environment and water quality. Following the basic water chemistry parameters—dissolved oxygen, water temperature, pH, and salinity—Kruskal–Wallis test did not show significant differences between sites (Table 1). As additional information, data on nutrient enrichment and turbidity was available from Water Framework Directive (WFD) samplings on four sites. Greater nutrients concentrations, but not significantly different, were encountered on sites Atl1 and Car2. Bay1 was distinct from other sites by significantly higher chlorophyll a concentration and turbidity. Finally, some sites were marked out according to some parameters and longer data time series for each parameter could help for better discrimination. Sites can at least be distinguished by environmental characteristics and wave force. Atl1 had the rougher sea conditions, followed by the site located on Car2. Each site was situated in a different environment, more or less influenced by continental inputs, and represented a different type of sea bottom.

Biological parameters

Results between sites were also examined regarding biofilm productivity—diatom valve density, biofilm dry weight—and diatom community structure. Evaluation of differences between site is important to determine whether sites can be grouped or not for further analysis. In case sites would show completely distinct biofilm growth, further analysis for substrate effect and optimal growth may have to be conducted distinctly for each site.

In Fig. 2, large standard errors for valve density and dry weight highlighted the variability of results between sites. To understand how valve density evolved for each site, results from week 1 to 8 on p were taken as an example (Fig. 3). Sites Atl1, Bay1, and Bay2 reached a valve density up to 200 × 103 valves cm−2 between weeks 3 and 4, while sites Car1 and Car2 needed a longer growing period to reach the same density. The period corresponding to optimal growth depended on the site and went from week 6 (Bay2, Car2) to week 8 (Car1). Kruskal–Wallis tests for differences between sites were done on valve density and biofilm dry weight in week 8 (Table 3). Week 8 was chosen in order to evaluate differences on the basis of a mature community. Valve density was the only parameter discriminating each sites, with both Car1 and Car2 being significantly different from Atl1 and Bay1.
Fig. 3

Diatom valve density at the five sites (Atl1, Bay1, Bay2, Car1, Car2) on Plexiglass® substrate (×103 cm−2) for each of the 8 weeks of growth. Boxplot: mean value (n = 3) with standard error (box) and min and max values (whiskers)

Table 3

Kruskal–Wallis multiple comparison between sites for all substrata at week 8. p values of valve density (n = 9), dry weight (n = 6)

Diatom valve density (valve × 103 cm−2)

Biofilm dry weight (mg cm−2)

 

Atl1

Bay1

Bay2

Car1

Car2

 

Atl1

Bay1

Bay2

Car1

Car2

Atl1

 

1.000

1.000

0.098**

0.008*

Atl1

 

0.510

1.000

0.280

1.000

Bay1

1.000

 

1.000

0.083**

0.007*

Bay1

0.510

 

0.822

1.000

1.000

Bay2

1.000

1.000

 

0.699

0.103

Bay2

1.000

0.822

 

0.473

1.000

Car1

0.098**

0.083**

0.699

 

1.000

Car1

0.280

1.000

0.473

 

1.000

Car2

0.008*

0.007*

0.103

1.000

 

Car2

1.000

1.000

1.000

1.000

 

*p < 0.05 (significant); **p < 0.1 (significant)

Figure 4 shows relative abundance of the dominant species (relative abundance ≥7 % in at least one sample) for each site in weeks 1, 5, and 8 on p. Regarding community only in week 8, considered as mature, we notice that most of the three first dominant species for each site were the same. Nitzschia sp38 is in the top three for three sites, while Nitzschia sp34, Nitzschia microcephala (NMIC), and Amphora pseudotenuissima (APTN) are in the top three for two sites. Sites Bay1and Car2 differed from the other sites by only one species of the top three.
Fig. 4

Relative abundances of diatom species at the five sampling sites (Atl1, Bay1, Bay2, Car1, Car2) for weeks 1, 5, and 8 with species having a relative abundance ≥7 % for at least one sampling date. (BNOT, B. notata; KOER, Koernerella sp.; NI38, Nitzschia sp38; NI34, Nitzschia sp34; CCPT, Cyclophora tenuis; APTN, A. pseudotenuissima; SRET, Stauroneis retrostauron; NILG, N. longissima; NI37, Nitzschia sp37; NI31, Nitzschia sp31; BPTU, B. paxillifera var. tumidula; NMIC, N. microcephala; EPDU, Entomoneis pseudoduplex; AICR, A. incrassata; LBIS, Lunella bisecta; MDEC, M. decipiens; ATNI, A. tenerrima)

As the analysis of differences showed no clear distinction between each of the five sites, we considered that analysis of the following sections could be conducted on all sites grouped together.

Comparison of biofilm development on the three different substrata

Biofilm growth on the three different substrata was compared regarding the mean values (of the five sites) of biofilm dry weight, diatom valve density, and diatom richness, for three key phases of development: establishment on substratum (week 1), exponential phase (week 5), and mature biofilm (week 8) (Table 4). For each phase (1, 5, 8), the Kruskal–Wallis test revealed no significant difference of valve density and species richness between substrata. Results for dry weight are more contrasted as values for g and t were significantly lower than values for p in week 5, while values for g were significantly lower than values for p and t in week 8. Those differences are graphically visible in Fig. 2.
Table 4

Kruskal–Wallis multiple comparisons p values of valve density (mg × 103 × cm−2) (n = 15), biofilm dry weight (mg cm−2) (n = 10), and diatom species richness (n = 10) for Plexiglass (p), glass (g), and tiles (t), for 1, 5, and 8 weeks of growth

Week 1

Week 5

Week 8

 

p

g

T

p

g

T

P

g

t

Valve density

  P

 

1.000

1.000

 

0.457

1.000

 

1.000

0.992

  G

1.000

 

1.000

0.457

 

1.000

1.000

 

1.000

  T

1.000

1.000

 

1.000

1.000

 

0.992

1.000

 

Dry weight

  P

 

1.000

1.000

 

0.037*

0.012*

 

0.049*

0.668

  G

1.000

 

1.000

0.037*

 

1.000

0.049*

 

0.713

  T

1.000

1.000

 

0.012*

1.000

 

0.668

0.713

 

Species richness

  P

 

1.000

1.000

 

0.668

0.443

 

1.000

1.000

  G

1.000

 

1.000

0.668

 

1.000

1.000

 

1.000

  T

1.000

1.000

 

0.443

1.000

 

1.000

1.000

 

*p < 0.05 (significant)

Substrata were photographed before scraping to show colonization at the end of the experiment (week 8) (Fig. 5). Coverage was highly variable between sites but colonization was worse on g for most of the sites.
Fig. 5

Colonization differences between substrata before sampling, for the five sampling sites in week 8 (p = Plexiglass®, g = frosted glass, t = enameled tile)

Dynamics of biofilm growth on Plexiglass

Results of the previous section revealed that biofilm growth was independent of substratum type. In order to present concise results, we have chosen to limit to p the temporal evolution interpretation of valve density, dry weight, species richness, and species relative abundances.

Valve density of biofilm increased progressively up to 6 weeks on p, reaching a mean value of 496,000 valves cm−2 (248,000 cells cm−2) (Fig. 6a). Beyond this period, valve density remained high thus showing a plateau phase in the diatom community development. Differences between the five sampling sites were highlighted by large standard deviations. Dry weight evolution was similar, beginning slowly during the first colonization weeks and reaching its maximum (5.6 mg cm−2) after a period of 5 weeks (Fig. 6b). Standard deviation was larger for the last weeks, suggesting that biofilm development was more variable between sites. Species richness had an exponential growth during the first 3 weeks of colonization followed by an asymptotic phase characterized by small variations in species richness (Fig. 6c). Maximal mean richness was 47.8 species, which occurred after 5 weeks. The ratio between valve density in biofilm and biofilm dry weight can be an indicator of diatom population development (Fig. 7). Biofilm structure fluctuated with an optimal ratio in weeks 3 and 6 and a minimal in week 5.
Fig. 6

Diatom valve density (×103 cm-2) (a), species richness (b), dry weight (mg cm−2) (c) for each of the 8 weeks of growth, on Plexiglass® (mean values of sites ± standard deviation, n = 15 for valve density, n = 10 for richness and dry weight)

Fig. 7

Evolution of diatom community (valves cm−2) within total biofilm (mg cm−2) during colonization period, on Plexiglass®

An assemblage characterized by unchanging species indicates a stable community. Species relative abundances between each week for p were compared by means of an ANOSIM multiple comparison (Table 5). Communities from weeks 1 to 4 and from weeks 5 to 8 were significantly similar within each group. This result emphasizes two important steps in diatom population development, with a shift from the first to the second step occurring around the fourth and fifth week of biofilm growth. Species composition of diatom population varies depending on site and on period of biofilm development (Fig. 4). The low total relative abundance of the species represented in Fig. 4 (relatives abundance ≥7 %) reflects the high diversity of the community on one site, while the long list of species indicates the variability between sites. On week 1, more than 50 % of the relative abundance was given by three to five species, for three of the five sites. Population was dominated by either Bleakeleya notata, Koernerella sp., or Nitzschia longissima. They are large diatoms, the first two forming long chains. On weeks 5 and 8, relative abundance of dominant species (≥7 %) covered less than 30 % of the total abundance. On week 5, small species of Nitzschia and Amphora became dominant: Nitzschia sp38, Nitzschia sp31, N. microcephala (NMIC), and Amphora tenerrima (ATNI), along with Mastogloia decipiens (MDEC). On week 8, Nitzschia sp38, N. microcephala (NMIC), and A. tenerrima (ATNI) kept their dominance on three sites. In site Bay1, Bacillaria paxillifera var. tumidula (BPTU) increased its position in the biofilm and became dominant. In site Car1, Amphora incrassata (AICR) appeared as a new dominant species. Each site presents its own characteristics, but the global trend was to have a dominant species for week 1 and a different dominant species for weeks 5 and 8. Large and chain-forming forms encountered in the first phase of biofilm development were progressively replaced or accompanied by smaller and solitary forms
Table 5

ANOSIM multiple comparison between weeks (1–8). p values of species relative abundances for all sites, substrata p (n = 5)

 

1

2

3

4

5

6

7

8

1

 

0.194

0.076

0.009**

0.006**

0.008**

0.01*

0.007**

2

0.194

 

0.287

0.06

0.009**

0.007**

0.008**

0.008**

3

0.076

0.287

 

0.072

0.056

0.025*

0.016*

0.006**

4

0.009**

0.06

0.072

 

0.212

0.097

0.032*

0.001**

5

0.006**

0.009**

0.056

0.212

 

0.727

0.501

0.290

6

0.008**

0.007**

0.025*

0.097

0.727

 

0.761

0.463

7

0.01*

0.008**

0.016*

0.032*

0.501

0.761

 

0.904

8

0.007**

0.008**

0.006**

0.001**

0.2902

0.463

0.904

 

*p < 0.05 (significant); **p < 0.1 (significant)

Simpson index (1 − D) and evenness (J′) values were low in the young biofilm (week 1) (Fig. 8) and stabilized at values between 0.8 and 1 from week 5.
Fig. 8

Simpson diversity index (1 − D) and evenness (J′) of diatom community on Plexiglass® for the colonization period (mean values of sites ± standard deviation, n = 5)

Discussion

The present study aimed to identify the best conditions for the development of diatom community on artificial substrata, in tropical marine oligotrophic environment. The use of artificial substrata to study biofilms is widespread in the marine environment (Hillebrand and Sommer 2000b; Brandini et al. 2001; Lewis et al. 2002). But the purpose here was to establish a standardized protocol for the use of artificial substrata in further research on marine diatom bioindication abilities. The appreciation of the best substratum type and colonization time can either be based on a comparison of the diatom community composition between artificial and natural substrata (Lane et al. 2003) or based on a theoretical evolution curve (Biggs 1988). In the marine environment, natural and artificial communities cannot easily be compared due to two major reasons. First, marine diatom communities have been proven to vary depending on the nature of their natural substratum such as sediment grain size in case of epipelic or epipsammic species (Cahoon et al. 1999) or macroalgae taxa in case of epiphytic diatoms (Almeida and Beltrones 2008). While sampling on natural substrata, it is not easily possible to get rid of this influence, especially in tropical environment where macroalgae quickly cover any hard substratum (Diaz-Pulido and McCook 2002). Second, light penetration, modified by water turbidity and depth, influences the species composition of assemblages (Munda 2005; Patil and Anil 2005a; Titlyanov et al. 2008). At sites with high turbidity, community structure may not be the same on artificial substrata placed near the surface compared to the community on bottom substrata. Because a comparison with natural community is not clearly relevant for marine environments, this study focused on theoretical conditions to reach a stable community. Sampling a mature community is a crucial point for bioindication. The notion of community in a climax stage indicates a level of interspecific organization based on ecological requirements for each species. This community on its way to reach the climax will be influenced by the physical and chemical factors of its environment, leading to variable biofilm growth according to the characteristics of the growing site. This variability, instilled by the environment, represents one interesting aspect of community studies, particularly for bioindication aspects.

The artificial substrata were not chosen to represent any natural substratum. Moreover, they may not influence the presence of any specific species by their chemical composition (Edyvean et al. 1985; Penna et al. 2003; Sekar et al. 2004). In our study, substrata p, g, and t were chosen as inert substrata and the use of either of them appeared to offer no particular advantage: all three artificial substrata provided globally similar results. Similarly, numerous studies comparing artificial substrata (Lane et al. 2003; Patil and Anil 2005a; Cattaneo and Amireault 1992) have revealed no differences between stable diatom community structures which developed on different neutral artificial substrata. Fiberglass and glass are compared by Patil and Anil (2005a) in a coastal environment, glass slides and clay tiles by Lane et al. (2003) in coastal dune lakes, while Cattaneo and Amireault (1992) reviewed studies using artificial substrata. Sekar et al. (2004) demonstrated a greater attachment on Plexiglass® compared with glass. Furthermore, periphytometers used by the Environmental Protection Agency procedures in the USA are composed of acrylic substrata (common name for Plexiglass®) (Lewis et al. 2002). Patil and Anil (2005a) in their study comparing fiberglass and glass substrata, found that the highest diatom abundance and species diversity were often encountered on fiberglass. They also observed higher diatom recruitment, due to in situ growth, on fiberglass. This result indicates either an active choice of substratum by the diatoms or an influence of the substratum physicochemistry on diatom settlement. Differences in the substratum microtopography may also cause variable growth in the early stage of colonization. Blinn et al. (1980) have demonstrated that in steam environments surfaces of substrata are quickly modified by the accumulation of organic aggregates which provide a similar attachment surface for microbial invasion. After the first week of colonization, the community structure became similar among the substrata. In the sampling protocol of our study, all substrata had a rough surface with however a higher roughness for t than for g and p. This difference did not lead to significant differences in growth between the substrata at the end of the first week of development. As the biofilm growth and the species composition were not really impacted by the substrate nature, p could be selected as the more convenient one to use.

Diatom community optimal growth time to reach a mature community on artificial substrata has not been studied for the marine environment, mainly because water quality monitoring does not use diatoms as yet. Periphyton growth is characterized for stream environments by a sigmoidal curve, with three phases: (a) colonization, (b) logarithmic growth phase, (c) upper asymptote indicating equilibrium between production and grazing and sloughing (Tilley and Haushild 1975). In the marine environment, biofilm development can vary depending on physical conditions such as wave action, currents, and nature of the bottom, and on water chemistry which includes nutrient concentrations, salinity variations, and turbidity. There is no example, for the tropical environment, of water chemistry effect on diatom growth on artificial substrata. However, Frankovich et al. (2006) demonstrated that Florida Bay epiphytic diatom assemblages were structured by salinity and nutrient availability. Our sampling sites were chosen to be representative of the various environments that can be found on a tropical island coastal area, in order to integrate variability of biofilm development and thus establish a standardized colonization time. Despite the few significant differences in chemical parameters between sites, they showed distinct characteristics and distinct biofilm development. Patil and Anil (2005a) observed a seasonal effect on diatom development on artificial substrata in a monsoon-influenced tropical estuary. Differences in water movement due to waves and currents could be indicated by changes in turbidity and salinity. They studied newly established biofilm (4 days) and pointed out that its composition was dependent on the resuspension of benthic diatoms in the water column. Our study did not include seasonal variability, but we assume that main phases of biofilm development remain similar despite the fact that the species composition of the biofilm differs.

Maximum species richness was reached very early in the colonization process while the density reached its optimum after 6 weeks. According to Hillebrand and Sommer (2000a), the rate of colonization by new species may be important only for a short period following the exposure of the substrata. Afterwards, differential biofilm development may result from the growth rates of each species forming the community rather than by new species integrating in the biofilm. Thus, cell density and diversity indices of the diatom community are accurate parameters to identify the colonization time required for optimal community development, while species richness is less informative. For our study, the maximal valve density was reached between weeks 6 and 8 depending on the site. New colonization was indicated by an exponential species richness growth, which reached the mean value of 43 species in week 3. Patil and Anil (2005a) found up to 62 diatoms species on glass substrata after only 4 days of growth on the west coast of India. This high richness was associated with a cell density of around 0.01 × 103 cell cm−2, which is lower than the one we found after 1 week, i.e. 0.24 × 103 cells cm−2. These results confirm the idea that an elevated richness is not necessarily representative of a mature biofilm. The ratio between valve density and biofilm dry weight can provide information on the biofilm formation process. It can indicate for our results: (1) a biofilm dominated by diatoms up to week 3; (2) between weeks 3 and 5, a appearance of other forms which minimize the places of diatoms in the biofilm; (3) between weeks 5 and 6, a growth of diatom population while the global biofilm is constant (see Fig. 6). This interpretation must be taken with care because it reflects only the proportion between diatoms and other organisms forming the biofilm, mainly macroalgae but also small invertebrates. No parameter has been studied here to reflect the effect of grazers in the process of reaching equilibrium. Nevertheless, Fig. 7 shows the presence of a cycle and week 5 appeared to be the beginning of a second phase. Species relative abundances indicate a shift in diatom community between weeks 4 and 5, and thus a higher stability in the community beyond 5 weeks. Dominant species at weeks 5 and 8 belonged mainly to the genera Nitzschia and Amphora and are small and solitary cells. These species are suspected to be motile and adnate (i.e., growing closely attached to substrate) forms, but no scanning electron microscopy was done on the biofilm to confirm this hypothesis. Several studies on epilithic communities recognized that adnate and motile forms appear early in the colonization process together with erect species, while the last phase of colonization is characterized by the massive growth of stalk-producing species creating a complex three-dimensional community (Tuji and Hino 2000; Korte and Blinn 1983). Similarly to our results, Totti et al. (2007) observed the dominance of motile forms on 7-week-old biofilm developed on artificial substrata in the Adriatic Sea.

In stream environments, a 4-week period is usually considered to be sufficient time for a representative diatom community to be established (Kelly et al. 1998; Lane et al. 2003; Tuchman and Stevenson 1979). In marine environment, there is no reference on the appropriate colonization time to study mature biofilm. A 5-week period was used by Edyvean et al. (1985) to study, in a temperate sea environment, the effect of substrate type on colonization. Our results revealed that the plateau phase of growth curves (dry weight, valve density) and the optimal organization between species (species relative abundance and diversity)—both indicating the climax stage—were reached from the fifth week of substratum colonization. However, some sites showed later progression. As pointed out by Cattaneo and Amireault (1992), experiments with a too long exposure time are often impractical because they increase the chance of losing the substrata because of vandalism or bad weather conditions. Results of the present study allow the definition of a colonization time on artificial substrata placed vertically at 3 m depth, which can be applied to tropical oligotrophic coastal environments. A different geographic zone as well as a different substratum position can lead to substantially different results. Diatom cell density observed on artificial substrata in summer in a coastal area of the Adriatic Sea was between 8 and 25 × 103 cells cm−2 after a 6-week vertical immersion period (Totti et al. 2007). In contrast, an experiment in a Brazilian mudflat area conducted on glass slides placed horizontally near the bottom gave a cell density of around 25 × 108 cells cm−2 after a single week of colonization (Brandini et al. 2001). Diatom densities for our study were about 0.24 × 103 cells cm−2 after 1 week and around 248 × 103 cells cm−2 after 6 weeks. It seems that the bottom nature and the substratum position have great influence on the development of the diatom community on substrata. Conditions encountered in the Mediterranean Sea are indeed closer to those of our study, with rocky environments and oligotrophic water. As silicate concentrations are lower in the Mediterranean (Vermeulen et al. 2012) than in the Caribbean, a higher density in our study was justified.

In conclusion, using artificial substrata instead of natural substrata to study stable diatom assemblages in marine coastal waters offers repeatability of sampling protocols between sites by using the same substratum, depth, and growing time. On the contrary, sampling natural substrata induces variability of biofilm due to the nature of the substrata (i.e., rocks with macroalgae, coral, sand grains) and sampling depth. Our study established the best substratum type and growth time to get a community representing its surrounding environment. Marine benthic diatoms bioindication abilities will be the object of further research, which will focus on species relative abundances related to a precise description of water chemistry and sites characteristics. In order to use artificial substrata properly and efficiently in research on bioindication in tropical oligotrophic environments, the present study proposes the use of frosted Plexiglass ®immersed vertically for a minimum of 5 weeks.

Notes

Acknowledgments

This work was funded by the European Regional Development Fund and supported by the French Direction de l'Environnement, de l'Agriculture et du Logement de Martinique and the Office de l'Eau Martinique.

References

  1. Acs E, Kiss KT (1993) Colonization processes of diatoms on artificial substrates in the River Danube near Budapest (Hungary). Hydrobiologia 269/270:307–315CrossRefGoogle Scholar
  2. AFNOR (2007) NF T90-354: Détermination de l'indice biologique diatomées (IBD)Google Scholar
  3. Agatz R, Asmus M, Deventer B (1999) Structural changes in the benthic diatom community along a eutrophication gradient on a tidal flat. Biomed Life Sci 53:92–101Google Scholar
  4. Almeida OUH, Beltrones DAS (2008) Variations in the structure of epiphytic diatom assemblages in subtropical macroalgae. Hidrobiologica 18:51–61Google Scholar
  5. Biggs BJF (1988) Artificial substrate exposure times for periphyton biomass estimates in rivers. N Z J Mar Freshwat Res 22:507–515CrossRefGoogle Scholar
  6. Blinn DW, Fredericksen A, Korte V (1980) Colonization rates and community structure of diatoms on three different rock substrata in a lotic system. Eur J Phycol 15:303–310CrossRefGoogle Scholar
  7. Brandini FP, Da Silva ET, Pellizzari FM, Fonseca ALO, Fernandes LF (2001) Production and biomass accumulation of periphytic diatoms growing on glass slides during a 1-year cycle in a subtropical estuarine environment (Bay of Paranagua, southern Brazil). Mar Biol 138:163–171CrossRefGoogle Scholar
  8. Brown H (1976) A comparison of the attached algal communities of a natural and an artificial substrate. J Phycol 12:301–306Google Scholar
  9. Cahoon LB, Nearhoof JE, Tilton CL (1999) Sediment grain size effect on benthic microalgal biomass in shallow aquatic ecosystems. Estuaries 22:735–741CrossRefGoogle Scholar
  10. Cattaneo A, Amireault MC (1992) How artificial are artificial substrata for periphyton. J N Am Benthol Soc 11:244–256CrossRefGoogle Scholar
  11. CEN (2003) EN 13946 Water quality–Guidance standard for the routine sampling and pretreatment of benthic diatoms from rivers. European StandardsGoogle Scholar
  12. Cooksey B, Cooksey KE, Miller CA, Paul JH, Rubin RW, Webster D (1984) The attachment of microfouling organisms. In: Costlow JD, Tipper RC (eds) Marine biodeterioration: an interdisciplinary study. US Naval Institute, Annapolis, pp 167–171CrossRefGoogle Scholar
  13. Desrosiers C, Leflaive J, Eulin A, Ten-Hage L (2013) Bioindicators in marine waters: benthic diatoms as a tool to assess water quality from eutrophic to oligotrophic coastal ecosystems. Ecol Indic 32:25–34CrossRefGoogle Scholar
  14. Diaz-Pulido G, McCook LJ (2002) The fate of bleached corals: patterns and dynamics of algal recruitment. Mar Ecol Prog Ser 232:115–128CrossRefGoogle Scholar
  15. Edyvean RGJ, Rands GA, Moss BL (1985) A comparison of diatom colonization on natural and artificial substrata in seawater. Estuar Coast Shelf Sci 20:233–238CrossRefGoogle Scholar
  16. Frankovich TA, Gaiser EE, Zieman JC, Wachnicka AH (2006) Spatial and temporal distributions of epiphytic diatoms growing on Thalassia testudinum Banks ex König: relationships to water quality. Hydrobiologia 569:259–271CrossRefGoogle Scholar
  17. Hein MK, Winsborough BM, Sullivan MJ (2008) Bacillariophyta (diatoms) of the Bahamas. In: Lange-Bertalot (ed) Iconographia diatomologica Vol. 19. Koeltz, KoenigsteinGoogle Scholar
  18. Hendey NI (1951) Littoral diatoms of Chichester harbour with special reference to fouling. J Roy Microsc Soc 71:1–86CrossRefGoogle Scholar
  19. Hillebrand H, Sommer U (1997) Response of epilithic microphytobenthos of the Western Baltic Sea to in situ experiments with nutrient enrichment. Mar Ecol Prog Ser 160:34–46CrossRefGoogle Scholar
  20. Hillebrand H, Sommer U (2000a) Diversity of benthic microalgae in response to colonization time and eutrophication. Aquat Bot 67:221–236CrossRefGoogle Scholar
  21. Hillebrand H, Sommer U (2000b) Effect of continuous nutrient enrichment on microalgae colonizing hard substrates. Hydrobiologia 426:185–192CrossRefGoogle Scholar
  22. Kelly MG (2000) Identification of common benthic diatoms in rivers. Field Stud 9:583–700Google Scholar
  23. Kelly MG, Whitton BA (1995) The trophic diatom index: a new index for monitoring eutrophication in rivers. J Appl Phycol 7:433–444CrossRefGoogle Scholar
  24. Kelly MG, Cazaubon A, Coring E, Dell'Uomo A, Ector L, Goldsmith B, Guasch H, Hürlimann J, Jarlman A, Kawecka B, Kwandrans J, Laugaste R, Lindstrøm E-A, Leitao M, Marvan P, Padisák J, Pipp E, Prygiel J, Rott E, Sabater S, van Dam H, Vizinet J (1998) Recommendations for the routine sampling of diatoms for water quality assessments in Europe. J Appl Phycol 10:215–224CrossRefGoogle Scholar
  25. Kelly M, Juggins S, Guthrie R, Pritchard S, Jamieson J, Rippey B, Hirst H, Yallop M (2008) Assessment of ecological status in U.K. rivers using diatoms. Freshwater Biol 53:403–422Google Scholar
  26. Kolkwitz R, Marsson M (1902) Grundsätze für die biologische Beurteilung des Wassers nach seiner Flora und Fauna (Principles for the biological assessment of water bodies according to their flora and fauna). Mitt Königlischen Prüfungsanstalt für Wasserversorgung und Abwässerbeseitigung zu Berlin 1:33–72Google Scholar
  27. Korte VL, Blinn DW (1983) Diatom colonization on artificial substrates in pool and riffle zones studied by light and scanning electron microscopy. J Phycol 19:332–341CrossRefGoogle Scholar
  28. Krammer K, Lange-Bertalot H (1991) Bacillariophyceae 3.Teil: Centrales, Fragilariaceae, Eunotiaceae, vol 2/3. Süβwasserflora von Mitteleuropa. Spektrum Akademischer Verlag, BerlinGoogle Scholar
  29. Krammer K, Lange-Bertalot H (1997a) Bacillariophyceae 1.Teil: Naviculaceae, vol 2/1. Süβwasserflora von Mitteleuropa. Spektrum Akademischer Verlag, BerlinGoogle Scholar
  30. Krammer K, Lange-Bertalot H (1997b) Bacillariophyceae 2.Teil: Bacillariaceae, Epithemiaceae, Surirellaceae, vol 2/2. Süβwasserflora von Mitteleuropa. Spektrum Akademischer Verlag, BerlinGoogle Scholar
  31. Lane C, Taffs K, Corfield J (2003) A comparison of diatom community structure on natural and artificial substrata. Hydrobiologia 493:65–79CrossRefGoogle Scholar
  32. Lange-Bertalot H (1979) Pollution tolerance of diatoms as a criterion for water quality estimation. Nova Hedwigia 64:285–304Google Scholar
  33. Lecointe C, Coste M, Prygiel J (1993) ‘Omnidia’ software for taxonomy, calculation of diatom indices and inventories management. Hydrobiologia 269/270:509–513CrossRefGoogle Scholar
  34. Lenoir A, Coste M (1996) Development of a practical diatom index of overall water quality applicable to the French National Water Board Network. In: Whitton B, Rott E (eds) Use of algae for monitoring rivers II. Universität Innsbruck, Innsbruck, pp 29–43Google Scholar
  35. Lewis MA, Weber DL, Moore JC (2002) An evaluation of the use of colonized periphyton as an indicator of wastewater impact in near-coastal areas of the Gulf of Mexico. Arch Env Contam Toxicol 43:11–18CrossRefGoogle Scholar
  36. Lobban CS, Schefter M, Jordan RW, Arai Y, Sasaki A, Theriot EC, Ashworth MP, Ruck EC, Pennesi C (2012) Coral-reef diatoms (Bacillariophyta) from Guam: new records and preliminary checklist, with emphasis on epiphytic species from farmer-fish territories. Micronesica 43:237–479CrossRefGoogle Scholar
  37. Lopez Fuerte FO, Siquieros Beltrones DA, Navarro JN (2010) Benthic diatoms associated with mangroves environments in the northwest region of México. 1st edn. Siquieros Beltrones, D.A.Garcia Gomez, R.E., MexicoGoogle Scholar
  38. Munda IM (2005) Seasonal fouling by diatoms on artificial substrata at different depths near Piran (Gulf of Trieste, Northern Adriatic). Acta Adriat 46:137–157Google Scholar
  39. Patil JS, Anil AC (2005a) Biofilm diatom community structure: influence of temporal and substratum variability. Biofouling 21:189–206PubMedCrossRefGoogle Scholar
  40. Patil JS, Anil AC (2005b) Quantification of diatoms in biofilms: standardisation of methods. Biofouling 21:181–188PubMedCrossRefGoogle Scholar
  41. Penna A, Magnani M, Fenoglio I, Fubini B, Cerrano C, Giovine M, Bavestrello G (2003) Marine diatom growth on different forms of particulate silica: evidence of cell/particle interaction. Aquat Microb Ecol 32:299–306CrossRefGoogle Scholar
  42. Potapova M, Charles DF (2007) Diatom metrics for monitoring eutrophication in rivers of the United States. Ecolog Indic 7:48–70CrossRefGoogle Scholar
  43. Prygiel J, Coste M (1996) Les diatomées et indices diatomiques dans les réseaux de mesure de la qualité des cours d’eau Français: Historique et Avenir. B Fr Pêche Piscic 341–342:65–79CrossRefGoogle Scholar
  44. Riaux-Gobin C, Romero OE, Compère P, Al-Handal AY (2011) Small-sized Achnanthales (Bacillariophyta) from coral sands off Mascarenes (Western Indian Ocean). Bibl Diatomol 57:1–234Google Scholar
  45. Sekar R, Venugopalan VP, Satpathy KK, Nair KVK, Rao VNR (2004) Laboratory studies on adhesion of microalgae to hard substrata. Hydrobiologia 512:109–116CrossRefGoogle Scholar
  46. Siver P (1977) Comparison of attached diatom communities on natural and artificial substrata. J Phycol 13:402–406Google Scholar
  47. Tilley LJ, Haushild WL (1975) Use of productivity of periphyton to estimate water quality. J Water Pollut Contr Fed 47:2157–2171Google Scholar
  48. Titlyanov EA, Titlyanova TV, Chapman DJ (2008) Dynamics and patterns of algal colonization on mechanically damaged and dead colonies of the coral Porites lutea. Bot Mar 51:285–296CrossRefGoogle Scholar
  49. Totti C, Cucchiari E, de Stefano M, Pennesi C, Romagnoli T, Bavestrello G (2007) Seasonal variations of epilithic diatoms on different hard substrata in the northern Adriatic Sea. J Mar Biol Assoc U K 87:649–658CrossRefGoogle Scholar
  50. Tuchman ML, Stevenson RJ (1979) Comparison of clay tile, sterilized rock, and natural substrate diatom communities in a small stream in Southeastern Michigan, USA. Hydrobiologia 75:73–79CrossRefGoogle Scholar
  51. Tuji A, Hino S (2000) Observation of developmental processes in loosely attached diatom (Bacillariophyceae) communities. Phycol Res 48:75–84CrossRefGoogle Scholar
  52. Vermeulen S, Sturaro N, Gobert S, Bouquegneau JM, Lepoint G (2011) Potential early indicators of anthropogenically derived nutrients: a multiscale stable isotope analysis. Mar Ecol Prog Ser 422:9–22CrossRefGoogle Scholar
  53. Vermeulen S, Lepoint G, Gobert S (2012) Processing samples of benthic marine diatoms from Mediterranean oligotrophic areas. J Appl Phycol 24:1253–1260CrossRefGoogle Scholar
  54. Witkowski A, Lange-Bertalot H, Metzeltin D (2000) Diatom flora of marine coasts. In: Lange-Bertalot (ed) Iconographia diatomologica. Vol.7. Koeltz, KoenigsteinGoogle Scholar
  55. Zelinka M, Marvan P (1961) Zur Präzisierung der biologischen Klassifikation des Rheinheit fliessender Gewässer. Arch Hydrobiol 57:389–407Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  • Catherine Desrosiers
    • 1
    • 2
  • Joséphine Leflaive
    • 1
  • Anne Eulin
    • 2
  • Loïc Ten-Hage
    • 1
  1. 1.Laboratory of Functional Ecology and EnvironmentUniversité Paul SabatierToulouse Cedex 9France
  2. 2.Asconit Consultants, ZI ChampignyDucosFrance

Personalised recommendations