//img.uscri.be/pth/b8780134b06ac651997467063d71e23f785c2180
La lecture en ligne est gratuite
Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres
Télécharger Lire

Species assemblages and diets of Collembola in the organic matter accumulated over an old tar deposit

De
18 pages
In: European Journal of Soil Biology, 2005, 41 (1-2), pp.39-44. An oil refinery was abandoned in 1964 at Merkwiller-Pechelbrorm (Alsace, France). We investigated the food diets and species assemblages of collembolan communities living in a thick pasty tar deposit overlaid by an organic soil, by comparison with an adjacent unpolluted plot. At the polluted plot the species richness of the springtail community was low (nine species, versus 27 at the unpolluted. plot) but its total abundance was at the same level in both plots. Most springtails consumed more fungi than bacteria at the polluted plot, contrary to the unpolluted plot. Changes in habitat features and food resources might explain the observed decrease in local biodiversity, rather than direct PAH toxicity.
Voir plus Voir moins
Species assemblages and diets of Collembola in the organic matter accumulated
over an old tar deposit
Email address:jeanfrancois.ponge@wanadoo.fr
habitat features and food resources might explain the observed decrease in local
24
AbstractAn oil refinery was abandoned in 1964 at MerkwillerPechelbronn (Alsace,
14
6
10
1
20
fungi than bacteria at the polluted plot, contrary to the unpolluted plot. Changes in
15
19
total abundance was at the same level in both plots. Most springtails consumed more
18
Keywords:Collembolan communities; Gut contents; Polycyclic Aromatic Hydrocarbons
Museum National d’Histoire Naturelle, CNRS UMR 5176, 4 avenue du Petit Château,
communities living in a thick pasty tar deposit overlaid by an organic soil, by
comparison with an adjacent unpolluted plot. At the polluted plot the species richness
France). We investigated the food diets and species assemblages of Collembolan
16
17
* Corresponding author. Fax: +33160465009.
Running title:Collembola over tar
4
2
3
5
of the springtail community was low (9 species, versus 27 at the unpolluted plot) but its
7
9
8
91 800 Brunoy, France
biodiversity, rather than direct PAH toxicity.
Servane Gillet, JeanFrançois Ponge*
23
22
21
12
13
11
25
26
1. Introduction
-1-
or benzo[a]pyrene, are toxic aromatic compounds [2,24] which are nonpolar,
17
incorporation of litter to mineral matter through earthworm activity [12]. We used a
humus profiles, we observed that the organic matter which accumulated over the 40
21
impact of this type of pollution on soil fauna by a comparison between a plot polluted by
patchy pollution of soil and ponds by PAHs. The aim of our work was to study the
19
communitylevel approach to evaluate the impact of PAHs on the saprophagous
20
18
yearold tar deposits was permeated by a large population of mycorrhizal fungi, in
and the flexibility of its alimentary habits [11,15,21]. Population indices and food bolus
22
23
24
indirect effects, for instance through changes in habitat and food quality, are probable
2
14
10
6
compartment of the ecosystem. The collembolan community (Arthropoda: Hexapoda)
25
their vertical distribution [1,10]. Laboratory bioassays have demonstrated toxic and
waste deposition, even long after site abandonment. Field studies revealed that the
15
16
5
presence of PAHs decreases the biodiversity of soil animal communities and changes
temperature industrial processes, which induce pollution by accidental oil spilling and
linear, step, or cluster arrangement. Nowadays, their presence in the soil results mainly
hydrophobic and neutral molecules composed of two or more fused benzene rings in
Polycyclic Aromatic Hydrocarbons (PAHs), such as naphthalene, phenanthrene,
anaesthetic effects of PAHs on soil animals [7,5,8], as well as bioaccumulation [17] and
contrast to the adjacent unpolluted plot which was characterized by the rapid
avoidance [18].
1
PAHs and a nearby unpolluted control plot. In a parallel study on the composition of
from industrial activities such as coke production, petroleum refining and other high
27
3
[3,11].
4
8
9
7
The direct toxicity of pollutants to soil animals has been widely reported and
13
11
12
28
26
Our study site was a petroleum refinery abandoned in 1964. It still displays a
-2-
was chosen for its great abundance and diversity in a wide variety of soil types [19,20]
andTaraxacum officinale, a shrub layer composed ofFraxinus excelsior,Rubus
pendula andQuercus robur. The unpolluted,control plot was located 15 m south of
18
17
buildings now included in a great variety of naturally established ecosystems
19
tar, as ascertained by the examination of a number of adjacent (polluted and
composed ofHedera helix,Geranium robertianum,Carex pilosa,Solidago canadensis
deposits. Two plots (polluted versus unpolluted used as control), were sampled in
22
October 2002. They were considered to be representative of the patchy occurrence of
analyses were used to follow changes in function and biodiversity of collembolan
14
15
populations under the influence of PAH pollution.
composed of recent fluvial deposits overlaying 1400 m thick sediments above the
granitic substratum [26].
16
13
The study was conducted at the site of the petroleum refinery of Merkwiller
9
2.1. Study site
6
24
20
2
4
3
7
12
11
1
5
sand) on the western edge of the Rhine rift valley. The superficial soil (5 m) is
10
abandoned. The site is located on the MerkwillerPechelbronn oil field (bituminous
8
2. Materials and methods
intensive activity until 1964, afterwards it was progressively dismantled then totally
Pechelbronn, about 50 km north of Strasbourg (Alsace, France). The refinery had an
23
21
a thick pasty tar patch overlaid by a 35 cm organic layer, a discontinuous field layer
28
26
27
25
Nowadays, the site area (20 ha) is characterised by the presence of old
2 unpolluted) soil profiles. Thepolluted plot, approximately 20 m , was characterised by
2 the polluted plot, and its surface was arbitrarily fixed to 20 m , for the sake of
fruticosus andSalix capraea and a tree layer composed ofAcer campestre,Betula
(woodland, grassland, ponds) with zones polluted by hydrocarbons, in particular by tar
-3-
identification. The animals were sorted under a dissecting microscope, then mounted in
1
7
8
abundance, the the species richness, the Shannon Index and the equitability. We also
identified to species level using a phase contrast microscope at 500x and several keys
21
comparison, although it did not differ from the rest of the site. It was characterised by a
22
26
-4-
much greater plant biodiversity and a good earthworm mull humus. The field layer was
24
25
composed ofHedera helix,Arum maculatum,Carex pilosa,Fragaria vesca,Geranium
robertianum, Geum urbanum,
15
14
canadensis andTaraxacum officinale, the shrub layer was composed ofAcer
11
Rubus fruticosus, and the tree layer was composed ofAcer campestre,Prunus avium
10
12
chlorallactophenol (25 ml lactic acid, 50 g chloral hydrate, 25 ml phenol). They were
calculated the total species richness of the five samples taken at each plot.
deposit and in the adjacent unpolluted plot. At each plot, five soil cores, 5 cm diameter
2
monogyna,Fraxinus excelsior,Ligustrum vulgare,Prunus avium,Rosa canina and
2.3. Observation of gut contents
Collembola were sampled on the organic matter accumulated over the tar
4
3
19
5
17
and diagnoses [4,6,13,16,23,27,28,29]. We calculated in each sample the total
Potentilla reptans,Stachys sylvatica,Solidago
18
by the dryfunnel method [9] and preserved in 95% (v/v) ethyl alcohol until sorting and
and 10 cm depth (litter included), were collected with a core sampler and, after
andQuercus robur.
20
23
16
13
discarding tar pieces, stored in plastic bags. Springtails were extracted from soil/litter
6
2.2. Sampling and identification of animals
9
pseudoplatanus,Carpinus betulus,Cornus mas,Cornus sanguinea,Crataegus
1
Samples were collected in April 2003 to determine the amount of PAHs from the
Gut contents from each mounted Collembola were observed under phase
-5-
19
20
Holorganic humus (amorphous organic matter)
Mineral particles
17
16
5% in each animal gut then averaged by species and by site.
were taken from each plot, then kept in glass jars and rapidly transported to the
laboratory. Then, soil samples were homogenised and sieved at 1 cm then kept in
25
27
26
23
22
validity of our control plot, samples were collected in May 2003 in the park of the
at 1020 cm depth and (3) the OF horizon overlying the tar deposit. To check the
Hyaline hyphae
Pollen grains
12
13
earthworm mull at neutral pH) under deciduous woody vegetation. Composite samples
Fungal spores
Melanized hyphae
14
10
8
9
6
4
5
3
laboratory (Brunoy, IledeFrance, France), in the 10 top cm of a similar soil (rich
7
11
15
The proportion in volume of the different categories was visually estimated to the next
24
contrast at 500x magnification and classified in twelve categories:
2
2.4. Chemical analyses
18
21
Microalgae
Bacteria
Empty guts
Hemorganic humus (mixture of organic matter and mineral particles)
EPA list [14] in (1) the 10 top cm of the A horizon of the control soil (2) the tar deposit
detector, column LCPAH Supelco). After sieving the contaminated soil, which was
20
17
benzo(ghi)perylene, the others being under detection level (0.02 µg/g for
9
2
mm.
acetone (50/50) for soil and acetonitrile for tar. The extract was concentrated under
2.5. Statistical treatment
After extraction of fauna, airdried soil was stored in plastic bags for pH
24
7
8
Power Liquid Chromatography) with UV detection (alliance 2690 chain, PDA 996
glass jars at 18C° until analysis. Each sample was defrozen, dried, then sieved at 2
mainly made of badly decomposed tree litter, only a small amount of fine matter was
12
-6-
21
22
23
11
10
14
15
16
an estimate of exchange acidity.
PAHs were extracted from each sample with the automatic system ASE 200
measurements were done in triplicate.
13
6
µg/g for contaminated soil, 0.1 µg/g for tar). All
uncontaminated soil, 0.04
19
18
1
glass electrode. The difference between these two measurements (pH) was taken as
5
3
4
(Accelerated Solvent Extraction) DIONEX, using a mixture of dichloromethane and
naphthalene, phenanthrene, fluoranthrene, pyrene, benzo(a)anthracene, chrysene and
available for analysis, and only 7 major PAHs could be analysed on this material:
forced air with a TuroVap LV (Zymark), then PAHs were separated by HPLC (High
v:v) for pH H20 and pH KCl, respectively. Each suspension was shaken for five
measurement. The soil was suspended in deionized H20 and 1M KCl (1:5 soil:water
25
27
26
minutes, then pH was measured in the supernatant after sedimentation for 2 h with a
Nonparametric MannWhitney tests were performed to compare chemical
parameters (pH) and population parameters (abundance, species richness, diversity
16
18
17
3. Results
from the park of the laboratory (Table 1). As a consequence we estimated that the plot
12
8
7
chosen as a control at MerkwillerPechelbronn was valid. The total amount of the
were quite similar in the control soil from the MerkwillerPechelbronn site and in the soil
The pH measured in water and potassium chloride was neutral and did not
24
27
26
25
the total abundance of springtails did not differ between plots.
control, while the amount of fluoranthrene was only three times higher.
11
10
3
5
4
equitability were higher at the control than at the contaminated plot (Table 2). However,
23
3.2. Collembolan communities
differ between both sites at MerkwillerPechelbronn (Table 1). However, mannWhitney
µg/g). The concentration of benzo(ghi)perylene was eleven times higher than the
and equitability of collembolan communities) between sites, using samples as
seven PAHs analysed in the soil over the tar deposit at MerkwillerPechelbronn (5.95
1
14
-7-
site.
20
21
polluted zone. The species richness per sample, the Shannon Index and the
The distribution of the 16 PAHs of the EPA list and the total amount of PAHs
At the unpolluted plot 27 species were found whereas only 9 were found in the
µg/g) was six times higher than the corresponding amount at the control plot (0.93
13
15
22
19
U test revealed that exchange acidity, expressed bypH, was higher on the unpolluted
3.1. Chemical analyses
replicates (five in each site).
2
6
9
samples taken at the control plot. The abundance of this species (together with that of
guts was higher at the polluted plot forP. notabilisandS. aureus.
bacteriophagous speciesminimus M.  contained 68% bacteria, and this species was
the low number of specimens (3) collected over the tar deposit. The gut of the
mycelium in the polluted plot. However the latter observation could be an artifact due to
algae, pollen, humus, bacteria and fungal spores and mycelia, but only fungal
was present in all samples taken at the polluted plot while it was absent from all
-8-
candida,Parisotoma notabilis,Sminthurides malmgreni,Sminthurides parvulus and
27
23
24
only present at the unpolluted site, where it was found in all samples. The isotomidP.
26
25
22
17
16
Sminthurinus aureus. Among these species,S. aureuswas significantly more abundant
represented byProisotoma minutaand (43%) Parisotoma notabilis (40%).P. minuta
Sminthurinus aureus was compared between plots (Figure 1). Fungal mycelium was
inF. candida guts which showed bacteria in equal proportion in both sites but more
more common at the polluted plot, while the contrary was observed for bacteria, except
1
2
Paristoma notabilisin most samples from the tar deposit) explains why the total
3.3. Diets of Collembola
Only five species were common to both polluted and unpolluted plots:Folsomia
8
10
9
20
19
21
abundance of Collembola did not differ between plots in spite of a much lower number
hyaline hyphae and hemorganic humus over the tar deposit. The proportion of empty
14
13
12
6
5
in the control soil (x 4). Over the tar deposit, 83% of the total abundance was
4
18
15
The food diet of the three speciesParisotoma notabilis,Folsomia candida and
of species at the polluted plot (Table 2).
11
The diet ofS. aureuswas more diverse at the unpolluted plot, containing micro
7
3
5
6
use of food resources, since fungi were favoured to the detriment of bacteria. This
influence of pollution by hydrocarbons. This shift can be explained by a change in the
8
9
20
metals [11], the species richness of springtail communities was observed to decrease
19
18
cases, this could be explained by an increase in free habitats liberated by most
control soil (Table 2). It should be noticed that sieving of the soil to 2 mm before PAH
16
15
-9-
27
28
minuta, which was present only at the polluted site where it dominated the community,
showed a high proportion of empty guts (67%), 21% mycelium and few bacteria (2%).
speciesProisotoma minutaandParisotoma notabilis.
In our study, changes were observed in the diet of springtails under the
control plot. Three years after application, toxicity was still present given the absence of
Sendstad [25] who found 2% of total Collembola in a polluted plot compared to a
12
13
11
could be affected by i) the presence of pollutants in the soil, through direct effects of
concentration in the soil over tar. The toxicity of crude oil for Collembola was shown by
observed by Blakely et al. [3]. Here this group was represented by the two dominant
PAHs or other toxic compounds on the most sensitive species and/or ii) indirect effects
sensitive species. Isotomids seemed to be less sensitive to PAH pollution, as was also
under the influence of pollution, at constant or even higher total abundance. In both
4. Discussion
observed between the organic soil accumulated over the tar deposit and the adjacent
17
such as changes in food resources. Like in a previous study on a site polluted by heavy
21
14
extraction (see Materials and methods) could have led to underestimate PAH
the tar deposit (Table 1), a great dissimilarity in collembolan communities was
26
10
7
4
recolonisation by fauna of the polluted zone. In our study, the collembolan community
25
In spite of a slight distance between plots and a low concentration of PAHs over
24
22
23
1
3
2
possibility reflected a change in the composition of microbial communities in the favour
ectomycorrhizal root mat caused significant changes in the environment of animal as
humus form (and associated foodwebs) than by the direct toxicity of PAHs to
15
by
References
PAH analyses.
National de Recherches sur les Sites et Sols Pollués (CNRSSP, Douai, France) for
2
1
present at the unpolluted plot. Micromorphological analyses showed a high amount of
16
activity)
22
of fungi, which could explain why the bacteriophagousMegalothorax minimuswas only
to
23
5
4
with a poor earthworm activity, while a rapid incorporation of organic matter to mineral
ascertained
well as microbial communities. Given the shift from mull (dominated by earthworm
change in humus form [22].
We found strong changes in invertebrate communities and alimentary habits of
We thank the Agence de l’Environnement et de la Maîtrise de l’Energie (ADEME) for
activity)
18
20
19
earthworm populations to colonize pollution spots, and the development of a superficial
28
25
26
27
microbial communities in favour of fungi could be better explained by a change in
matter through earthworm activity was registered in the control soil [12]. Shifts in
enchytraeid
soil animals forty years after abandonment of industrial activity. Despite a low PAH
11
9
8
13
7
12
moder
micromorphological analysis [12], the observed decrease in biodiversity and change in
(dominated
concentration in the soil accumulated over the tar deposit (Table 1), the inability of
microorganisms [3].
17
24
21
Acknowledgements
6
3
-10-
ectomycorrizal fungal hyphae in the organic matter accumulated over the tar deposit
10
financial support and Vasilica HAMMADE and Laurence BELKESSAM from the Centre
by
the food diet of the Collembolan community can be ascribed, at least partly, to a
14
[1]
2
G.R. Best, J.V. Nabholz, J. Ojasti, D.A. Crossley Jr, Response of microarthropod
[3]
[2]
6
21
10
[6]
7
8
9
25
[7]
22
26
23
27
24
3
4
5
13
12
A. Bispo, M.J. Jourdain, M. Jauzein, Toxicity and genotoxicity of industrial soils
K. Christiansen, M.M. Da Gama, P. Bellinger, A catalogue of the species of the
P. Chenon, A. Rousset, Y. Crouau, Genetic polymorphism in nine clones of a
17
[5]
[4]
14
contamination, Appl. Soil Ecol. 21 (2002) 7188.
(2000) 103110.
parthenogenetic collembolan used in ecotoxicological testing, Appl. Soil Ecol. 14
1
Naturkundemuseums Görlitz 71 (1999) 1318.
genusPseudosinella, Ciência Biol 5 (1983) 1331.
16
Isotomidae) for the bioassay of xenobiotic substances and soil pollutants, Appl.
15
Soil Ecol. 12 (1999) 103111.
J.K. Blakely, D.A. Neher, A.L. Spongberg, Soil invertebrate and microbial
G. Bretfeld, Synopses on Palaearctic Collembola. II. Symphypleona, Abh. Ber.
polluted by polycyclic aromatic hydrocarbons (PAHs), Org. Geochem. 30 (1999)
11
19
20
18
Y. Crouau, P. Chenon, C. Gisclard, The use ofFolsomia candida (Collembola,
947952.
populations to naphthalene in three contrasting habitats, Pedobiologia 18 (1978)
189201.
communities and decomposition as indicators of polycyclic aromatic hydrocarbon
-11-