Résumés
Résumé
A partir de références bibliographiques encore réduites, les auteurs ont mis au point un protocole expérimental d'évaluation de la toxicité des sédiments de rivière, après extraction au dichlorométhane et mesure de la toxicité par le lest de Bioluminescence Bactérienne.
Ce protocole simple appliqué à quarante-trois prélèvements correspondant à des situations de pollution diversifiées, a permis d'établir et de proposer une typologie des sédiments en fonction de leur niveau de toxicité.
Correspondant à une approche nouvelle (tests biologiques de toxicité sur sédiments), la méthodologie apparaît particulièrement intéressante pour la gestion patrimoniale des cours d'eau (réseaux de surveillance, mise en évidence de situations de pollutions), et devrait être intégrée dans les opérations de mesure de l'impact des pollutions toxiques sur les écosystèmes aquatiques.
Mots-clés:
- Sédiments,
- toxicité,
- bioluminescence,
- microtox,
- rivières
Abstract
As toxic substances in river water are not easy to detect and to measure, sediments able to precipitate them have been userd, successfully, to quaintly metals and a limited number of organics. Actually, analytical determinations, very expansive, time consuming and never exhaustive, are not an adequate means for a general estimation of the presence of toxic organic substances.
Another developing approach consists in determining effects not the substances themselves (contained in the sediments) but their toxic effects on biotests applied directly on all the sediments or on the extracts. This approach has been used by relatively few teams, principally in North-America and in The Netherlands, in limited areas, to evaluate the local impacts of industrial effluents on sea and-or river waters. On the contrary, the object of this work is to measure the interest of this kind of approach, at a large Water Basin, first to establish the toxic profiles of rivers with areas of pollution, and second to constitute a memorization, susceptible of being compared at regular intervals for an estimation of general depollution policies .Based on data taken from the literature, an organic extraction of sediments with dichloromethane as solvent (without pretreatment) and photobacterium phosphoreum luminescence inhibition test were chosen. Actually, according to the literature, dichloromethane is the must convenient solvant and toxic activities of extracts are well correlated with organic toxic contents in the few cases where, exceptionally, the two approaches have been carried out simultaneously. Similarly, the bacterial luminescence test, Microtox (commercial name), quick and inexpensive, is well correlated with results from other biotests when, occasionally, both were utilised. Moreover, Microtox was well tested in the Agence laboratory.
Different conditions of extraction were carefully tested, before the adoption of the definite protocol : whole sediments (10-g) are mixed (slow agitation in rolling flasks) for 6 hours with dichloromethane (100-mL) and sodium sulphate (50 g). Dichloromethane is then eliminated and the remaining solid materials are washed twice with fresh dichloromethane (2 x 20 mi). Dichloromethane extract and washings are collected, dried over sodium sulphate and concentrated (below 40 °C) to 1 mL, further adjusted to 3 mL. A portion measured accurately of this concentrate is added to ethanol and concentrated to remove dichloromethane.
Bacterial bioluminescence tests, in duplicate, are performed on ethanol extracts, diluted (1 %) in salified (sodium chloride 2 %) distilled water. Estimates of the EC 50 (concentration causing a 50 % reduction in bioluminescence) obtained using linear regression analysis are converted into Toxic Units and referred to the net organic weight, measured separately, at 550 °C. This expression of results is particularly relevant when sediments with different organic net weights are compared.
The protocol has been applied to sediments collected at 43 sites in the principal rivers of Rhône Méditerranée Corse Basin and in special distant from significant domestic and industrial activities.
The toxic amplitude observed (ratio of l/200 between lowest and highest toxic units obtained with the panel) and the good reproductibility of the whole process (10 % to 25 %) were consistent with a trial subdivision in 5 classes as follow :
Class 1 : Out of pollution - Class 2 : Moderate toxicity - Class 3 : Important toxicity - Class 4 : Very important toxicity - Class 5 : Exceptional toxicity.
The relevancy of this classification was estimated when sediments were classed according to their toxic content. The classification proves to be correct in must cases. Exceptionally toxic sediments were collected in sites affected by large industries and agglomerations as (CHASSE and SAINT-VALLIER) on River Rhone near Lyon and important chemical plants, or GIVORS, on the small River GIER, markedly affected by many various industries. On the contrary sediments of moderate toxicity corresponded to sites relatively far from sources of important pollutions, e.g., ARLES on the River RHONE near the sea and a number of sites on the River SAONE. Moreover, when repeated sampling was performed on the same sites, the results were consistent, belonging to the same class of toxicity.
The general object of this work was not to determine a definite classification of sediment toxicities, but to assess the relevancy of this approach. In our opinion, it was proved that bioassays on sediment extracts, and specially the bacterial bioluminescence assay, are a valuable tool, before the material impossibility to determine the toxic substances content. Moreover this approach could be used at the Basin. More investigations are necessary to define more accurately the number and level of toxic classes and e.g. to establish correlations with perturbations of benthic communities living in the sediments. It is also a contribution to the general knowledge and action based on biotests performed on effluents and different compartments of rivers.
Keywords:
- Sediments,
- toxicity,
- bioluminescence,
- microtox,
- rivers
Veuillez télécharger l’article en PDF pour le lire.
Télécharger