Les effets de l'oxygénation dans un lac eutrophe étudiés par la méthode de l'«enceinte»

Le lac de Comabhio (Lombardie, Italie du Nord) est un lac peu profond (z- = 4,4 m) en comparaison à sa surface (3,44 km2). L'eutrophisation naturelle du lac a été accélérée par l'apport d'effluents domestiques pas traités. Pour améliorer la situation, les effluents conduits par un collecteur circumlacustre, ont été recueillis à l'extérieur du bassin versant.Les effets seront visibles seulement dans plusieurs années et ces interventions ne seront pas suffisantes pour ramener le lac à l'état de mésotrophie. Afin d'accélérer l'assainissement du lac, mais surtout pour éviter la forte mortalité du poisson périodiquement à la fin de l'été, on a pris la décision d'oxygéner artificiellement l'hypolimnion. Pour prévoir les effets de l'oxygénation on a fait une expérience dans le lac avec deux enceintes en PVC (diamètre : 40 m; profondeur : 6 m; volume : 7000 m3). On a gardé une enceinte comme témoin; dans l'autre a été insufflé de l'oxygène pur. On a recueilli, à certains intervalles de temps, des échantillons d'eau et de plancton du lac et des deux enceintes. Les éléments nutritifs ont été analysés; la température, la transparence, la conductivité et le pH ont été mesurés sur place. L'effet le plus visible du traitement consistait dans une augmentation de la concentration de l'oxygène visant à produire des conditions normales pour la vie des poissons. Les effets de l'oxygénation sur les caractéristiques chimiques, physiques et biologiques du milieu sont discutés.Lake Comabbio (Lombardy, Northern Itaty) is a shallow Lake (zmax = 8.0 m; z-= 4.4 m; surface = 3.44 km2). Its naturally high trophic level has been increased by the huge nutrient loading from domestic effluents. Each year (except in winter, when commonly the lake is covered with ice) more phytoplankton blooms are observed and in the late summer oxygen depletion, with a consequent mass mortality of fish, occurs.At present, to reduce the nutrient charge, the effluents are collected in a channel and diverted from the lake watershed. Because of the significant nutrient release from the sediment ("internal eutrophication") the reduction of the external loading may slow the eutrophication rate, but it is not sufficient to restore the lake to an acceptable mesotrophic state within a reasonable time.To prevent the periodic mass mortality of fish and possibly to accelerate the restoration process of the ecosystem, the Lake Comabbio Protection Committee decided to oxygenate the hypolimnetic layer artificially. In order to avoid undesirable effects of aeration (e.g. nitrogen enrichment of lake water), dissolved oxygen injection was preferred. As there is no general agreement on the ecological effects of the oxygen (or air) addition it was decided, before applying oxygenation to the whole lake, to carry out experiments using the "enclosure" method. One experiment was carried out from 26th September to 9th December 1986 and this was replicated in 1987 from 22nd April to 30th November. This paper is concerned with the first experiment, as data treatment of the second is still in progress. Two cylindrical "enclosures" of PVC (diameter = 40.0 m; height = 6.0 m; volume = 7000 m3) were settled in Lake Comabbio to isolate a water column with its sediments. One enclosure was treated with oxygen from 15th November to 6th December, the other was kept as control. The water aspirated from the enclosure was oxygenated (20.3 mg O2/l) and injected in the same enclosure at 4 m depth at a rate of 4 m3/h. During the enclosure settlement the fish escaped from the enclosures. To simulate a complete ecosystem, the plan was to add to each enclosure 60 kg of the most abundant fish species (Scardinius erythrophtalmus). By mistake, fish were added to the enclosure that was to be oxygenated but not to the control. Water and zooplankton were sampled simultaneously from both the enclosures as well as from the Lake. Zooplankton was collected by vertical hauls from the bottom to the water surface. The following parameters were measured : temperature, transparency, pH, electrical conductivity, dissolved oxygen, alkalinity, phosphorus and nitrogen compounds, chlorophyll-a and pheopigments. The injection of water rich in oxygen did not resuspend the sediments and the absence of bubbles showed that the oxygen had completely dissolved.Soon after oxygenation the difference in oxygen concentration in the treated enclosure and in the control was progressively reduced until 26th November, when both the enclosures had the same oxygen concentration. At the end of the experiment (9th December) the oxygen concentration in the control was about 4 mg/l, whereas in the treated enclosure it attained a concentration of 6 mg/l, although in the latter the oxygen consumption was greater than in the control, owing to fish respiration and the increased activity of aerobic microorganisms. The higher concentrations of total phosphorus (TP), soluble reactive phosphorus (SRP) and ammonium (N-NH4) measured in the lake were probably due to the external loading of nutrients. The nutrient concentrations in the treated enclosure were similar to those of the control. The greater concentration of nutrients was probably the cause of more abundant phytoplankton in the lake, which was also demonstrated by the lower values of water transparency, when compared with those measured in both the enclosures. The pH values of the control were similar to those of the treated enclosure. The pattern of abundance and species composition of the zooplankton in the treated enclosure did not differ from that of the control. The Entomostraca populations decreased in the lake as well as in both the enclosures from the beginning to the end of the experiment.From this experiment, carried out when oxygen depletion affected the whole water column, we may conclude that oxygenation can prevent fish mortality, without significantly influencing the structure and biomass of the planktonic community. The variations in the plankton and chemical and physical characteristics of the water (except oxygen concentration) were essentially due to the season and mot to the artificial oxygenation. In addition, oxygenation, by abolishing the periodical mortality of fish, may also prevent the release of a huge amount of organic substances from the dead fish into the water