Nouvelles trappes à sédiment destinées aux milieux peu profonds vidangeables New sediment traps for drainable shallow freshwater systems

Résumé

Les mesures de taux de sédimentation en milieux aquatiques peu profonds sont rares et sont souvent réalisées à l'aide de méthodes inadaptées. Les trappes à sédiment utilisées ont au minimum 25 cm de haut. Par conséquent, pour beaucoup de milieux peu profonds, plus de 25 % de la colonne d'eau ne sont pas échantillonnés. Nous avons pallié ce problème en développant un réceptacle mis en place dans les sédiments et destiné à recevoir des trappes à sédiment cylindriques. Le sommet des trappes peut alors être situé à moins de 5 cm de la surface des sédiments. Ce système est utilisable pour des milieux de profondeur inférieure à cinq mètres. Nos résultats montrent que les trappes à sédiments généralement utilisées sous-estiment de 35 à 79 % du taux de sédimentation. Les particules négligées proviennent des flux sédimentaires primaire et secondaire.

Mots-clés:

• Trappes à sédiments,
• sédimentation,
• milieux aquatiques peu profonds

Abstract

Sediment traps are a unique tool that can be used to investigate particle settling flux throughout the water column, whereas other methods such as sediment dating can only measure accumulation rates of bottom sediments. Several works on trapping efficiency have shown that cylindrical traps with height/diameter ratio greater than to 5 (10 in turbulent systems) are the more appropriate instruments to correctly measure the downward settling flux of particulate matter. Furthermore, traps with a diameter narrower than 5 cm should be avoided. It is well documented that bottle-type vessels overestimate the settling sediment whereas funnels and flat containers underestimate it. All this support the idea that an ideal trap must be at least 25 cm high, and in this sense, numerous studies investigating shallow aquatic systems have neglected a large proportion of the water column. Consequently, mechanical and biological processes occurring in this layer of the water column have not been taken into account

We have overcome this problem with a structure composed of two parts (figure 2).

The first part is a receptacle (bucket) buried in the sediment and intended to receive cylindrical traps. The top of the receptacle is placed 2 cm above the sediment. A guide made of a rope covered with a PVC tube is placed in the centre of the receptacle. This receptacle is intended to receive cylindrical traps whose tops stand less than 5 cm higher than the surface of the sediments.

The second part is composed of seven cylinders (height/diameter ratio=10) which are fixed in a PVC disc 600 mm in diameter and 15 mm deep. The cylinders are placed around the central axis of the PVC disc. The bottom of the cylinders is closed with a removable polyethylene cap. Another cylinder, through which the guide can slide, is placed on the central axis. The bottom part of this last cylinder is ballasted with concrete. The stability of this second part, during both deposition and removal steps, is ensured by the low density of the PVC disc, the ballast at the bottom of the central cylinder, and the symmetry of the structure. This removable part may be lifted from the receptacle with three 2 mm diameter ropes attached to the PVC disc and fixed to a float. This apparatus may be used as deep as five meters.

The sediment traps were tested in two extensively-managed fish ponds in North-Eastern France. Our investigations showed that tubes with a diameter between 26 and 140 mm could be efficiently used to estimate the sedimentation rate, whereas cylinders with a narrower diameter missed a large amount of particles. The use of tubes with a diameter above 50 mm, which is preferable for the study of turbulent systems, seems to allow the collection of sufficient sediment during a short period of time. We selected tubes with a diameter of 57 mm, which made it possible to handle them easily during the removal. Our investigations showed that in turbulent systems and for high sedimentation rates (> 5 kg·m^-2 ·month^-1), the top of the cylinders must be placed 1 cm above the top of the PVC disc. When considering low sedimentation rates (< 5 kg·m^-2 ·month^-1), we did not observe any significant differences between the cylinders placed 0 and 1 cm above the PVC disc.

Complementary investigations were conducted in order to compare sedimentation rates estimated 1) by the apparatus we designed, in which the top of the cylinders was placed 5 cm above the sediment surface, 2) by traditional traps (57 mm in diameter and a height/diameter ratio of 5) in which the top of the cylinders was 28.5 cm above the sediment surface.

These results showed that in shallow systems (1.2 m deep), traditional traps underestimate the downward settling flux of particulate matter by 35 to 79%. Furthermore, we compared the organic matter content of the sediment collected by the two types of traps with the organic matter content of bottom sediment, suspended particles, and submerged macrophytes. Results showed that the underestimation of particles was not only due to the resuspension of bottom sediment, but also to the sedimentation of phytoplankton and submerged macrophyte fragments which are not collected by the traditional traps.

Keywords:

• Sediment traps,
• sedimentation,
• shallow freshwater-systems