A dedicated trawling experiment was performed at three sites on the Gulf of Lion continental shelf, with the aim of assessing the resuspension of particulate and dissolved matter triggered by different types of trawls on muddy sediments. The different configurations were: (i) bottom trawl, with bobbin for ground rope (Rockhopper): (ii) bottom trawl, without bobbin (Medits); and (iii) pelagic trawl, towed at 1 and 10m above the seabed. The plumes of resuspended sediment were measured using the acoustic backscattered intensity, from a towed ADCP. Concomitant profiles of particle size-distribution, light transmission and water samples were collected, outside and inside the plumes. The analysis of the data enabled derivation of the major physical and chemical characteristics of the plumes generated by the trawls; likewise, and to quantify the resuspension fluxes of sediment, particulate (PN, POC) and dissolved (nutrients) elements. The residence time and dispersal of the plumes were monitored and modelled, considering the settling velocity of the particulate matter and the near-bottom turbulence. The results indicate that the bottom trawls produce significant resuspension, whilst the near-bottom and mid-water pelagic trawls have no impact upon the sediment. The sediment clouds at several hundreds metres astern of the bottom trawls are 3-6m high and 70-200m, wide; they were generated both by the otter doors and the net. The average suspended sediment concentrations measured in the plumes reach 50 mg l(-1). Resuspension fluxes of sediment along the path of the trawls range from 190 g m(-2) s(-1), for the coarsest sediment (clayey silt) to 800 g m(-2) s(-1) for the finest sediment (silty clay). Whilst the resuspended loads of dissolved elements (nutrients) within the plume segment suggest a release of porewater, present at least in the first few centimetres of sediment, the particulate matter load only resulted from the resuspension of less than 1 mm thickness of the sediment bed. This discrepancy shows that a very small fraction of the sediment ploughed by the trawl is effectively injected into the water column. The monitoring of the settling of the plumes indicates a rapid decay of the sediment load, during the first hour after its generation. Some of the sediment (about 10-15% of the initial load) remains in suspension; this is due, probably, to the near-bottom turbulence that prevents the redeposition of the fine particles and aggregates. Lateral spreading of the plume is strongly dependent upon the variability of horizontal currents.