For environmental and economical issues, it is relevant to map the underwater vegetation distribution. Acoustic techniques seem to be the more efficient and cost effective method for that. Many observations were performed on submerged aquatic vegetation detection by acoustic echosounder. However few general studies were performed to quantitatively understand the acoustic process of scattering from underwater vegetation. In the first place, the effect of vegetation meadows on the temporal signal of the echo sounder is studied. Real data are used. They come from different sites and types of submerged aquatic vegetation. In most cases, the vegetation signal has a relatively strong backscatter signal before the bottom detection time and a less abrupt leading edge. However different species dependant behaviors are found. To explain this phenomenon, it is necessary to know more accurately the interaction between the acoustic wave and the environment made up of underwater vegetation and sea water. Therefore a model is developed. It recreates the temporal backscatter signal from an echosounder. The seabed is composed of two layers, the vegetation and the sediment layer. For sediments, existing statistical and physical models are used. For the vegetation, the model takes inspiration from the previous approaches and the geoacoustic parameters of this environment are fitted to the ground truth data.