Numerical modelling of bacteriologic impacts in the Laïta estuary (France)

The Laïta estuary is l ocated on the French Atlantic coast. Degraded for a long time by different sources (agriculture, industry, waste water treatment plant (WWTP) outfall...), the bacteriological quality of Laïta has improved, confirmed by a recent decree allowing mussels farm ing downstream the estuary. However, local authorities wish to increase knowledges on contamination sources in order to decrease drastically the main bacteriological releases in the watershed. In this context, a three dimensional model has been set up to b etter understand the behaviour and impact of the various sources of contamination. The study also aims to prioritise contamination flows and evaluate a series of proposals for improvements. The model Telemac 3D is set up in order to represent maritime and river processes. Various data allow calibrating and validating the model hydraulically. In order to model faecal coliform concentrations, it is necessary or even essential to study the mortality of bacteria. Classically, a first order of decay law is used where a value of T90 is imposed (duration to obtain 90% mortality). However, for an estuarine problem where salinity and temperature in particular may change, the use of a constant T90 value may be incorrect. In fact, the T90 bacteria vary from a few hours to a few days, but for the majority of the hydrodynamic conditions, the transit of bacteria from the river waters to the marine waters (which can greatly increase the mortality rate of bacteria) lasts less than one tidal cycle. It was therefore decided to use a formula [ to calculate the degradation coefficient (k) bacteria based on key parameters affecting bacteria mortality such as salinity, temperature, depth and light intensity. For the calibration scenarios, the flows imp osed at the affluent borders are either the result of monitoring survey or from parametric analysis. Effluent discharges from WWTP outfalls and any accidental pollution are also included in the modelling. They are represented by a concentration of micro ba cteria and a characteristic discharge This paper presents firstly the calibration of the numerical bacteriological dispersion model, the prioritization of faecal coliform fluxes, a brief analysis on impacts and a discussion on possible solutions to improv e the bacteriological qu ality