Recovering the salinity intensity of distributed sources in the river using simulation-optimization approach

In recent years, the issue of identifying polluting sources in rivers has been one of the most important topics in scientific researches in the field of water. In the main researches, the pollutant sources have been considered as the point sources and in order to recover pollutant concentration, it is necessary to have an observation point for each source. In this study, the places where groundwater enters to river are considered as several distributed sources with known location and length and goal is to recover the intensity of sources, using only one observation point. The sources which considered in this research are distributed sources with constant loading and significant distance from each other. The existence of significant distance among sources prevents the complete mixing of concentration at observation point. This matter and also the constant intensity of loading, makes it possible to recover several distributed sources using only one observation point. For this purpose, the inverse solution of the advection-dispersion equation is done using the simulation-optimization approach. To design the backward model, MIKE11, linked with genetic algorithm in MATLAB. Considering one observation point for recovering the salinity intensity of several distributed sources and providing a general framework to recovery of distributed sources are advantages of the present study. The performance of the developed approach is evaluated under different examples and measurement error conditions. The computational results indicated the backward model can recover the salinity intensity of several distributed sources with high accuracy using only one observation point, even with noise.