Evaluation and mathematical modeling of the removal of particulate matter in a real-scale lawn channel

Abstract

A mathematical model was applied to describe the removal of particulate matter in a real-scale lawn-lined channel. For this purpose different initial concentrations (Co) of particulate matter were investigated, as follows: 65; 131; 196; 262 and 327 mg/L. The hydrographs investigated, containing suspended solids, were discharged into a 100 m long channel with a 2% slope. The particulate matter was measured by analyzing turbidity and total suspended solids (TSS) at 26 sites throughout the length of the channel. The experimental data were fitted to the model using nonlinear regression parameters (k(d)) and (C*), which correspond to the decay kinetics constant and minimum limiting values of the removal curve, respectively. Fractions removed and mathematical fitting were assessed and performed for each isolated event and also for the set of results. On average, the magnitude of the relative error ranged from 0.83% to 5.11%, with the highest average value obtained for the lowest concentration of particulate matter (65 mg L-1). Regarding the values obtained for the decay constants (k(d)), which represent the frequency of removal of particulate matter, it is noted that it tends to a minimum value for the higher initial concentrations of particles (Co). This fact indicates that the higher the initial concentration of particulate matter (Co), the greater the extent of travel required for its removal, with a maximum observed at around 80 m, for concentrations above 196 mg.L-1