Application of first-order kinetic removal models on constructed wetlands under Mediterranean climatic conditions

Abstract

The objective of the study was to verify the applicability of P-k-C* model to evaluate the treatment response of horizontal CWs (H-CWs) and to assess key design parameters for the model optimization in Mediterranean climate conditions. The experimental study was carried out in 4H-CWs located in the province of Catania (Eastern Sicily, Italy). The parameters P, kA20 and θ were optimized to minimize the sum of the square of the errors (SSE) between observed and modeled outlet concentrations for COD, BOD5 and NO−3 quality parameters. Once the minimization process was carried out by an iterative, nonlinear procedure, then data were divided into calibration and validation sets for each CW. The calibration was set by fixing a P value from the literature range and varying kA20 and θ and then by using the obtained values of kA20 and θ and by varying P. This procedure was repeated until two successive P calibrated values differed less than 2%. The performance of the model was very good in most of the cases, as also evaluated in terms of R2, NSE and RMSE. In particular, considering the three investigated pollutants, the calculated kA20 values were generally higher than those from CWs located in temperate climate zone (65–157.1 m year−1), while the θ values were slightly lower than 1 (0.88–0.91). This study can be considered a first contribute for P-k-C* model application in semi-arid regions characterized by Mediterranean weather conditions.