Simulation of carbon, nitrogen and sulphur conversion in batch-operated experimental wetland mesocosms

Abstract

A simulation model based on Constructed Wetland Model No. 1 (CWM1) using the AQUASIM mixed reactor compartment as a platform was built to study the dynamics of key processes governing COD and nutrient removal in wetland systems. Data from 16 subsurface-flow wetland mesocosms operated under controlled greenhouse conditions with three different plant species (Typha latifolia, Carex rostrata, Schoenoplectus acutus) and an unplanted control were used for calibration and validation in this mechanis-tic model. Mathematical equations for plant related processes (growth, physical degradation, decay, and oxygen leaching), physical re-aeration, as well as adsorption and desorption processes for COD and ammo-nium were included and implemented alongside CWM1 in the AQUASIM software, while some CWM1 parameters were adjusted to better fit the model predictions to experimental data during calibration. The simulation results showed that the model was able to describe the general trend of COD (R 2 = 0.97-0.99), ammonium (R 2 = 0.85-0.97) and sulphate (R 2 = 0.71-0.93) removal in the wetland mesocosms and also in their controls (unplanted) through the experimental temperature range of 12-24 • C. Oxygen transfer by physical re-aeration was found to be 0.05 and 0.09 g m −2 d −1 at 12 • C and 24 • C, respectively. The amount of root oxygen transfer was the highest for the planted mesocosms at 12 • C at rates of 1.91, 0.94, and 0.45 g m −2 d −1 in the Carex, Schoenoplectus and Typha mesocosms, respectively, indicating that COD of the bulk wastewater was removed mainly by anaerobic processes under the specific experimental situations. Measured COD removal was better in the planted mesocosms than in the control; differences were effectively modelled by varying the bacteria concentration. The sorption process was found to be important in simulating COD and ammonia removal under these experimental conditions.

Subsurface Flow Constructed Wetland Models: Review and Prospects

“…According to these criteria, six models are selected for review: PHWAT (Brovelli et al 2007(Brovelli et al , 2009a, FITOVERT (Giraldi et al 2009(Giraldi et al , 2010, HYDRUS-2D-CW2D (Langergraber and Š imůnek 2005), HYDRUS-2D-CWM1 (Langergraber and Š imůnek 2012), CWM1-RETRASO (Llorens et al 2011a, b), BIO_PORE (Samsó and García 2013a, b) and AQUASIM-CWM1 (Mburu et al 2012). …”

“…One of the advantages of the implementation of AQUASIM-CWM1 is that the AQUASIM platform itself can be downloaded free of charge (although it is not open source), and so the distribution of the implementation of the CWM1 made by Mburu et al (2012) only depends on the author's willingness to share it. Likewise, the equations describing the different physico-chemical and biological processes are easily customizable.…”

“…HYDRUS-2D-CWM1 was also used to recreate the simulations performed by Llorens et al (2011a, b), obtaining different results. Mburu et al (2012) implemented the biokinetic model CWM1 into AQUASIM software (Reichert 1998) to simulate the fate of organic matter, nitrogen and sulphur within 16 batch-operated subsurface flow wetland mesocosms planted with three different plant species (for a detailed description of the mesocosms, the reader is referred to Allen et al (2002) and Stein et al (2006)). The mixed reactor compartment configuration in AQUASIM was used and the mesocosms were described as constantly stirred tank reactors (CSTR).…”

Effect of key design parameters on bacteria community and effluent pollutant concentrations in constructed wetlands using mathematical models

“…In any case, the performed 500-days simulations had been long enough for the stabilization of the bacteria communities. On the other hand, bacteria communities distribution in CWs is still mainly based on assumption and circumstantial evidence (Krasnits et al, 2009;Mburu et al, 2012;Ojeda et al, 2008), since there are only a few studies on the subject.…”

“…The recent development of mathematical models simulating the complex functioning of CWs (e.g. Ojeda et al (2008), Llorens et al (2011b), Langergraber and Šimunek (2012), Mburu et al (2012)) also provides a useful tool to improve the knowledge about their internal dynamics, making it possible to assess the influence of design parameters on the treatment efficiencies of CWs. Nevertheless, most existing models have been employed so far to match the simulated effluent concentrations and those measured experimentally, while the description of the internal functioning of the CWs has received less attention (Samsó and García, 2014).…”