Comparison of interannual removal variation of various constructed wetland types María Hijosa-Valsero a, ⁎, Ricardo Sidrach-Cardona b , Eloy Bécares a a Department of Biodiversity and Environmental Management, Faculty of Biologic and Environmental Sciences, University of León, Campus de Vegazana s/n, E-24071 León, Spain b Environment Institute, University of León, C/La Serna 58, E-24007 León, Spain abstract article info Article history: Received 5 August 2011 Received in revised form 27 April 2012 Accepted 28 April 2012 Available online 28 May 2012 Keywords: Constructed wetlands Urban wastewater Temporal removal changes RDA analysis Seven mesocosm-scale (1 m 2 ) constructed wetlands (CWs) of different configurations were operated out- doors for thirty-nine months under the same conditions to assess their ability to remove organic matter and nutrients from urban wastewaters. CWs differed in some design parameters, namely the presence of plants, the species chosen (i.e., Typha angustifolia or Phragmites australis), the flow configuration (i.e., surface flow or subsurface flow) and the presence/absence of a gravel bed. It was observed that, in general, removal efficiencies decreased with the aging of the system and that seasonality had a great influence on CWs. A com- parison was made in order to figure out which kind of CW was more efficient for the removal of every pol- lutant in the long term. Planted systems were clearly better than unplanted systems even in winter. Efficiency differences among CWs were not extremely great, especially after a few years. However, some types of CWs were more adequate for the removal of certain pollutants. The effect of the aging on the main parameters involved in pollutant removal in CWs (temperature, pH, conductivity, dissolved oxygen concentration and redox potential) was assessed. The efficiency of CWs should not be evaluated based on short monitoring periods (1–2 years) after the start-up of the systems, but on longer periods. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Constructed wetlands (CWs), used for treatment of urban sewage from small communities and of various kinds of industrial wastewa- ters, have been employed and studied for several decades (Kadlec and Wallace, 2009; Vymazal, 2009; Zhang et al., 2009). Their design is a function of the wastewater nature, its pollutant load, the available area to build the wetland and the climatic conditions of the site. The removal efficiency of CWs is usually assessed based on the data from relatively short experimental or sampling campaign pe- riods (up to 2 years). However, CWs are intended to treat wastewater during decades. Therefore, the application of those preliminary data (obtained during a brief period) to the design and/or maintenance of real CWs can result in an unexpected decrease in the long-term system performance. Bulc (2006) monitored the performance of a CW for landfill leachate treatment during seven years. Liikanen et al. (2006) studied seasonal and temporal changes in a boreal CW used to purify peat mining runoff waters 5 and 15 years after its construc- tion. Mitsch and Wilson (1996) studied the creation and restoration of new wetlands for mitigation of lost wetland habitat and proposed a period of at least 15 years to start judging the success of a CW. Stefanakis and Tsihrintzis (2009, 2012) assessed during 3 years the performance of experimental scale CWs treating synthetic wastewa- ter. However, the temporal behaviour of CWs treating raw urban wastewater could be different from that described by those authors, due to the different nature of the treated waters. It has been observed that the configuration and nature of a CW affects its performance (García et al., 2004; Hijosa-Valsero et al., 2010a,b). In general, physicochemical parameters (temperature, pH, dissolved oxygen concentration, redox potential, etc.) and the pres- ence of plants (which can modify some of the previous parameters) influence pollutant removal in CWs (Hijosa-Valsero et al., 2011), be- cause many biological removal processes (like microbiological degra- dation, plant uptake, biofilm adsorption, etc.) and abiotic removal processes (photodegradation, adsorption, chemical degradation, etc.) are controlled to a lesser or greater extent by these parameters. It is known that physicochemical parameters in CWs suffer spatial and temporal changes (Imfeld et al., 2009), even at a monthly or daily scale (Wießner et al., 2005), thus affecting and modifying the removal of organic matter. In this work, the evolution of seven types of mesocosm-scale (1 m 2 ) CWs was assessed during a 39-month period under the same environmental conditions and using the same urban wastewater to feed all the systems. The monitoring of CWs during such long periods is not very common in the field of urban wastewater treatment. These treatment systems differed in some design parameters, namely, the presence or absence of plants, their species (Typha angustifolia or Phragmites australis) and the flow configuration and the presence or Science of the Total Environment 430 (2012) 174–183 ⁎ Corresponding author at: Instituto de Diagnóstico Ambiental y Estudios del Agua (IDAEA), CID, CSIC, C/Jordi Girona 18‐26, 08034 Barcelona, Spain. Tel.: +34 93 400 61 00x1305; fax: +34 93 204 59 04. E-mail addresses: maria.hijosa@idaea.csic.es (M. Hijosa-Valsero), rsidm@unileon.es (R. Sidrach-Cardona), ebecm@unileon.es (E. Bécares). 0048-9697/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2012.04.072 Contents lists available at SciVerse ScienceDirect Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv