Environmental Pollution 144 (2006) 967e975 www.elsevier.com/locate/envpol
Accumulation of nutrients and heavy metals in Phragmites australis (Cav.) Trin. ex Steudel and Bolboschoenus maritimus (L.) Palla in a constructed wetland of the Venice lagoon watershed* Claudia Bragato a, Hans Brix b, Mario Malagoli a,* b
a Department of Agricultural Biotechnology, University of Padova, Agripolis, 35020 Legnaro (PD), Italy ˚ rhus C, Denmark Department of Biological Sciences, Plant Biology, University of Aarhus, Ole Worms Alle´, Building 1135, DK-8000 A
Received 26 September 2005; received in revised form 20 December 2005; accepted 26 January 2006
Heavy metal shoot concentration in Phragmites australis and Bolboschoenus maritimus increased significantly at the end of the growing season. Abstract A recently constructed wetland, located in the Venice lagoon watershed, was monitored to investigate growth dynamics, nutrient and heavy metal shoot accumulation of the two dominating macrophytes: Phragmites australis and Bolboschoenus maritimus. Investigations were conducted over a vegetative season at three locations with different distance to the inlet point to assess effects on vegetation. The distance from the inlet did not affect either shoot biomass or nutrients (N, P, K and Na) and heavy metals (Cr, Ni, Cu and Zn) shoot content. With the exception of Na, nutrient and heavy metal concentrations were higher in shoots of P. australis than in B. maritimus. Heavy metal concentration in the incoming water and in the soil was not correlated to the plant content of either species. Shoot heavy metal concentrations were similar to those reported in the current literature, but accumulation generally increased towards the end of the growing season. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Bulrush; Common reed; Heavy metals; Nutrients; Phytoremediation
1. Introduction The utilisation of wetland areas as natural filters for the abatement of pollutants transported by water in rivers or lakes is considered to be an effective, low-cost, cleanup option to ameliorate the quality of surface waters. Indeed, wetlands have been extensively utilised in last the decades to clean pollutant water almost all over the world (Gopal, 2003). The vegetation covering the wetland areas plays an important role in sequestering large quantitative of nutrients (Kadlec and Knight, 1996; Knight, 1997; Cronk and Fennessy, 2001) * This paper is dedicated to the memory of the late Professor Giuseppe Bendoricchio. * Corresponding author. Tel.: þ39 049 827 2908; fax: þ39 049 827 2929. E-mail address:
[email protected] (M. Malagoli).
0269-7491/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2006.01.046
and metals (Karpiscak et al., 2001; Mays and Edwards, 2001; Stoltz and Greger, 2002; Baldantoni et al., 2004) from the environment by storing them in the roots and/or shoots. Wetland plants have high remediation potential for macronutrients because of their general fast growth and high biomass production. Wetland plants also take up heavy metals from the environment but tend mainly to accumulate these in the belowground tissues (Peverly et al., 1995; Stoltz and Greger, 2002; Weis et al., 2004). Restriction of shoot translocation is believed to be the strategy of metal tolerance for non-hyperaccumulators. In so doing, the plants avoid the potential effects of high metal concentrations on the photosynthetic tissue. It has been shown that heavy metals accumulation is responsible for the decrease of total chlorophyll concentration and negatively affects the Chl a/Chl b ratio (Abdel-Basset et al., 1995; Manios et al., 2003). However, the capacity to accumulate
C. Bragato et al. / Environmental Pollution 144 (2006) 967e975
968
heavy metals in the aboveground plant tissues represents a central point for the suitability of the plants for metals phytoextraction (Salt and Kra¨mer, 2000). The amount of metals accumulated in the aerial part may vary during the growing season as a consequence of the inherent growth dynamics of the plant, as well as in response to variations in the heavy metals levels and availability in the surrounding water and soil (Larsen and Schierup, 1981; Schierup and Larsen, 1981; Hardej and Ozimek, 2002). It is therefore important to evaluate the seasonal and spatial variations in plant accumulation in wetland systems in order to assess the potential for nutrient and metal removal by plant uptake and harvesting. In the present study we monitored the seasonal and spatial variation of plant shoot biomass, leaf chlorophyll concentration, and shoot content of nutrients and heavy metals of two emergent macrophytes, Phragmites australis (Cav.) Trin. ex Steudel and Bolboschoenus maritimus (L.) Palla, in a constructed wetland of the Venice lagoon watershed. 2. Materials and methods 2.1. Site description Ca di Mezzo wetland is located in the Venice lagoon watershed (Codevigo, Padova, Northeast Italy) between the River Bacchiglione and the Canal Morto. The free water surface wetland was constructed in 2000 in order to reduce the amount of pollutants in the water discharged into Venice lagoon. The wetland has a total area of 29 ha and is divided into three basins (Fig. 1). About twothird of the wetland is covered by water which has a mean depth of 0.8 m. The mean inlet flow is 0.4 m3 s1 and the mean residence time of the water in the wetland is estimated to be 2.5e3 days. Phragmites australis (Cav.)
Trin. ex Steudel was the only plant species originally planted at a density of 1 plant m2. Although at the time of the study Phragmites was the predominant plant species in the marsh, other macrophytes had colonised the area, such as Bolboschoenus maritimus (L.) Palla (syn. Scirpus maritimus), Typha latifolia (L.), Cyperus glomeratus (L.), Aster squamatus (Spreng), Echinochloa crus-galli (L.), and Juncus spp.
2.2. Sampling procedure The sampling was carried out between June and December 2001 in the first basin of the wetland, only (Fig. 1). Six shoots of the two most dominating plant species, P. australis and B. maritimus, were randomly collected from three sampling sites of the basin, near the inlet (In), in the middle (Mid) and near the outlet (Out) (Fig. 1). It was impossible to sample B. maritimus plants in the area near the inlet, and P. australis did not appear at the Mid site until after July. The shoots were cut off 3 cm above ground level, gently cleaned with paper towel and then quickly transferred in plastic bags to the laboratory for analysis. Fresh and dry (80 C for 48 h) weights were measured for each sample. Some fresh leaf samples were immediately analysed for pigment contents using the procedure described below, while the remaining fresh plant material was frozen in liquid N2 and then stored at 80 C. Water samples were collected every month and kept at 4 C in 1-L clean polyethylene bottles until analysis. Sediment samples (approx. 20 cm depth) were collected by a 15cm-diameter sediment corer from each sampling point in June, September and December.
2.3. Photosynthetic pigment analysis Fresh leaf samples (approx. 500 mg) were extracted with ethanol (96% v/v) and the contents of chlorophyll a (Chl a), chlorophyll b (Chl b) and carotenoids were analysed spectrophotometrically (UVKON 922, Kontron) following the procedure described by Welburn and Lichtenthaler (1984).
Fig. 1. Planimetry of Ca di Mezzo wetland, with the three sampling locations (In, Mid, Out) of the first basin.
C. Bragato et al. / Environmental Pollution 144 (2006) 967e975
2.4. Nutrient and heavy metal analysis Plant total nitrogen (TN) concentration was determined after digestion of powdered dry shoot material (500 mg) in a Digesdahl flask (HACH) with 4 ml H2SO4 (98% v/v). The digest was amended with 10 ml of H2O2 (40% v/v), brought to a final volume of 100 ml with deionised water and adjusted to neutral pH with 6 M NaOH. After addition of 2 ml of Nessler’s reagent (Titolchimica) the absorbance was read at 425 nm using a spectrophotometer (UVKON 922, Kontron). The nitrate (NO 3 ) content was determined after extraction of 100 mg of powdered dry plant material in 10 ml hot water at 85 C for 2 h. After centrifugation (Centrikon T-124, Kontron), the supernatant was removed, diluted, filtered (0.22 mm Millipore) and analysed by HPLC using IonPac AS14 column/AG14 guard column (Dionex), with NaHCO3/Na2CO3 eluent (1 mM:3.5 mM) and a flow rate of 0.9 ml min1. Phosphorus, potassium, sodium and metals were analysed after mineralisation of 400 mg dry shoot material in a microwave oven (Milestone Ethos 1600) with 5 ml of nitric acid (69% v/v), 5 ml deionised water and 2 ml H2O2 (30% v/v). The digest was made to 25 ml final volume with deionised water, filtered (0.45 mm, Millipore) and then analysed for P, K, Na and heavy metals using ICP/OES (Spectro CirosCCD). Water samples were analysed, after filtration (0.22 mm, Millipore), for total 3 nitrogen (TKN), nitrate (NO 3 ) and phosphate (PO4 ) according to CNR-IRSA (1994), and for heavy metals, using ICP/OES (Spectro CirosCCD). Dried sediment samples were passed through a 1-mm-diameter sieve. Approximately 100 mg dry sediment was digested with 3 ml HNO3 (69% v/v) and 2 ml HCl (37% v/v) in a microwave oven (Milestone Ethos 1600). After mineralisation, the samples were diluted, filtered and analysed for heavy metals by ICP/OES (Spectro CirosCCD).
969
metals (Cr, Mn, Fe, Co, Ni, Cu, Zn, Cd, Mo, Pb) but only Cr, Cu, Ni and Zn were detected. The mean concentrations of the four heavy metals were relatively constant, with the exception of Zn in August (63.5 mg/L).
3.2. Plant growth dynamics No significant differences in the shoot biomass of P. australis were found among locations during the sampling period (Fig. 2a). Shoot biomass was highest in July and September and decreased in the following months at all sites. Shoot biomass of B. maritimus declined during the season, but no differences were observed between plants grown at Mid and Out locations (Fig. 2b). The shoot dry weight/fresh weight ratio (DW/FW ratio) of P. australis ranged from 0.27 in June near the outlet (Out) to 0.61 in October at the inlet (In) location, while B. maritimus
2.5. Data analysis The KruskaleWallis non-parametric test was used to evaluate the significance of differences between groups with a level of significance set to p < 0.05. When the KruskaleWallis test gave a significant result a pairwise ManneWhitney U-test was carried out to evaluate differences between each pair.
3. Results 3.1. Incoming water quality The mean total nitrogen content in the incoming water for the whole period was about 2.11 mg L1 with a maximum peak (3.05 mg L1) measured in September (Table 1). The 1 concentration of NO 3 -N ranged between 0.71 and 1.55 mg L , 3 and PO4 -P concentration was low, with a mean value of 0.04 mg L1. Incoming water was also analysed for heavy Table 1 3 Total nitrogen (TN), nitrate (NO 3 -N), phosphate (PO4 -P) and heavy metals (Cr, Cu, Ni and Zn) concentrations of the incoming water during the period JuneeDecember 2001 Month
TN
NO 3 -N
PO3 4 -P
Cr
Cu
Ni
Zn
June July August September October November December
1.80 2.25 1.62 3.05 1.67 2.20 2.15
1.05 1.50 1.15 0.71 0.77 1.45 1.55
0.03 0.08 0.03 0.03 0.02 0.04 0.04
