Skip to main content

Phosphate uptake and translocation in a tropical Canna-based constructed wetland



Considering the problem of eutrophication of the water bodies, phosphate removal from water has emerged as a research of topical interest. The present study aims to investigate the efficacy of Canna lily-based constructed wetland to remove phosphate from wastewater. The translocation of phosphate in plant tissue and its biochemical transformation in sediments is also studied to understand its accumulation and recirculation within the system.


The removal of phosphate stabilized at around 50% in the present study and plant uptake was found to be the major removal mechanism. Average removal was 167 mg/m2 day for total phosphate and 84 mg/m2 day for available phosphate for an initial loading rate of 200 mg/m2 day and 85 mg/m2 day, respectively, at a HRT of 24 h. Most of the phosphate concentrated in above ground tissue of plant and its relative accumulation was maximum in flowers. Fractionation of phosphate in sediments confirmed removal by sediments with an accumulation of apatite phosphate (Ca and Mg bound), but release of non-apatite form (Fe and Al bound).


The study concludes that Canna-based constructed wetland can be an effective tool for phosphate removal from wastewater and sediments particularly under tropical conditions. Regular harvesting of above ground tissue of Canna can result in nutrient export from the system, whereas autochthonous addition may result in recirculation.


Phosphate is a critical parameter to establish the wholesomeness of water in respect of its quality and treatment strategy. Conventional treatment of wastewater removes a significant fraction of suspended impurities and organic carbon, but the removal of nutrients like nitrogen and phosphorus remains limited. Advanced treatment techniques like chemical precipitation, ion exchange, and reverse osmosis are either energy or cost intensive or result in production of secondary sludge (Joshi and Srivastava 2006; Haritash et al. 2015). In such a case, discharge of partially treated wastewater may result in nutrient excess and eutrophication in water bodies (Yadav et al. 2015). Phosphate being an integral component of food and plant waste, fertilizers, detergents etc. is continuously discharged in wastewater, and it reaches to water bodies through various routes leading to a eutrophic state of water body every 10–15 years (Jeppesen et al. 2005). Measures towards restoration of such eutrophic water bodies include chemical application of lime, alum, and ferric salt and introduction of planktons, macrophytes (Gao et al. 2009), and higher organisms to restore a functional biological cycling of phosphorus (Jeppesen et al. 2007). Application of constructed wetlands (CWs) for stabilization of secondarily treated wastewater (Vymazal 2007), and bioretention systems for stormwater, prior to discharge in a water body, has resulted in substantial removal and mobilization of nutrients and restoration of functional nutrient cycling (Sack 2013; Hseih and Davis 2005). Regular harvesting of fish or plants from an under-restoration water body results in export of nutrients, but on-site death and decomposition may lead to recirculation of nutrients in the same environment. Phosphorus being a key limiting nutrient for algal growth in freshwater regulates the biochemical health of a water body (Padma and Nair 2010). Most of the studies report that total phosphorus concentration may not be adequate to determine the ecological risks associated with it. The retention or release of phosphate and its bioavailability are regulated by the chemical state of phosphorus and environmental conditions (Zhang et al. 2012). The release of bound phosphorus to free soluble form, as regulated by pH and redox conditions, is reported in several studies (Kim et al. 2003). Many authors have reported phosphate removal from wastewater by Phragmites, Typha, Eichhornia, etc., but the ultimate treatment/disposal of such plants by composting is difficult. Whereas Phragmites and Typha are hard-stem plants, Eichhornia is an invasive weed and is difficult to harvest. In such a case, Canna lily, a plant with soft tissue, is identified as a potential candidate for phosphate removal without compromising the treatment efficiency, aesthetics, and option of its composting. Some studies on Canna spp. have reported its ability to remove nitrogen and phosphorus efficiently (Polomski et al. 2007), as a result of its higher evapo-transpiration rates against other ornamental plants. The rate of dry weight accumulation was also reported high owing to high rate of nutrient accumulation in plant tissue. Analysis of different plant parts confirmed maximum storage in shoots and roots. Other important factors responsible for nutrient removal are sediments and microorganisms. Gravel-based sediments and plant roots offer suitable substrate for microbial growth. Root exudates in the form of carbohydrates, amino-acids, enzymes, etc. offer oxidizing conditions and support nitrifying conditions during decomposition of organic matter (Gersberg et al. 1986). The chemistry of sediments too compliments the removal of nutrients, especially phosphate. Fragmented limestone waste used as packing material in a phragmites-based CW demonstrated removal efficiency of around 62% for phosphate (Mateus et al. 2012). Calcium and magnesium present in sediments can bind to phosphate to immobilize it temporarily under alkaline conditions, whereas iron and aluminium can dominantly remove phosphate at neutral pH (VanBeusekom and DeJonge 1997). Thus, pH plays an important role in regulating availability of phosphate in an aquatic system. Reducing/acidic conditions may favour resolubilization of phosphate, making a water body prone to eutrophication. Therefore, it is important to monitor the total load, and fractionation of phosphate in a water body to comment upon its trophic status, and to determine the actual efficiency of plants towards removal of phosphate. The present study, therefore, aimed to study the removal of phosphate by plants and sediments, to characterize the different fractions of phosphate in sediments and water, and to quantify the distribution of phosphate in different plant tissues during phytoremediation.


The study was carried out in a pilot-scale vertical-flow constructed wetland cell constructed in the campus of Delhi Technological University during April–May 2013 (summer season). The temperature profile during the period was monitored along with other meteorological conditions (Table 1) to relate the uptake of nutrients to prevalent weather conditions. The CW cell was packed with a 35-cm-thick layer of sand-gravel substrate. The cell had dimensions of length 1.1 m, width 0.8 m, and depth 0.45 m and rectangular tank made up of brick and masonry. The influent was fed from the top, and effluent was collected from an outlet placed opposite to the inlet at the bottom of the tank (Fig. 1). The supporting matrix consisted of sand- and gravel-packed bed with a height of 0.35 m corresponding to a total volume of 0.325 m3. The bulk density of the packing medium was 1.82 g/cm3, and d10, d30, and d60 computed from particle size distribution were 0.42, 2.36, and 9.0 mm, respectively. The supporting matrix was classified as well-graded medium with a value of 21.43 for coefficient of uniformity (C u ) and 1.47 for coefficient of curvature (C c ). The void ratio was determined as 0.77 with porosity of 0.44. The classification of packing medium represented conditions suitable for easy percolation of water and penetration of plant roots. The CW cell was vegetated with 90 (stem count) healthy plants of Canna lily with an average shoot length of 43.2 cm. The initial concentration of phosphate in different tissues of plant (roots, stem, leaves, and flowers) was determined to compare it with the final plant-tissue phosphate level to confirm uptake by plants and subsequent accumulation as a process of removal. The CW cell was initially flushed with water at a hydraulic loading rate (HLR) of 400 L/day for 5 days, and later, the plants were irrigated, on daily basis, with HLR of 30 L/day of simulated wastewater. Synthetic wastewater was prepared by dissolving 21.74 mg of potassium-dihydrogen-orthophosphate in groundwater to obtain a concentration of 5.0 mg/L of PO4 3−-P. The Canna-based CW cell was fed with this synthetic wastewater on daily basis (for 30 days) maintaining the hydraulic retention time (HRT) of 24 h on plug-flow basis. The concentration of available (AP) and total phosphate (TP) was determined regularly (for 30 days) in the influent and effluent using standard methods (APHA) (Eaton et al. 1995). Phosphate present in dissolved form was determined as available phosphate, and the sample was acid-digested for extraction of TP. Different fractions of phosphate, i.e. apatite phosphate (bound to Ca and Mg), non-apatite inorganic phosphate (NAIP bound to Fe, Al, and Mn), organic phosphate (OP-bound to biomass), were determined in the sediments, in triplicate (n = 3), before initiating the experiments and after the completion using standard prescribed procedure (Williams et al. 1980). The total phosphate accumulated in different plant tissues was also determined initially and finally in five (n = 5) samples of each tissue. All the experiments were carried out using analytical (AR) grade chemicals and double glass distilled water in triplicates. The glassware and plasticware used in the experiments were soaked in 0.3% HCl and rinsed with deionized water prior to use. The meteorological observations during the study were obtained from the weather station of Delhi Pollution Control Committee (DPCC). Statistical analysis of the obtained results was done over MS-Excel software to calculate mean, standard deviation, removal rate, etc.; to plot the normalized curve for removal efficiency; and to plot the graphs.

Table 1 Concentration of total and available phosphate and its removal rate in Canna-based constructed wetland
Fig. 1
figure 1

Schematic description of configuration of constructed wetland cell used in the study

Results and discussion

Based on the values of ambient temperature and solar insolation during the study, it was observed that the study area represents tropical to sub-tropical weather conditions which favour optimum activity and growth of vegetation. Average ambient temperature was 30 °C ranging from 24.5 °C (minimum) to 40.2 °C (maximum) during the study. Such conditions promote good rate of uptake of nutrients and hence maximum efficiency towards removal. In order to highlight the role of vegetation and sediments in removal of phosphate from wastewater, this paper discusses the uptake, transformation, and translocation of different chemical forms of phosphate in detail.

Phosphate removal from wastewater

Significant removal of phosphate from wastewater in a Canna-based constructed wetland was observed in the present study. The percent removal of TP was slightly higher than the removal of available phosphate (AP) (Table 1). Average removal of TP was around 68% as against 57% for AP. Percent TP removal was initially higher and went decreasing to stabilize at around 55% in the later stage (after 20 days) (Fig. 2a). On the other hand, the percent removal of AP ranged from 39 to 89% with an average value of 57%. It was low for first 2 days but increased to about 86% on the third day and fluctuated around the same value for 10 days. Later, it decreased and stabilized around 45% by the 20th day (Fig. 2b). Chemical transformation of bound phosphate to orthophosphate (AP) by microorganisms and enzymes secreted by plant roots results in higher removal efficiency for TP. Percent removal of P has been reported in the similar range (24.1–61.55) in another study on nutrient removal from wastewater by 15 wetland species including Canna (Jiang et al. 2011). In the present study, average removal rate of TP (167 mg/m2 day) was higher than that of AP (84 mg/m2 day) as against the loading rate of 200 mg/m2 day for TP and 85 mg/m2 day for AP. In a similar study on treating dairy industry wastewater using a Canna-based CW (DeBusk et al. 1995), removal rate of 173 mg/m2 day has been reported. The present study reports slightly higher removal rates owing to the climatic differences and chemical composition of wastewater in the study. The major mechanisms for phosphate removal are reported as uptake by plants and adsorption over supporting matrix in the study and its efficiency is dependent on plant-uptake rate and sediment binding. In the present study, initially higher percent removal of phosphate could be attributed to higher activity of plants and sediments in the initial phase as reported in other studies too (Haritash et al. 2015). Later, the decrease in removal efficiency is observed in wetlands and bio-retention systems with saturation of sediments (Kadlec and Knight 1996).

Fig. 2
figure 2

Percent removal (curve represents normalized removal) of total phosphate (a), and available phosphate (b) from wastewater in Canna-based constructed wetland

Phosphate in plant tissue

Since removal of phosphate depends on plant uptake, it is important to study the accumulation of phosphate in various plant tissues. It was determined in various parts (roots, shoots, stem, and flowers) of Canna to ascertain its point of accumulation and investigate the possibility of recirculation. Maximum increase in phosphate concentration was observed in leaves, followed by flowers, roots, and stem respectively. The tissue phosphate concentration increased to about 1.7, 1.2, 22, and 4.2 times that of initial phosphate concentration in roots, shoots, leaves, and flowers, respectively (Fig. 3). It confirms that phosphate is taken up by the plants and is accumulated throughout the tissue via metabolism. Similar results have been reported by Polomski et al. (2007) in a study to remove nutrients from nursery run-off in a lab-scale gravel-based system employing seven garden plants individually. The study reported 40% more accumulation of P in shoots than roots at an influent level of 6.77 mg/L phosphate. In a similar study (Jiang et al. 2011), 15 wetland plants were observed to accumulate higher concentration of phosphate in shoots than roots with maximum accumulation by Canna among the plants. If relative fraction of initial and final phosphorus in different tissues is compared, flowers accumulated relatively more concentration, and it would result in recirculation of phosphate in the form of organic phosphate (OP) with the fall of flowers into the wetland system. Observations similar to the fact have been reported by Menon and Holland (2014) in a vegetated constructed wetland. In the present study, the higher concentration and overall accumulation of phosphate is observed in stem, leaves, and flowers (above ground tissue), and its recirculation via leaf fall, plant death, and degradation can be avoided by harvesting the above ground tissue at regular intervals, thus facilitating nutrient export from the wetland. Since the nutrients taken up by the plant take part in growth and metabolism, it is important to monitor the physical growth of wetland plants. In the present study, the root length, shoot length, and plant density were found to increase by a factor of 3.69, 3.67, and 2.04, respectively. The shoot-root ratio remained almost the same throughout the study indicating that there was no significant effect over the plant health. It stresses on application of Canna-based wetland systems for the removal of nutrients from wastewater.

Fig. 3
figure 3

Average concentration of phosphorus (mgP/g) in different plant tissues of Canna before and after the treatment (n=5)

Phosphate in sediments

Since adsorption of phosphate by sediments is a major removal pathway, its removal and fractionation was studied in the supporting matrix before initiating the experiments and after the completion of study. An increase in the concentration of total phosphate (TP) was observed from 5.14 to 5.88 mg/g confirming to its adsorption by sediments (Fig. 4). Since the root system of Canna is strong and extensive, it removes phosphate from sediments too. As a result, the increase of phosphorus in sediments is not as significant. This strengthens the idea that the removal of P in Canna-based wetland is predominantly by plant uptake, and sediments too play a role in it. Apart from it, the chemical composition and mineral profile of sediments may also be the reason for relatively less adsorption over sediments. The fraction of organic phosphate (OP) was observed to increase from 1.0 to 1.37 mg/g indicating to the role of microorganisms and organic matter in phosphate removal via biomass accumulation. Some addition of biomass as a result of leaf fall or dead tissue of old roots may also be the reason for an increase of OP in sediments. The concentration of inorganic phosphate (IP) too increased in sediments indicating chemical binding/precipitation of phosphate with its binding to Ca and Mg (AP) or Fe, Al, and Mn (NAIP). The Ca and Mg bound phosphate (AP) increased during the study indicating the binding of phosphate to Ca and Mg present in sediments. Similar observations have been presented by Gu and Dreschel (2008) in a constructed wetland in Florida. The NAIP concentration decreased during the study from 2.37 to 0.76 mg/g. It may be attributed to release of phosphate bound to Fe, Al, and Mn in sediments during higher oxygen demand of wetland for respiration. During high oxygen demand, ferric bound phosphate is released to water with reduction of Fe3+ to Fe2+ because iron is an alternate terminal electron acceptor during deficiency of dissolved oxygen (Kortmann and Rich 1994). Studies have confirmed that AP will be trapped in sediments for a longer duration and is regarded as non-available. It will not be released to water easily. On the other hand, NAIP can be released to water column easily particularly during summers depending on the redox conditions present (Zhang et al. 2012). Apart from it, the phosphorus concentration in sediments in rooted macrophyte-based system decreases since the plants use sediment too as source of phosphorus (Carignan and Kalff 1980). It further strengthens the objective of using a Canna-based wetland since it not only removes phosphate from wastewater but also removes a fraction of NAIP. Based on the results obtained, it was observed that major plant uptake of phosphate is from wastewater and sediment NAIP during oxygen-stressed conditions.

Fig. 4
figure 4

Average concentration (mg/g) of different fractions of phosphate in sediments before and after the treatment (n=3)


Based on the observations of the present study, Canna lily-based constructed wetland is found to be an effective treatment option for phosphate from wastewater. The removal of sediment bound phosphate is also facilitated by Canna, but the available phosphate accumulates in sediments which may be released in hypolimnion if pH is acidic or conditions are reducing. Apart from it, though phosphate is translocated to all the tissues, its relative accumulation is observed to be more in flowers which may cause recirculation. Most of the fraction of removed phosphate is accumulated in above ground tissue, and hence, regular harvesting of aerial tissue can result in nutrient export. Considering the nutrient accumulation of harvested tissue, it may be used as raw material for composting if ultimate disposal is aimed at. Within the sediments, AP is accumulated whereas NAIP is released. It confirms easy release of Fe and Al bound phosphate to water though the redox conditions play a crucial role in it. The study concludes that Canna lily removes phosphorus not only from wastewater but also from sediments. Hence, Canna lily-based wetlands are effective in phosphate removal from wastewater especially under tropical conditions.


  • Carignan R, Kalff J (1980) Phosphorus sources for aquatic weeds: water or sediments. Science 207:987–989

    Article  CAS  Google Scholar 

  • Debusk TA, Peterson JE, Rameshreddy K (1995) Use of terrestrial and aquatic plants for removing phosphorus from dairy wastewaters. Ecol Eng 5:371–390

    Article  Google Scholar 

  • Eaton AD, Clesceri LS, Greenberg AE (1995) Standard methods for the examination of water and wastewater, 19th edn. American Public Health Association (APHA), New York

    Google Scholar 

  • Gao J, Xiong Z, Zhang J, Zhang W, Mba FO (2009) Phosphorus removal from water of eutrophic lake Donghu by five submerged macrophytes. Desalination 242:193–204

    Article  CAS  Google Scholar 

  • Gersberg RM, Elkins BV, Lyon SR, Goldman CR (1986) Role of aquatic plants in wastewater treatment in artificial wetlands. Water Res 20:363–368

    Article  CAS  Google Scholar 

  • Gu B, Dreschel T (2008) Effects of plant community and phosphorus loading rate on constructed wetland performance in Florida, USA. Wetlands 28(1):81–91

    Article  Google Scholar 

  • Haritash AK, Sharma A, Behal K (2015) The potential of Canna lily for wastewater treatment under Indian conditions. Int J Phytoremediation 17(10):999–1004

    Article  CAS  Google Scholar 

  • Hseih C, Davis AP (2005) Evaluation and optimization of bioretention media for treatment of urban storm water runoff. J Environ Eng 131:1521–1531

    Article  Google Scholar 

  • Jeppesen E, Meerhoff M, Jacobsen BA, Hansen RS, Sondergaard M, Jensen JP, Lauridsen TL, Mazzeo N, Branco CWC (2007) Restoration of shallow lakes by nutrient control and manipulation—the successful strategy varies with lake size and climate. Hydrobiol 581(1):269–285

    Article  CAS  Google Scholar 

  • Jeppesen E, Sondergaard M, Jensen JP, Havens K, Anneville O, Carvalho L, Covene MF, Deneke R, Dokulil MT, Foy B, Gerdeaux D, Hampton SE, Hilt S, Kangur K, Kohler J, Lammens E, Lauridsen TL, Manca M, Miracle R, Moss B, Noges P, Perrson G, Phillips G, Portielje R, Romo S, Schelske CL, Straile D, Tatrai I, Willen E, Winder M (2005) Lakes’ response to reduced nutrient loading—an analysis of contemporary data from 35 European and North American long term studies. Freshw Biol 50:1747–1771

    Article  CAS  Google Scholar 

  • Jiang FY, Chen X, Luo AC (2011) A comparative study on growth and nitrogen and phosphorus uptake characteristics of 15 wetland species. Chem Ecol 27(3):263–272

    Article  CAS  Google Scholar 

  • Joshi M, Srivastava RK (2006) Removal of iron from waste water by using non-conventional ecofriendly material. Pollut Res 25(1):185–188

    CAS  Google Scholar 

  • Kadlec RH, Knight RL (1996) Treatment wetlands. Lewis Publishers, Boca Raton

    Google Scholar 

  • Kim LH, Choi E, Stenstrom MK (2003) Sediment characteristics, phosphorus types and phosphorus release rates between river and lake sediments. Chemosphere 50:53–61

    Article  CAS  Google Scholar 

  • Kortmann RW, Rich PH (1994) Lake ecosystem energetics: the missing management link. Lake Reservoir Manage 8(2):77–97

    Article  Google Scholar 

  • Mateus DMR, Vaz MMN, Pinho HJO (2012) Fragmented limestone wastes as a constructed wetland substrate for phosphorus removal. Ecol Eng 41:65–69

    Article  Google Scholar 

  • Menon R, Holland M (2014) Phosphorus release due to decomposition of wetland plants. Wetlands 34:1191–1196

    Article  Google Scholar 

  • Padma P, Nair SM (2010) Assessing the phosphorus fractionation in a tropical river-estuarine system of South India. Chem Ecol 26(6):453–466

    Article  CAS  Google Scholar 

  • Polomski FR, Bielenberg DG, Whitwell T, Taylor MD, Bridges WC, Klaine SJ (2007) Nutrient recovery by seven aquatic garden plants in laboratory-scale subsurface-constructed wetland. Hortic Sci 42(7):1674–1680

    CAS  Google Scholar 

  • Sack C (2013) Landscape architecture and novel ecosystems: ecological restoration in an expanded field. Ecol Process 2(35):1–9

    Google Scholar 

  • Vanbeusekom JEE, Dejonge VN (1997) Transformation of phosphorus in the Wadden sea: apatite formation. Ger J Hydrogr 49:297–305

    Google Scholar 

  • Vymazal J (2007) Removal of nutrients in various types of constructed wetlands. Sci Total Environ 380:48–65

    Article  CAS  Google Scholar 

  • Williams JDH, Mayer T, Nriagu JO (1980) Extractability of phosphorus from phosphate minerals common in soils and sediments. Soil Sci Soc Am J 44:462–465

    Article  CAS  Google Scholar 

  • Yadav A, Kataria A, Singh K, Mathur K, Goswami S, Haritash AK (2015) Seasonal assessment of trophic status of a palustrine water body. Int J Eng Res Technol 4(3):37–40

    Google Scholar 

  • Zhang B, Fang F, Guo J, Chen Y, Li Z, Guo S (2012) Phosphorus fractions and phosphate sorption-release characteristics relevant to the soil composition of water-level-fluctuating zone of Three Gorges Reservoir. Ecol Eng 40:153–159

    Article  Google Scholar 

Download references

Authors’ contributions

AKH designed the study, participated in the experimental work, and drafted the manuscript. SD carried out the analysis of phosphate in wastewater and plant. AS did the analysis of sediments and helped prepare the wetland cell for the study. All authors have read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Anil Kumar Haritash.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Haritash, A.K., Dutta, S. & Sharma, A. Phosphate uptake and translocation in a tropical Canna-based constructed wetland. Ecol Process 6, 12 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: