Dw

Xa

b

a

ARRAA

KANPUNH

1

ci2Kt(e

3

h0

Ecological Engineering 101 (2017) 107–119

Contents lists available at ScienceDirect

Ecological Engineering

jo ur nal home p age: www.elsev ier .com/ locate /eco leng

ominating aquatic macrophytes for the removal of nutrients fromaterways of the Indian River Lagoon basin, South Florida, USA

iaohong Zhoua,b, Zhenli Hea,∗, Kimberly D. Jonesa, Liguang Lia, Peter J. Stoffellaa

Indian River Research and Education Center, Institute of Food and Agricultural Sciences, University of Florida, Fort Pierce, FL, 34945-3138, USASchool of the Environment and Safety Engineering, Jiangsu University, Zhenjiang, Jiangsu, 212013, China

r t i c l e i n f o

rticle history:eceived 3 May 2016eceived in revised form 5 January 2017ccepted 5 January 2017vailable online 22 January 2017

eywords:quatic plantsitrogenhosphorusptakeutrients storagearvesting

a b s t r a c t

Aquatic macrophytes have an important role in cleaning eutrophic runoff water from agriculture andurban areas. However, minimal information is available regarding the quantity of nutrients and/or pol-lutant they can remove each year from the water and sediments in waterways. This study investigatedthe biomass productivity of eight dominating aquatic plant species, the concentrations of nutrients inplant tissues and their capacity to absorb and store nutrients. Samples of plant, water, and sedimentwere collected from October 30 to November 19, 2014 at 22 representative sites in the waterways ofthe Indian River Lagoon basin, South Florida, USA. The biomass yield of the plant species decreasedin the order: cattail (Typha orientalis)> pickerelweed (Pontederia cordata)> water lettuce (Pistia stra-tiotes)> hydrilla (Hydrilla verticillata)> maidencane (Panicum hemitomon)> spatterdock (Nuphar advena)>pondweed (Potamogeton spp.)> salvinia (Salvinia spp.). Cattail had the highest biomass productivity, butonly a small part (35.8%) of the total biomass productivity was harvestable, whereas, water lettuce andhydrilla were mostly harvestable and could contribute almost 100% to harvestable biomass. Concentra-tion of nutrients in plant varied significantly among the eight plant species and with the sampling sites,suggesting that in addition to genetic differences, physicochemical parameters of overlying water andsurface sediment influenced uptake of nutrients by the plants. Among the eight plant species, cattailhad the highest total nitrogen (N) (23.4 g N m−2) and phosphorus (P) (1.59 g P m−2) storage but waterlettuce and hydrilla exhibited the highest total N (14.6 g N m−2) and P (1.04 g P m−2) net storage capac-ity in this survey. In addition, the highest N and P uptake per year occurred with water lettuce and

−1 −1 −1 −1

hydrilla, with the peak of 146 kg N ha y and 10.4 kg P ha y , respectively. The results also indicatethat multiple harvests of biomass are necessary to realize the removal potential of nutrients/pollutant bythe aquatic plants, as the harvesting (cutting) practice can enhance plant growth and prevent release ofnutrients/pollutant back into water from plant residue decomposition, which are estimated at 1.87 × 103

to 72.4 × 103 kg N and 0.07 × 103 to 4.80 × 103 kg P per year in the IRL basin.© 2017 Elsevier B.V. All rights reserved.

. Introduction

Water pollution, caused by excessive input of N and P from agri-ulture, urbanization, and industrial discharge has become a globalssue (Sherwood and Qualls, 2001; Dodds, 2007; Yang et al., 2008a,008b; Schindler and Hecky, 2009; Liu et al., 2011; Dai et al., 2012;earney and Zhu, 2012). Therefore, removal of N and P is an effec-

ive method for mitigating the serious situation of eutrophicationLiu et al., 2011; Li et al., 2015). Aquatic macrophytes are consid-red to be the crucial biological components in the fresh surface

∗ Corresponding author at: 2199 South Rock Road, Fort Pierce, Florida, 34945-138, USA.

E-mail address: zhe@ufl.edu (Z. He).

ttp://dx.doi.org/10.1016/j.ecoleng.2017.01.006925-8574/© 2017 Elsevier B.V. All rights reserved.

water ecosystems and have important roles in water purification(Reddy, 1983; Lu et al., 2010, 2011; Wu et al., 2011; Zhu et al.,2011; Zhao et al., 2012a, 2012b; Li et al., 2015). Aquatic plants arewidely used for ecological remediation of eutrophic lake, pollutedriver, and other water bodies. They directly absorb N and P fromeutrophic water for growth and reproduction (Ellis et al., 1994).Ultimately, the biomass is harvested to remove the N and P from thewater systems. Therefore, plant uptake is an important mechanismof water remediation (Gottschall et al., 2007). However, the capac-ity of plant uptake is related to nutrient concentration in plant andbiomass yield. Potential of plants for removing nutrients/pollutant

from water systems varies with species and is affected by envi-ronmental factors such as climate and hydrological conditions andnutrient enrichment in water and sediment (Jampeetong and Brix,2009; Zhang et al., 2009).

1 nginee

diwOtesitbrt2mrocnwsiaw(e

racPwwdrtwa

tt

2

2

otFeF

2

at

c2lbsc

08 X. Zhou et al. / Ecological E

In Florida, owing to heavy irrigation and/or frequent rainfall,issolved nutrients such as N and P from fertilized soils, are read-

ly lost into waterways which results in eutrophication of surfaceater systems including the Indian River Lagoon (IRL) and Lakekeechobee (He et al., 2006; Yang et al., 2008b, 2009). Eutrophica-

ion of IRL has been a major public concern in the last decade (Lut al., 2010). Agricultural practices and urban development are con-idered as major nonpoint sources of pollution. Runoff water thats laden with nutrients delivers a large amount of nutrients (N, P)o the IRL through waterways of farm ditches and canals in the IRLasin. Previous studies indicated that concentrations of N and P inunoff water were considerably attenuated in the waterways dueo biological assimilation (He et al., 2006; Yang et al., 2008a, 2008b,009). In addition to microbial fixation, nutrient uptake by plantsay have significantly contributed to the purification of eutrophic

unoff water, as these waterways are often fully-grown with vari-us ecotypes of aquatic plants including hydrilla, water lettuce, andattail, and other aquatic plants, due to warm weather and adequateutrient supply. Aquatic vegetation takes up nutrients from theater, rendering the waterways like a natural phytoremediation

ystem. However, to date, minimal information is available regard-ng the amounts of nutrients/pollutant that can be removed byquatic macrophytes and their potential impact on water quality,hich is critical for the development of best management practices

BMPs) to minimize impact of agriculture and urbanization on thenvironment, particularly water quality.

The objectives of this study were to evaluate the potential ofepresentative aquatic macrophytes in the removal of nutrients (Nnd P) from waterways in the IRL basin, South Florida, USA. The spe-ific objectives were to: 1) determine the concentrations of N and

in plant of dominant aquatic macrophytes and their correlationsith N and P concentrations in water and sediment in the water-ays of the IRL basin; 2) determine the annual biomass yield of theominant aquatic plant species so that the amounts of nutrientsemoved by each plant species can be quantified; and 3) evaluatehe potential of the aquatic plants in removing N and P from water-ays of the IRL basin on the condition that plant biomass could be

dequately harvested.The information is expected to improve our understanding of

he aquatic plants’ contribution to water quality improvement ando facilitate the development of BMPs in the Indian River area.

. Materials and methods

.1. Sampling sites and aquatic plants survey

Field surveys were conducted from August 10 to October 20f 2014 to determine species distribution and composition pat-erns of aquatic macrophytes in waterways of the IRL basin, Southlorida, USA. Based on survey results, 22 representative sites werestablished for sampling plant, water and sediments (Table 1 andigs. 1 and 2).

.2. Sample collection and pre-treatments

Triplicate samples of the overlying water, surface sedimentsnd macrophytes were collected from each site from October 30o November 19 of 2014.

Overlying water was collected and stored into 500 mL pre-leaned polyethylene bottle. Surface sediment samples (from 0 to0 cm depth) were collected using a grab sampler from the same

ocation of the water samples and were sealed in plastic ziplockags. Meanwhile, triplicate 0.25–0.30 m2 quadrats of each plantpecies were collected from the same site. Percentage of surfaceoverage by each dominant plant species was also estimated by

ring 101 (2017) 107–119

visual observation. All the samples were placed in iced chest to thelaboratory within 4 h. In the laboratory, pH and electricity conduc-tivity (EC) of water samples were immediately determined usinga pH/ion/conductivity meter (Model 220, Denver Instrument Inc.,CO, USA). Subsamples of the water were stored at −20 ◦C for furtheranalyses. Sediment samples were air-dried, ground, sieved througha 1-mm sieve and stored at 4 ◦C for further analyses. Plants werewashed thoroughly with tap water to remove any adhered impu-rities. Pickerelweed (Pontederia cordata) and spatterdock (Nupharadvena) were separated into roots (plus rhizomes), stems, leavesand flower; maidencane (Panicum hemitomon) was separated intoroots, stems and leaves; cattail (Typha orientalis) was separated intoroots (plus rhizomes) and leaves; water lettuce (Pistia stratiotes)was separated into roots and leaves (plus stems); and hydrilla(Hydrilla verticillata), pondweed (Potamogeton spp.) and salvinia(Salvinia spp.) were kept as whole plant. All plant tissue sam-ples were oven dried at 70 ◦C for 7 days to constant weight, anddry biomass was recorded. The oven-dried plant samples werepowdered <1 mm prior to chemical analyses. All samples were pre-treated in three replicates.

2.3. Physicochemical analysis

Water samples were filtered through a 0.45 �m membrane fil-ter for measuring the concentrations of dissolved organic carbon(DOC), total dissolved P (DP), ortho-phosphate (PO4

3−-P), ammo-nia (NH4

+-N), nitrate (NO3−-N), and total Kjeldahl N (TKN). Both

unfiltered and filtered water samples were digested for TKN byacidified cupric sulfate and potassium sulfate, and the concentra-tions of NH4

+-N and NO3−-N in water and digested samples were

determined using a discrete auto-analyzer (EasyChem, Systea Sci-entific, IL, USA) (US EPA 350.1) Total N was calculated as the sum ofTKN and NO3

−-N in the unfiltered samples; particulate N (PN) wascalculated as the difference in total N between the unfiltered andfiltered samples; organic N (ON) was calculated as the differencebetween TKN and NH4

+-N in the unfiltered samples; dissolved N(DN) was calculated as the sum of TKN and NO3

—N in the filteredsamples. Total P (TP) in the filtered and unfiltered water samples isdetermined by the molybdenum-blue method using U-3010 Spec-trophotometer after digestion with acidified ammonium persulfate(US EPA 365.1). Concentration of reactive P in water samples wasdetermined with the colorimetric method and total dissolved P inwater was calculated by the difference in total P between the fil-tered and unfiltered water samples. The concentration of DOC wasdetermined using Liquid TOC analyzer (liquid TOC trace, ElementalAnalysensystem GmbH, Hanau, Germany). Triplicate samples wereperformed for each measurement.

Sediment pH and EC were determined in water at a solid:water ratio of 1:1 and 2:1 using a pH/ion/conductivity meter(Model 220, Denver Instrument Inc., CO, USA), respectively. Totalorganic C and total N of sediment samples were determinedusing C/N-Analyzer (vario MAX CN elementar AnalysensystemeGmbH, Hanau, Germany). NH4

+-N and NO3−-N in the sediment

were extracted with 2M KCl and then measured using a discreteauto-analyzer (EasyChem, Systea Scientific, IL, USA). Extractable P(Exc-P) in the sediments samples was extracted with Mehlich 3reagent according to previous studies (He et al., 2006; Yang et al.,2013) and P concentration in the extracts was determined usingan inductively coupled plasma optical emission spectrometer (ICP-OES, Ultima, JY Horiba group, Edison, NJ, USA). All measurementswere performed in triplicates.

Total N concentration in dry plant samples was determined

using TOC analyzer (vario MAX CN elementar AnalysensystemeGmbH, Hanau, Germany). For measurement of total P, plant sam-ples (0.4 g each) was soaked with 5 mL concentrated nitric acid(HNO3) overnight, digested at 80 ◦C for 180 min, and 140 ◦C for

X. Zhou et al. / Ecological Engineering 101 (2017) 107–119 109

Table 1Major aquatic plant species in the waterways of Indian River Lagoon basin.

Scientific name Common name Ecotype Relative abundancea (%) Coverage (%)

Hydrilla verticillata Hydrilla Submerged 18.18 90–100Potamogeton spp. Pondweed Submerged 4.55 50–70Salvinia spp. Salvinia Free-floating 4.55 90–100Pontederia cordata Pickerelweed Emergent 13.62 60–100Pistia stratiotes Water lettuce Free-floating 31.82 70–100Typha orientalis Cattail Emergent 13.64 80–100Nuphar advena Spatterdock Floating-leaved 9.09 30–80Panicum hemitomon Maidencane Emergent 4.55 50–80

a Relative abundance was calculated using the proportion of each species sample numbers in the total species sample numbers.

erway

1(ds

2

todvvss

3

3s

sit(cse

Fig. 1. Photograph of major aquatic macrophytes in the wat

40 min on an A.I. digestion block with a programmable controllerAIM 500, A.I. Scientific Inc., Australia). Concentrations of P in theigested solutions were determined using the ICP-OES. All mea-urements were performed in three replicates.

.4. Statistics analysis

Statistical analyses were performed using SPSS 19.0. Correla-ions were performed between the pollutant concentrations in theverlying water or surface sediments and those in plants. Standardeviation (SD.) was calculated using the Excel 2007. Coefficient ofariable (CV) was calculated using the ratio of SD value to meanalue. In addition, one-way ANOVA was performed to determineignificant differences among the different sampling sites and plantpecies, with significant levels at P < 0.05 or 0.01 for all the analyses.

. Results

.1. Nutrient concentrations in the overlying water and surfaceediment

The concentrations of nutrients in overlying water and surfaceediment samples collected from the waterways were presentedn Table 2. The TN and TP concentrations were 1.41 and 5.25imes higher than the respective critical level for this ecoregion

0.9 mg L−1 for total N and 40 �g L−1 for total P) (US EPA, 2000), indi-ating that runoff water in the waterways is a potential nonpointource of pollution to the IRL. Coefficient of variation (CV) was high-st for PO4

3-P (109.9%), followed by TP (93.8%),NO3−-N (75.1%), EC

s of Indian River Lagoon basin (August to October of 2014).

(72.6%), ON (59.5%), NH4+-N (50.9%), and TN (49.1%), DOC (18.7%)

and pH (6.08%), suggested a large spatial variation in water qual-ity parameters among the sampling sites, except for pH and DOC.There were significant differences in pH, EC, concentrations of TN,NH4

+-N, NO3−-N, ON, DOC, TP, PO4

3−-P, C/N, and N/P among the22 sampling sites (P < 0.01). Concentration and availability of N andP vary largely in different waterways of the IRL basin. pH, EC, TOC,TN and extractable-P in the surface sediments also had consider-able spatial variation. The TN, EC, and TOC content ranged from0.11 to 6.20 g kg−1, 79.2 to 2957 �S cm−1 and 2.00 to 83.1 g kg−1

with a CV value of 127%, 103% and 120%, respectively. ExtractableP, NH4

+-N and NO3−-N also varied with sampling sites, with a range

of 15.58–243.5 mg kg−1, 1.11–8.29 mg kg−1 and0.49–3.59 mg kg−1,respectively. Differences in extractable P, C/N and N/P concentra-tions was significant among the 22 sampling sites (P < 0.01).

3.2. Biomass productivity

Total biomass significantly varied among the eight plantspecies (P < 0.01) (Table 3), from 118 to 2515 g m−2, anddecreased in the following order: cattail > pickerelweed > water let-tuce > hydrilla > maidencane > spatterdock > pondweed > salvinia.There were significant differences in the distribution of biomassbetween above and belowground for each plant species. Thebelowground biomass (root +rhizome) and leaves of cattail was

the highest, amounting to 1615 g m−2 and 901 g m−2, respectively,whereas the highest values of stems and flower biomass occurredin pickerelweed and spatterdock, at 191 g m−2 and 8.20 g m−2,respectively. The proportion of above ground biomass (stems,

110 X. Zhou et al. / Ecological Engineering 101 (2017) 107–119

Fig. 2. Sampling locations within the Indian River Lagoon basin (IRL), South Florida, USA.

Table 2Physicochemical parameters of overlying water and surface sediment.

Water Sediment

Properties Mean ± SD Properties Mean ± SD

pH 7.91 ± 0.48 pH 7.06 ± 0.49EC (�S cm−1) 498± 361 EC (�s cm−1) 860 ± 883Total N (mg L−1) 1.27 ± 0.62 Total N (g/kg) 1.04± 1.32NH4

+-N (mg L−1) 0.22 ± 0.11 NH4+-N (mg kg−1) 3.26 ± 1.71

NO3−-N (mg L−1) 0.22 ± 0.16 NO3

−-N (mg kg−1) 1.89± 1.10Organic N (mg L−1) 0.83 ± 0.49 TOC(g kg−1) 14.7 ± 17.6Total P (mg L−1) 0.21 ± 0.20 Available P (mg kg−1) 79.9 ± 59.9PO4

3−-P (mg L−1) 0.14 ± 0.15 C/N 14.9 ± 3.11DOC (mg L−1) 19.7 ± 3.69 N/P 15.1 ± 15.7C/N 18.1 ±6.76N/P 11.9± 11.1

SD.: standard deviation; n = 22.

Table 3Biomass (mean ± SD) of each species (g m−2 DW).

Species Organ biomass Total biomass Harvestable biomassb

Root + Rhizome Stem Leaf Flower

Hydrillaa NA NA NA NA 579 ± 235 579 ± 235Pondweed a NA NA NA NA 152 ± 32.9 152 ± 32.9Salvinia a NA NA NA NA 118 ± 28.9 118 ± 28.9Water lettuce 167 ± 169 NA 552 ± 552 NA 719 ± 718 719 ± 718Cattail 1615 ± 358 NA 901 ± 78.0 NA 2515 ± 436 901 ± 78.0Maidencane 43.8 ± 12.7 167 ± 52.6 87.3 ± 4.36 NA 298 ± 69.7 254 ± 57.0Pickerelweed 804 ± 218 191 ± 47.5 38.5 ± 11.4 8.01 ± 6.63 1039 ± 186 237 ± 57.8Spatterdock 132 + 28.8 70.5 ± 8.72 51.2 ± 49.1 8.20 ± 1.04 258 ± 92.4 130 ± 63.6

ND: no data.a Root, stem and leaf of hydrilla, pondweed and salvinia were not separated, therefore, all the tissues were combined.b Harvestable biomass of hydrilla, pondweed, salvinia and water lettuce were equal to the whole plant; harvestable biomass of maidencane, pickerelweed, cattail,

spatterdock was the sum of aboveground parts including stems, leaves and flower.

nginee

l0abamnehtpfsplt

thuw(uwblEpsiPattei

3

sw1mlaslehFair

sa

3

pvidt

nia.

X. Zhou et al. / Ecological E

eaf and flower) to below ground biomass (root + rhizome) were.30, 0.56, 0.98 and 5.81, for pickerelweed, cattail, spatterdocknd maidencane, respectively, indicating that the abovegroundiomass of pickerelweed, cattail and spatterdock contributed to

small part of the total biomass productivity, whereas that ofaidencane contributed more to the total biomass. However,

ot all the plant biomass is harvestable, for example, the roots ofmergent plants such as cattail are difficult to take out, therefore,arvestable biomass is used to represent plant productivity. Inhis research, plant species varied significantly in harvestableroductivity (P < 0.05), with cattail being the highest (901 g m−2),ollowed by water lettuce (719 g m−2), hydrilla (579 g m−2), andalvinia being the lowest (118 g m−2). Floating and submergedlant species, such as hydrilla, pondweed, salvinia and water

ettuce were mostly harvestable and could contribute almost 100%o harvestable biomass (Table 3).

Apart from species variation, plant biomass also had a large spa-ial variability over sampling sites. The highest value of total andarvestable biomass occurred at site 13, whereas the lowest val-es occurred at sites 3 and 21, respectively (Fig. 3). Total biomassas positively correlated with EC, PN or PO4

+-P in overlying waterP < 0.05) (Table 4). For emergent aquatic and floating-leaved plants,nderground biomass (root + rhizome) were positively correlatedith EC, PO4

+-P in overlying water (P < 0.05), whereas, stemsiomass was significantly related to NH4

+-N, C/P, or N/P in over-ying water (P < 0.05); leaf biomass was positively associated withC, TP, PP and PO4

+-P in overlying water (P < 0.05); For submergedlants, total biomass was positively associated with available P inediment (P < 0.05). For free-floating plants, total biomass was pos-tively associated with EC in overlying water, TN, TOC and available

in sediment (P < 0.05); root biomass was positively related to TNnd TOC in sediment (P < 0.05); leaf biomass was positively relatedo EC in overlying water, TN, TOC in sediment but negatively relatedo DN in water (P < 0.05); indicating that physicochemical param-ters of overlying water and surface sediment have a significantnfluence on the aquatic plant biomass production.

.3. Nutrient concentrations in plant

Plant nutrient concentrations varied among all the studied plantpecies (P < 0.05) (Table 5). Mean N and P concentration in thehole plant ranged from 9.35 to 27.4 g kg−1 and from 0.64 to

.71 g kg−1, respectively. The highest N and P concentration wereeasured in spatterdock and hydrilla, respectively, whereas, the

owest valued were in cattail. N/P ratio ranged from 14.9 to 26.2nd decreased in the order of salvinia (26.2)> maidencane (21.3)>patterdock (20.8)> pickerelweed (20.6)> pondweed (17.6)> waterettuce (16.7)> hydrilla (15.6)> cattail (14.9). In addition, the high-st N concentration in root was measured in water lettuce and theighest N in stems, leaf and flower were measured in spatterdock.urthermore, spatterdock had the highest P concentration in rootnd leaf, whereas, pickerelweed had the highest P concentrationn stems and flower, indicating that N and P concentrations in theoot, stem, leaf and flower vary significantly with plant species.

Plant nutrient concentrations varied significantly among all theampling sites (P < 0.05), as affected by N and P availability in waternd sediment and other hydrological (Tables 6,7).

.4. Nutrients storage in aquatic plants tissues

Plant storage capacity for N and P, as calculated by multi-lying biomass with N, P concentrations in plant, significantly

aried among the plant species (P < 0.01). Mean N storagen the whole plant ranged from 2.64 to 23.4 g N m−2 andecreased in the order: cattail > pickerelweed > water let-uce > hydrilla > spatterdock > maidencane > pondweed > salvinia,

ring 101 (2017) 107–119 111

whereas the mean P storage in the wholeplant ranged from 0.10 to 1.59 g P m−2 anddecreased in the order: cattail > hydrilla > water let-tuce > pickerelweed > spatterdock > maidencane > pondweed > salviCattail had the highest while salvinia had the lowest N and P storagecapacity (Table 5).

Storage by root plus rhizome of cattail reached a peak of12.8 g N m−2 and 1.04 g P m−2, and pickerelweed had the secondhighest N storage in roots (8.41 g N m−2). As cattail and pickerel-weed store more N and P in roots than leaves and stems, so theseplants have potential to fix nutrients belowground. Net N and Pstorage, as calculated by multiplying harvestable biomass withmean N, P concentrations in plant tissues, ranged from 2.64 to 14.6 gN m−2 and 0.10 to 1.04 g P m−2, respectively (Table 5). Water lettucehad the highest amounts of N net storage, with a peak value of 14.6 gN m−2, followed by hydrilla and cattail, at 12.6 and 8.40 g N m−2,respectively. In comparison hydrilla had the highest net P storage,with a peak value of 1.04 g P m−2, followed by water lettuce andcattail, at 0.97 and 0.57 g P m−2, respectively. The remaining fivespecies had similar net N and P storage in their tissues, ranging from2.64 to 4.70 g N m−2, and from 0.10 to 0.22 g P m−2, respectively(Table 5).

Total N storages were positively associated with total biomass,harvestable biomass and N concentrations in plants tissues(P < 0.05), and total P storages were positively associated with totalbiomass and harvestable biomass (P < 0.05). However, net N, P stor-ages was only significantly related to harvestable biomass, butunexpectedly not to N, P concentrations in plant tissues (P > 0.05)(Table 8).

3.5. Estimation of annual N and P removal values

The amount of nutrients annually removed by each plant specieswas calculated as follows:

Nutrient removal amount (kg) = nutrient net storage in plant (g m−2)

× distribution area (m2) × 10−3 (1)

Total amount of nutrient removal

=∑

Nutrient removal amount of each species (2)

According to Eqs. (1) and (2), the amount of N and P uptakewere estimated per hectare water area. Among the eight domi-nant aquatic plant species, water lettuce had the greatest annualN uptake (146 kg N ha−1 y−1), followed by hydrilla and cattail,126 kg N ha−1 y−1 and 84.0 kg N ha−1 y−1, whereas hydrilla had thegreatest annual P uptake (10.4 kg P ha−1 y−1), followed by water let-tuce and cattail, 9.67 kg P ha−1 y−1 and 5.73 kg P ha−1 y−1, (Table 9).

The total amounts of nutrients annually removed by each plantspecies from the studied water surface area (15.6 km2) were esti-mated at 1.87 × 103–72.4 × 103 kg N, 0.07 × 103–4.80 × 103 kg P,respectively. Water lettuce had the greatest amounts of N and Premoval, with a peak value of 72.4 × 103 kg N, and 4.80 × 103 kg Pper year, followed by hydrilla, 35.6 × 103 kg N and 2.95 × 103 kg Pper year, then by cattail, 17.9 × 103 kg N and 1.22 × 103 kg P peryear. The total amounts of nutrients annually removed by all theeight dominant plant species were estimated at 146 × 103 kg N and9.94 × 103 kg P from the waterways of the IRL basin (Table 9).

3.6. Potential of plant harvesting in removing nutrients from

waterways

In order to evaluate the potential of plant harvesting for theremoval of nutrients from the waterways, the biomass of each

112

X.

Zhou et

al. /

Ecological Engineering

101 (2017)

107–119

Table 4Pearson’s correlation coefficients (r) of biomass with physico-chemical parameters in water and sediment.

Environmental factors Total(n = 22)

Emergent aquatic and floating-leaved plants Submerged plants Free-floating plants

Total (n = 9) Root + Rhizomes (n = 9) Stems (n = 6) Leaves (n = 9) Total (n = 5) Total biomass (n = 8) Root + Rhizomes (n = 7) Leaves (n = 7)

pH-water 0.00 0.01 −0.02 −0.67 0.17 0.76 −0.4 −0.69 −0.64EC-water 0.60** 0.78* 0.74* −0.76 0.82** −0.24 0.82* 0.74 0.81*NH4

+-N-water 0.40 0.33 0.35 0.83* 0.14 −0.06 −0.08 −0.13 −0.17NO3

−-N-water 0.02 0.32 0.44 0.49 −0.01 −0.74 −0.23 −0.16 −0.19Total N-water 0.12 0.12 0.18 0.76 −0.12 −0.12 −0.62 −0.66 −0.66Dissolved N-water −0.13 −0.09 −0.03 0.69 −0.31 −0.22 −0.73* −0.76* −0.79*Organic N-water 0.06 0.01 0.06 0.71 −0.19 0.13 −0.58 −0.67 −0.63Particulate N-water 0.46* 0.40 0.45 0.74 0.17 0.15 0.09 0.03 0.16Dissolved organic N-water −0.27 −0.33 −0.29 0.56 −0.45 0.11 −0.53 −0.63 −0.64Total P-water 0.13 0.68* 0.64 −0.63 0.73* −0.70 −0.46 −0.65 −0.54Dissolved P P-water 0.14 0.60 0.57 0.02 0.62 0.51 −0.38 −0.54 −0.48Particulate P-water 0.07 0.64 0.60 −0.56 0.71* −0.83 −0.15 −0.20 −0.10Ortho P-water 0.48* 0.72* 0.67* −0.64 0.77* 0.50 −0.36 −0.64 −0.52Dissolved organic carbon-water −0.14 0.19 0.19 0.29 0.11 −0.26 −0.57 −0.63 −0.61C/N-water −0.12 −0.27 −0.29 −0.85* −0.08 −0.12 0.59 0.60 0.58N/P-water −0.03 −0.30 −0.22 0.85* −0.55 0.38 0.12 0.24 0.11pH-sediment −0.49* −0.44 −0.45 −0.08 −0.37 −0.76 −0.43 −0.46 −0.34EC-sediment 0.49* 0.53 0.48 −0.29 0.61 0.54 0.62 0.64 0.55Total N-sediment 0.26 0.39 0.38 0.69 0.30 0.67 0.79* 0.86* 0.81*NH4

+-N −sediment 0.13 −0.16 −0.17 −0.45 −0.09 0.77 0.67 0.63 0.63NO3

−-N −sediment −0.07 0.23 0.33 0.64 −0.04 −0.47 −0.25 −0.09 −0.06Total organic C −sediment 0.27 0.36 0.35 0.71 0.25 0.66 0.83* 0.89** 0.83*Avail P-sediment 0.28 0.10 0.10 −0.50 0.18 0.88* 0.71* 0.70 0.67C/N-sediment −0.17 −0.21 −0.24 −0.15 −0.14 −0.71 0.18 0.21 0.06N/P-sediment 0.02 −0.16 −0.16 0.55 −0.22 0.46 0.40 0.75 0.68

*and **indicate significant level at P < 0.05, 0.01 level, respectively.

X. Zhou et al. / Ecological Engineering 101 (2017) 107–119 113

Fig. 3. Biomass of major aquatic plant species of the sampling sites.

Table 5Averaged nutrient concentration (mean ± SD) and nutrient storage in plant tissues.

Species name Tissue Nutrient concentration Total nutrient storagea Total net storageb

TN (g kg−1) TP (g kg−1) N/P N storage (g N m−2) P storage (g P m−2) N storage (g N m−2) P storage (g P m−2)

Water lettuce Root 23.7 ± 5.28 1.38 ± 0.31 17.7 ± 5.08 3.60 ± 2.69 0.22 ± 0.19 14.6 ± 9.84 0.97 ± 0.78Leaves 22.2 ± 6.08 1.59 ± 0.52 16.2 ± 10.8 10.4 ± 6.46 0.76 ± 0.57Whole 23.0 ± 5.35 1.49 ± 0.37 16.7 ± 7.71 14.6 ± 9.84 0.97 ± 0.78

Pickerelweed Root 11.2 ± 4.52 0.29 ± 0.04 39.4 ± 17.4 8.41 ± 1.00 0.23 ± 0.07 3.58 ± 0.67 0.17 ± 0.02Stems 9.78 ± 4.10 0.81 ± 0.29 12.8 ± 4.97 1.74 ± 0.27 0.15 ± 0.06Leaves 23.9 ± 3.62 0.87 ± 0.13 27.4 ± 0.48 0.90 ± 0.17 0.03 ± 0.01Flower 18.0 ± 0.45 1.27 ± 0.06 14.1 ± 0.30 0.10 ± 0.12 0.01 ± 0.01Whole 15.5 ± 1.42 0.76 ± 0.13 20.6 ± 2.05 16.0 ± 3.01 0.79 ± 0.21

Cattail Root 8.02 ± 0.51 0.66 ± 0.16 12.5 ± 2.42 12.8 ± 2.20 1.04 ± 0.17 8.40 ± 0.42 0.57 ± 0.08Leaves 10.7 ± 0.81 0.62 ± 0.10 17.7 ± 3.76 9.61 ± 0.76 0.56 ± 0.12Whole 9.35 ± 0.56 0.64 ± 0.10 14.9 ± 2.62 23.4 ± 3.00 1.59 ± 0.30

Spatterdock Root 13.5 ± 3.93 1.77 ± 0.17 7.56 ± 1.50 1.85 ± 0.91 0.24 ± 0.07 3.39 ± 1.51 0.17 ± 0.10Stems 23.3 ± 4.02 0.65 ± 0.50 55.3 ± 49.3 1.62 ± 0.08 0.05 ± 0.04Leaves 47.5 ± 2.56 1.69 ± 0.14 28.2 ± 0.84 2.37 ± 2.20 0.08 ± 0.08Flower 21.9 ± 0.35 1.19 ± 0.30 18.4 ± 0.21 0.09 ± 0.13 0.01 ± 0.01Whole 27.4 ± 1.88 1.33 ± 0.12 20.8 ± 3.33 6.99 ± 2.05 0.35 ± 0.15

Madicance Root 23.4 ± 0.28 1.34 ± 0.05 17.5 ± 0.19 1.03 ± 0.08 0.06 ± 0.00 4.70 ± 0.72 0.22 ± 0.04Stems 11.2 ± 7.86 0.47 ± 0.07 23.9 ± 13.5 1.88 ± 2.34 0.08 ± 0.02Leaves 20.9 ± 0.37 0.80 ± 0.42 26.1 ± 19.2 1.82 ± 0.11 0.07 ± 0.13Whole 18.5 ± 2.83 0.87 ± 0.16 21.3 ± 2.17 5.51 ± 0.85 0.26 ± 0.05

Hydrilla Whole 22.0 ± 5.87 1.71 ± 0.67 15.6 ± 10.0 12.6 ± 6.54 1.04 ± 0.67 12.6 ± 6.54 1.04 ± 0.67

Pondweed Whole 22.2 ± 0.35 1.26 ± 0.22 17.6 ± 2.89 3.37 ± 0.05 0.19 ± 0.03 3.37 ± 0.05 0.19 ± 0.03

Salvinia Whole 22.4 ± 0.06 0.85 ± 0.58 26.2 ± 34.7 2.64 ± 0.01 0.10 ± 0.07 2.64 ± 0.01 0.10 ± 0.07

meanean N

stlw72((tteo

emcii

a Total nutrient storages were calculated by multiplying of total biomass and theb Total net storages were calculated by multiplying of harvestable biomass and m

pecies were estimated after each harvest based on the followinghree hypothesis: 1) the whole plant of free-floating plant (waterettuce and salvinia) and submerged plant (hydrilla and pondweed)

ould be harvested 1–3 times per year to maintain 30%, 50% and0% plant coverage of the total space during the growth season;) the aerial parts (stems, leaves, and flower) of emergent plantcattail, pickerelweed and maidencane) and floating-leaved plantspatterdock) would be harvested 1–3 times per year to main-ain 30%, 50% and 70% plant coverage of the total space duringhe growth season; and 3) the whole plant or aerial parts of theight plant species would be completely harvested before the endf growing season.

According to these assumptions, the total amounts of nutri-nts removed annually by plant harvesting were determined by

ultiplying total harvestable biomass per year with the mean con-

entration of N, P in plant tissues. The amounts of N, P removalncreased with harvesting times and intensity (Table 10). Fornstance, when water lettuce was harvested once at 30, 50 and

N, P content in plant tissues., P content in plant tissues.

70% harvest intensity, the amount of nutrients removed wouldincrease by 30, 50 and 70%; and by 60, 100 and 140% if harvestedtwice, and by 90, 150, and 210% if harvested three times. Similarly,others species, including hydrilla, pondweed, salvinia, etc. couldalso increase their nutrient removal by periodic harvesting. Theseaquatic plant species have great potential of removing nutrientsfrom the waterways in the IRL basin if adequate harvest of biomassis managed.

4. Discussion

4.1. Biomass productivity of aquatic plants

Biomass productivity is one of the important parameters for

evaluating potential of aquatic macrophytes in removing nutrientsfrom water. Tanner (1996) reported that TN removal was linearlycorrelated to plant total biomass. Liu et al. (2012) reported thatthe TN and TP accumulation ability of plant is determined by its

114 X. Zhou et al. / Ecological Engineering 101 (2017) 107–119

Table 6Pearson’s correlation coefficients (r) of N concentrations in plant tissue with physico-chemical parameters of water and sediment.

Environmental factors Mean (n = 22) Emergent aquatic and floating-leaved plants Submerged plants Free-floating plants

Mean(n = 9)

Root+ Rhizomes(n = 9)

Stems(n = 6)

Leaves(n = 9)

Mean(n = 5)

Mean(n = 8)

Root+ Rhizomes(n = 7)

Leaves(n = 7)

pH-water 0.10 0.24 −0.05 0.62 0.25 −0.46 0.20 0.34 0.07EC-water −0.19 −0.47 −0.43 0.70 −0.42 0.02 −0.16 0.05 −0.37NH4

+-N-water −0.38 −0.50 −0.06 −0.74 −0.46 0.46 0.65 0.46 0.76*NO3

−-N-water −0.05 −0.47 −0.49 −0.57 −0.35 −0.55 0.55 0.30 0.74Total N-water −0.30 −0.39 −0.09 −0.77 −0.34 0.45 0.38 0.09 0.59Dissolved N-water −0.10 −0.18 0.04 −0.70 −0.11 0.28 0.44 0.15 0.64Organic N-water −0.28 −0.29 0.01 −0.73 −0.26 0.60 −0.11 −0.22 −0.01Particulate N-water −0.52* −0.61 −0.26 −0.75 −0.59 0.78 −0.04 −0.12 0.05Dissolved organic N-water −0.01 0.07 0.24 −0.58 0.11 0.90 −0.19 −0.26 −0.13Total P-water −0.16 −0.37 −0.47 0.76 −0.34 −0.17 0.17 0.25 0.07Dissolved P P-water −0.26 −0.43 −0.27 −0.14 −0.41 −0.79 −0.05 −0.03 −0.09Particulate P-water −0.03 −0.28 −0.57 0.71 −0.24 0.09 0.58 0.71 0.42Ortho P-water −0.38 −0.42 −0.48 0.78 −0.39 −0.72 0.12 0.13 0.09Dissolved organic carbon-water 0.03 −0.12 −0.08 −0.23 −0.01 0.20 −0.10 −0.11 −0.09C/N-water 0.27 0.65 −0.05 0.95** 0.66 −0.53 −0.60 −0.32 −0.80*N/P-water −0.06 −0.04 0.22 −0.84* −0.01 0.43 0.16 −0.06 0.37pH-sediment 0.38 0.31 0.41 0.04 0.27 −0.19 0.33 0.29 0.39EC-sediment −0.04 −0.34 −0.05 0.17 −0.36 0.69 −0.16 0.01 −0.34Total N-sediment 0.01 −0.42 −0.02 −0.55 −0.40 0.61 −0.57 −0.37 −0.69NH4

+-N −sediment 0.17 0.39 −0.04 0.64 0.39 0.47 −0.60 −0.56 −0.69NO3

−-N −sediment −0.17 −0.47 −0.44 −0.62 −0.38 −0.77 0.23 0.09 0.52Total organic C −sediment 0.01 −0.41 0.05 −0.60 −0.39 0.62 −0.57 −0.39 −0.67Avail P-sediment −0.16 0.17 −0.27 0.64 0.15 −0.43 −0.64 −0.55 −0.89**C/N-sediment 0.18 0.23 0.36 −0.02 0.24 0.06 0.11 −0.12 0.29N/P-sediment 0.04 −0.02 0.35 −0.48 0.01 0.80 −0.34 −0.20 −0.49

*and **indicate significant level at P < 0.05, 0.01 level, respectively.

Table 7Pearson’s correlation coefficients (r) of P concentrations in plant tissue with physico-chemical parameters of water and sediment.

Environmental factors Emergent aquatic and floating-leaved plants Submerged plants Free-floating plants

Mean(n = 22)

Mean(n = 9)

Root+ Rhizomes(n = 9)

Stems(n = 6)

Leaves(n = 9)

Mean(n = 5)

Mean(n = 8)

Root+ Rhizomes(n = 7)

Leaves(n = 7)

pH-water 0.40 0.43 0.43 0.44 0.34 0.64 0.76* 0.48 0.67EC-water −0.03 −0.23 −0.09 0.48 −0.29 0.64 0.09 0.29 −0.39NH4

+-N-water −0.49* −0.48 −0.66 0.13 −0.51 0.79 −0.22 −0.18 −0.48NO3

−-N-water −0.38 −0.05 −0.77* 0.49 −0.45 −0.39 −0.59 −0.47 −0.56Total N-water −0.29 −0.23 −0.63 0.04 −0.46 0.67 −0.04 −0.39 0.16Dissolved N-water −0.12 0.03 −0.47 0.25 −0.25 0.63 −0.00 −0.38 0.16Organic N-water −0.13 −0.17 −0.50 −0.07 −0.39 0.73 0.46 −0.08 0.74Particulate N-water −0.46* −0.54 −0.71* −0.25 −0.68* 0.61 −0.10 −0.16 0.07Dissolved organic N-water 0.19 0.20 −0.18 0.14 −0.03 0.77 0.50 −0.06 0.71Total P-water 0.06 −0.51 −0.10 −0.70 −0.26 −0.04 0.53 0.20 0.59Dissolved P P-water 0.18 −0.35 −0.20 0.25 −0.40 0.52 0.40 −0.10 0.51Particulate P-water −0.03 −0.58 −0.01 −0.68 −0.10 −0.19 0.29 0.78* 0.15Ortho P-water −0.07 −0.50 −0.12 −0.68 −0.30 0.61 0.53 0.05 0.55Dissolved organic carbon-water 0.26 −0.02 −0.21 0.41 −0.06 0.67 0.38 −0.06 0.63C/N-water 0.33 0.10 0.67* −0.30 0.77* −0.54 0.14 0.21 0.01N/P-water −0.42* 0.04 −0.48 0.26 −0.16 −0.36 −0.74* −0.72 −0.78*pH-sediment 0.18 0.37 0.15 0.69 0.15 −0.66 −0.18 −0.21 0.10EC-sediment 0.20 0.02 0.11 0.06 −0.19 0.69 0.05 0.20 −0.37Total N-sediment 0.22 −0.42 −0.38 −0.04 −0.31 0.55 −0.19 −0.03 −0.33NH4

+-N −sediment 0.40 −0.20 0.34 −0.60 0.53 0.70 0.19 −0.13 0.02NO3

−-N −sediment −0.22 −0.23 −0.75* 0.35 −0.42 −0.54 −0.57 −0.33 −0.27Total organic C −sediment 0.20 −0.35 −0.38 0.04 −0.34 0.52 −0.23 −0.12 −0.41Avail P-sediment 0.12 −0.40 0.23 −0.73 0.31 0.62 0.32 0.10 0.01C/N-sediment −0.16 0.70* 0.36 0.36 0.16 −0.69 −0.47 −0.76* −0.70

.39

*

bvsa(fw

N/P-sediment 0.03 0.07 −0.20 0

and **indicate significant level at P < 0.05, 0.01 level, respectively.

iomass productivity. In the present study, both total and har-estable biomass varied among the eight dominant aquatic plantpecies in the IRL basin. Cattail had the highest total biomass

nd harvestable biomass, followed by water lettuce and hydrillaTable 3). These three aquatic plant species merit attention foruture application in surface water cleaning. In addition, thereas significant spatial variation in biomass production for each

−0.05 0.53 −0.49 −0.02 −0.38

plant species. Total biomass of submerged plant was associatedwith available P in sediment (P < 0.05) (Table 4). Similar resultswere reported in literature, indicating that biomass productivity

of aquatic plants is affected by nutrient availability in the waterand sediment, hydrological conditions, intrinsic species, possibleecotype as well as growth characteristics and other environmen-tal factors (Brisson and Chazarenc, 2009; Wu et al., 2011). For each

X. Zhou et al. / Ecological Engineering 101 (2017) 107–119 115

Table 8Pearson’s correlation coefficients (r) of N, P storages with nutrients concentration and biomass.

N storages P storages

Total N storages Net N storages Total P storages Net P storages

N concentrations in plants tissues −0.71* −0.04 −0.64 −0.03P concentrations in plants tissues −0.20 0.57 −0.03 0.62Total biomass 0.92** 0.31 0.89** 0.30Harvestable biomass 0.73* 0.82* 0.86** 0.81*

*and **indicate significant level at P < 0.05, 0.01 level, respectively.

Table 9Amounts of N and P uptake by aquatic plants per hectare and the total amount of N and P removed in the study area.

Species Uptake by aquatic plants (kg ha−1 y−1) The total amount removed in the studied area (kg y−1)a × 103

N P N P

Hydrilla 126 10.4 35.6 2.95Pondweed 33.7 1.91 2.39 0.14Salvinia 26.4 1.01 1.87 0.07Water lettuce 146 9.67 72.4 4.80Cattail 84.0 5.73 17.9 1.22Maidencane 47.1 2.21 3.34 0.16Pickerelweed 35.8 1.73 7.62 0.37Spatterdock 33.9 1.71 4.81 0.24

Total / / 146 9.94

a The area of water surface is 15.6 km2 in the study region (https://en.wikipedia.org/wiki/Fort Pierce, Florida).

Table 10Estimated values of N, P removal (g m−2) from waterways if plant were harvested at different intensities.

Species Harvested intensity** Harvesting frequency during the plant growth period* Harvestingin decline phase

Once Twice Three times

N P N P N P N P

Hydrillaa 30% 3.81 0.30 7.63 0.60 11.4 0.8950% 6.36 0.50 12.7 0.99 19.1 1.4970% 8.90 0.69 17.8 1.39 26.7 2.08100% NA NA NA NA NA NA 12.7 0.99

Pondweeda 30% 1.01 0.06 2.02 0.11 3.03 0.1750% 1.68 0.10 3.37 0.19 5.05 0.2970% 2.36 0.13 4.71 0.27 7.07 0.40100% NA NA NA NA NA NA 3.37 0.19

Salviniaa 30% 0.79 0.03 1.58 0.06 2.38 0.0950% 1.32 0.05 2.64 0.10 3.96 0.1570% 1.85 0.07 3.70 0.14 5.55 0.21100% NA NA NA NA NA NA 2.64 0.10

Water lettucea 30% 4.95 0.32 9.91 0.64 14.9 0.9650% 8.26 0.54 16.51 1.07 24.8 1.6170% 11.56 0.75 23.12 1.50 34.7 2.25100% NA NA NA NA NA NA 16.5 1.07

Pickerelweedb 30% 1.10 0.05 2.20 0.11 3.30 0.1650% 1.84 0.09 3.67 0.18 5.51 0.2770% 2.57 0.13 5.14 0.25 7.71 0.38100% NA NA NA NA NA NA 3.67 0.18

Cattailb 30% 2.53 0.17 5.06 0.35 7.58 0.5250% 4.21 0.29 8.43 0.58 12.6 0.8770% 5.90 0.40 11.8 0.81 17.7 1.21100% NA NA NA NA NA NA 8.43 0.58

Spatterdockb 30% 1.07 0.05 2.14 0.10 3.20 0.1650% 1.78 0.09 3.56 0.17 5.34 0.2670% 2.49 0.12 4.98 0.24 7.48 0.36100% NA NA NA NA NA NA 3.56 0.17

Maidencaneb 30% 1.41 0.07 2.82 0.13 4.23 0.2050% 2.35 0.11 4.70 0.22 7.06 0.3370% 3.29 0.15 6.59 0.31 9.88 0.46100% NA NA NA NA NA NA 4.70 0.22

ND: no data.a All plant organs could be periodically harvested during their growth period.b Aerial parts could be periodically harvested during their growth period.* Indicates plants being harvested 1, 2, or 3 times during their growth period.

** 30%, 50%, 70% means 70%, 50%, or 30% of plant coverage was maintained when the plants were harvested. 100% means all harvestable plant tissues were removed beforethe end of growing season.

1 nginee

safacor

4

ms2OcsNwHgacpsoot(ps

o2p(Hs(o1egeiMtrplcwflw(lbwsf

4a

n

16 X. Zhou et al. / Ecological E

pecies, for example, the amounts of biomass could vary season-lly and yearly with the change of external environmental/climaticactors, such as temperature, rainfall, and nutrient levels in waternd sediments. Therefore, further long-term research is needed toomprehensively evaluate the aquatic plants yield in waterwaysf this study area, especially in extreme conditions, such as largeainfall events, mild temperature and high available nutrients, etc.

.2. Nutrient concentrations in plants

Nutrient concentrations in aquatic plants are influenced byany factors such as water conditions, nutrient availability, plant

pecies, and growth stage (Lombardo and Cooke, 2003; Vymazal,007; Wu et al., 2011; Liu et al., 2012; Steffenhagen et al., 2012).ur results confirm the observations to some extent. Plant nutrientoncentration varied significantly among the eight aquatic plantpecies (P < 0.05) and among the 22 sampling sites (P < 0.05). Mean, P concentration in whole plant had a significant correlationith PN, or NH4

+ concentration, N/P in water column (Tables 6,7).owever, the relationships varied with plant species. For emer-ent aquatic and floating-leaved species, plant P was significantlyssociated with C/N in sediment (P < 0.05), whereas no signifi-ant correlation occurred between plant N and physicochemicalarameters of overlying water and surface sediment (P > 0.05). Forubmerged species, no significant correlation occurred between Nr P concentration in plant and physico-chemical variables of waterr sediment (P > 0.05). For free-floating species, mean P concentra-ion in plant was significantly associated with pH and N/P in waterP < 0.05), whereas no significant correlation occurred betweenlant N and physicochemical parameters of overlying water andurface sediment (P > 0.05).

Aquatic plants take up nutrients from either water or sedimentr both (Scheffer et al., 1993; Lacoul and Freedman 2006; Bal et al.,011; Dai et al., 2012; Jampeetong et al., 2012). Free-floating macro-hytes generally obtain nutrients from water column through rootsCedergreen and Madsen, 2002; Fang et al., 2007; Tanner andeadley, 2011). Emergent macrophytes have dense fibrous root

ystems, large surface areas and well developed rihizome tissuesCardwell et al., 2002; Bonanno and Lo Giudice, 2010) and theybtain nutrients mostly from the sediment (Brix and Schierup,989; Ran et al., 2004; Iamchaturapatr et al., 2007; Kantawanichkult al., 2009; Jampeetong et al., 2012); Submerged macrophytes areenerally acquire nutrients from both water and sediment (Dait al., 2012), therefore, submerged macrophytes have importantmplications in nutrient cycling within fresh water systems (Flindt

ogens et al., 1999). Floating-leaved macrophytes petioles holdhe leaves floating on the water surface and they are generallyooted in the sediment (Wu et al., 2007; Seto et al., 2013). In theresent study, there was a significant positive correlation between

eaf N and NH4+-N in water (P < 0.05) and a significant negative

orrelation between leaf N and C/N in water, but no relationshipas found between root N and environmental variables for free-oating macrophytes. However, root P was significantly correlatedith PP in water and C/N in sediment for free-floating macrophytes

P < 0.05). For emergent and floating-leaved macrophytes, no corre-ation was found between root N and water or sediment propertiesut root P was significantly correlated with NO3

−-N, PN or C/N inater and NO3

−-N in sediment (P < 0.05). Obviously, the relation-hips between nutrient concentrations in plant and environmentalactors are complicated.

.3. Potential of nutrient uptake and annual removal quantity by

quatic macrophytes

Nutrient uptake and storage by plants is an important mecha-ism of N and P removal from water in waterways (Gottschall et al.,

ring 101 (2017) 107–119

2007). Total storage of a substance is called “standing stock” andoften calculated by the multiplying nutrient concentration in planttissues and dry biomass productivity per unit area and expressedas mass per unit area, such as g m−2 or kg ha−1 (Vymazal, 2011;Wu et al., 2011; Zhu et al., 2011; Borin and Salvato, 2012). As it isa function of dry biomass per unit of area and nutrient concen-tration in the materials, so a plant species with a high biomassyield and nutrient concentration would have great potential foruptake and storage of nutrients (Reddy and DeBusk, 1987; Vymazal,2007). In the present study, plant N and P uptake and storage var-ied largely among the eight aquatic species, with cattail havingthe greatest potential for N and P storage. Similar results werepreviously reported. For example, Vymazal (1995) reported above-ground N standing stock in the range of 22–88 g N m−2 for 29species; Maddison et al. (2009) indicated that N stored in the above-ground and belowground tissues of Typha latifolia ranged from17.0 g m−2 to 32.3 g m−2 and 11.6 g m−2 to 19.4 g m−2, respectively.Borin and Tocchetto (2007) reported that the N standing stock in tis-sues of Phragmites australis peaks at 111 g m−2. Plant P storage had asimilar variation trend with N, and aboveground P standing stock ofemergent macropyhtes has been reported to be within the range of0.1–19 g P m−2 (Brix and Schierup, 1989; Johnston, 1991; Vymazal,1995, 2011). In addition, for the same species, nutrient uptake alsoevidently differs. Borin and Salvato (2012), for example, comparednutrient uptake by cattail and the results indicated that cattail aerialbiomass productivity was 5.65 kg m−2 in Morocco and 2.54 kg m−2

in Estonia, and the corresponding N uptake values were 92.2 and34.0 g m−2. Maddison et al. (2009) reported that N storage in theaboveground and belowground tissues of Typha latifolia rangedfrom 17.0 to 32.3 g m−2 and 11.6 to 19.4 g m−2, respectively. Ourresearch found that the total N standing stock by cattail changedfrom 19.99 to 24.45 g m−2. These results indicate that plant uptakeand storage for nutrients varies widely among the different speciesand is affected by age, physiological state and environmental con-ditions (Gottschall et al., 2007; Vymazal, 2007; Borin and Salvato,2012; Liu et al., 2012).

Compared with total nutrient uptake by the whole plant, it maybe more important to consider the allocation of nutrients betweenthe above and belowground tissues (Borin and Salvato, 2012). Pre-vious studies indicated that N and P uptake by stem, leaf andflower of aboveground parts represent definitive nutrient removalfrom the treatment units because they could be harvested, whereasthe belowground parts only represent a temporary immobilizationbecause it is subjected to a turnover and could release nutrientsduring decomposition, leading to medium-term to long-term sec-ondary water quality deterioration (Sartoris et al., 2000; Borin andSalvato, 2012). Therefore, the potential rate of direct N and P uptakeby plants is related to their net productivity and the concentrationin the plant tissue (Reddy and DeBusk, 1987; Vymazal, 2007; Wuet al., 2011). So for net productivity, their calculation method variedwith plant species. For emergent aquatic and floating-leaved plant,net productivity only included steam, leaf and flower of plant aerialparts, while for floating and submerged plant, the total biomassrepresents definitive net productivity because of all organs couldbe fully harvested. In this case, the net N and P storage is calculatedbased on net biomass (harvestable biomass) and mean N, P concen-tration in plant tissues. In this study, net accumulation of nutrientsin plants varied widely among the different species, from 2.64 to14.6 g N m−2, and 0.10 to 1.04 g P m−2 (Table 5). Although cattailhad the highest potential for N and P storage, their net accumula-tions of N and P were lower than water lettuce and hydrilla, as theformer allocated only 35.9% of the total nutrients in the aerial part,

which is harvestable, while 100% of the nutrients in the latter can beremoved by harvesting. This is consistent with previous researchresults. Under natural conditions, the belowground/abovegroundnutrient standing stock ratio is usually very high owing to high

nginee

bfaabwtt2lttPpb1

iw2tmoLoZtowv(vaPla

rphR1rs1wPuePnlrpbwu2tshtuatv

X. Zhou et al. / Ecological E

elowground biomass and high element concentrations in the rootsor emergent species (Vymazal, 1995, 2011; Wetzel, 2001; Vymazalnd Kröpfelová, 2008). Therefore, water lettuce and hydrilla have

greater potential of removing N and P from waterways in the IRLasin. A similar result was reported by Reddy and DeBusk (1987),ho observed that nutrients standing stock for emergent species in

he range of 14–156 g N m−2 and 1.4–37.5 g P m−2, whereas, morehan 50% of this amount stored in the belowground. Vymazal (1995,001) also reported that water hyacinth (Eichhornia crassipes), simi-

ar to water lettuce, could accumulate 250 g N and 45 g P m−2 owingo its high productivity. The amount of nutrients removed via mul-iple biomass harvesting could be up to 600 g N m−2 y−1 and 126 g

m−2 y−1 (Vymazal, 1995, 2001). Therefore, the capability of alant for nutrient assimilation and storage is dependent on its highiomass yield and high nutrient concentration (Reddy and DeBusk,987; Vymazal, 2007).

Previous studies suggested that biomass yield might be moremportant than nutrient concentration in nutrient removal from

ater (Reddy and DeBusk, 1987; Vymazal, 2007; Vymazal et al.,010; Wu et al., 2011). For example, Zhu et al. (2011) reportedhat the differences in N and P accumulation between the species

ainly resulted from the differences in their biomass productionwing to relatively smaller differences in N and P concentrations.iu et al. (2012) showed that the TN and TP accumulation abilityf plants is primary determined by their biomass yield. Therefore,heng et al. (2013) suggested that high potential of biomass produc-ion should be considered an important criterion for the selectionf plant species. However, our research found that total N storagesere significantly positively associated with total biomass, har-

estable biomass, and N concentrations in plant tissues (P < 0.05)Table 8). Species cattail had the highest total biomass and har-estable biomass, whereas their net accumulations capacity of Nnd P were lower than water lettuce and hydrilla, because N and

concentrations in cattail tissues were lower than those in waterettuce and hydrilla. Our results indicate that both biomass yieldnd nutrient concentration are important for nutrient removal.

In addition, previous research studies had estimated the annualemoval potential by aquatic plants (Lu et al., 2010). For exam-le, Boyd (1970) reported that 1980 kg N ha−1 year−1 and 322 kg Pa−1 year−1 could be removal by invasive species water hyacinths.eddy and Tucker (1983) found that 5350 kg N ha−1 year−1 and260 kg P ha−1 year−1 could be removed by this species. In ouresearch, the annual amount of nutrient uptake varied with plantpecies. Water lettuce had the highest value of N uptake up to45.8 kg N ha−1 y−1, followed by the Hydrilla (125.5 kg N ha−1 y−1),hereas Hydrilla had the highest value of P uptake up to 10.41 kg

ha−1 y−1, followed by water lettuce ( 9.67kg P ha−1 y−1). These fig-res were similar to those reported by Desmet et al. (2011), whostimated approximately 100 kg N per ha and approximately 10 kg

per ha removed by aquatic vegetation, including Potamogetonatans L., Callitriche sp., Ceratophyllum demersum L., Elodea nuttal-

ii (Planch) St John, Stucenia pectinatus (L.) Boerner, etc., from theiver Aa, a tributary of the Kleine Nete, situated in the northernart of Belgium. In our research, the annual amounts of N uptakey water lettuce are lower than that reported by Lu et al. (2010),ho reported that water lettuce, which were planted in two nat-ral ponds in south Florida, could remove 190–329 kg N ha−1 and4.6–34.1 kg P ha−1 annually, respectively. Annual removal poten-ial of water lettuce in our research were dramatically inferior topecies water hyacinths, the research is that water hyacinth has aigher nutrient uptake and biomass yield potential than water let-uce (Lu et al., 2011). For cattail, the annual amounts of nutrient

ptake were 84.0kg N ha−1 y−1 and 5.73 kg P ha−1 y−1, which arelso lower than that reported by Maddison et al. (2009), who foundhat the annual amounts of N and P uptake values by cattail shootsaried from 140 to 304 kg N ha−1 y−1 in autumn in wastewater

ring 101 (2017) 107–119 117

treatment wetlands of Estonia. This discrepancy is likely attributedto the difference in nutrient enrichment among habitats, climaticfactors and the genetic predisposition for each species. It is impor-tant to note that the amounts of biomass and nutrient levels of eachspecies was determined just once in this study, which could varyseasonally and yearly with the change of external environment,such as temperature, rainfall, and nutrient levels in water and sedi-ments. Therefore, further long-term research is needed to evaluatethe potential of aquatic plants in waterways in removing nutrientsfrom runoff/storm water.

4.4. Influence of harvesting intensity on nutrient removal

Enormous storage of nutrients by aquatic plants had beendemonstrated, though on a temporary basis. As plant biomass iswilting and decaying, and N, P in biomass are released back tothe water system again (Kuehn and Suberkropp, 1998; Longhiet al., 2008). Plant biomass harvesting is an effective and obligatorymethod to avoid the secondary pollution of nutrients (Brix, 1997;Vereecken et al., 2006; Liu et al., 2011; Xu et al., 2014). Nutrientremoval by harvesting and subsequent effects on plant productiv-ity and nutrient cycling in wastewater treatment systems had beenwell documented (Vymazal et al., 2010; Liu et al., 2011). For exam-ple, Masi (2009) and Meuleman et al. (2003) reported that properharvesting of wetland plants at suitable time could not only preventN, P release to water body, but also recycle the nutrients benefi-cially. In the present study, total nutrient removal was determinedaccording to three harvesting modes. The results demonstrated thatthe nutrient removal substantially increased if plant biomass wasperiodically harvested during their growth period (Table 10). Forinstance, the amounts of N and P removed by water lettuce couldbe increased by 1.80, 3.00 and 4.20 times, respectively if three har-vestings were conducted when maintained 70%, 50% and 30% plantcoverage of the total space during the growth period as comparedto one harvest at the declining stage. These results are consistentwith previous reports, confirming that multiple harvestings of plantbiomass can significantly enhance the removal of nutrients fromthe treatment systems or waterways (Karathanasis et al., 2003;Edwards et al., 2006; Ságová-Marecková et al., 2009; Vymazal et al.,2010; Zhu et al., 2011; Borin and Salvato, 2012). Periodic harvest-ing of plant biomass could stimulate regeneration and growth ofyoung plants (Riis et al., 2000; Mareckova et al., 2009; Ságová-Marecková et al., 2009). Grimshaw et al. (1997) reported thatregular harvesting reduces densely shaded areas, which directlyaffects light competition in a community, thereby promoting thephotosynthetic activity of plants in the system. Greenway andWoolley (1999) found that periodic cutting of common reed stim-ulates its growth and increases nutrient content of new shoots.In South Florida owing to favorable climate conditions and abun-dant nutrient supply waterways are often fully grown with aquaticplants, such as water lettuce, hydrilla, and cattail, those plantscould be harvested several times a year. Periodic harvesting of plantbiomass would facilitate plant growth and consequently increaseplant uptake and storage capacity for nutrients, thus benefiting towater cleaning and improving hydrological efficiency of the water-ways.

Apart from promoting continuous nutrient removal by plantuptake, harvesting effectively prevents the release of N and P fromplant tissues into the water during decomposition by plant senes-cence, wilting and decay in their decline phase and ultimatelyreduces the harmful effects to the treatment systems (Kuehn andSuberkropp, 1998; Longhi et al., 2008; Zhou and Wang, 2010;

Liu et al., 2011). Kadlec (1999) reported that the majority of theplant storage of N and P is temporary because plant tissues re-release nutrients to the water treatment systems when they dieand decompose. If not harvested, most of the nutrients that have

1 nginee

bw2trIpgg(dtIt

5

SpebNttytecurfh

A

DODMh

R

B

B

B

B

B

B

B

B

C

C

D

18 X. Zhou et al. / Ecological E

een incorporated into the plant tissue may be returned to theater during the decomposition processes (Brix, 1997; Vymazal,

011; Xu et al., 2014). It is estimated that a total of 1.87 × 103

o 72.37 × 103 kg N, 0.07 × 103 to 4.80 × 103 kg P per year may bee-release into the waterways if harvesting is not exploited in theRL basin. Harvesting of the aquatic plants has become a commonractice in Florida, but occurs only once per year at the end of therowing season. In fact, multiple harvesting is possible during therowth period of a year in the warm tropical and subtropical regionsOkurut, 2001; Vymazal, 2011), where biological activity and pro-uctivity are high (Kivaisi, 2001). The present study indicates thathe 2 times more N and 1.5 times more P can be removed from theRL waterways if harvestings of plant biomass are performed threeimes per year during the growing season.

. Conclusions

Eight aquatic species grown in the waterways of the IRL basin,outh Florida, USA, display significant difference in the biomassroductivity, N, P concentrations and storage. Cattail has the high-st total biomass productivity and harvestable biomass, followedy water lettuce and hydrilla. Although cattail has the highest total

and P storage, water lettuce and hydrilla exhibit the highestotal N and P net storage owing to their highest nutrient concen-rations in plants tissues, as well as higher harvestable biomassield. Biomass productivity, nutrient concentration and accumula-ion in each plant species varied with sampling sites, indicating thatnvironment factors such as nutrient enrichment and hydrologicalonditions in the waterways influence plant growth and nutrientptake. Multiple harvestings of harvestable biomass are stronglyecommended by this study as the amount of nutrients removedrom the waterways could substantially increase with increasingarvesting times during the growth period of a year.

cknowledgments

This study was, in part, supported by a scholarship awarded tor. Xiaohong Zhou from the Jiangsu Government Scholarship forverseas Studies, a grant from South Florida Water Managementistrict, and University of Florida. The authors would like thankingr. Wei Sun for his assistance in sampling and Mr. Brian Cain for

is assistance with sample analyses.

eferences

al, K., Struyf, E., Vereecken, H., Viaene, P., De Doncker, L., de Deckere, E., Mostaert,F., Meire, P., 2011. How do macrophyte distribution patterns affect hydraulicresistances? Ecol. Eng. 37, 529–533.

onanno, G., Lo Giudice, R., 2010. Heavy metal bioaccumulation by the organs ofPhragmites australis (common reed) and their potential use as contaminationindicators. Ecol. Indic. 10, 639–645.

orin, M., Salvato, M., 2012. Effects of five macrophytes on nitrogen remediationand mass balance in wetland mesocosms. Ecol. Eng. 46, 34–42.

orin, M., Tocchetto, D., 2007. Five years water and nitrogen balance for aconstructed surface flow wetland treating agricultural drainage waters. Sci.Total Environ. 380, 38–47.

oyd, C.E., 1970. Vascular aquatic plants for mineral nutrient removal frompolluted waters. Econ. Bot. 23, 95–103.

risson, J., Chazarenc, F., 2009. Maximizing pollutant removal in constructedwetlands: should we pay more attention to macrophyte species selection? Sci.Total Environ. 47, 3923–3930.

rix, H., Schierup, H.H., 1989. The use of macrophytes in water pollution control.Ambio 18, 100–107.

rix, H., 1997. Do macrophytes play a role in constructed treatment wetlands.Water Sci. Technol. 35, 11–17.

ardwell, A.J., Hawker, D.W., Greenway, M., 2002. Metal accumulation in aquaticmacrophytes from southeast Queensland. Australia.Chemosphere 48, 653–663.

edergreen, N., Madsen, T.V., 2002. Nitrogen uptake by the floating macrophyteLemna minor. New Phytol. 155, 285–292.

ai, Y.R., Jia, C., Liang, R., Hu, W., Wu Z.B, S.H., 2012. Effects of the submergedmacrophyte Ceratophyllum demersum L. on restoration of a eutrophicwaterbody and its optimal coverage. Ecol. Eng. 40, 113–116.

ring 101 (2017) 107–119

Desmet, N.J.S., Van Belleghem, S., Seuntjens, P., Bouma, T.J., Buis, K., Meire, P., 2011.Quantification of the impact of macrophytes on oxygen dynamics and nitrogenretention in a vegetated lowland river. Phys. Chem. Earth. 36, 479–489.

Dodds, W.K., 2007. Trophic state, eutrophication and nutrient criteria in streams.Trends Ecol. Evol. 22, 669–676.

Edwards, K.R., Cizková, H., Zemanová, K., et al., 2006. Plant growth and microbialprocesses in a constructed wetland planted with Phalaris arundinacea. Ecol.Eng. 27, 153–165.

Ellis, J.B., Shutes, R.B., Revitt, D.M., Zhang, T.T., 1994. Use of macrophytes forpollution treatment in urban wetlands. Resour. Conserv. Recycl. 11, 1–12.

Fang, Y.Y., Yang, X.E., Chang, H.Q., et al., 2007. Phytoremediation ofnitrogen-polluted water using water hyacinth. J. Plant Nutr. 30, 1753–1765.

Flindt Mogens, R., Pardal Â, M., Lillebø, A.I., Martins, I., Marques, J.C., 1999. Nutrientcycling and plant dynamics in estuaries: a brief review. Acta Oecol. 20,237–248.

Gottschall, N., Boutin, C., Crolla, A., Kinsley, C., Champagne, P., 2007. The role ofplants in the removal of nutrients at a constructed wetland treatingagricultural (dairy) wastewater Ontario, Canada. Ecol. Eng. 29, 154–163.

Greenway, M., Woolley, A., 1999. Constructed wetlands in Queensland:performance efficiency and nutrient bioaccumulation. Ecol. Eng. 12, 39–55.

Grimshaw, H.J., Wetzel, R.G., Brandenburg, M., et al., 1997. Shading of periphytoncommunities by wetland emergent macrophytes: decoupling of algalphotosynthesis from microbial nutrient retention. Archiv fur Hydrobiologie139, 17–27.

He, Z.L., Zhang, M., Stoffella, P.J., Yang, X.E., Banks, D.J., Calvert, D.V., 2006.Phosphorus concentrations and loads in surface runoff from sandy soils undercrop production. Soil Sci. Soc. Am. J. 70, 1807–1816.

Iamchaturapatr, J., Yi, S.W., Rhee, J.S., 2007. Nutrient removals by 21 aquatic plantsfor vertical free surface-flow (VFS) constructed wetland. Ecol. Eng. 29, 287–293.

Jampeetong, A., Brix, H., 2009. Nitrogen nutrition of Salvinia natans: effects ofinorganic nitrogen form on growth, morphology, nitrate reductase activity anduptake kinetics of ammonium and nitrate. Aquat. Bot. 90, 67–73.

Jampeetong, A., Brix, H., Kantawanichkul, S., 2012. Effects of inorganic nitrogenforms on growth, morphology, nitrogen uptake capacity and nutrientallocation of four tropical aquatic macrophytes (Salvinia cucullata, Ipomoeaaquatica, Cyperus involucratus and Vetiveria zizanioides). Aquat. Bot. 97, 10–16.

Johnston, C.A., 1991. Sediments and nutrient retention by freshwater wetlands:effects on surface water quality. CRC Crit. Rev. Environ. Control. 21, 491–565.

Kadlec, R.H., 1999. Chemical, physical and biological cycles in treatment wetlands.Water Sci. Technol. 40, 37–44.

Kantawanichkul, S., Kladprasert, S., Brix, H., 2009. Treatment of high-strengthwastewater in tropical vertical flow constructed wetlands planted with Typhaangustifolia and Cyperus involucratus. Ecol. Eng. 35, 238–247.

Karathanasis, A.D., Potter, C.L., Coyne, M.S., 2003. Vegetation effect on fecalbacteria, BOD, and suspended solid removal in constructed wetlands treatingdomestic wastewater. Ecol. Eng. 20, 157–169.

Kearney, M.A., Zhu, W.X., 2012. Growth of three wetland plant species under singleand multi-pollutant wastewater conditions. Ecol. Eng. 47, 214–220.

Kivaisi, A.K., 2001. The potential for constructed wetlands for wastewatertreatment and reuse in developing countries-a review. Ecol. Eng. 16, 545–560.

Kuehn, K.A., Suberkropp, K., 1998. Decomposition of standing litter of the freshwater emergent macrophyte Juncus effusus. Freshwater Biol. 40, 717–727.

Lacoul, P., Freedman, B., 2006. Environmental influences on aquatic plants infreshwater ecosystems. Environ. Rev. 14, 89–136.

Li, J.H., Yang, X.Y., Wang, Z.F., Shan, Y., Zheng, Z., 2015. Comparison of four aquaticplant treatment systems for nutrient removal from eutrophied water.Bioresour. Technol. 179, 1–7.

Liu, W.J., Zeng, F.X., Jiang, H., Yu, H.Q., 2011. Total recovery of nitrogen andphosphorus from three wetland plants by fast pyrolysis technology. Bioresour.Technol. 102, 3471–3479.

Liu, X., Huang, S.L., Tang, T.F.Z., Liu, X.G., Scholz, M., 2012. Growth characteristicsand nutrient removal capability of plants in subsurface vertical flowconstructed wetlands. Ecol. Eng. 44, 189–198.

Lombardo, P., Cooke, D.G., 2003. Ceratophyllum demersum—phosphorusinteractions in nutrient enriched aquaria. Hydrobiologia 497, 79–90.

Longhi, D., Bartoli, M., Viaroli, P., 2008. Decomposition off our macrophytes inwetland sediments: organic matter and nutrient decay and associated benthicprocesses. Aquat. Bot. 89, 303–310.

Lu, Q., He, Z.L., Graetz, D.A., Stoffella, P.J., Yang, X.E., 2010. Phytoremediation toremove nutrients and improve eutrophic stormwaters using water lettuce(Pistia stratiotes L.). Environ. Sci. Pollut. R. 17, 84–96.

Lu, Q., He, Z.L., Graetz, D.A., Stoffella, P.J., Yang, X.E., 2011. Uptake and distributionof metals by water lettuce (Pistia stratiotes L.). Environ. Sci. Pollut. R. 18,978–986.

Maddison, M., Mauring, T., Remm, K., et al., 2009. Dynamics of Typha latifolia L.populations in treatment wetlands in Estonia. Ecol. Eng. 35, 258–264.

Mareckova, M.S., Petrusek, A., Kvet, J., 2009. Biomass production and nutrientaccumulation in Sparganium emersum Rehm. after sediment treatment withmineral and organic fertilizers in three mesocosm experiments. Aquat. Ecol.43, 903–913.

Masi, F., 2009. Water reuse and resources recovery: the role of constructed

wetlands in the Ecosan approach. Desalination 246, 27–34.

Meuleman, A.F.M., van Logtestijn, R., Rijs, G.B.J., Verhoeven, J.T.A., 2003. Water andmass budgets of a vertical-flow constructed wetland used for wastewatertreatment. Ecol. Eng. 20, 31–44.

nginee

O

R

R

R

R

R

S

S

S

S

S

S

S

T

T

U

V

V

V

Oenanthe javanica growth and decay in the ecological floating bed system. J.Environ. Sci-CHINA 22, 1710–1717.

Zhu, L.D., Li, Z.H., Ketola, T., 2011. Biomass accumulations and nutrient uptake ofplants cultivated on artificial floating beds in China’s rural area. Ecol. Eng. 37,1460–1466.

X. Zhou et al. / Ecological E

kurut, T.O., 2001. Plant growth and nutrient uptake in a tropical constructedwetland. In: Vymazal, J. (Ed.), Transformations of Nutrients in Natural andConstructed Wetlands. Backhuys Publishers, Leiden, The Netherlands, pp.451–462.

an, N., Agami, M., Oron, G., 2004. A pilot study of constructed wetlands usingduchweed (Lemna gibba L.) for treatment of domestic primary effluent inIsrael. Water Res. 38, 2241–2248.

eddy, K.R., DeBusk, W.F., 1987. Nutrient storage capabilities of aquatic andwetland plants. In: Reddy, K.R., Smith, W.H. (Eds.), Aquatic Plants for WaterTreatment and Resource Recovery. Magnolia Publishing, Orlando, FL, USA, pp.337–357.

eddy, K.R., Tucker, J.C., 1983. Productivity and nutrient uptake of water hyacinth,Eichhornia crassipes I. effect of nitrogen source. Econ. Bot. 37, 237–247.

eddy, K.R., 1983. Fate of nitrogen and phosphorus in a wastewater retentionreservoir containing aquatic macrophytes. J. Environ. Qual. 12, 137–141.

iis, T., Jensen, K.S., Vestergaard, O., 2000. Plant communities in lowland Danishstreams: species composition and environmental factors. Aquat. Bot. 66,255–272.

ágová-Marecková, M., Petrusek, A., Kvet, J., 2009. Biomass production andnutrient accumulation in Sparganium emersum Rehm after sedimenttreatment with mineral and organic fertilizers in three mesocosmexperiments. Aquatic Ecol. 43, 903–913.

artoris, J.J., Thullen, J.S., Barber, L.B., et al., 2000. Investigation of nitrogentransformations in a southern California constructed wastewater treatmentwetland. Ecol. Eng. 14, 49–65.

cheffer, M., Hosper, S.H., Meijer, M.L., Jeppesen, E., Moss, B., 1993. Alternativeequilibria in shallow lakes. Trends Ecol. Evol. 8, 275–279.

chindler, D.W., Hecky, R.E., 2009. Eutrophication: more nitrogen data needed.Science 324, 721–722.

eto, M., Takamura, N., Iwasa, Y., 2013. Individual and combined suppressiveeffects of submerged and floating-leaved macrophytes on algal blooms. J.Theor. Biol. 319, 122–133.

herwood, L.J., Qualls, R.G., 2001. Stability of phosphorus withina wetland soilfollowing ferric chloride treatment to control eutrophication. Environ. Sci.Technol. 35, 4126–4131.

teffenhagen, P., Zak, D., Schulz, K., Timmermann, T., Zerbe, S., 2012. Biomass andnutrient stock of submersed and floating macrophytes in shallow lakes formedby rewetting of degraded fens. Hydrobiologia 692, 99–109.

anner, C.C., Headley, T.R., 2011. Components of floating emergent macrophytetreatment wetlands influencing removal of stormwater pollutants. Ecol. Eng.37, 474–486.

anner, C.C., 1996. Plants for constructed wetland treatment systems-a comparisonof the growth and nutrient uptake of eight emergent species. Ecol. Eng. 7, 9–83.

.S. Environmental Protection Agency. 2000. Ambient water quality criteriarecommendations. Information supporting the development of state and tribalnutrient criteria. Rivers and streams in nutrient ecoregion XII. EPA822-B-00-021. U.S. Environmental Protection Agency, Office of water.Washington. D.C.

ereecken, H., Baetens, J., Viaene, P., Mostaert, F., Meire, P., 2006. Ecologicalmanagement of aquatic plants: effects in lowland streams. Hydrobiologia 570,205–210.

ymazal, J., Kröpfelová, L., 2008. Wastewater Treatment in Constructed Wetlandswith Horizontal Sub-Surface Flow. Springer, Dordrecht.

ymazal, J., Kröpfelová, L., Svehla, J., et al., 2010. Can multiple harvest ofaboveground biomass enhance removal of trace elements in constructedwetlands receiving municipal sewage? Ecol. Eng. 36, 939–945.

ring 101 (2017) 107–119 119

Vymazal, J., 1995. Algae and Element Cycling in Wetlands. Lewis Publishers,Chelsea, Michigan, pp. 698.

Vymazal, J., 2001. Types of constructed wetlands for wastewater treatment: theirpotential for nutrient removal. In: Vymazal, J. (Ed.), Transformations ofNutrients in Natural and Constructed Wetlands. Backhuys Publishers, Leiden,The Netherlands, pp. 1–93.

Vymazal, J., 2007. Removal of nutrients in various types of constructed wetlands.Sci. Total Environ. 380, 48–65.

Vymazal, J., 2011. Plants used in constructed wetlands with horizontal subsurfaceflow: a review. Hydrobiologia 674, 133–156.

Wetzel, R.G., 2001. Limnology, 3rd ed. Academic Press, San Diego, pp. 1006.Wu, Z.H., Dan, Y., Tu, M.H., Qiang, W., We, n.X., 2007. Interference between two

floating-leaved aquatic plants: Nymphoides peltata and Trapa bispinosa. Aquat.Bot. 86, 316–320.

Wu, H.M., Zhang, J., Li, P.Z., Zhang, J.Y., Xie, H.J., Zhang, B., 2011. Nutrient removalin constructed microcosm wetlands for treating polluted river water innorthern China. Ecol. Eng. 37, 560–568.

Xu, W.W., Hu, W.P., Deng, J.C., Zhu, J.G., Li, Q.Q., 2014. Effects of harvestmanagement of Trapa bispinosa on an aquatic macrophyte community andwater quality in a eutrophic lake. Ecol. Eng. 64, 120–129.

Yang, X.E., Wu, X., Hao, H.L., He, Z.L., 2008a. Mechanisms and assessment of watereutrophication. J. Zhejiang Univ. Sci. B 9, 197–209.

Yang, Y.G., He, Z.L., Lin, Y.J., Phlips, E.J., Yang, J., Chen, G.C., Stoffella, P.J., Powell,C.A., 2008b. Temporal and spatial variations of nutrients in the Ten Mile Creekof South Florida, USA and effects on phytoplankton. J. Environ. Monit. 10,508–516.

Yang, Y.G., He, Z.L., Lin, Y.J., Phlips, E.J., Stoffella, P.J., Powell, C.A., 2009. Temporaland spatial variations of Cu, Cd, Pb, and Zn in Ten Mile Creek in South Florida,USA. J. Water Environ. Res. 81, 40–50.

Yang, Y.G., He, Z.L., Wang, Y.B., Fan, J.J., Liang, Z.B., Stoffella, P.J., 2013. Dissolvedorganic matter in relation to nutrients (N and P) and heavy metals in surfacerunoff water as affected by temporal variation and land uses-A case study fromIndian River Area, south Florida, USA. Agr. Water Manage. 118, 38–49.

Zhang, D., Gersberg, R.M., Kate, T.A.S., 2009. Constructed wetlands in China. Ecol.Eng. 35, 1367–1378.

Zhao, F.L., Xi, S., Yang, X.E., Yang, W.D., Li, J.J., Gu, B.H., He, Z.L., 2012a. Purifyingeutrophic river waters with integrated floating island systems. Ecol. Eng. 40,53–60.

Zhao, F.L., Yang, W.D., Zeng, Z., Li, H., Yang, X.E., He, Z.L., Gu, B.H., Rafiq, M.T., Peng,H.Y., 2012b. Nutrient removal efficiency and biomass production of differentbioenergy plants in hypereutrophic water. Biomass Bioenerg. 42, 212–218.

Zheng, Z.C., Li, T.X., Zeng, F.F., Zhang, X.Z., Yu, H.Y., Wang, Y.D., Liu, T., 2013.Accumulation characteristics of and removal of nitrogen and phosphorus fromlivestock wastewater by Polygonum hydropiper. Agr. Water Manage. 117,19–25.

Zhou, X.H., Wang, G.X., 2010. 2010: Nutrient concentration variations during