Top Banner
Self-purification ability of a resurgence stream Roberta Vagnetti a , Paola Miana b , Mario Fabris b , Bruno Pavoni a, * a Dipartimento di Scienze Ambientali, Universit a Ca’Foscari di Venezia, Calle Larga S. Marta, 2137-30123 Venice, Italy b VESTA S.p.A., Venezia Servizi Territoriali Ambientali, Palazzo Bonfadini, Cannaregio 462-30121 Venezia, Italy Received 5 July 2002; received in revised form 10 April 2003; accepted 16 April 2003 Abstract The self-purification ability of a resurgence stream has been investigated by taking samples along the course of a channeled tract made up of a first part in beaten soil (3.3 km) and a second in concrete (7.2 km). The study has been conducted by statistically processing pre-existent data, acquired monthly by analyzing waters at the beginning and at the end of the whole canal for 6 years, from 1995 to 2000 (historic data), and by performing specific experiments (recent data) to evaluate differently the self-purification capacity of the beaten soil section and that in concrete. A significant abatement of concentrations has been observed from historic data for ammonium, phosphates, turbidity, heavy metals and bacteria. From the recent data, all these parameters seem to decrease in the beaten soil tract. Whereas significant further decreases in the concrete tract were observed only for ammonium, phosphates and bacteria. For other pa- rameters, e.g. pH, dissolved oxygen, chlorides, fluorides, sodium, and sulfates, a significant increase was observed from the historic data. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Self-purification; Resurgence stream; Nutrients; Metals; Turbidity; Bacteria 1. Introduction The water environment reacts to the input of pol- luting substances by means of a number of mechanisms aiming to restore its original conditions. This process, referred to as self-purification, actually consists of a re- cycling of materials (Vismara, 1998). More precise defi- nition for self-purification could be: ‘‘self-purification means the partial or complete restoration, by natural processes, of a stream pristine condition following the introduction (usually through the agency of man) of foreign matter sufficient in quality and quantity to cause a measurable change in physical, chemical and/or bio- logical characteristics of the stream’’ (Benoit, 1971). This transformation produces compounds having a less neg- ative impact than the starting ones. The natural self- purification process is therefore consisting of various complex phenomena involving numerous physical, chem- ical and biological factors acting and interacting more or less effectively. 1.1. Physical processes Dilution is an important component of self-purifica- tion, for it allows the achievement of suitable concen- trations for biological assimilation (Vismara, 1998). Adsorption is the binding of molecules and ions which are present in solution to solid particles. During the adsorption process other ions are displaced from the solid matrix into the solution (Benoit, 1971). Clays and other colloidal particles (e.g. oxides–hydroxides of Fe and Mn) can adsorb several organic and/or inorganic solutes. Furthermore the solid phase can be a proper support for bacterial degradation (Vismara, 1998). In Chemosphere 52 (2003) 1781–1795 www.elsevier.com/locate/chemosphere * Corresponding author. Tel.: +39-41-234-8522; fax: +39-41- 2348582. E-mail address: [email protected] (B. Pavoni). 0045-6535/03/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0045-6535(03)00445-4
15

Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Aug 23, 2014

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Chemosphere 52 (2003) 1781–1795

www.elsevier.com/locate/chemosphere

Self-purification ability of a resurgence stream

Roberta Vagnetti a, Paola Miana b, Mario Fabris b, Bruno Pavoni a,*

a Dipartimento di Scienze Ambientali, Universit�aa Ca’Foscari di Venezia, Calle Larga S. Marta, 2137-30123 Venice, Italyb VESTA S.p.A., Venezia Servizi Territoriali Ambientali, Palazzo Bonfadini, Cannaregio 462-30121 Venezia, Italy

Received 5 July 2002; received in revised form 10 April 2003; accepted 16 April 2003

Abstract

The self-purification ability of a resurgence stream has been investigated by taking samples along the course of a

channeled tract made up of a first part in beaten soil (3.3 km) and a second in concrete (7.2 km). The study has been

conducted by statistically processing pre-existent data, acquired monthly by analyzing waters at the beginning and at

the end of the whole canal for 6 years, from 1995 to 2000 (historic data), and by performing specific experiments (recent

data) to evaluate differently the self-purification capacity of the beaten soil section and that in concrete. A significant

abatement of concentrations has been observed from historic data for ammonium, phosphates, turbidity, heavy metals

and bacteria. From the recent data, all these parameters seem to decrease in the beaten soil tract. Whereas significant

further decreases in the concrete tract were observed only for ammonium, phosphates and bacteria. For other pa-

rameters, e.g. pH, dissolved oxygen, chlorides, fluorides, sodium, and sulfates, a significant increase was observed from

the historic data.

� 2003 Elsevier Ltd. All rights reserved.

Keywords: Self-purification; Resurgence stream; Nutrients; Metals; Turbidity; Bacteria

1. Introduction

The water environment reacts to the input of pol-

luting substances by means of a number of mechanisms

aiming to restore its original conditions. This process,

referred to as self-purification, actually consists of a re-

cycling of materials (Vismara, 1998). More precise defi-

nition for self-purification could be: ‘‘self-purification

means the partial or complete restoration, by natural

processes, of a stream pristine condition following the

introduction (usually through the agency of man) of

foreign matter sufficient in quality and quantity to cause

a measurable change in physical, chemical and/or bio-

logical characteristics of the stream’’ (Benoit, 1971). This

transformation produces compounds having a less neg-

*Corresponding author. Tel.: +39-41-234-8522; fax: +39-41-

2348582.

E-mail address: [email protected] (B. Pavoni).

0045-6535/03/$ - see front matter � 2003 Elsevier Ltd. All rights res

doi:10.1016/S0045-6535(03)00445-4

ative impact than the starting ones. The natural self-

purification process is therefore consisting of various

complex phenomena involving numerous physical, chem-

ical and biological factors acting and interacting more or

less effectively.

1.1. Physical processes

Dilution is an important component of self-purifica-

tion, for it allows the achievement of suitable concen-

trations for biological assimilation (Vismara, 1998).

Adsorption is the binding of molecules and ions which

are present in solution to solid particles. During the

adsorption process other ions are displaced from the

solid matrix into the solution (Benoit, 1971). Clays and

other colloidal particles (e.g. oxides–hydroxides of Fe

and Mn) can adsorb several organic and/or inorganic

solutes. Furthermore the solid phase can be a proper

support for bacterial degradation (Vismara, 1998). In

erved.

Page 2: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

1782 R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795

particular, heavy metals are frequently adsorbed onto

the surface of suspended particulate matter instead of

being dissolved in the liquid phase as free or complexed

ions. This process plays a very important role in the

cycles of natural materials in the environment. Many

species are accumulated in the suspended solids or in

sediments, and partitioning between the solid and liquid

phase influences the transport and fate of contaminants

in the water bodies (Suzuky, 1997).

Sedimentation is one of the most important self-

purification mechanisms, especially in lakes, lagoons

and lentic waters. By this means suspended particles are

removed along with their adsorbed ions and molecules,

thus removing soluble materials from the water column.

Pollutants generally remain on the bottom, but can be

remobilized during period of increased water flow or

turbulence, with the concurrent activities of desorption

and ionic exchange, which are mainly favored by an-

aerobic conditions (Vismara, 1998). Sediment and sus-

pended particles are depositories of large amounts of

heavy metals, which can be present as distinct com-

pounds or associated with clays, insoluble humic sub-

stances or iron and manganese oxides. Accumulation of

heavy metals in bottom sediments is one of the most

effective factors in self-purification. However, since this

is a reversible process, metal accumulation can be con-

sidered a constant potential for secondary pollution,

caused by resuspension and release phenomena (Linnik

and Zubenko, 2000).

Volatilization (transfer of pollutants from water to

the gas phase) causes a permanent removal of com-

pounds from the liquid phase. From the vapor pressure

of the compounds, its tendency to volatilize can be in-

ferred, however its solubility and other features can also

be important (Brusseau and Bohn, 1996).

1.2. Chemical processes

Acid–base reactions maintain the natural water pH

by neutralizing acid and base pollutants. Water buffer

capacity is strongly related to alkalinity which is the

water�s content of carbonate and hydroxide species

(Vismara, 1998).

Within the very high number of redox reactions, very

important are the oxidation of organic matter which

consumes oxygen and the oxidation of ammonia to ni-

trate, which is important for nitrogen to be assimilated

by plants and therefore removed. It must be considered

that many of the most important redox reactions are

catalyzed by microorganisms or governed by other bio-

logical processes (Manahan, 1994).

Precipitation reactions, depending on the solubility

product of several compounds, are very important for

removing ions from the liquid phase. Many precipitation

reactions, such as the formation of salts of phosphates

or carbonates, involve the removal of cations from so-

lution.

Processes leading to the formation of aggregates

from colloidal suspensions are also very important and

result in sediment formation and water clarification

(Manahan, 1994). These processes are coagulation and

flocculation, which are complex phenomena of a physical

and chemical nature, and are not yet well understood.

1.3. Biological processes

Bacterial degradation is the most important removal

process for organic substances and some inorganic sub-

stances, especially nitrogen and phosphorus compounds.

Bacteria obtain the necessary energy for their survival

by redox reactions, which transform organic matter and

nutrients.

Plants can protect an ecosystem by assimilation and

removal of a portion of the macronutrients present in

the water body (Chambers and Prepas, 1994; Volterra

and Mancini, 1994; Cunningham and Davi, 1996; Borin

and Marchetti, 1997). Mostly nitrogen and phosphorus

compounds are involved, but many plants can accu-

mulate heavy metals and toxic substances. It is necessary

to distinguish between the phytoremediation of inor-

ganic elements and toxic organic compounds (Meagher,

2001): elemental pollutants are essentially immutable

by any biological and physical process, whereas organic

substances are mineralized into relatively non-toxic

constituents. Assimilation is higher in summer, when

nutrients are stored in expectation of the winter season

(Volterra and Mancini, 1994). Finally it must be con-

sidered that the rhizosphere hosts a very abundant

population of microorganisms and that vegetation itself

can act as a filter for the suspended particulate matter.

From a preliminary examination of historic analyti-

cal results, a considerable decrease of some substances

was observed in water samples taken at the beginning

and at the end of a canal. The aim of the present in-

vestigation was to establish, by means of a statistical

treatment of pre-existent data (historic data) and some

ad hoc confirmation experiments (recent data), for which

parameters the abatement is significant and to formulate

possible interpretations.

2. Materials and methods

2.1. Study site

The stream investigated in this study is an artificial

canal (averagely 2 m deep, 5 m large with a flow rate of

about 2.2 m3/s) feeding a municipal purification plant

for household (285 000 people served) and industrial use.

Main treatments carried out are coagulation, activated

carbon adsorption, clarification, filtration and disinfec-

Page 3: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795 1783

tion. The canal conveys water from the resurgence river

Sile, in Quarto d�Altino to Favaro Veneto (Veneto Re-

gion, North–East of Venice, Italy). Surrounding area is

only used for agriculture and along the course of the

canal (10.5 km) some water can be derived for irrigation.

The first section of the canal (3.3 km) is in beaten soil

and the second section (7.2 km) is in concrete. Various

plants species are present, mainly in the beaten soil tract:

Apium nodiflorum (abundance 0–1), Callitriche stagnalis

(ab. 2–3), Ceratophillum demersus (ab. 1–2), Elodea

canadensis (ab. 3–4), Potamogeton natans (ab. 3–4),

Potamogeton pectinatus (ab. 3–4), Ranunculus fluitans

(ab. 3–4), Vallisneria spiralis (ab. 1–2) (Marconato,

2003). The extended biotic index value for the canal

(beginning of the canal) is 8, i.e. the water is slightly

polluted (Marconato, 2003). Fish species present are

Anguilla anguilla, Cobitis taenia, Esox lucius, Padogobius

martensii, Rutilus erythrophthalmus, Scardinius erythro-

phthalmus (Marconato et al., 2000).

2.2. Statistical analyses

The historic data set consisted of pre-existent ana-

lytical results of water samples collected at stations lo-

cated at the beginning and at the end of the canal during

6 years (1995–2000) with a monthly frequency. These

data were processed with the Statistical t Test using

paired data (Piccolo, 2000) to discriminate significant

differences between the pairs of values obtained for any

single parameter at the beginning and at the end of the

canal in the same sampling session. A 5% significance

level was used. A parametric method was chosen after

checking data normality.

2.3. Sampling design

To validate the results obtained by means of the

statistical data processing, some additional experiments

were carried out. Sampling sessions took place on 27th

March, 27th June, and 12th September 2001 (recent

data).

The sampling stations were the following:

1. Quarto d�Altino (Head of the canal).

2. End of canal tract # 1 (Beaten soil).

3. Siphon.

4. Favaro Veneto (End of the canal).

A map of the canal is shown in Fig. 1.

Sampling times were established according to the

water velocity. Sampling station 2 was selected in order

to possibly distinguish the self-purification ability of

the canal tract bordered by beaten soil from that framed

by concrete. Sampling station 3 was chosen to detect

the influence of a siphon along the course of the canal

enabling the crossing of the river Zero. During the

samplings it was noticed that the siphon was partly

plugged with sediment and that an intense resuspension

was evident in the tract downstream. In order to es-

tablish the sampling times necessary to collect the same

water parcel at the beginning and at the end of its

travel along the canal, the water velocity was prelimi-

narily estimated. During each sampling session the flow

rate was measured by means of a float moving at a

water depth of about 20 cm below the surface to avoid

wind influence. Water samples (one sample for micro-

biological analyses, one for chemical analyses) were

collected by dipping the containers about 1 m below

the surface. Containers for microbiological analyses

were previously sterilized, whereas those for chemical

analyses were previously cleaned in the laboratory and

then simply rinsed with the water to be sampled.

Samples were then transferred to the laboratory by

means of a portable refrigerated container and kept at

4 �C before analyses, which were performed the same

or the following day.

The parameters selected for the analyses were those

showing a significant abatement. In addition, nitrates

and nitrites were also analyzed to estimate the total

inorganic nitrogen budget. In this study the organo-

chlorine compounds, such as 1,1,1-trichloroethane and

tetrachloroethylene were not included because they

were considered to be clearly lost to the atmosphere

by evaporation.

All microbiological and chemical analyses (historic

and recent data) were performed according to the pro-

cedures reported in the manuals ‘‘Standard Methods for

the Examination of Water and Wastewater’’ (AWWA,

1998) and ‘‘Analytical Methods for Waters’’ (CNR,

1994).

2.4. Microbiological analyses

Water samples were filtered through a membrane

with a pore size of about 0.45 lm, which traps most of

the bacteria on its surface. The membrane was then

placed in a pad saturated with a medium selected to

favor growth and differentiation of organisms. Media

used were M-endo agar for total coliforms, M-FC agar

for fecal coliforms, KF-streptococcus agar for fecal

streptococci and plate count agar for heterotrophic count

at 22 and 36 �C. The bacterial number was then reported

as colony forming units (CFU) per unit of water volume.

Precisions for microbiological analyses, calculated as

coefficient of variations from six replicates, were: 8% for

heterotrophic count at 22 �C, 13% for heterotrophic

count at 36 �C, 21% for total coliforms, 3% for fecal

coliforms, 2% for fecal streptococci and 15% for clo-

stridia spores. Increases and decreases (percent), for

each parameter, were considered significant when higher

than coefficients of variations.

Page 4: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Fig. 1. Map of the canal (North–East of Venice). Station 1 (head of the canal), station 2 (end of beaten soil tract), station 3 (siphon)

and station 4 (end of the canal, purification plant).

1784 R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795

2.5. Physical–chemical analyses

Turbidity was measured by the Nephelometric

Method using a HACH 2100 AN instrument, provided

with a tungsten source and a photoelectric detector.

Ammonium concentration was measured by the phe-

nol-hypochlorite method, using a spectrophotometer

UV–VIS Lambda 2, Perkin Elmer, operating at 660 nm.

Nitrates concentration was determined spectropho-

tometrically at 220 nm, after sample acidification to

eliminate carbonates.

Nitrites concentration was determined through for-

mation of a reddish purple azo dye (measured spectro-

photometrically at 540 nm) produced at a pH 2.0–2.5 by

coupling diazotized sulfanilamide with N-(1-naphtyl)-

ethylenediamine dihydrochloride.

Page 5: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795 1785

Phosphates concentration was measured after trans-

formation of polyphosphates into orthophosphates

with sulfuric acid and potassium persulphate. Ortho-

phosphates concentration was measured spectrophoto-

metrically at 650 nm, after reaction with ammonium

molybdate and potassium antimonyl tartrate in acid

medium to form phosphomolybdic acid, and further

reduction to the intensely colored molybdenum blue by

ascorbic acid.

Dissolved organic carbon (DOC) concentration was

determined potentiometrically as CO2 using a selective

Table 1

Average and percent abatement of the parameters analyzed at the be

data)

Average difference

pH 0.12

Conductivity )2 lS/cmTurbidity )3.3 NTU

Alkalinity 1 mg/dm3 CaCO3

Ammonium )0.16 mg/dm3

Sodium 0.70 mg/dm3

Potassium 0.04 mg/dm3

Calcium 0.49 mg/dm3

Magnesium )0.19 mg/dm3

Total hardness )0.5 mg/dm3 CaCO3

Dissolved oxygen 0.3 mg/dm3

Fluorides 0.01 mg/dm3

Chlorides 0.55 mg/dm3

Nitrites 0.05 mg/dm3

Nitrates 0.01 mg/dm3

Total phosphates )0.04 mg/dm3

Sulfates 0.57 mg/dm3

D.O.C. 0.1 mg/dm3

Al )65.9 lg/dm3

Cr )0.4 lg/dm3

Fe )38.2 lg/dm3

Pb )0.2 lg/dm3

Mn )3.9 lg/dm3

Ba )3.0 lg/dm3

B 0.8 lg/dm3

Cu )1.1 lg/dm3

Zn )3.8 lg/dm3

Chloroform )0.28 lg/dm3

1,1,1-Trichloroethane )0.07 lg/dm3

Trichloroethylene )0.01 lg/dm3

Dichlorobromomethane )0.11 lg/dm3

Tetrachloroethylene )0.18 lg/dm3

Dibromochloromethane )0.04 lg/dm3

Bromoform )0.01 lg/dm3

Heterotrophic count at 22 �C )8001 UFC/cm3

Heterotrophic count at 36 �C )6580 UFC/cm3

Total coliforms )19 661 UFC/100 cm3

Fecal coliforms )4382 UFC/100 cm3

Fecal streptococci )371 UFC/100 cm3

Sulfite-reducing clostridia spores )3 UFC/100 cm3

Negative difference¼ abatement; positive¼ increase. Results of t-test (5value.

electrode. Samples were previously filtered, acidified to

remove carbonates/bicarbonates and finally oxidized

with persulphate and UV irradiation.

Heavy metals concentration was determined by an

Atomic Absorption Spectrophotometer (3030, Perkin

Elmer) after acidification of the non-filtered sample with

nitric acid. The following wavelengths were used: Al,

309.3 nm; Fe, 248.3 nm; Mn, 279.5 nm; Pb, 324.7 nm;

Zn, 213.19 nm. Precisions for chemical analyses, calcu-

lated as coefficient of variations from six replicates,

were: 2% for turbidity, 2% for ammonium, 3% for

ginning and at the end of the canal from 1995 to 2000 (historic

% abatement n t Significance

0.15 81 6.57 Significant

)0.4 81 )1.41 Non-sign.

)61 80 )7.26 Significant

0.5 79 0.26 Non-sign.

)69 81 )13.25 Significant

10 65 13.40 Significant

3 65 1.28 Non-sign.

0.7 81 0.52 Non-sign.

)0.8 81 )1.62 Non-sign.

)0.2 80 )0.64 Non-sign.

3 73 1.67 Non-sign.

10 79 2.16 Significant

6.6 80 9.20 Significant

31 80 6.21 Significant

0.07 80 0.40 Non-sign.

)25 77 )6.51 Significant

1.3 80 3.72 Significant

8 73 0.36 Non-sign.

)58.3 75 )4.41 Significant

)27 74 )1.79 Non-sign.

)50.2 74 )5.73 Significant

)25 74 )2.33 Significant

)43 70 )4.49 Significant

)6 22 )1.25 Non-sign.

2 19 0.42 Non-sign.

)46 31 )2.49 Significant

)35 28 )2.35 Significant

)82 72 )1.20 Non-sign.

)35 71 )6.40 Significant

)11 72 )1.78 Non-sign.

)92 72 )1.40 Non-sign.

)30 72 )10.64 Significant

)80 72 )1.26 Non-sign.

)4 72 )0.28 Non-sign.

)71.1 75 )5.10 Significant

)77.2 76 )3.32 Significant

)86.8 76 )10.20 Significant

)90.8 76 )10.02 Significant

)90.5 75 )11.11 Significant

)60 69 )3.84 Significant

% significance level): n ¼ number of observations, t ¼ statistic t

Page 6: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

1786 R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795

nitrites, 1% for nitrates, 18% for phosphates, 2% for

DOC and 2% for heavy metals. Increases and decreases

(percent), for each parameter, were considered signifi-

cant when higher than coefficients of variations.

3. Results

Average abatement and the percent abatement for

historic data are shown in Table 1. For a clear identifi-

cation of the most important parameters involved in the

self-purification of the canal, these data were represented

in a Box Whisker Plot (Fig. 2). By means of the t Test

using paired data, the following parameters were found

as having a significant decrease between the beginning

and the end the canal (5% significance level; t values are

reported in Table 1): turbidity, ammonia, total phos-

phates, Al, Fe, Pb, Mn, Cu, Zn, 1,1,1-trichloroethane,

tetrachloroethylene, heterotrophic count at 22 and 36

�C, total and fecal coliforms, fecal streptococci, sulfite-

reducing clostridia spores. For some others (pH,

chlorides, fluorides, nitrites, sodium, and sulfates) a

significant increase was observed. Also dissolved oxygen

showed a significant increase with a 10% significance

level.

Statistical results are evident also in Fig. 3, which

shows the trends of pH (Fig. 3(a)), oxygen (Fig. 3(b)),

nitrates (Fig. 3(c)), nitrites (Fig. 3(d)) and phosphates

(Fig. 3(e)), from historic data (1995–2000). Dissolved

oxygen, nitrites concentration and pH were generally

higher at station 4 than at station 1 (particularly in

Box Whisker Plot o

-30

-20

-10

0

10

20

pH*1

0

Tur

bidi

ty

Am

mon

ium

Sod

ium

*10

Pot

assi

um*1

0

Oxy

gen

Flu

orid

e*10

0

Chl

orid

e*10

Nitr

ites*

100

Pho

spha

tes*

100

Sul

fate

s

Alu

min

um/1

0

Means+SDMeans-SD

Means+SEMeans-SE

Means

Fig. 2. Box Whisker Plot of differences for routinely acquired histori

Table 1 for units.

summer for nitrites). Phosphates were lower at station 4

than at station 1 and nitrates were also lower, but only

in summer. However for nitrates and nitrites a seasonal

trend was more evident.

Analytical results (recent data) are reported in Tables

2–4, and plotted in Figs. 4–6.

For ammonium, phosphates, turbidity, heavy metals

and bacteria a net abatement was observed from station

1 to station 2. Only ammonium, phosphates and bac-

teria decreased from station 2 to station 4. Whereas for

turbidity and heavy metals an increase was observed

from station 2 to station 4 and also from station 2 to

station 3 on 12th September. Concentration of nitrites

grew significantly from station 1 to station 4. Nitrates

and dissolved organic carbon did not show a particular

trend.

An estimation of the budget for all species of inor-

ganic nitrogen was attempted, in order to establish if

these species were simply transforming from one species

into another or if a total inorganic nitrogen decrease

also occurred. Concentrations of inorganic nitrogen for

historic data is reported in Table 5 and for recent data in

Table 6.

4. Discussion

4.1. Nitrogen

From historic data, a systematic decrease of total

inorganic nitrogen during the summer period (especially

f differences

Iron

/10

Man

gane

se

Cop

per

Zin

c

1,1,

1Tric

hlor

oet*

100

Tet

rach

loro

eth*

10

Het

erC

ount

22˚C

/100

0

Het

erC

ount

36˚C

/100

0

Tot

alC

olifo

rm/1

0000

Fec

alC

lifor

m/1

000

Fec

alS

trep

t./10

0

Clo

strid

iaS

pore

s

c data: negative difference¼ abatement; positive¼ increase. See

Page 7: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Fig. 3. Trends of pH (a), oxygen (b), nitrates (c), nitrites (d) and phosphates (e) from 1995 to 2000 (historic data obtained from

monthly analyses) at stations 1 and 4 (see Fig. 1, map of the canal).

R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795 1787

Page 8: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Table 3

Concentrations of selected parameters obtained by analyzing water samples during sampling session II (27th June), at stations 1 (Quarto d�Altino), 2 (End soil tract) and 4 (Favaro

Veneto) (recent data)

St. Turbidity (NTU) Ammonium (mg/dm3) Nitrites (mg/dm3) Nitrates (mg/dm3) Total phosphates (mg/dm3) DOC (mg/dm3)

1 9.4 0.19 0.15 18.99 0.13 1.9

2 7.4 0.08 0.19 16.92 0.11 1.8

4 14.4 0.08 0.18 17.05 0.09 1.7

Aluminum (lg/dm3) Iron (lg/dm3) Lead (lg/dm3) Manganese (lg/dm3) Copper (lg/dm3) Zinc (lg/dm3)

1 42.5 87.6 <0.1 9.9 2.8 6.6

2 48.8 98.1 <0.1 11.0 2.5 4.7

4 266.0 207.0 <0.1 11.8 3.1 7.6

Heterotrophic count

at 22 �C (CFU/cm3)

Heterotrophic count

at 36 �C (CFU/cm3)

Total coliforms

(CFU/100 cm3)

Fecal coliforms

(CFU/100 cm3)

Fecal streptococci

(CFU/100 cm3)

Clostridia spores

(CFU/100 cm3)

1 80 000 20 000 97 400 7560 400 30

2 40 000 4000 39 000 3960 44 22

4 20 000 3000 24 800 780 19 20

Table 2

Concentrations of selected parameters obtained by analyzing water samples during sampling session I (27th March), at stations 1 (Quarto d�Altino), 2 (End soil tract) and 4 (Favaro

Veneto) (recent data)

St. Turbidity (NTU) Ammonium (mg/dm3) Nitrites (mg/dm3) Nitrates (mg/dm3) Total phosphates (mg/dm3) DOC (mg/dm3)

1 7.4 0.26 0.10 15.6 0.27 2.20

2 2.8 0.12 0.20 15.5 0.30 2.32

4 8.0 0.07 0.19 15.5 0.24 2.16

Aluminum (lg/dm3) Iron (lg/dm3) Lead (lg/dm3) Manganese (lg/dm3) Copper (lg/dm3) Zinc (lg/dm3)

1 162 145.0 <0.1 10.1 2.5 14.8

2 111 53.0 <0.1 6.0 1.8 10.2

4 352 147.5 <0.1 9.4 7.4 11.4

Heterotrophic count

at 22 �C (CFU/cm3)

Heterotrophic count

at 36 �C (CFU/cm3)

Total coliforms

(CFU/100 cm3)

Fecal coliforms

(CFU/100 cm3)

Fecal streptococci

(CFU/100 cm3)

Clostridia spores

(CFU/100 cm3)

1 150 000 100 000 33 400 6000 246 42

2 17 300 11 900 4750 700 71 15

4 50 400 50 000 10 000 360 3 30

1788

R.Vagnetti

etal./Chem

osphere

52(2003)1781–1795

Page 9: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Table

4

Concentrationsofselected

para

metersobtained

by

analyzing

watersa

mplesduring

sampling

session

III(12th

Sep

tember),

atstations1

(Quarto

d�A

ltino),

2(E

nd

soil

tract),

3(S

iphon)and

4(F

avaro

Ven

eto)(recen

tdata)

St.

Turb

idity(N

TU)

Ammonium

(mg/dm

3)

Nitrites(m

g/dm

3)

Nitra

tes(m

g/dm

3)

Totalphosp

hates(m

g/dm

3)

DOC

(mg/dm

3)

16.0

0.16

0.13

17.6

0.14

1.9

23.3

0.07

0.17

17.5

0.13

1.8

36.2

0.07

0.18

17.6

0.13

1.8

49.7

0.06

0.17

17.5

0.12

1.8

Aluminum

(lg/dm

3)

Iron

(lg/dm

3)

Lea

d(l

g/dm

3)

Manganese(l

g/dm

3)

Copper

(lg/dm

3)

Zinc(l

g/dm

3)

1103

93

<0.1

7.7

1.0

6.7

2100

76

<0.1

5.5

0.6

3.1

3332

155

<0.1

7.7

0.7

4.9

4487

238

<0.1

10.5

2.9

7.9

Heterotrophic

countat

22�C

(CFU/cm

3)

Heterotrophic

countat

36�C

(CFU/cm

3)

Totalco

lifo

rms

(CFU/100cm

3)

Fecalco

lifo

rms

(CFU/100cm

3)

Fecalstrepto

cocci

(CFU/100cm

3)

Clostridia

spores

(CFU/100cm

3)

13760

13200

82600

6000

750

83

22000

2440

12000

2440

100

56

32320

4680

5900

780

300

26

41680

2100

5100

550

70

32

R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795 1789

in June) was observed. This phenomenon can last until

the early autumn depending on the particular year. This

is evident from the data reported in Table 5. During the

other periods of the year, the inorganic nitrogen budget

can be considered closed: in other words the concen-

tration of total inorganic nitrogen remains constant.

This trend has been confirmed by the results of the ex-

periments (Table 6). The only total inorganic nitrogen

decrease was monitored in the sampling session of 27th

June. In the other sampling sessions the concentration at

station 1 is not different from the concentration at sta-

tion 4. The summer abatement of nitrogen species was

observed by several investigators (Elosegui et al., 1995;

Bratli et al., 1999; Jing et al., 2001) and it can be rea-

sonably ascribed to a greater assimilation by living or-

ganisms, particularly plants. In fact, in this season the

most efficient nutrient and pollutant uptake occurs, be-

cause higher air and water temperatures and a more

intense light irradiation favor higher plant productivity

(Volterra and Mancini, 1994). The vegetale species pre-

sent in the canal include also some which are used for

phytoremediation: e.g. Elodea canadensis, Potemogeton

natans, Ceratophyllum demersus (Volterra and Mancini,

1994). In addition, nitrogen can be definitively removed

by denitrification, a biological process which is depen-

dent on temperature and favored in the summer (Bratli

et al., 1999; De Crespin De Billy et al., 2000; Haag and

Kaupenjohann, 2001). It is evident from the recent data

(Table 6) that this reduction of nitrogen in summer is

mostly due to a reduction of nitrate concentrations. In

fact, nitrates accounted for more than 98% of the total

nitrogen species.

Considering the nitrogen species separately, a sig-

nificant ammonium abatement was observed both in the

historic data and in the experimental data. Considering

these latter, it can also be observed that this decrease is

more evident at the end of the beaten soil tract. This

phenomenon is considered normal in natural oxygen-

ated waters, and it was observed in all considered self-

purification studies (Elosegui et al., 1995; Lam-Leung

et al., 1996; Bratli et al., 1999; Jing et al., 2001). It

is caused by several mechanisms, including oxidation

(nitrification) and biological assimilation.

In parallel to an ammonium drop, a significant

increase of nitrites was observed both in the historic

data and in the experiments. In this case, the more

important increase of nitrites was monitored between

the beginning of the canal and the end of beaten soil

tract. It is known that the overall process of nitrogen

oxidation from ammonia to nitrate can be considered

a two step process, i.e. the oxidation from ammonia

to nitrite and from nitrite to nitrate. The concentra-

tion of the intermediate species nitrite depends on the

relative rates of the two steps. When the step from

nitrite to nitrate is somehow slowed down, the nitrite

concentration increases. According to Von der Wiesche

Page 10: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Sampling session I - Stations 1,2,4 - Nutrients

0.04

0.08

0.12

0.16

0.20

0.24

0.28

0.32

Station 1 Station 2 Station 4

NH4 (mg/dm3)NO2 (mg/dm3)NO3 (mg/dm3)*10-2

PO4 (mg/dm3)DOC (mg/dm3)*10-1

(a)

Sampling session I - Stations 1,2,4 - Turbidity and metals

-5

0

5

10

15

20

25

30

35

40

Station 1 Station 2 Station 4

Turbidity (NTU)Al (mg/dm3)*10-1

Fe (mg/dm3)*10-1

Mn (mg/dm3)Cu (mg/dm3)Zn (mg/dm3)

(b)

Sampling session I - Stations 1,2,4 - Bacteria

0

50

100

150

200

250

300

350

Station 1 Station 2 Station 4

Heter. count 22°C (CFU/cm3)*10-3

Heter. count 36°C (CFU/cm3)*10-3

Total coliforms (CFU/100cm3 )*10-2

Fecal coliforms (CFU/100cm3 )*10-1

Fecal streptococci (CFU/100cm3)Clostridia spores (CFU/100cm3)

(c)

Fig. 4. Recent data obtained by analyzing water samples during sampling session I (27th March) at stations 1 (Quarto d�Altino), 2

(End soil) and 4 (Favaro Veneto): (a) concentrations of ammonium, dissolved organic carbon, nitrates, nitrites and phosphates, (b)

turbidity, aluminum, copper, iron, manganese, and zinc, (c) heterotrophic count at 22 and 36 �C, total coliforms, fecal coliforms, fecal

streptococci, clostridia spores.

1790 R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795

Page 11: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Sampling session II - Station 1,2,4 - Nutrients

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

Station 1 Station 2 Station 4

NH4(mg/dm3)

NO2 (mg/dm3)

NO3 (mg/dm3)*10-2

PO4 (mg/dm3)

DOC (mg/dm3)*10-1

Sampling session II-Stations 1-2-4- Turbidity and metals

0

6

12

18

24

30

Station 1 Station 2 Station 4

Turbidity (NTU)Al (µg/dm3)*10-1

Fe (µg/dm3)*10-1

Mn (µg/dm3)Cu ( µg/dm3)Zn (µg/dm3)

Sampling sessionII-Stations1-2-4-Bacteria

-10

10

30

50

70

90

110

Station1 Station2 Station4

Heter. count 22°C (CFU/cm3)*10-3

Heter. count 36°C (CFU/cm3)*10-3

Total coliforms (CFU/100cm3)*10-3

Fecal coliforms (CFU/100cm 3)*10-2

Fecal streptococci (CFU/100cm3)*10-1

Clostridia spores (CFU/100cm3)

(a)

(b)

(c)

Fig. 5. Recent data obtained by analyzing water samples during sampling session II (27th June) at stations 1 (Quarto d�Altino), 2 (End

soil) and 4 (Favaro Veneto). Same parameters as Fig. 4.

R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795 1791

Page 12: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Sampling session III - Stations 1,2,3,4 - Nutrients

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

Station 1 Station 2 Station 3 Station 4

NH 4 (mg/dm3)

NO2 (mg/dm3)

NO3 (mg/dm3)*10-2

PO4 (mg/dm3)

DOC (mg/dm3)*10-1

Sampling session III - Stations 1,2,3,4 - Turbidity and metals

0

10

20

30

40

50

Station 1 Station 2 Station 3 Station 4

Turbidity (NTU)Al (µg/dm3 )*10-1

Fe (µg/dm3)*10-1

Mn (µg/dm3)Cu ( µg/dm3)Zn (µg/dm3)

Sampling session III-Stations 1,2,3,4 - Bacteria

0

20

40

60

80

100

120

140

Station 1 Station 2 Station 3 Station 4

Heter. count 22°C (CFU/cm3)*10-2

Heter. count 36°C (CFU/cm3)*10-2

Total coliforms (CFU/100cm3)*10-3

Fecal coliforms (CFU/100cm3)*10-2

Fecal streptococci (CFU/100cm3)*10-1

Clostridia spores (CFU/100cm3)

(a)

(b)

(c)

Fig. 6. Recent data obtained by analyzing water samples during sampling session III (12th September) at stations 1 (Quarto d�Altino),

2 (End soil), 3 (Additional point) and 4 (Favaro Veneto). Same parameters as Fig. 4.

1792 R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795

Page 13: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

Table 5

Total inorganic nitrogen (N–NH4 + N–NO2 + N–NO3) at the beginning and at the end of the canal (station 1 and 4) from 1995 to 2000

(historic data)

Date Tempera-

ture (�C)

Total nitrogen

(mg/dm3)

Difference

station 4� 1

(mg/dm3)

Date Tempera-

ture (�C)

Total nitrogen

(mg/dm3)

Difference

station 4� 1

(mg/dm3)1 4 1 4

16/01/95 8.1 4.17 4.15 )0.02 15/12/97 9.3 3.57 3.41 )0.1613/02/95 10.6 3.93 3.78 )0.15 12/01/98 10.1 3.64 3.46 )0.1827/02/95 10.3 4.09 3.92 )0.16 09/02/98 10.0 3.65 3.36 )0.2913/03/95 11.0 3.43 3.21 )0.23 09/03/98 13.2 3.75 3.50 )0.2510/04/95 14.0 4.12 3.93 )0.19 06/04/98 13.7 3.35 3.26 )0.0922/05/95 15.0 4.06 4.13 0.07 19/05/98 16.3 3.16 2.79 )0.3605/06/95 16.3 3.61 3.37 )0.24 01/06/98 17.4 3.24 2.95 )0.2903/07/95 18.3 4.09 3.46 )0.63 30/06/98 19.9 3.26 3.00 )0.2624/07/95 19.8 4.01 3.68 )0.33 27/07/98 19.6 3.33 3.31 )0.0328/08/95 18.0 4.25 3.56 )0.69 24/08/98 16.9 3.58 3.36 )0.2325/09/95 16.5 4.33 4.10 )0.23 21/09/98 15.7 3.53 3.26 )0.2730/10/95 13.6 4.50 4.46 )0.04 19/10/98 14.1 3.86 3.65 )0.2120/11/95 10.5 3.65 3.70 0.05 14/12/98 8.7 3.62 3.39 )0.2318/12/95 10.4 3.42 3.58 0.16 08/02/99 8.5 3.47 3.29 )0.1815/01/96 11.2 3.73 3.83 0.10 08/03/99 10.9 3.97 3.77 )0.1912/02/96 8.0 4.08 3.87 )0.21 12/04/99 13.9 3.50 3.37 )0.1311/03/96 10.1 3.96 3.88 )0.08 03/05/99 16.0 3.29 2.96 )0.3309/04/96 11.9 3.50 3.01 )0.48 31/05/99 19.2 3.04 2.61 )0.4306/05/96 14.0 3.33 3.02 )0.32 28/06/99 18.0 3.25 3.00 )0.2503/06/96 17.4 3.32 2.92 )0.40 26/07/99 18.5 3.42 3.00 )0.4101/07/96 17.4 3.30 2.98 )0.32 23/08/99 19.1 3.39 3.17 )0.2229/07/96 18.2 3.77 3.46 )0.31 18/10/99 13.0 3.73 3.13 )0.6026/08/96 18.0 3.66 3.27 )0.39 15/11/99 11.2 3.72 3.55 )0.1723/09/96 14.5 3.68 3.41 )0.27 13/12/99 9.6 3.71 3.50 )0.2121/10/96 13.7 3.68 2.81 )0.88 10/01/00 8.7 3.58 3.21 )0.3818/11/96 13.2 3.42 3.07 )0.36 07/02/00 11.4 3.56 3.51 )0.0516/12/96 9.9 3.51 3.32 )0.19 06/03/00 14.4 3.96 3.72 )0.2414/01/97 9.5 3.80 3.59 )0.21 17/04/00 16.5 3.26 3.14 )0.1217/02/97 9.6 3.65 3.35 )0.31 15/05/00 17.2 3.31 3.19 )0.1210/03/97 13.7 3.59 3.38 )0.21 05/06/00 18.6 3.50 3.32 )0.1905/05/97 16.3 3.32 2.89 )0.44 10/07/00 16.6 3.71 3.40 )0.3102/06/97 14.3 3.27 2.79 )0.48 07/08/00 15.9 3.46 3.44 )0.0201/07/97 18.5 3.11 2.92 )0.19 02/10/00 13.7 3.52 3.25 )0.2728/07/97 19.2 3.41 3.18 )0.22 30/10/00 11.9 4.20 4.11 )0.1025/08/97 18.2 3.55 3.35 )0.20 07/11/00 12.9 3.62 3.92 0.30

22/09/97 21.3 3.48 3.09 )0.40 08/11/00 12.3 4.10 3.67 )0.4317/11/97 11.7 3.58 3.34 )0.24 20/11/00 11.9 4.25 4.23 )0.02

Table 6

Nitrogen concentration in ammonium, nitrites, nitrates species and total inorganic nitrogen concentration (mg/dm3) from recent data

St. 27th March 27th June 12th September

N–NH4 N–NO2 N–NO3 Tot N–NH4 N–NO2 N–NO3 Tot N–NH4 N–NO2 N–NO3 Tot

1 0.20 0.03 3.52 3.76 0.15 0.05 4.29 4.48 0.13 0.04 3.98 4.14

2 0.09 0.06 3.50 3.66 0.06 0.06 3.82 3.94 0.05 0.05 3.95 4.06

3 0.05 0.05 3.98 4.08

4 0.05 0.06 3.50 3.61 0.06 0.05 3.85 3.97 0.05 0.05 3.95 4.05

R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795 1793

and Wetzel (1998), who observed a similar trend, this

phenomenon can occur when the temperature exceeds

the range 10–17 �C, in which the balance between the

two forms is maintained. At higher temperatures,

inhibition of nitrate formation is greater than the in-

hibition of nitrite formation. In the studied canal, tem-

peratures were often above that range in summer (Table

5). Observing the seasonal trend of nitrite concentra-

tion, we can notice winter minima and summer maxima

(Fig. 3(d)).

Page 14: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

1794 R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795

Considering all of the data, no significant difference

of nitrate concentrations was observed between the be-

ginning and the end of the canal, apart from the summer

period when a significant decrease of concentrations was

detected (Fig. 3(c)).

4.2. Phosphates

From the statistical processing of historic data (Table

1), it is evident that total phosphate concentrations

significantly decrease along the course of the canal (Fig.

3(e)). This is a typical phenomenon frequently observed

in natural purification processes (Benka-Coker and

Ojior, 1995; Elosegui et al., 1995; Lam-Leung et al.,

1996; Bratli et al., 1999; De Crespin De Billy et al., 2000;

Jing et al., 2001). For phosphorus removal the physical–

chemical processes, e.g. precipitation with calcium, ad-

sorption and sedimentation, are dominant. Phosphates

can be eliminated from water also by uptake by primary

producers or microbial decomposers (Elosegui et al.,

1995; Bratli et al., 1999). The abatement, for recent data,

was significant (percent higher than coefficient of varia-

tions) only on 27th June: in fact in the summer period

biologic assimilation is greater.

4.3. Bacteria

The concentration of bacteria in the water sharply

dropped from the beginning to the end of the canal as

a consequence of sedimentation and natural decay, but

an important abatement factor is also the filtration by

aquatic plants (mainly by rhizosphere) (Benka-Coker

and Ojior, 1995; Elosegui et al., 1995; Borin and

Marchetti, 1997).

4.4. Turbidity and heavy metals

The observed behavior for turbidity and heavy metal

concentrations was difficult to explain. A net abatement

was observed from the data collected for six years

(1995–2000). This decrease was attributed to sedimen-

tation or to a combination of adsorption and sedimen-

tation (Linnik and Zubenko, 2000). However data

investigated from the second half of 2000 only, showed

remarkable increases of these parameters. This phe-

nomenon non-remarkable overall, being limited to the

last period, is confirmed by all recent data. From the

analytical results of the experiments, it can be seen that

a decrease occurs in the beaten soil tract, whereas in

the sample corresponding to the siphon located 1 km

downstream at the end of this tract (station 3), some

resuspensions occur that causes turbidity and heavy

metal concentrations to exceed the values at the begin-

ning of the canal. The above observation is further

emphasized by a remarkable drop in concentrations

observed at the end of the beaten soil tract for other

parameters more sensitive to the self-purification pro-

cess, namely ammonia, phosphates and all bacteria.

For other parameters e.g. pH, chlorides, fluorides,

sodium, sulfates and dissolved oxygen, a significant in-

crease from the beginning to the end of the canal was

observed from the historic data. Whereas it is easy to

explain the increase of pH and dissolved oxygen (Fig.

3(a) and (b)) due to more photosynthesis occurring in

the canal, the explanation for increases of chlorides,

fluorides, sodium and sulfates is unclear. It can be rea-

sonably hypothesized that these parameters, which are

typical of marine waters, would increase as a conse-

quence of deposition of marine spray (Wilson, 1975)

transported from the Venice Lagoon, which is less than

5 km from the canal.

5. Conclusions

Along the course of the canal, a self-purification

process was observed to occur which was significant for

ammonium, total phosphates and bacteria. The largest

extent of this process took place in the first tract of the

canal (beaten soil) in which a net abatement of turbidity

and metals occurred as well. Evidence of an increase of

turbidity and metals in the second tract and the re-

markable increase of some parameters such as chlorides,

fluorides, sodium, sulfates in the whole tract are under

further investigation. It can be therefore concluded that

a more natural condition, such as that in the beaten soil

tract, can significantly favor natural self-purification

processes leading to improved water quality.

References

AWWA, 1998. Standard Methods for the Examination of

Water and Wastewater, 20th ed. APHA, Washington.

Benka-Coker, M.O., Ojior, O.O., 1995. Effect of slaughterhouse

wastes on the water quality of Ikpoba River, Nigeria.

Bioresource Technology 52, 5–12.

Benoit, R.J., 1971. Self-purification in natural waters. In:

Ciaccio, L.L. (Ed.), Water and Water Pollution Handbook,

vol. 1. Dekker, New York.

Borin, M., Marchetti, C., 1997. Sistemi di depurazione delle

acque basati sull�uso di vegetazione macrofita. Ambiente

Risorse Salute 55, 7–13.

Bratli, J.L., Skiple, A., Mjelde, M., 1999. Restoration of lake

Borrevannet––self-purification of nutrients and suspended

matter through natural reed-belts. Water Science and

Technology 40 (3), 325–332.

Brusseau, M.L., Bohn, H.L., 1996. Chemical processes affecting

contaminant fate and transport in soil and water. In:

Pollution Science. Academic Press, San Diego, pp. 61–78.

Chambers, P.A., Prepas, E.E., 1994. Nutrient dynamics in river-

beds: the impact of sewage effluent and aquatic macro-

phytes. Water Research 28 (2), 453–464.

C.N.R.––Istituto di Ricerca sulle Acque, 1994. Metodi analitici

per le acque. Istituto poligrafico e Zecca dello Stato, Roma.

Page 15: Vagnetti Et Al., 2002. Self-Purification Ability of a Resurgence Stream

R. Vagnetti et al. / Chemosphere 52 (2003) 1781–1795 1795

Cunningham, S.D., Davi, W.O., 1996. Promises and prospect

of phytoremediation. Plant Physiology 110, 715–719.

De Crespin De Billy, V., Reyes-Marchant, P., Lair, N.,

Valadas, B., 2000. Impact of agricultural practices on a

small headwater stream: terrestrial and aquatic character-

istics and self-purifying processes. Hydrobiologia 421, 129–

139.

Elosegui, A., Arana, X., Basaguren, A., Pozo, J., 1995. Self-

purification processes along a medium-sized stream. Envi-

ronmental Auditing 19 (6), 931–939.

Haag, D., Kaupenjohann, M., 2001. Landscape fate of

nitrate fluxes and emissions in Central Europe––a critical

review of concepts, data, and models for transport and

retention. Agriculture, Ecosystems and Environment 86

(1), 1–21.

Jing, S.R., Lin, Y.F., Lee, D.Y., Wang, T.W., 2001. Nutrient

removal from polluted river water by using constructed

wetlands. Bioresource Technology 76 (2), 131–135.

Lam-Leung, S.Y., Cheung, M.T., He, Y.Q., Zheng, Q.H., Wen,

W.Y., Zhang, Y.Y., 1996. Natural removal of added

nutrients, reactive phosphorus, crude oil, and heavy metals

from the water phase in a simulated water/sediment system.

Environment International 22 (2), 195–212.

Linnik, P.M., Zubenko, I.B., 2000. Role of bottom sediments in

the secondary pollution of aquatic environments by heavy-

metal compounds. Lakes and Reservoirs: Research and

Management 5 (1), 11–21.

Manahan, S.E., 1994. Environmental Chemistry. Lewis Pub-

lishers by CRC Press, Boca Raton, FL.

Marconato, E., Aquaprogram S.r.l., 2003. Private communica-

tion of unpublished results obtained during the survey: La

fauna ittica della provincia di Venezia. Provincia di Venezia.

Marconato, E., Maio, G., Salviati, S., 2000. La fauna ittica

della provincia di Venezia. Provincia di Venezia, Assessor-

ato alla Caccia, Pesca e Polizia Provinciale.

Meagher, R.B., 2001. Phytoremediation of toxic elemental and

organic pollutants. Current Opinion in Plant Biology 3 (2),

153–162.

Piccolo, D., 2000. Statistica. Il Mulino, Bologna.

Suzuky, M., 1997. Role of adsorption in water environment

processes. Water Science and Technology 35 (79), 1–11.

Vismara, R., 1998. Ecologia Applicata. Hoepli U, Milano.

Volterra, L., Mancini, L., 1994. Fitodepurazione e riuso

successivo delle macrofite. Ambiente Risorse Salute, 30–36.

Von der Wiesche, M., Wetzel, A., 1998. Temporal and spatial

dynamics of nitrite accumulation in the river Lahn. Water

Research 32 (5), 1653–1661.

Wilson, T.R.S., 1975. Salinity and the major elements of sea

water. In: Riley, J.P., Skirrow, G. (Eds.), Chemical Oceano-

graphy, 2nd ed., vol. 1. Academic Press, London.