Environmental Pollution 133 (2005) 265–274

www.elsevier.com/locate/envpol

Influence of temperature and salinity on heavy metal

uptake by submersed plants
A˚ . Fritioff*, L. Kautsky, M. Greger
Department of Botany, Stockholm University, S-106 91 Stockholm, Sweden

Received 21 November 2003; accepted 31 May 2004

Metal concentrations increase with increasing temperature and decreasing

salinity in two aquatic plants.

Abstract

Submersed plants can be useful in reducing heavy metal concentrations in stormwater, since they can accumulate large amounts
of heavy metals in their shoots. To investigate the effects of water temperature and salinity on the metal uptake of two submersed
plant species, Elodea canadensis (Michx.) and Potamogeton natans (L.), these plants were grown in the presence of Cu, Zn, Cd, and
Pb at 5, 11, and 20 ○C in combination with salinities of 0, 0.5, and 5&. The metal concentrations in the plant tissue increased with
increasing temperature in both species; the exception was the concentration of Pb in Elodea, which increased with decreasing
salinity. Metal concentrations at high temperature or low salinity were up to twice those found at low temperature or high salinity.
Plant biomass affected the metal uptake, with low biomass plants having higher metal concentrations than did high biomass plants.
© 2004 Elsevier Ltd. All rights reserved.

Keywords: Elodea canadensis; Potamogeton natans; Phytoremediation; Stormwater; Cd; Cu; Zn; Pb

1. Introduction pH, redox potential, and surrounding metal concen-

trations as well as the temperature and salinity of the
Submersed plants may be useful in reducing heavy
surrounding stormwater. The chemistry of stormwater is
metal concentrations in stormwater, since they have the
highly variable depending on its origin, on rain intensity,
ability to take up heavy metals directly from water and
and on the season; its pH, however, is often around 7
accumulate them in their shoots (St-Cyr et al., 1994;
(Pettersson, 1999). In a temperate climate, like that of
Coquery and Welbourn, 1995; Greger et al., 1995; Rai
Sweden, the temperature and salinity of stormwater
et al., 1995; Jackson, 1998). Stormwater comprises
vary with the season: the water temperature may exceed
rainwater and meltwater draining from hard surfaces,
20 ○C in the summer and can decrease to below 0 ○C in
such as roads and roofs in urban areas, and often
the winter. Salinity is very low most of the year, but
contains heavy metals, in particular Cu, Zn, Cd, and Pb
salinities up to 5& have been measured (Wittman, 1979;
(Morrison, 1989; Sansalone et al., 1996; Boller, 1997;
Sandersson, 1997; Christer La¨nnergren, pers. comm.)
Pettersson et al., 1999). The metal uptake of submersed
during winter when NaCl is used for deicing roads.
plants may be influenced by several factors, including
During summer, slightly elevated salinities up to 0.5&
* Corresponding author. Tel.: C46 8 161348; fax: C46 8 165525. can be measured (Wittman, 1979; Sandersson, 1997;
E-mail address: fritioff@botan.su.se (A˚ . Fritioff).
Christer La¨nnergren, pers. comm.), probably due to the
resuspension of sediment (Odum, 2000).
0269-7491/$ - see front matter © 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.envpol.2004.05.036
26 A˚ . Fritioff et al. / Environmental Pollution 133 (2005) 265–274
6
Water temperature may influence water chemistry,
metal solubility, metal uptake by plants, and plant
growth. According to Zumdahl (1992), seasonal varia-
tion in water temperature has no direct effect on the
solubility of metal in water. However, cool water
contains more dissolved oxygen than does warm water.
Thus, metal concentration in the interstitial water of the
sediment may decrease with decreasing temperature, as
more metals are bound to sediment colloids at high
rather than low redox potentials (Fo¨rstner, 1979). In
addition, temperature has a profound effect on plant
growth rates and higher temperatures will thus result in
greater biomass production and distribution of sub-
mersed macrophyte communities (Marschner, 1995;
Rooney and Kalff, 2000). A plant of relatively high
biomass may have a greater metal uptake capacity; this
results from lower metal concentration in its tissue
because of a growth rate that exceeds its uptake rate
(Ekvall and Greger, 2003). Changes in temperature
further change the composition of the plasma membrane
lipids (Lynch and Steponkus, 1987). This alters the plant
membrane fluidity, resulting in lower membrane perme-
ability at low temperatures and lower metal uptake
(Marschner, 1995). In the aquatic environment, the Cu
adsorption to the alga Dunaliella tertilecta increases
with increasing temperature (Gonzalez-Davila et al.,
1995). Further, in the lichen Peltigera horizontalis
(Huds.)
(Beckett and Brown, 1984) and the liverwort
Dumortiera hirsuta (Sw.) (Mautsoe and Beckett, 1996),
both intra- cellular and extracellular uptake of Cd was
stimulated by an increase in temperature. Similarly,
several studies of terrestrial plants grown at high root
temperatures found higher uptakes of Zn, Pb, Ag, Cr,
Sb, and Cd than was the case with plants grown at low
root temperatures (Hooda and Alloway, 1993; Macek et
al., 1994; Baghour et al., 2001; Albrecht et al., 2002).
Therefore, a general increase of metal uptake with
increasing temperature seems likely.
The salinity of stormwater may affect the plant
growth rate and plant metal uptake through the toxic
effects of both the NaC and Cl— ions. NaC ions may
release Cd from the sediment to the water, thereby
increasing the Cd concentration in the water (Greger
et al., 1995). In seawater, chloride commonly forms
complexes with Zn, Cd, and, to a lesser extent, Cu,
but not with Pb, and thus the free ion concentration of
the former metals will be reduced (Fo¨rstner, 1979;
Williams et al., 1994). Fewer free Cd ions in water of
higher salinity correlates with a lower uptake of Cd
by Potamogeton pectinatus (L.) growing in higher
salinity water (Greger et al., 1995). At a higher
salinity, the increased NaC concentration reduced both
intracellular
and extracellular uptake of Cd in the free-floating plant,
Lemna polyrhiza (L.) (Noraho and Gaur, 1995). Not
only in submersed (Greger et al., 1995) and free-floating
(Noraho and Gaur, 1995) plants, but also in
macro-
algae, such as the seaweed Fucus vesiculosus (L.), did the
uptake of metal (Zn) decrease with increasing salinity
( Munda and Hudnik, 1988 ). Thus, a general
decrease in metal uptake with increasing salinity could
be expected. Variations in the temperature and salinity
of storm- water result in different combinations of
these two factors over the course of a year (Wittman,
1979; Sandersson, 1997). In winter/early spring, low
temper- ature is combined with low to high (5&)
salinity, while in summer, high or medium temperatures
are combined with low to medium (0.5&) salinity.
Furthermore, in late autumn/early winter both
temperature and salinity may be low. To the best of our
knowledge, no investigation has as yet determined the
effects of temperature in combination with salinity on
the heavy metal uptake of submersed plants. This study
evaluates the combined effects of temperature and
salinity on the uptake of a set of heavy metalsdCu,
Zn, Cd, and Pbd
in two morphologically different submersed aquatic
plant species, Elodea canadensis and Potamogeton
natans. In a previous study (Fritioff and Greger, 2003),
these two species were shown to accumulate higher
concentrations of heavy metals when grown in storm-
water ponds than were emergent species sampled.
Although these are known as freshwater species, they
also grow in somewhat brackish water (Luther, 1951).
Furthermore, both species grow in ponds and wetlands
until an ice layer covers the water surface. Our
hypotheses were that metal concentrations and accu-
mulations in submersed plant species would: (1) in-
dependent of salinity, increase with increasing water
temperature and (2) independent of temperature, de-
crease with increasing salinity.
2. Materials and methods
2.1. Plant material
The two studied species commonly grow in storm-
water treatment ponds and wetlands. Although both are
submersed, E. canadensis (L.) and P. natans (Michx.)
have quite different morphologies. Elodea has very thin
submersed leaves, almost lacking a cuticle, arranged
around the stem, giving a high surface-to-volume ratio.
In contrast, Potamogeton has long petioles and leaves
that are several cell layers thick floating on the water
surface. These leaves are covered with a comparatively
thick cuticle, resulting in a much lower surface-to-
volume ratio. Both studied species grow and remain
green until an ice layer covers the wetland (during cold
winters), causing a die back. Below the ice,
Potamogeton over-winters by means of its hardy
rootstocks, while Elodea over-winters by means of short
vegetative shoots.
Whole plants of Potamogeton were collected
in August 2001 from an unpolluted pond called
Ska˚ bydammen, located south of Stockholm, Sweden.
Plants were gently cleaned with tap water to remove the
sediment, planted in a greenhouse pond containing tap
water and minimally contaminated sediment, and left
growing for 4 weeks until the start of the experiment.
For Elodea, 10-cm-long vegetative shoots were
collected from a greenhouse pond. The shoots were
planted in a new pond containing the same types of
sediment and water as did the pond containing the
Potamogeton plants, and grown for about 5 weeks until
new vegetative shoots had developed.
2.2. Growth conditions
The two species were grown in a greenhouse
equipped with supplementary lamps. During the pre-
experimental phase, Osram daylight lamps (HQI-BT
400W) were used for a light period of 12 h. Temperature
in the pond water was 18 G 1 ○C. The acclimatization
and experi- mental phases were carried out in a climate-
controlled
chamber with an air temperature of 10 G 1 ○C.
Halogen lamps (Osram Poverstar HQI-E) were used
to create light intensity of 50 G 5 mmol photons m—2 s
—1
for a light period of 18 h.
2.3. Experiment
To study the effects of salinity and water temperature
on the heavy metal uptake of Potamogeton and Elodea,
a two-factorial design, 32 (Sokal and Rohlf, 1995), was
set up with temperature and salinity as the two factors.
Each Potamogeton plant (with 1–5 leaves) was thor-
oughly cleaned under running tap water to remove
sediment and other particles and then rinsed in
redistilled water before being put in a container
containing 2 L of 2% Hoagland medium (Eliasson,
1978) with a pH of 5.0 G 0.2; the medium was not
aerated. In the experiment with Elodea, four vegetative
shoots (one each of 10, 15, 20, and 25 cm in length)
were
cut and placed together in each container, a single
container comprising one replicate. Seven replicates
were used for each treatment and each species. Contain-
ers with plants were randomly placed in large, water-
filled tubs. Water in one-third of the tubs was chilled to
5 ○C using a compressor (TECO RA 680, Italy),
thermostats (Ja¨ger, Type LZRH, Wu¨stenrot) to keep it
at 20 ○C, heated the water of another third of the tubs
and the water of the remaining third of the tubs was
kept at 11 ○C. To reduce evaporation, all containers
were covered with transparent plastic film. After 10 days
of acclimatization, the fresh weights (FW) of the
individuals of both species were measured after first
gently shaking the plants for 10 s to remove water.
Additionally, the lengths of the Elodea plants were
measured. Plants were then again set in containers of
fresh medium kept at the same temperatures as before.
After another day of acclimatization, plants from three
containers at each temperature were used to determine
the baseline heavy metal content in the plants. There-
after, heavy metals and NaCl were added to all
containers. To simulate a storm event, when a flush of
highly polluted water originating from rainfall enters the
pond, the plants were treated for 48 h with metal levels
of 1.5, 20, 1, and 4 mM (equal to 95, 1307, 112, and
836 mg L—1) of Cu, Zn, Cd, and Pb, respectively
(Morrison, 1989; Sansalone et al., 1996; Boller, 1997;
Pettersson et al., 1999). Then, NaCl was added, which
resulted in final salinity levels of 0, 0.5, and 5&. After
48 h, the experiment was ended and the fresh weights of
plants were measured as well as the length of the Elodea
plants. The pH remained at 5.0 G 0.1 for the duration of
the experiment.
All plants of both species were rinsed twice in
redistilled water, once in 20 mM EDTA, and twice in
redistilled water, for half a minute each time. They
were dried at 105 ○C for 24 h and then weighed to
determine the plant dry weights (DW). All plant
materials from each container was wet digested in
concentrated HNO3:HClO4 (7:3, v:v). Metal concen-
tration was measured using a flame atomic absorption
spectrophotometer (SpectrAA-100, Varian, Spring-
valve, Australia). Standards were added to the samples
to eliminate the interaction of the sample matrix.
Lagharosiphon major was used as a reference material
(BCR No 60 N ○ 675, Commission of the European
communities).
2.4. Calculations and statistics
Dry weight-to-fresh weight ratios (DW/FW) were
calculated for both species. To determine the growth
rate, the increase in fresh weight of Potamogeton and the
increase in length of Elodea during the 48-h experimen-
tal period were calculated as g FW (24 h) —1 and cm
(24 h)—1, respectively. Metal accumulations during the
experiment were calculated by subtracting control
concentrations from the concentrations found in the
treated plants.
The effects of the fixed factors, temperature and
salinity, on Zn, Cu, Cd, and Pb concentrations in
Potamogeton and Elodea were analyzed in a general
linear model which used biomass ( g DW L —1) as a
covariable, in order to sort out the effects of the various
factors on the metal concentrations in the two plant
species. Insignificant interactions with the con- tinuous
covariable were dropped from the model using a
backward elimination procedure. Analysis of the metal
accumulations was preformed in the same way. Tukey’s
honest significant difference was used as a post hoc test
(Statistica software, StatSoft Inc., ’99 edition, kernel
release 5.5 A).
3. Results than did the treated plants. Concentrations of each
metal in the control plants did not differ significantly
Growth rate, measured as increase in length
among temperature treatments.
(cm ! 24 h—1), increased significantly with
The metal concentrations in plants increased with
increasing temperature for Elodea (Table 1).
increasing temperature in most cases (Fig. 2, Tables 3
Furthermore, Elodea plants grew significantly longer
and 4). This was found to be significant in Elodea for
(cm) in containers kept at 20 ○C than in containers
Cu, Zn, and Cd and in Potamogeton for Cu, Cd, and Pb,
kept at colder temperatures. The DW/FW ratio (%)
and a tendency was found in Potamogeton for Zn.
was significantly higher in
Further- more, the concentrations of Cu, Zn, and Cd
Elodea subject to colder rather than warmer
increased with decreasing salinity: this was significant in
treatments. The dry and fresh weights of Elodea per
both species for Zn and Cd and in Elodea for Cu, and a
liter did not differ significantly among treatments. For
tendency was found in Potamogeton for Cu. Lead con-
Potamogeton, no significant differences in dry or
centration, however, was unaffected by salinity.
fresh weights, DW/ FW ratios, or growth rates ( g FW
Analysis of metal accumulations (not shown) revealed
! 24 h—1) were found the same patterns as were found for metal
among treatments. The plant fresh weights were slightly
concentrations.
higher for Potamogeton than for Elodea, while the plant
Biomass affected the metal uptake in plants, smaller
dry weights and DW/FW ratios were more than twice as
plants having higher metal concentrations and accumu-
high for Potamogeton as for Elodea.
lations than larger plants, significantly so for Cu and Zn
Because of the higher biomass (DW l —1) of Potamo-
in Elodea and for all the metals measured in Potamo-
geton than of Elodea (Table 1), total metal accumu-
geton (Fig. 3, Tables 3 and 4).
lations (mg container—1) in the plant material of
Several interactions were seen. Temperature and
Potamogeton in some cases exceeded those of Elodea
salinity interacted both with each other and with the
(Fig. 1). Thus, compared with Elodea, Potamogeton
biomass. Between temperature and salinity, two inter-
removed more of the added Zn from the medium in all
actions were found (Fig. 4, Tables 3 and 4). First,
treatments and more Cu and Cd from the medium in
Potamogeton plants growing in medium salinity
the 5 ○C treatment. However, the metals added to the
water (0.5&) accumulated more Zn than did plants
nutrient solution were never depleted in any of the
growing in high salinity water (5&) at 5 and 20 ○C but
treatments.
not at 11 ○C. Second, the temperature effect on Pb
The metal concentrations in the plants at the end of
concentration in Potamogeton was significant only
the experiment are shown in Table 2. Elodea had, with
when plants were growing in medium salinity water
a few exceptions in the case of Zn, higher concentrations
(0.5&).
of heavy metals than did Potamogeton in both treated
The different temperatures had stronger effects on the
and control plants. Except for Cu in Elodea at 5 ○C and
metal concentrations and accumulations in plants with
5& salinity, controls had lower metal concentrations
lower biomass rather than in plants with higher biomass,

Table 1
Treatment averages of dry and fresh weights, DW/FW ratio (dry weight/fresh weight ratio), length, and growth rate for Elodea and Potamogeton
treated for 48 h with metal levels of 1.5 mM Cu, 20 mM Zn, 1 mM Cd, and 4 mM Pb, temperatures of 5, 11, or 20 ○C, and salinities of 0, 0.5, or 5&
Species Treatment Dry weight Fresh weight DW/FW ratio Total length Growth rate Growth rate
( g DW L—1) ( g FW L—1) (%) (cm) (cm (24 h)—1) ( g FW (24 h)—1)
Elodea Temperature ( ○C)
5 0.43 G 0.01 a 8.1 G 0.08 a 5.2 G 0.03 a 76.76 G 0.106 b 0.23 G 0.01 b n.d.
11 0.43 G 0.00 a 8.7 G 0.06 a 5.0 G 0.02 b 75.10 G 0.059 b 0.32 G 0.01 ab n.d.
20 0.44 G 0.01 a 9.0 G 0.08 a 4.8 G 0.02 b 83.67 G 0.251 a 0.43 G 0.01 a n.d.
Salinity (&)
0 0.45 G 0.01 a 8.7 G 0.08 a 5.1 G 0.03 a 78.50 G 0.26 a 0.31 G 0.01 a n.d.
0.5 0.44 G 0.01 a 8.7 G 0.08 a 5.0 G 0.02 a 78.82 G 0.286 a 0.32 G 0.01 a n.d.
5 0.41 G 0.00 a 8.3 G 0.06 a 4.9 G 0.02 a 78.21 G 0.16 a 0.35 G 0.01 a n.d.
Potamogeton Temperature ( ○C)
5 1.32 G 0.03 a 11.1 G 0.23 a 11.8 G 0.1 a n.d. n.d. 0.00 G 0.01 a
11 1.49 G 0.04 a 12.4 G 0.26 a 11.8 G 0.1 a n.d. n.d. 0.12 G 0.02 a
20 1.11 G 0.03 a 10.3 G 0.27 a 10.5 G 0.1 a n.d. n.d. 0.02 G 0.01 a
Salinity (&)
0 1.35 G 0.04 a 11.7 G 0.28 a 11.2 G 0.1 a n.d. n.d. 0.03 G 0.02 a
0.5 1.39 G 0.03 a 11.7 G 0.24 a 11.7 G 0.1 a n.d. n.d. 0.00 G 0.00 a
5 1.19 G 0.03 a 10.3 G 0.25 a 11.2 G 0.1 a n.d. n.d. 0.06 G 0.01 a
Different letters indicate significant differences within each metal and treatment (n = 21) GSE. n.d. = no data.
 150 150
Cu Cd

µg Cu accumulated

µg Cd accumulated
100 100

50 50

0
Salinity (‰) 0
0 0.5 5 0 0.5 5 0 0.5 5 0 0.5 5 0 0.5 5 0 0.5 5
Temperature (C) 5 11 20 5 11 20

Zn 1000 Pb
1600
µg Zn accumulated

µg Pb accumulated
800
1200
600
800
400
400
200
0
Salinity (‰) 0 0.5 5 0 0.5 5 0 0.5 5 0
0 0.5 5 0 0.5 5 0 0.5 5
Temperature (C) 5 11 20 5 11 20

Fig. 1. Total accumulation of Cu, Zn, Cd, and Pb in mg container in Elodea (black bar) and Potamogeton (white bar), treated for 48 h with
—1

metal levels of 1.5 mM Cu, 20 mM Zn, 1 mM Cd, and 4 mM Pb, temperatures of 5, 11, or 20 ○C, and salinity of 0, 0.5, or 5&. Each container
contained 2 L of medium and on average 0.43 and 1.31 g of Elodea and Potamogeton, respectively (n = 7).

this being the case for Cu in both plant species and rather than higher biomasses. Furthermore, the Cd
for Cd and Pb in Potamogeton (Tables 3 and 4). con- centration and accumulation in Elodea increased
Biomass also interacted with salinity, and the with increasing biomass when grown at 11 ○C, while
concentration and accumulation of Cu in Elodea and at the other temperatures Cd concentrations
Zn in Potamogeton were more affected by differences decreased with increasing biomass (Table 3).
in salinity at lower

Table 2
Concentrations of Cu, Zn, Cd, and Pb in Elodea and Potamogeton both before (Controls) and after treatment for 48 h with 1.5 mM Cu, 20 mM Zn,
1 mM Cd, and 4 mM Pb, temperatures of 5, 11, or 20 ○C, and salinities of 0, 0.5, or 5& (n = 7, for controls n = 2–3, GSE)
Species Temperature ( ○C) Salinity (&) Metal accumulation (mg( g DW)—1)
Cu Zn Cd Pb
Elodea 5 0 340 G 15 1165 G 41 145 G 11 1732 G 289
0.5 311 G 24 1137 G 114 117 G 9 1838 G 336
5 246 G 7 548 G 37 38 G 4 1495 G
Control 271 G 81 387 G 91 3G2 184
145 G 55
11 0 345 G 10 1168 G 65 259 G 43 1447 G 208
0.5 354 G 15 1128 G 50 181 G 27 1528 G 300
5 302 G 15 626 G 62 85 G 19 2037 G
Control 116 G 12 208 G 62 3G1 225
16 G 7
20 0 444 G 15 2193 G 118 269 G 16 1771 G 152
0.5 406 G 30 1919 G 138 244 G 24 1694 G 134
5 348 G 25 1354 G 276 212 G 23 2190 G 168
Control 154 G 19 178 G 21 1G1 53 G 33
Potamogeton 5 0 145 G 22 1381 G 122 94 G 12 127 G 40
0.5 118 G 13 1258 G 264 70 G 8 74 G 19
5 137 G 25 560 G 64 32 G 4 110 G 20
Control 23 G 0 48 G 6 5G1 5G5
11 0 133 G 19 1297 G 132 108 G 12 88 G 17
0.5 111 G 22 873 G 139 74 G 13 75 G 9
5 70 G 10 672 G 74 32 G 3 113 G 39
Control 17 G 3 69 G 9 3G2 7G3
20 0 193 G 60 1679 G 255 151 G 33 151 G 40
0.5 160 G 40 1311 G 194 100 G 18 227 G 68
5 142 G 26 693 G 84 63 G 27 162 G 42
Control 21 G 5 61 G 6 2G1 7G0
 Cd
Table 3
400 Cu Cu Cd
Analysis of variance of heavy metal concentrations in Elodea exposed
to Cu, Zn, Cd, and Pb, three different salinities, and three different
water temperatures
300
µg Me / g DW

Response Source of df MS F p
variable variation
200
Cu Temperature 2 17784 15.7 !0.001***
Salinity 2 9548 8.4 !0.001***
Biomass 1 54589 48.1 !0.001***
100 Temperature 4 1034 0.9 0.4655
! Salinity
Temperature 2 6564 5.8 0.0056**
0 !
5 11 20 5 11 20 0 0.5 5 0 0.5 5 Biomass 2 4143 3.6 0.0334*
Salinity
2500 !
Pb Zn Biomass
Zn Error 49 1136
Pb
Zn Temperature 2 5282023 53.4 !0.001***
2000
Salinity 2 2357154 23.8 !0.001***
Biomass 1 462852 4.7 0.0351*
µg Me / g DW

1500 Temperature 4 40936 0.4 0.7981

! Salinity
1000 Error 53 98979
Cd Temperature 2 33081 11.7 !0.001***
500 Salinity 2 57252 20.2 !0.001***
Biomass 1 1139 0.4 0.5288

0 Temperature 4 6665 2.4 0.0663

5 11 20
! Salinity
5 11 20 0 0.5 5 0 0.5 5 Temperature 2 22256 7.9 0.0011**
Temperature (C) Salinity (‰) ! Biomass
Error 51 2833
Fig. 2. Treatment averages of Cu, Zn, Cd, and Pb concentrations in
Elodea (black bar) and Potamogeton (white bar) treated for 48 h with Pb Temperature 2 343788 1.0 0.3889
metal levels of 1.5 mM Cu, 20 mM Zn, 1 mM Cd, and 4 mM Pb, temp- Salinity 2 584252 1.6 0.2049
eratures of 5, 11, or 20 ○C, and salinities of 0, 0.5, or 5& (n = 21) GSE. Biomass 1 1355437 3.8 0.0569
Temperature 4 314637 0.9 0.4823
! Salinity
4. Discussion Error 53 357601
Plants were allowed to acclimatize to the temperatures for 10 days
This research found that metal concentration and before exposure to salinity treatments and heavy metals. Only single
accumulation in the two studied submersed macrophyte factors (temperature and salinity), the covariable (biomass, as dry
weight), interactions between main factors (temperature and salinity),
species generally increased with increasing temperature
and signi- ficant interactions are shown. Error, within group variance;
but decreased with increasing salinity (Fig. 2, Tables 3 df, degrees of freedom; MS, between group variance; F, ratio between
and 4). The influences of temperature and salinity were MS and Error.
not always independent of each other (Fig. 4, Tables 3 *p ! 0.05, **p ! 0.01, ***p ! 0.001.
and 4). Furthermore, metal concentrations and accu-
mulations increased with decreasing biomass in most metal accumulations (mg container—1) increased with
cases (Fig. 3, Tables 3 and 4). Interactions between increasing temperature (Fig. 1). One likely explanation
temperature and biomass or salinity and biomass were for this is that increased temperature will increase the
sometimes found as well (Tables 3 and 4). biomass and thus the absorption area of the plant; this, in
Because of the higher biomass (DW l —1) of Potamo- turn, will increase the metal uptake. This was found to
geton than of Elodea (Table 1), total metal accumu- be the case for Cd uptake by Pinus silvestris (L.) (Ekvall
lations (mg container—1) in Potamogeton plant material and Greger, 2003). In the present study, Elodea length
in some cases exceeded those of Elodea (Fig. 1). Thus, increased significantly more in containers at 20 ○C than
compared with Elodea, Potamogeton removed more in those subject to colder treatments (Table 1). However,
added Zn from the medium in all treatments and more biomass did not differ between treatments; thus the
Cu and Cd from the medium in the 5 ○C treatment. effect of temperature on metal uptake was direct, and an
In line with the first hypothesis, metal concentrations explanation other than growth rate must be found.
and accumulations increased with increasing water Earlier investigations suggested that the extracellular
temperature for Cu, Zn, and Cd in Elodea (Fig. 2, Table concentration increases with increasing temperature
3) and for Cu, Zn, Cd, and Pb in Potamogeton (Fig. 2, because there is a change in the equilibrium between
Table 4), independent of salinity. Furthermore, total the cell wall exchange sites and the metal in solution
Table 4
Analysis of variance of heavy metal concentrations in Potamogeton
lular uptake of Cu, Cd, and Zn might also be the frac-
exposed to Cu, Zn, Cd, and Pb, three different salinities, and three tions most affected by temperature in the present study.
different water temperatures Slightly increased extracellular uptake could facilitate
Response Source of df MS F p intracellular uptake by concentrating metals close to the
variable variation membrane. Furthermore, a higher temperature changes
Cu Temperature 2 16893 6.7 0.0026** the lipid composition of the plasma membrane (Lynch
Salinity 2 6306 2.5 0.0926 and Steponkus, 1987) and thereby its fluidity, which
Biomass 1 182345 72.1 !0.001*** may facilitate both passive and active metal flux through
Temperature 4 1894 0.7 0.5632 the membrane. In addition, increased temperature
! Salinity
Temperature 2 13555 5.4 0.0077**
increases both metabolism (Marschner, 1995; Nilsen and
! Orcutt, 1996) and protein synthesis (Nilsen and Orcutt,
Biomass 51 2528 1996), and this may result in higher metal uptake at
Error
additional uptake sites on membranes or an increased
Zn Temperature 2 204246 2.6 0.0876
Salinity 2 1888943 23.6 !0.001***
release of molecules facilitating metal uptake.
Biomass 1 4417721 55.3 !0.001*** The second hypothesis, that metal concentration and
Temperature 4 216076 2.7 0.0406* accumulation decrease with increasing salinity, was
! supported by the results for Cu, Zn, and Cd in both
Salinity 2 510731 6.4 0.0033** species (Fig. 2, Tables 3 and 4), independent of
Salinity
!
temperature. E. canadensis and P. natans are freshwater
Biomass 51 79944 species and have been found growing in the field in
Error salinities up to 3& and 2&, respectively (Luther, 1951).
Cd Temperature 2 11998 11.3 !0.001*** Even at the highest salinity used in this study, 5&, no
Salinity 2 25012 23.6 !0.001***
toxic effect was observed in these species (not shown).
Biomass 1 42135 39.7 !0.001***
Temperature 4 1033 1.0 0.4305 The likely reason for the decreased accumulation with
! Salinity increasing salinity is the decreased availability of the
Temperature 2 7696 7.2 0.0017** metals in the growth medium because of the complexes
! formed between chloride and metals (Fo¨rstner, 1979).
Biomass 51 1062
Error This has, for example, been shown to depress Cd uptake
Pb Temperature 2 36547 6.8 0.0024** in P. pectinatus (Greger et al., 1995). In addition,
Salinity 2 1807 0.3 0.7152 increased competition with sodium ions at uptake sites,
Biomass 1 191946 35.8 !0.001*** both on the plasma membrane and in apparent free
Temperature 4 16666 3.1 0.0229* space in the cell walls, may account for the decreased
! Salinity
metal accumulation at higher salinities (Noraho and
Temperature 2 19125 3.6 0.0354*
! Gaur, 1995). That salinity has no influence on Pb
Biomass 51 5355 accumulation (Fig. 2, Tables 3 and 4) may be because
Error Pb does not form a complex with chloride (Fo¨rstner,
Plants were allowed to acclimatize to the temperatures for 10 days 1979), is retained briefly in water, and has a high affinity
before exposure to salinity treatments and heavy metals. Only single
factors (temperature and salinity), the covariable (biomass, as dry
to bind to organic matter (Fo¨rstner, 1979) such as plant
weight), interactions between main factors (temperature and salinity), surfaces (Va´squez et al., 1999). Therefore, despite
and significant interactions are shown. Error, within group variance; different salinities, the same amount of Pb will bind to
df, degrees of freedom; MS, between group variance; F, ratio between and be taken up by plants.
MS and Error. Biomass did not differ among treatments, but rather
*p ! 0.05, **p ! 0.01, ***p ! 0.001. varied within each treatment (Table 1), and plants with
lower biomasses had higher metal accumulations and
(Gonzalez-Davila et al., 1995; Mautsoe and Beckett, concentrations than did plants with higher biomasses
1996). However, if this was the only factor, temperature (Fig. 3, Tables 3 and 4). This may be because a propor-
should affect Pb accumulation in Elodea in the same tionally larger absorption area in relation to the external
way it affects accumulation of the other three metals, metal concentration may result in a dilution effect and
which was not the case (Fig. 2, Table 3). Instead, it is thus a low internal metal concentration. Ekvall and
suggested that the absence of a temperature effect on Pb Greger (2003) showed that this was the case for Cd in
accumu- lation is due to a high proportion of Pinus sylvestris. Besides different amounts of biomass,
extracellular binding sites for Pb in Elodea. This was different metal accumulations may also be due to
earlier shown by Va´squez et al. (1999) to be the case in different proportions of stems to leaves. Preliminary
two aquatic bryophyte species, Fontinalis antipyretica results for P. natans indicate that the uptake by its leaves
Hedw. and Scapania undulata (L. Dum.). Extracellular is almost 10 times higher than the uptake by its stems
binding of metal is less temperature dependent than is (Fritioff and Greger, unpublished).
intracellular uptake, according to Beckett and Brown
(1984), Marschner (1995), and Mautsoe and Beckett
(1996). The intracel-
 (µg Me / g DW)

(µg Me / g DW)
2000 2000

1000

1000
A 0 B 0
0 0.25 0.5 0.75
400 400 0123
Biomass (g DW)
Biomass (g DW)
(µg Me / g DW)

(µg Me / g DW)
200 200

0
0
0 0.25 0.5 0.75 0 1 2 3
Elodea biomass (g DW)
Potamogeton biomass (g DW)

Fig. 3. The effect of biomass on Cu, Zn, Cd, and Pb concentrations in Elodea (A) and Potamogeton (B) treated for 48 h with metal levels of 1.5
mM Cu, 20 mM Zn, 1 mM Cd, and 4 mM Pb, temperatures of 5, 11, or 20 ○C, and salinities of 0, 0.5, or 5&. Only trend lines are shown; the line
pattern indicates which metal is represented: Cu (d – d – d –), Zn (––– ––– ––– –––), Cd (– – – – – –), Pb (ddd). n = 21.

The concentrations of the various metals tested in accordance with other studies done both in labora-
were, in almost all cases, higher in Elodea than in tory-scale wetlands (Nyquist and Greger, 2003) and in
Potamogeton (Fig. 2). The Zn concentration was about the field (Yurukova and Kochev, 1994). First, Potamo-
the same in the two species, while Cu and Cd geton plants had higher biomasses than did the Elodea
concentrations were about twice as high in Elodea plants (Table 1); as previously discussed, within
than in Potamogeton and the Pb concentration was a species, plants with higher biomass will have lower
over 10 times higher in Elodea. This is metal concentrations when exposed to the same
metal
concentration in the medium. Second, Elodea had
A 5 ºC
a higher water content than did Potamogeton, as
11 ºC
20 ºC shown by low DW:FW ratio (Table 1). This may
300
facilitate metal uptake, by presenting larger vacuoles
250
for storage and because a higher water content in the
µg Pb / g DW

200
cell wall facilitates diffusion into the apparent free
150
space by dilution. Other reasons for the higher metal
100
accumula- tion in Elodea than in Potamogeton (Fig. 2)
50
may be species differences in uptake capacities and
0
0 0.5 5
differences in the proportion of stems to leaves.
Salinity (‰) Sculthorpe (1967) described how Elodea had a higher
leaf uptake of solutes than did Potamogeton, because
B 0‰ of their differences in morphology. Elodea’s high
0.5 ‰ surface-to-volume ratio as well as its high proportion
5‰
2000
of leaves relative to total plant biomass may facilitate
metal uptake by diffusion, while the thick waxy cuticle
µg Zn / g DW

1500 of Potamogeton may obstruct metal uptake.

1000 Submersed plants often display high metal
accumu- lations in the field (Galiulin et al., 2001). In
500
the present laboratory experiment, Elodea had in most
0 cases higher
metal concentrations than did Potamogeton. In some
5 11 20
Temperature (°C ) cases, however, Potamogeton removed more of the
added metals in total from the containers than did
Fig. 4. Concentrations (mg ! g —1 ! DW—1) of Pb and Zn in Elodea. Depending on the treatment, Elodea removed
Potamogeton (A and B, respectively), showing interactions
between temperature and salinity after treatment for 48 h with at most 70, 30, 51, and 56% and at least 13, 5, 8, and
1.5 mM Cu, 40% of the added Cu, Zn, Cd, and Pb, respectively,
20 mM Zn, 1 mM Cd, and 4 mM Pb, temperatures of 5, 11, or 20 ○C, during the 48-h treatment with elevated levels of
and salinity of 0, 0.5, or 5& (n = 7) GSE. metals.
Potamogeton removed at most 75, 64, 50, and 11% and
at least 50, 22, 14, and 5% of the added Cu, Zn, Cd, and
Pb, respectively, during the same 48-h treatment. The
average weight of plant material in each container was
0.43 g DW for Elodea and 1.3 g DW for Potamogeton
(Table 1). Thus, the high removal might depend on
a high plant biomass per volume of solution. Both metal
concentration in plant material and total uptake were
highly dependent on plant biomass; therefore, biomass
in combination with metal load was important for
overall metal removal during the experiment. In fact,
Potamogeton removed at least twice as much Zn as did
Elodea in this experiment. However, Nyquist and
Greger (2003) found the opposite in a laboratory-scale
wetland experiment. They reported that Elodea removed
64% of the total amount of added Zn, while Potamo-
geton removed just 23%. The present study found that
Elodea and Potamogeton did accumulate heavy metals,
although the quantity depended on the biomass, in-
dicating that Elodea and Potamogeton can both remove
metals from stormwater during storm events.
5. Conclusions
Although Elodea and Potamogeton are most effective
at accumulating heavy metals discharged in stormwater
at high temperatures, they have still good heavy metal
accumulation capacity down to at least 5 ○C. They may
therefore be useful in treating stormwater in temperate
climates. Since plant biomass is low in early spring,
metal accumulation per unit of biomass can be expected
to be comparatively high, because of generally higher
metal accumulation when a small, compared to a large,
biomass is exposed to a given amount of metal in the
water. Increased salinity resulting from road deicing will
reduce metal accumulation in both Elodea and Potamo-
geton more than temperature will. Low salinity levels in
the summer lower Cu, Zn, and Cd accumulation by
plants, while salinity peaks in the winter may reduce the
accumulations by plants by over 50%. Unlike the other
metals, Pb accumulation is generally unaffected or very
little affected by temperature and/or salinity, and will
thus remain relatively stable throughout all seasons and
possible salinity levels. Although during the experiment
Potamogeton accumulated more Zn and sometimes
more Cu and Cd from the medium than did Elodea,
Potamogeton more often had a lower metal concentra-
tion than did Elodea; this is possibly because Potamo-
geton had a higher biomass per liter than did Elodea.
Based on our results, we suggest that since in the field
Elodea often occurs in more dense stands, grows faster,
and often has higher metal concentrations than does
Potamogeton, it is the better of the two tested species for
use in stormwater treatment facilities. However, further
in situ investigations are needed, and studies of the
storage capacity of heavy metals and the fate of metals
accumulated in biomass decomposing over the winter
merit further investigation.
Acknowledgements
We gratefully thank the Swedish Council for Forestry
and Agricultural Research for financing this study.
Thanks are also due to Stefan Dahlgren for collecting P.
natans and to Patrik Dinnetz for help with the statistical
calculations.
References
Albrecht, A., Schultze, U., Liedgens, M., Flu¨hler, H., Frossard, E.,
2002. Incorporating soil structure and root distribution into plant
uptake models for radionuclides: toward a more physically based
transfer model. Journal of Environmental Radioactivity 59 (3),
329–350.
Baghour, M., Moreno, D.A., Herna´ndez, J., Castilla, N., Romero, L.,
2001. Influence of root temperature on phytoaccumulation of As,
Ag, Cr and Sb in potato plants (Solanum tuberosum L. var. spunta).
Journal of Environmental Science and Health. Part A. Toxic/
Hazardous Substances & Environmental Engineering 36 (7), 1389–
1401.
Beckett, R.P., Brown, D.H., 1984. The control of cadmium uptake in
the lichens genus Peltigera. Journal of Experimental Botany 35,
1071–1082.
Boller, M., 1997. Tracking heavy metals reveals sustainability deficits
of urban drainage systems. Water Science and Technology 35 (9),
77–87.
Coquery, M., Welbourn, P.M., 1995. The relationship between metal
concentration and organic matter in sediments and metal
concentration in the aquatic macrophyte Eriocaulon septangulare.
Water Research 29 (9), 2094–2102.
Ekvall, L., Greger, M., 2003. Effects of environmental biomass-
producing factors on Cd uptake in two Swedish ecotypes of Pinus
sylvestris. Environmental Pollution 121, 401–411.
Eliasson, L., 1978. Effects of nutrients and light on growth and
root formation in Pisum sativum cuttings. Physiologia Plantarum
43, 13–18.
Fritioff, A˚ ., Greger, M., 2003. Heavy-metal uptake by aquatic and
terrestrial plant species sampled in stormwater-treatment areas.
International Journal of Phytoremediation 5 (3), 211–224.
Fo¨rstner, U., 1979. Metal transfer between solid and aqueous phases.
In: Fo¨rstner, U., Wittman, G.T.W. (Eds.), Metal Pollution in the
Aquatic Environment. Springer-Verlag, Berlin, pp. 197–270.
Galiulin, R.V., Bashkin, V.N., Galiulina, R.R., Birch, P., 2001. A
critical review: protection from pollution by heavy metals –
phytoremediation of industrial wastewater. Land Contamination
& Reclamation 9 (4), 349–357.
Gonzalez-Davila, M., Santana-Casiano, J.M., Perez-Pen˜ a, J., Millero,
F.J., 1995. The binding of Cu(II) to the surface and exudates of the
alga Dunaliella tertiolecta in seawater. Environmental Science and
Technology 29 (2), 289–301.
Greger, M., Kautsky, L., Sandberg, T., 1995. A tentative model of Cd
uptake in Potamogeton pectinatus in relation to salinity. Environ-
mental and Experimental Botany 35 (2), 215–225.
Hooda, P.S., Alloway, B.J., 1993. Effects of time and temperature on
the bioavailability of Cd and Pb from sludge-amended soils.
Journal of Soil Science 44 (1), 97–110.
Jackson, L.J., 1998. Paradigms of metal accumulation in rooted
aquatic vascular plants. The Science of the Total Environment 21,
223–231.
Luther, H., 1951. Verbreitung und O ¨kologie der ho¨heren Wasserp-
flanzen im Brackwasser der Ekena¨s-Gegend in Su¨dfinnland. Acta
Botanica Fennica 49, 1–231 (50, pp. 1–370).
Lynch, D.V., Steponkus, P.L., 1987. Plasma membrane lipid
alterations associated with cold acclimation of winter rye seedlings
(Secale cereale L. cv Puma). Plant Physiology 83, 761–767.
Macek, T., Kotrba, P., Suchova, M., Skacel, F., Demnerova, K.,
Ruml, T., 1994. Accumulation of cadmium by hairy-root cultures
of Solanum nigrum. Biotechnology Letters 16 (6), 621–624.
Marschner, H., 1995. Mineral Nutrition of Higher Plants. Academic
Press Limited, London.
Mautsoe, P.J., Beckett, R.P., 1996. A preliminary study of the factors
affecting the kinetics of cadmium uptake by the liverwort
Dumortiera hirsute. South African Journal of Botany 62 (6), 332–
336.
Morrison, G.M.P., 1989. Bioavailable metal uptake rate in urban
stormwater determined by dialysis with receiving resins. Hydro-
biologia 176/177, 491–495.
Munda, I.M., Hudnik, V., 1988. The effects of Zn, Mn, and Co
accumulation on growth and chemical composition of Fucus
vesiculosus L. under different temperature and salinity conditions.
Marine Ecology 9 (3), 213–225.
Nilsen, E.T., Orcutt, D.M., 1996. Physiology of Plants under Stress
Abiotic Factors. John Wiley & Sons, Inc., New York, United
States of America.
Noraho, N., Gaur, J.P., 1995. Effect of cations, including heavy
metals, on cadmium uptake by Lemna polyrhiza L. Biometals 8,
95–98.
Nyquist, J., Greger, M., 2003. Phytoremediation of metal containing
water using submersed plants. In: COST Action 837 on Phy-
toremediation of Toxic Metals, Stockholm, Sweden, 12–15 June
2003.
Odum, H.T., 2000. Heavy Metals in the Environment: Using Wetlands
for Their Removal. CRC Press LLC, Boca Raton, Florida.
Pettersson, T.J.R., 1999. The effects of variations of water quality
on the partitioning of heavy metals in a stormwater pond. In:
Proceedings of the Eighth International Conference on Urban
Storm Drainage, Sydney, Australia, 30 August–3 September
1999, vol. 4, pp. 1943–1946.
Pettersson, T.J.R., German, J., Svensson, G., 1999. Pollutant removal
efficiency in two stormwater ponds in Sweden. In: Proceedings of
the Eighth International Conference on Urban Storm Drainage,
Sydney, Australia, 30 August–3 September 1999, vol. 2, pp. 866–
873.
Rai, U.N., Sinha, S., Triphati, R.D., Chandra, P., 1995. Waste water
treatability potential of some aquatic macrophytes: removal of
heavy metals. Ecological Engineering 5 (1), 5–12.
Rooney, N., Kalff, J., 2000. Inter-annual variation in submerged
macrophyte community biomass and distribution: the influence of
temperature and lake morphometry. Aquatic Botany 68, 321–335.
Sandersson, M., 1997. Norr Ma¨larstrand lokalt omha¨ndertagande av
trafikdagvatten. Vatten, Stockholm.
Sansalone, J.J., Buchberger, S.G., Al-Abed, S.R., 1996. Fractionation
of heavy metals in pavement runoff. The Science of the Total
Environment 189/190, 371–378.
Sculthorpe, C.D., 1967. The Biology of Aquatic Vascular Plants.
Edward Arnold, London.
Sokal, R.R., Rohlf, F.J., 1995. Biometry. W.H. Freeman and
Company, New York, United States of America.
St-Cyr, L., Campbell, P.G.C., Guertin, K., 1994. Evaluation of the role
of submerged plant beds in the metal budget of a fluvial lake.
Hydrobiologia 291, 141–156.
Va´squez, M.D., Lo´pez, J., Carballeira, A., 1999. Uptake of heavy
metals to the extracellular and intracellular compartments in three
species of aquatic bryophyte. Ecotoxicology and Environmental
Safety 44, 12–24.
Williams, T.P., Bubb, J.M., Lester, J.N., 1994. Metal accumulation
within salt marsh environments: a review. Marine Pollution
Bulletin 28 (5), 277–290.
Wittman, G., 1979. Toxic metals. In: Fo¨rstner, U., Wittman, G.T.W.
(Eds.), Metal Pollution in the Aquatic Environment. Springer-
Verlag, Berlin, pp. 3–70.
Yurukova, L., Kochev, K., 1994. Heavy metal concentration in
freshwater macrophytes from the Aldomirovsko swamp in the
Sofia District, Bulgaria. Bulletin of Environmental Contamination
and Toxicology 52 (4), 627–632.
Zumdahl, S.S., 1992. Chemical Principles. D.C. Heath and Company,
Canada.