Content uploaded by Marianne V. Moore
Author content
All content in this area was uploaded by Marianne V. Moore on Jan 20, 2018
Content may be subject to copyright.
Verh. Internat. Verein. Limnol. 27 1–4 Stuttgart, October 2000
©2000 E. Schweizerbart’sche Verlagsbuchhandlung, D-70176 Stuttgart
0368-0770/00/0027-01 $ 1.00
Urban light pollution alters the diel vertical migration of Daphnia
Marianne V. Moore, Stephanie M. Pierce, Hannah M. Walsh, Siri K. Kvalvik and Julie D. Lim
Introduction
Light is the fundamental factor controlling the diel
vertical migration (DVM) of zooplankton (RINGEL-
BERG 1987, HANEY 1993). It not only serves as the
proximate cue triggering the ascent of zooplankton,
but it also reduces the amplitude of migration if light
levels are sufficiently high at night. For example, the
light of a full moon reduces the amplitude of Daph-
nia (GLIWICZ 1986, DODSON 1990) and chaoborid
(SMITH et al. 1992) migrations. Night-time light
intensities, however, are influenced not only by
moonlight but also by artificial outdoor lighting,
particularly in urban areas. Light pollution, or the
sky glow produced by inefficient outdoor lighting, is
prevalent in metropolitan areas (LOCKWOOD et al.
1990), and these areas often border freshwater lakes,
coastal ecosystems, or both. Urban water quality
may be influenced indirectly by light pollution,
because zooplankton grazing influences water quality
and the depth distribution of many zooplankters is
affected by light.
We tested the hypothesis that light pollution asso-
ciated with urban areas reduces the amplitude and
magnitude of zooplankton vertical migration. A field
experiment manipulating underwater light intensity
at night was performed in a suburban lake bordering
a large metropolitan area.
Methods
An enclosure experiment was performed on 7–8 July,
1997, 3 nights after the new moon, in Lake Waban
(area 0.4 km2, mean depth 4.8 m), a dimictic kettle
lake, located 16 km southwest of Boston in Norfolk
County, MA. Lake Waban's watershed (~36 km2)
circumscribes dense residential and commercial areas
as well as major highways (MOORE et al. 1998).
According to Carlson's Trophic State Index, this sub-
urban lake falls between the boundary of mesotro-
phy and eutrophy (see MOORE et al. 1998 for
limnological details). Resident fish include bluegill
(Lepomis macrochirus), pumpkinseed sunfish (Lepo-
mis gibbosus), golden shiners (Notemigonos crysoleu-
cas), yellow perch (Perca flavescens), black crappie
(Pomoxis nigromaculatus), white perch (Morone amer-
icana), largemouth bass (Micropterus salmoides),
chain pickerel (Esox niger), brown bullhead (Ictalurus
nebulosus), and common carp (Cyprinus carpio). Low
densities (0.3 ± 0.1 ind./L) of the planktonic preda-
tor, Chaoborus punctipennis, also occur in the lake
(FISCHER & MOORE 1993).
Field enclosures (diameter, 56.6 cm; length, 4.5
m), constructed from collapsible, coiled metal
frames, were covered with either thick black (6 mm)
or clear (4 mm) sheet plastic to shield or expose por-
tions of the water column, respectively, to night-time
downwelling irradiance. All enclosures were open at
the bottom, but closed at the surface with lids made
from circular plastic hoops (diameter, 80 cm) and
covered with the appropriate plastic sheet. A slit, cut
in the middle of the plastic of each lid, allowed pas-
sage of sampling equipment. Black enclosures
blocked up to 96% of daytime downwelling irradi-
ance according to a comparison of light measure-
ments made with a Li-Cor quantum sensor in these
enclosures and the lake.
At midday, prior to the evening experiment, the
Secchi disk depth and underwater light measure-
ments were obtained with a Li-Cor quantum sensor.
The latter measurements were made per meter from
the surface to the bottom of the lake. Daytime depth
distributions of crustacean zooplankton were also
determined at depths of 2, 4, 6, 8, and 9 m by filter-
ing water samples collected with a Kemmerer bottle
(4.2 L) through a plankton net (10 µm mesh). All
zooplankton samples were preserved immediately
with 5% buffered, sucrose–Formalin.
Three experimental enclosures (black plastic) and
three control enclosures (clear plastic) were deployed
at the deepest point (11.5 m) of the lake 30 min
before sunset. In addition, three adjacent locations in
the lake served as controls for “enclosure effects”. A
total of nine sites for the enclosures and the lake
sampling were arranged in a grid pattern of three
rows with three sites per row. Treatments (black,
clear, and open lake) were blocked across rows, and
treatments were randomly assigned a site within each
row. Blocking of treatments across rows prevented
isolated light sources near shore from differentially
2 Biology of freshwater Crustacea
affecting some treatments but not others.
Immediately prior to sampling the enclosures,
temperature and dissolved oxygen were measured
each meter from the lake surface to the substrate
with a YSI model 58 oxygen meter. Night-time light
intensities were below the detection limit (0.01
µEinst/m2/s) of a Li-Cor quantum sensor so subse-
quent measurements were made with an Optec SPS-
3 photomultiplier on 9 October, 1997. On the night
of the experiment, between 23:45 and 03:45 h, three
replicate water samples were collected with a
Kemmerer bottle (4.2 or 3.2 L) at each of three
depths (2, 4, and 6 m) inside or directly below each
enclosure and in the open lake locations. Samples
were filtered and preserved as described previously.
Crustacean zooplankton were later identified to spe-
cies and counted using a stereo microscope (30 ).
Nauplii were sub-sampled with a 5-mL Hensen–
Stempel pipette when their densities 50 ind./mL.
The coefficient of variation for sub-sampling was
<0.02.
Proportions of zooplankton per taxon collected at
night were compared among depths and treatments
using two-way ANOVA. In the lake, zooplankton
proportions per taxon were also compared among
times (day vs. night) and depth to determine if taxa
exhibited DVM. All zooplankton proportions were
normalized using an arcsin transformation (SOKAL &
ROHLF 1981).
Results and discussion
The depth distribution of a single species,
Daphnia retrocurva, differed significantly
among treatments (Table 1; F4,68 = 4.11, P <
0.01). The movement of Daphnia was signifi-
cantly greater in both amplitude (2 m higher)
and magnitude (10–20% more individuals) in
the black enclosures than in control enclosures
or the lake (Fig. 1). An ‘enclosure effect’ did not
bias the results, because the depth distribution
of Daphnia was similar in the clear enclosures
and the lake. Also, because all enclosures were
open at the bottom, kairomones from predators
and the abundance and distribution of algal
food should have been similar among enclo-
sures. Interestingly, Daphnia did not exhibit
DVM in the lake. Its vertical distribution was
similar during the day and night in Lake
Waban (F2,30 = 12.26, P > 0.05).
None of the other cladoceran (Bosmina and
Diaphanosoma) or copepod species in Lake
Waban were affected by light pollution in the
enclosure experiment. Copepod nauplii and D.
retrocurva dominated numerically the crusta-
cean zooplankton community with mean densi-
ties of 41.4 and 37.3 ind./L, respectively, across
all sampling depths. Of the four copepod spe-
cies present, Diaptomus minutus was most
abundant (mean density, 5.7 ind./L across all
sampling depths).
Night-time light intensity resulting from
light pollution at the surface of Lake Waban
was less than that of full moonlight (i.e. 0.01
µEinst/m2/s; MOORE RODENHOUSE 1986) and
ten times less than that measured at the surface
of an urban lake near the center of Boston, MA.
Problems with the photomultiplier prevented
Table 1. Two-way ANOVA comparing proportional
abundance of Daphnia retrocurva among treatments
(black enclosures, clear enclosures and open lake)
and depths (2, 4, and 6 m) in Lake Waban, Massa-
chusetts on 8 July, 1997. Proportional abundance is
the proportion of total density over all depths sam-
pled.
Source df Sum of squares F P
Treatment 2 0.002 0.94 >0.05
Depth 2 0.179 99.96 <0.01
Treatment depth 4 0.015 4.11 <0.01
Error 68 0.061
Fig. 1. Mean (n = 3) percent abundance (% ± 1 SD)
of Daphnia retrocurva (per L) across all sampling
depths (2, 4, and 6 m) and treatments (Clear: clear
enclosure; Black: black enclosure; Lake: lake sam-
pling site) in Lake Waban, Massachusetts on 8 July
1997.
M. V. Moore et al., Urban light pollution and vertical migration of Daphnia 3
us from obtaining absolute measurements of
light intensity at night. At midday, 7 July 1997,
the Secchi disc depth of Lake Waban was 1.7
m, and the light extinction coefficient (k)
equaled 0.82/m. On the night of the experi-
ment, the metalimnion occurred between 4 and
10 m, and hypoxic or anoxic conditions
occurred at depths 5 m (Fig. 2).
Using the Secchi depth recorded for Lake
Waban and a regression model developed by
DODSON (1990), a migration amplitude of 2.1
m is predicted empirically for Daphnia under
normal conditions (i.e. no light pollution) in
Lake Waban. Daphnia’s upward movement of 2
m in the black enclosures, coupled with its
absence of DVM in the lake, suggests that light
pollution either eliminated Daphnia DVM or
reduced its amplitude to a distance too small
for detection with the experimental methods
used. Interestingly, the model of DODSON
(1990) also predicts that full moonlight reduces
the amplitude of Daphnia DVM by 2 m in
north temperate lakes. This prediction, in con-
junction with the results of this current study,
suggests that the ecological effects of light pol-
lution in this lake are comparable to those of
full moonlight.
Suppression of zooplankton DVM by light
pollution is probably most likely to occur in
lakes with fish and relatively clear water. In
lakes with fish, Daphnia genotypes most sensi-
tive to light (i.e. negative phototaxis) occur, and
these genotypes respond more strongly to the
induction of DVM by fish kairomones than
clones from fishless habitats (DE MEESTER
1993). Finally, the penetration of light pollu-
tion and its effects should be greater in clear
lakes with low concentrations of DOC and
algae.
The suppression of DVM by light pollution
may have consequences for both algae and
zooplankton. Algal mortality in the epilimnion
may be reduced due to lower rates of zooplank-
ton grazing. Alternatively, the lack of nutrient
regeneration by zooplankton in the upper sur-
face waters at night could actually slow algal
growth in lakes experiencing severe nutrient
limitation (STERNER & HESSEN 1994). If light
pollution confines zooplankton to metalim-
netic depths, individual growth and reproduc-
tion may decline markedly due to colder water
temperatures (LOOSE & DAWIDOWICZ 1994).
Conclusions
Diel vertical migration of Daphnia was signifi-
cantly reduced in both amplitude (2 m lower)
and magnitude (10–20% fewer individuals) by
urban light pollution in a suburban lake.
Reduced algal grazing by zooplankton at night
in epilimnetic waters could potentially contrib-
ute to enhanced algal biomass in lakes and
coastal waters near urban areas, thereby lower-
ing water quality.
Acknowledgements
We thank SUE KOHLER for her expert technical assis-
tance regarding light instrumentation and LEE
HAWKINS for the loan of the Optec SPS-3 photomul-
tiplier. Field assistance was kindly provided by MYNA
JOSEPH. This research was supported by a Brach-
man–Hoffman Fellowship to the first author. Co-
authors were supported by an NSF-REU site grant
(IBN 9424179) and grants from the Sherman Fair-
child Foundation and the Howard Hughes Medical
Institute.
Fig. 2. Temperature (°C) and dissolved oxygen
(mg/L) profiles for Lake Waban, Massachusetts on
7 July 1997 at 21:00 h.
4 Biology of freshwater Crustacea
References
DE MEESTER, L., 1993: Genotype, fish-mediated chemicals,
and phototactic behavior in Daphnia magna. – Ecology 74:
1467–1474.
DODSON, S., 1990: Predicting diel vertical migration of zoop-
lankton. – Limnol. Oceanogr. 35: 1195–1200.
FISCHER, J. M. & MOORE, M. V., 1993: Juvenile survival of a
planktonic insect: effects of food limitation and predation.
– Freshwat. Biol. 30: 35–45.
GLIWICZ, M. Z., 1986: A lunar cycle in zooplankton. – Ecology
67: 883–897.
HANEY, J. F., 1993: Environmental control of diel vertical
migration behaviour. – Arch. Hydrobiol. Beih. Ergebn. Lim-
nol. 39: 1–17.
LAMPERT, W., 1992: Zooplankton vertical migrations: Implica-
tions for phytoplankton–zooplankton interactions. – Arch.
Hydrobiol. Beih. Ergebn. Limnol. 35: 69–78.
LOCKWOOD, G. W., FLOYD, R. D. & THOMPSON, D. T., 1990:
Sky glow and outdoor lighting trends since 1976 at the
Lowell observatory. – Publ. Astron. Soc. Pac. 162: 481–491.
LOOSE, C. J. & DAWIDOWICZ, P., 1994: Trade-offs in diel verti-
cal migration by zooplankton: The costs of predator avoid-
ance. – Ecology 75: 2255–2263.
MOORE, M. V., ZAKOVA, P., SHAEFFER, K. A. & BURTON, R. P.,
1998: Potential effects of Canada geese and climate change
on phosphorus inputs to suburban lakes of the Northeast-
ern U.S.A. – J. Lake Reservoir Manage. 14: 52–59.
MOORE RODENHOUSE, M., 1986: Age-specific predation and pre-
dation risk of Chaoborus punctipennis. – Ph.D. dissertation,
Dartmouth College, Hanover.
RINGELBERG, J., 1987: Light induced behaviour in Daphnia. –
In: PETERS, R. H. & DE BERNARDI, R. (eds): Daphnia:
285–323. – Mem. Ist. Ital. Idrobiol.
SMITH, S. L., PIEPER, R. E., MOORE, M. V., RUDSTAM, L. G.,
GREENE, C. H., ZAMON, J. E., FLAGG, C. N. & WILLIAMSON,
C. E., 1992: Acoustic techniques for the in situ observation
of zooplankton. – Arch. Hydrobiol. Beih. Ergebn. Limnol.
36: 23–43.
SOKAL, R. R. & ROHLF, F. J., 1981: Biometry. – Second edition.
W. H. Freeman, San Francisco, 859 pp.
STERNER, R. W. & HESSEN, D. O., 1994: Algal nutrient limita-
tion and the nutrition of aquatic herbivores. – Annu. Rev.
Ecol. Syst. 25: 1–29.
Authors' addresses:
MARIANNE V. MOORE, STEPHANIE M. PIERCE, HANNAH
M. WALSH, SIRI K. KVALVIK, Department of Biologi-
cal Sciences, Wellesley College, Wellesley, Massachu-
setts, 02481-8203 USA.
JULIE D. LIM, Newton North High School, 360 Low-
ell Avenue, Newtonville, Massachusetts, 02460 USA.
