2007 NORTHEASTERN NATURALIST 14(2):235–250
Natural Communities in Catch Basins in
Southern Rhode Island
Mari Butler1,2,*, Howard S. Ginsberg2, Roger A. LeBrun1,
Alan D. Gettman3, and Fred Pollnak1
Abstract - Storm-water drainage catch basins are manmade structures that often
contain water and organic matter, making them suitable environments for various
organisms. We censused organisms inhabiting catch basins in southern Rhode Island
in 2002 in an effort to begin to describe these communities. Catch-basin inhabitants
were mostly detritivores, including annelids, arthropods, and mollusks that could
withstand low oxygen levels and droughts. Our results suggest that catch-basin
inhabitants were mostly washed in with rainwater, and populations increased over
the summer season as biotic activity resulted in increased nutrient levels later in the
summer. In contrast, mosquitoes and other Diptera larvae were abundant earlier in
the summer because the adults actively sought catch basins for oviposition sites.
Mosquito larvae were likely to be abundant in catch basins with shallow, stagnant
water that had relatively low dissolved oxygen and pH, and relatively high total
suspended solids, carbon, and nitrogen.
Introduction
Storm-water catch basins often contain stagnant, highly organic water
providing suitable breeding conditions for several species of mosquitoes and
a variety of other organisms (Ishii and Okubo 1989, Kikuchi 1992). Mosquito
species commonly found in catch basins in temperate regions include
Culex species such as C. pipiens Linnaeus and C. restuans Theobald (Covell
and Resh 1971, Gerry and Holub 1989, Ishii and Okubo 1989, Kikuchi 1992,
Knepper et al. 1992, McCarry 1996, Munstermann and Craig 1977, Siegal
and Novak 1997) and Aedes species such as A. vexans Meigen (Covell and
Resh 1971, Gerry Holub 1989) and A. albopictus Skuse (Ishii and Okubo
1989). Many of these species are known vectors of diseases including
arboviruses such as West Nile Virus. Considering the close proximity of
catch basins to humans, many municipalities target them for mosquito control.
Presently, methoprene slow-release pellets and briquets are popular
choices to use for mosquito control in these environments (Knepper et al.
1992, McCarry 1996, Schoeppner 1978). Other organisms inhabiting catchbasin
environments are also exposed to methoprene and other pesticides, yet
they have not been carefully examined. Information about biotic communities
living in catch basins, including biological, physical, and chemical
1University of Rhode Island, Kingston, RI 02881. 2Current address - Endicott College,
376 Hale Street, Beverly, MA 01915. 3USGS Patuxent Wildlife Research
Center, Kingston, RI 02881. 4Rhode Island Department of Environmental Management,
Mosquito Abatement Office, Kingston, RI 02881. *Corresponding author -
mbutler@endicott.edu.
236 Northeastern Naturalist Vol. 14, No. 2
parameters may also be useful in improving mosquito control tactics. For
example, understanding environmental conditions most likely to harbor
large numbers of mosquitoes and learning to take advantage of naturally
occurring mosquito predators may help target mosquito control efforts.
One objective of this study was to census organisms living in catch
basins, and to describe seasonal patterns in abundances of taxonomic
groups. The second objective was to model the abundance of mosquitoes, the
major group that actively oviposited in catch basins, as a function of environmental
factors (temperature, pH, dissolved oxygen, depth, conductivity,
total suspended solids, and the amount of carbon and nitrogen in the water).
Methods
Catch basin census
Thirty catch basins from two sites in Narragansett, RI, were sampled six
times between May 10 and November 7, 2002. One site consisted of 16 catch
basins all connected along a single street sharing an outflow pipe into
Narragansett Bay. The second site consisted of 14 catch basins on three
parallel roads that emptied into the Narrow River, Narragansett, RI. Each
catch basin covered an area of approximately 1.5 m2 with varying depths.
Biotic communities were sampled using a custom-made catch-basin sampler
capable of sampling the water column and the neustonic and benthic interfaces.
The sampler was constructed of a Plexiglas tube 15.24 cm in diameter
and 85 cm tall, opened at the bottom and sealed at the top except for a small
hole the size of a standard boat plug. Two poles were attached on opposite
sides to extend the reach of the two people lowering the sampler. The catchbasin
cover was removed, and the sampler was lowered vertically to the
bottom with the hole on top of the sampler left open. Once the sampler was
resting on the bottom, the boat plug was inserted into the hole using another
custom-designed pole. The sampler, containing a known volume of water
(area of sampler [m2] x depth of water [m]) that spanned from the air/
water interface to the water/sediment interface, was lifted until the bottom of
the sampler was just below the surface of the water. At this point a 164-m
sieve attached to another pole was held under the sampler and all were
slowly lifted above the surface of the water. As water drained out of the
sampler and through the sieve, particles greater than 164 m were captured
and rinsed into a sampling jar. The efficacy of the sampler was tested by
taking several samples with the sampler from test catch basins in the lab, and
then comparing abundances estimated by the sampler to abundances estimated
from the entire catch-basin contents. The proportions of different taxa
collected by our sampler did not differ significantly from samples taken by
subdividing the entire catch-basin contents (G < 7.1, P > 0.06).
In 16 of 180 sampling attempts, catch basins were completely dry, and
samples were not taken. In three of the 164 samples taken, water was
present, but too shallow to use the catch-basin sampler. In these cases, a
250-ml standard mosquito dipper was used to collect organisms. Samples
2007 M. Butler, H.S. Ginsberg, R.A. LeBrun, A.D. Gettman, and F. Pollnak 237
were stored on ice while the rest of the sampling was completed (usually
not more than 6 hours). Samples were brought back to the laboratory where
they were placed in near-boiling water to fix soft-bodied organisms, and
preserved in 70% ethanol. After 24 hours, the majority of the ethanol was
replaced with fresh ethanol. Organisms were later sorted from debris,
enumerated, and identified to the lowest possible taxonomic level. Because
annelids were often broken, their abundance was estimated as biomass.
Biomass of oligochaete worms was calculated from displacement volume,
and for other annelids (Hirudinea), a conversion factor (assuming all
leeches were of similar size) was calculated and applied to the numerical
abundance data. Abundances for all other taxonomic groups were calculated
as number per liter, and communities were characterized by plotting
dominance-diversity curves.
Modeling mosquito abundance
The number of mosquito larvae-per-liter living in each of the 30 catch
basins was monitored for six months as part of the catch-basin census.
Environmental factors included temperature, pH, conductivity, dissolved
oxygen, and water depth. Particulate carbon and nitrogen, and total suspended
solids greater than 0.7 m per liter, were also measured in the same
30 catch basins at the time the biological samples were taken. A Quanta®
environmental sensor probe was used to measure temperature, dissolved
oxygen, pH, and conductivity. Water depth was measured using a tape
measure. Total suspended solids and the amount of carbon and nitrogen per
liter were measured by filtering a known amount of water collected from the
catch basin thru a pre-weighed, pre-combusted glass fiber filter. The filters
were dried in a drying oven at 60 °C, weighed again, and the weight of
material collected on the filter calculated. Filters were then analyzed for
carbon and nitrogen content using a Carlo Erba EA1108 CHN analyzer
located at the Graduate School of Oceanography, University of Rhode Island.
Carbon and nitrogen were not measured in May and June due to a
problem with sampling methodology.
Step-wise regression analysis was used to model the number of mosquitoes
per liter based on independent variables consisting of the environmental
factors mentioned above (Tabachnick and Fidell 1989). Independent variables
were left in the model if P < 0.1 and removed from the model if P > 0.2
using Statistical Package for the Social Sciences (SPSS), and
multicolinearity was monitored and addressed if it significantly affected the
models (Pallant 2001).
Results
Catch-basin census
All animals collected were from the phyla Annelida, Mollusca, and
Arthropoda (Table 1), and total animal density (excluding Annelida) ranged
from an average of 5.4 per liter in June to an average of 43.0 per liter in
238 Northeastern Naturalist Vol. 14, No. 2
August (Fig. 1). The average number of different taxonomic groups in
individual catch basins ranged from 4.8 in October to 5.7 in August. Oribatei
soil mites were the most abundant organisms in May (2.9 per liter), when
overall average counts were low (9.5 organisms per liter), and they were
present throughout the sampling period (0.7 per liter in June to 3.8 per liter
in September). Copepods were the most abundant organisms in June (1.6 per
liter) and in September (11.3 per liter). They were consistently present
throughout the sampling period, sometimes in high numbers (86.4 per liter
from a single catch basin in August and 193.0 per liter from a single catch
Table 1. Taxa present in catch basins in southern Rhode Island.
Phyla, class, order Family, genus, and species
Phylum Arthropoda
Class Insecta
Order Diptera Family Culicidae
Culex restuans Theobald
Culex pipiens Linnaeus
Aedes japonicus Theobald
Unknown Culicid larvae
Famly Psychodidae
Family Chironomidae
Family Ceratopogonidae
Unknown Diptera larvae
Order Coleoptera Family Ptelalidae
Unknown Coleopteran larvae
Order Collembola Family Entomobryidae
Family Isotomidae
Family Onchyuridae
Family Poduridae
Family Sminthuridae
Unknown Collembola
Class Crustacea
Order Podocpoia (Ostracods)
Order Amphipoda Family Talitridae
Order Isopoda Family Assellidae (Caecidotea)
Order Copepoda Family Cyclopoida
Macrocyclops albidus Jurine
Paracyclops poppei Rehberg
Family Harpacticoid
Unidentified nauplii
Class Arachnida
Order Acarina
Suborder Hydracarina
Suborder Oribatidae
Phylum Mollusca
Class Gastropoda
Order Basommatophora Family Planorbidae
Family Physidae
Class Pelecypoda Family Bivalvia
Sphaerium occidentale Lewis
Phylum Annelida
Class Oligochaeta
Class Hirudinea
2007 M. Butler, H.S. Ginsberg, R.A. LeBrun, A.D. Gettman, and F. Pollnak 239
basin in September). They were also present in every catch basin in our
study at least once during the six-month study period. Mosquitoes dominated
the communities during July and August, but were present in all
months sampled. It is clear from the dominance-diversity curves (Fig. 1) that
when mosquitoes were abundant (July and August), they comprised a large
portion of the community. Hydracarina (water mites) were the most abundant
as fall approached. Because volume rather than number was measured
for Annelida (primarily Oligochaeta with a few Hirudinea), biovolume data
are shown separately (Fig. 2). Oligochaete worms were the only other
organisms that were present in all 30 catch basins at least once during the
sampling period.
Figure 1. Dominance-diversity plots of organisms caught each month. The Y-axes
are the number of organisms per liter, and the X-axes list the organisms from the
most abundant to least abundant. Numbers per liter shown above each graph are the
average numbers of organisms captured each month. The "n" is the number of catch
basins sampled.
240 Northeastern Naturalist Vol. 14, No. 2
Mosquitoes included both larval and pupal stages of C. pipiens, C.
restuans, and Aedes japonicus Theobald. The vast majority of the mosquito
larvae were Culex spp. Only 1 A. japonicus out of 1890 total mosquitoes was
positively identified for the month of July, and 8 out of 1390 total mosquitoes
from the month of August. Diptera included larval Diptera other than
mosquitoes such as Psychodidae, Chironomidae, Ceratopagonidae, and
other unidentified larvae. Coleopterans included larval Ptelalidae and other
unidentified larvae. Collembola included representatives from the following
families: Entomobryidae, Isotomidae, Poduridae, Sminthuridae, and
Onychiuridae. Gastropods included Physidae and Planorbidae. Ostracods
included a small number of specimens likely from the order Podocopida.
Bivalves were identified as Sphaerium occidentale Lewis (Harrington’s
fingernail clam), which is a new Rhode Island record (Jay Cordeiro,
NatureServe and the American Museum of Natural History, Boston, MA,
pers. comm.). Copepods included predominantly cyclopoids with a small
number of harpacticoids. Harpacticoids were likely Harpacticus spp. and the
two dominant cyclopoids were identified as Macrocyclops albidus Jurine
and Paracyclops poppei Rehberg (Janet Reid, Virginia Museum of Natural
History, Martinsville, VA, pers. comm.). Hydracarina (water mites) included
a variety of different stages and species of water mites. All darkcolored,
thick-skinned mites were classified as Oribatei (soil mites). Isopods
were identified as the family Asellidae and the genus Caecidotea. Amphipods
included mostly immature specimens, thought to be from the family
Talitridae and species Hyalella azteca Saussure (Maria Aliberti, University
of Rhode Island, Kingston, RI, pers. comm.).
In addition to the aquatic organisms addressed in this paper, a variety
of terrestrial arthropods were found in many of the catch basins, but not
included in these analyses since they were not truly members of the
aquatic community. Orders represented included: Protura, Diplura,
Figure 2. Mean volume of Annelida (± one standard error) caught each month. The
number of samples taken per month ranged from 25 to 30.
2007 M. Butler, H.S. Ginsberg, R.A. LeBrun, A.D. Gettman, and F. Pollnak 241
Diptera adults, Hymenoptera (predominantly Formicidae), Hemiptera,
Psocoptera, Coleoptera (including Carabidae, Scarabidae, Curculionidae,
and others), Arachnida, Lepidoptera larvae, Thysanaptera, Homoptera
(mostly aphids), Diplopoda (Millipedes), and Orthoptera.
As the summer went on and the water temperature increased, abundances
increased with peaks in most populations (Copepoda, Isopoda, Amphipoda,
and Mollusca) occurring in August or September. The Diptera, including the
mosquitoes, peaked earlier (July). Annelida populations were high in August
continuing into October, and the mites were always present in relatively low
numbers, increasing as the season progressed (Fig. 3).
Figure 3. Average numbers from each taxonomic group (± one standard error)
caught each month. The number of samples taken per month ranged from 25 to 30.
242 Northeastern Naturalist Vol. 14, No. 2
Environmental conditions and mosquito abundance
Catch basins experienced a range of environmental conditions over
the course of the season (Fig. 4). Water temperature spanned about 10 °C
in 6 months. Conductivity varied little with the exception of one catch
basin that emptied directly into the Narrow River, where a relatively high
conductivity was recorded repeatedly. pH was approximately neutral, except
for a single low reading of 3.97 taken in June and a single high
measurement of 11.61 taken in August from two different catch basins.
Dissolved oxygen (DO), total suspended solids per liter (TSS), the
amount of carbon per liter (C), and the amount of nitrogen per liter (N)
were variable, with a decline in DO and increase in particles per liter
through the summer months (Figs. 4 and 5).
The ability of environmental variables to predict mosquito abundance
was assessed using stepwise regression on the samples taken each month
(Table 2). Mosquito numbers were negatively related to DO in 3 of the 6
months and to pH levels in 2 months, and positively related to C in 2 months
and to N and TSS in one month each. In October, only 4 mosquito larvae
were caught. Correlations between environmental variables are shown in
Table 3.
Figure 4. Environmental factors measured each month. Symbols represent mean ±
standard error.
2007 M. Butler, H.S. Ginsberg, R.A. LeBrun, A.D. Gettman, and F. Pollnak 243
Figure 5. Indices
of available
food by month
(mean amount
± one standard
error). No carbon
or nitrogen
samples were
analyzed for
May and June.
Discussion
Catch-basins census
Very few researchers have examined organisms other than mosquitoes
living in catch basins. However, Kikuchi’s (1992) work evaluating nontarget
effects of methoprene in an urban drain in Tokushima, Japan provided some
information about other organisms living in this sort of environment. He
examined effects of methoprene on Syrphidae (hover flies), Assellus
hilgendorforii Bovallius (isopods), Cleon dipterum Linnaeus (mayfly),
Chironomidae (midges), and Hermetia illucens Linnaeus (black soldier
244 Northeastern Naturalist Vol. 14, No. 2
flies) living in catch basins. He also studied Physa fontalis Linnaeus (a
mollusk) and Assellus hilgendorforii in the laboratory, suggesting that they
too are found in catch basins (Kikuchi 1992). Ishii and Okubo (1989)
evaluated one hundred and fifty-six catch basins in Tokushima City in the
summer of 1988; of the 65% that were found flooded, 22% harbored Culicidae
(mosquitoes), 17% harbored Chironomidae (midges) and 8% harbored
Psychodidae (moth flies).
Many of the organisms that we found were from the same families as were
found in the studies mentioned above. Nearly all of the organisms captured in
our study can be classified as scavengers or detritivores, feeding chiefly on
dead plant or animal material as well as on fungi, algae, protozoans, periphyton,
and other organic detritus. Over the course of our sampling, we observed
increases in most organisms as the season progressed thru the summer
months and a decline in October (Fig. 3). Annelids (Fig. 2), however,
continued to increase into October. A similar trend was noted in organic
matter, the probable source of food for most of these organisms, which we
estimated by measuring TSS, C, and N (Fig. 4). Unfortunately, C and N data
for May and June were not available. We speculate that organisms and other
detritus were rinsed or blown into catch basins as the summer progressed,
causing peaks in late summer or early fall. The only groups that did not
follow this pattern were the mosquito larvae and other Diptera larvae. Their
populations peaked in July and then declined. They were perhaps the only
abundant organisms present due to the adults actively flying into the catch
basin to lay their eggs, in contrast to the other common organisms that arrived
predominantly through passive dispersal. This might account for their lack of
Table 2. Summary of environmental factors—dissolved oxygen (DO), the amount of carbon per
liter (C), the amount of nitrogen per liter (N), and total suspended solids per liter (TSS)—as
predictors of number of mosquito abundance. “n” is the number of catch basins sampled, and
the “(-)” or “(+)” next to the variables contributing to the model denote correlation with
mosquito abundance.
Independent
% catch variables
basins with included in the
Month n mosquitoes regression equationA R2 P
May 30 10% DO (mg/L)(-) 0.1 0.038
44
June 29 38% pH (-) 0.5 < 0.001
74
July 25 56% DO (mg/L) (-) 0.2 0.015
Depth (cm)(-) 54
August 26 54% C (mg/L) (+) 0.6 < 0.001
N (mg/L) (+) 25
September 27 33% C (mg/L) (+) 0.7 < 0.001
TSS (mg/L) (+) 87
pH (-)
DO (mg/L) (-)
October 27 7% ns ns ns
AStepwise analyses were added if P 0.1 and removed if P > 0.2.
2007 M. Butler, H.S. Ginsberg, R.A. LeBrun, A.D. Gettman, and F. Pollnak 245
synchronicity with the rest of the catch-basin inhabitants. Mosquitoes might
be responsible for transporting other organisms such as parasitic mites
among catch basins. This transport might contribute to the slight increase in
species richness that was seen in August and also the overall increasing
numbers displayed by the other organisms.
The structure of catch-basin communities is unique, but is similar in some
ways to those of temporary woodland pools. Early colonizers of temporary
pools are often detritivores that overwinter in resting stages and are capable
only of passive dispersal. They include Copepoda, Ostracoda, Oligochaeta,
Hirudinea, and Mollusca, as well as others not found in catch basins (Wiggins
et al. 1980). This group accounts for the majority of organisms identified in
Table 3. Correlation between independent variables used to predict mosquito abundances in
catch basins in 2002.
Month Related variables Pearson correlation
May pH and Temp r = 0.987, P < 0.0001
Depth and TSS r = -0.478, P = 0.004
June TSS and Depth r = -0.442, P = 0.007
July N and C r = 0.864, P < 0.0001
TSS and C r = 0.840, P < 0.0001
DO and Temp r = -0.644, P < 0.0001
TSS and N r = 0.578, P = 0.001
DO and N r = -0.567, P = 0.002
Depth and C r = -0.464, P = 0.010
Temp and N r = 0.425, P = 0.017
Depth and N r = -0.425, P = 0.017
DO and C r = -0.382, P = 0.030
Aug. N and C r = 0.996, P < 0.0001
Temp and DO r = -0.629, P < 0.0001
Temp and TSS r = -0.493, P = 0.005
DO and TSS r = 0.399, P = 0.022
September N and C r = 0.982, P < 0.0001
TSS and N r = 0.958, P < 0.0001
TSS and C r = 0.945, P < 0.0001
C and pH r = -0.502, P = 0.009
N and pH r = -0.489, P = 0.010
C and Depth r = -0.450, P = 0.018
TSS and pH r = -0.398, P = 0.033
TSS and Depth r = -0.382, P = 0.040
N and Depth r = -0.378, P = 0.041
TSS and Temp r = -0.372, P = 0.044
C and Temp r = -0.362, P = 0.049
October N and C r = 0.969, P < 0.0001
TSS and N r = 0.946, P < 0.0001
TSS and C r = 0.896, P < 0.0001
pH and DO r = 0.565, P = 0.001
N and Depth r = -0.463, P = 0.008
Depth and C r = -0.450, P = 0.009
TSS and Depth r = -0.393, P = 0.021
N and pH r = -0.363, P = 0.031
Conductivity and pH r = 0.362, P = 0.032
TSS and pH r = -0.328, P = 0.047
246 Northeastern Naturalist Vol. 14, No. 2
this study and was present early in the season increasing as the summer
progressed. The second group described by Wiggins (1980) consisted of
organisms capable of some dispersal such as Diptera, including mosquitoes.
Wiggins’ (1980) descriptions support our hypothesis that most of the organisms
found in catch basins arrived there passively with the exception of the
mosquitoes and other Diptera larvae.
Another similarity between communities of catch basins and temporary
pools is the taxa richness. Taxa richness in temporary pools was found to be
related to habitat size and permanence. Larger temporary pools that were
more permanent contained more species (Spencer et al. 1999). Catch basins
have a small surface area ( 1.5 m2), and the permanence of the water is
extremely variable. A temporary pool of about 1 m2 has slightly fewer than 5
species (Spencer et al. 1999). This is in close agreement with the average
taxa richness found in each catch basin, which ranged from 4.8 to 5.7 taxa.
Furthermore, Spencer et al. (1999) found that smaller and less permanent
pools had low proportions of predators. Perhaps the size or lack of permanence
of catch basins limits the number of predators that can live there.
Oribatei mites, Hydracarina, mosquito larvae, Copepoda, Collembola,
and Isopoda repeatedly were the most abundant organisms during our sampling
season. The dominance-diversity curves (Fig. 1) showed a community
evenly spread between several different taxonomic groups for most months.
The curves were relatively flat in May and June, while curves from August,
September, and October each showed one notably abundant taxonomic
group (mosquitoes, Copepoda, and Hydracarina respectively), with the remaining
groups still represented at numbers similar to other months. In July,
mosquitoes clearly dominated the community (Fig. 1).
The majority of catch-basin organisms can be classified at the trophic level
of detritivore (Pennak 1978). However, there is evidence that some of the
specific organisms identified might be involved in biological interactions of
interest with regard to mosquito control. For example, although amphipods
are usually considered benthic detritivores, Crangonyx shoemakeri Hubricht
and Mackin was found to capture and ingest mosquito larvae, suggesting that
amphipods may be facultative predators under some conditions (Schwartz
1992). In a laboratory study, amphipods left the benthos and consumed on
average 3.5 ± 1.0 mosquito larvae in 24 hours and 6.0 ± 0.1 after 72 hours
(Schwartz 1992). Under circumstances where other predators are lacking,
such as may be the case in catch-basin environments, species traditionally
thought to be benthic feeders may be at least facultative predators. Because
amphipods eat fungi, they have enzymes needed to digest chitin, a major
constituent of fungal cell walls and arthropod cuticle (Schwartz 1992). This
would make them capable of digesting mosquito larvae, but whether this
happens in catch basins has not yet been determined. We were unable to
positively identify to species the majority of the immature amphipods we
collected, which were present in 6 of the 30 catch basins sampled. One catch
basin in particular had amphipods present every time it was sampled, and in
2007 M. Butler, H.S. Ginsberg, R.A. LeBrun, A.D. Gettman, and F. Pollnak 247
six months, only one mosquito larva was caught in that particular catch basin.
This circumstantial evidence should be tested with laboratory feeding experiments
to determine if the amphipods cohabitating with mosquito larvae in
catch basins might be employed as biological control agents.
Copepods can also be involved in biological relationships that might be
of interest in mosquito control. Macrocyclops albidus and Paracyclops
poppei were the two species of copepods commonly found in catch basins.
M. albidus is a voracious predator of early instar mosquitoes (Calliari et al.
2003; Marten et al. 1994, 2000; Rey et al. 2004). M. albidus and P. poppei
can also act as carriers of mosquito pathogens, potentially reducing populations
of Culex spp., important vectors of West Nile Virus, thereby reducing
the need for larvicide application without introducing chemicals into these
environments. Certain species of copepods are intermediate hosts for species
of microsporidia, a protist pathogen. Microsporidians can have complex life
cycles, spending part in a copepod host and part in a mosquito host, usually
leading to the demise of both hosts. Transmission of these pathogens can be
both vertical to offspring and horizontal by mouth or via integument, increasing
the potential for infection once the pathogens are introduced.
Amblyospora salinaria Becnel and Andreadis n. sp. (Microsporidia: Amblyosporidae)
is a known pathogen of C. salinarius with M. albidus as the
intermediate host (Becnel and Andreadis 1998). Although C. salinarius was
not the species of Culex we found in catch basins, perhaps this pathogen or a
related one might also infect C. pipiens or C. restuans.
Paracyclops poppei, a species previously thought to be a subspecies of P.
fimbriatus (Fischer), was also found ubiquitously in catch basins. Presently,
P. poppei is accepted as a distinct species, and P. fimbriatus is not thought to
exist in North or South America since its redescription (Karaytug and
Boxshall 1998) (Janet Reid, pers. comm). A species of microsporidia,
Amblyospora camposi, has been described infecting the mosquito C. renatoi
(Lane & Ramalho) and the copepod Paracyclops fimbriatus fimbriatus living
in leaf axils in Argentina (Micieli et al. 2000). The species described as
P. fimbriatus fimbriatus may or may not be what is now described as P.
fimbriatus (Karaytug and Boxshall 1998), but regardless of the precise
taxonomy, P. poppei should be tested for susceptibility to A. camposi, as
should the pathogenicity of A. camposi to C. pipiens and C. restuans.
Copepods were found in every catch basin at least once during the
sampling period suggesting that these common mosquito habitats are quite
suitable for copepods. Biological control of catch-basin breeding mosquitoes
using endemic copepod species is worthy of further research.
Modeling mosquito abundance
Mosquitoes were the major group that actively entered catch-basin environments,
so we studied the conditions that fostered mosquito abundance in catch
basins in further detail. Mosquito species and abundances in catch basins have
been closely evaluated because of their importance in human disease transmission
(Kronenwetter-Koepel et al. 2005). Forty catch basins in Illinois sampled
248 Northeastern Naturalist Vol. 14, No. 2
twice per month during the summer months, yielded a total of 2147 mosquito
larvae, captured using standard mosquito dippers. Sixty-three percent were C.
pipiens and 37% were C. restuans (Gerry et al. 1989). Culex restuans were
more common earlier in the sampling (June) and C. pipiens were more
common later (July and Aug) both in Illinois and Kentucky (Covell and Resh
1971, Gerry and Holub 1989). Of the Culicidae observed in catch basins in
Tokushima City, 77% were C. pipiens, 32% were A. albopictus, and 9% were
C. halifaxii (Theobald) (Ishii and Okubo 1989). Other investigators examining
pesticide efficacy in catch basins have noted the presence of Culex spp., C.
restuans, and C. pipiens (Knepper et al. 1992, McCarry 1996, Munstermann
and Craig 1977, Schoeppner 1978, and Siegal and Novak 1997)
The mosquitoes we found were predominantly C. pipiens and C.
restuans, both of which are known to prefer water with high organic
content (Crans 2004). Low dissolved oxygen and indices of food availability
(C, N, and TSS) repeatedly contributed to regression models predicting
high mosquito numbers. Low oxygen was useful in predicting mosquito
numbers in May, July, and September, and low pH was the best predictor
in June. DO and pH are intimately related through the processes of respiration
and decomposition, so it was not surprising that they were correlated.
Decomposition of organic material uses up oxygen, makes water more
acidic, and puts more organic particles into the water. In addition, oak
leaves, rich in humic acid, are common in catch basins, no doubt contributing
to the decreased pH, especially as they decompose. We did not monitor
human factors such as the use of cleaning products, lawn chemicals, or
automobile products, all of which could significantly affect pH and other
environmental factors.
Particles in the water were measured as indices of food availability, and
predictors of mosquito abundance. Unfortunately, samples filtered for carbon
and nitrogen analyses in May and June were not usable. Nonetheless,
in August, either C or N successfully predicted where mosquito numbers
were high, and in September, all three indices of food availability predicted
mosquito numbers. In addition to the amount of food in the water,
the depth of the water can influence availability of food particles to mosquito
larvae. In July, depth was left in the model along with DO. In
laboratory experiments, water depth affected the development and accumulation
of caloric reserves in mosquito larvae (Timmermann and Briegel
1993). Since food particles often sink to the bottom, larval feeding may be
hindered when the water is deeper than a few centimeters (Timmermann
and Briegel 1993). It is generally accepted that C. pipiens and C. restuans
prefer water that has a high organic content (Crans 2004). Gerry and Holub
(1989) also observed that the amount of accumulated debris increased the
incidence of mosquito larvae in catch basins. The analytical tools employed
in this study clearly support these observations. The conditions
known to harbor Culex mosquitoes were the same conditions, i.e., characterized
by low DO, low pH, high TSS, high C, and high N.
2007 M. Butler, H.S. Ginsberg, R.A. LeBrun, A.D. Gettman, and F. Pollnak 249
In conclusion, we speculate that catch-basin communities are populated
by organisms that are washed or rinsed passively into the system. These
organisms tended to increase in abundance from May through September
(Figs. 2 and 3). During this same time period, total suspended solids in the
water as well as carbon and nitrogen gradually increased (Fig. 4). Mosquitoes
and other Diptera, on the other hand, were an exception to this rule since
they actively search out sites in which to lay their eggs. They peaked in
abundance earlier in July (Fig. 3), a month before the other organisms. Our
data showed the greatest numbers of Culex spp. larvae in shallow, putrid
water characterized by high amounts of organic material (C, N, and TSS)
and decomposition (low DO and low pH).
Acknowledgments
We are grateful to the following faculty, staff, and students at the University of
Rhode Island: Maria Aliberti, Katie Allen, Dr. Steven Alm, Dr. David Bengtson,
Adam Butler, Dr. Richard Casagrande, Charles Dawson, Caryn Debatt, Linda Green,
Mary McDougle, Dr. Neeta Pardanani, Carl Sawyer, Jesse Siligato, Cannsotha
Suom, and Jane Viera. In addition, we wish to thank Jay Cordeiro and Janet Reid for
help with identifying clams and copepods, respectively. We would also like to thank
two annonymous reviewers whose comments greatly improved the quality of this
manuscript. This work was funded by the Rhode Island Agriculture Experiment
Station Hatch Grant #RI00666, URI’s Coastal Research Fellowship Program, and the
Northeastern Mosquito Control Association John L. McColgan Grant In Aid.
Literature Cited
Becnel, J.J., and T.G. Andreadis. 1998. Amblyospora salinaria n. sp. (Microsporidia:
Amblyosporidae), parasite of Culex salinarius (Diptera: Culicidae): Its life cycle
stages in an intermediate host. Journal of Invertebrate Pathology 71:258–262.
Calliari, D., K. Sanz, M. Martinez, G. Cervetto, M. Gomez, and C. Basso. 2003.
Comparison of the predation rate of freshwater cyclopoid species on the larvae of
the mosquito Culex pipiens. Medical and Veterinary Entomology 17:339–342.
Covell, C.V., and V.H. Resh. 1971. Relative abundance of Culex pipiens and Culex
restuans in catch basins in Jefferson County, Kentucky. Mosquito News 31:73–76.
Crans, W.J. 2004. A classification for mosquito life cycles: Life cycle types for
mosquitoes of the northeastern United States. Journal of Vector Ecology 29:1–10.
Gerry, P.R., and R.E. Holub. 1989. Seasonal abundance and control of Culex spp. in
catch basins in Illinois. Journal of the American Mosquito Control Association
5(4):537–540.
Ishii, T., and S. Okubo. 1989. Mosquito survey of the catch basins in Tokushima City
in 1988 summer. Journal of Science, University of Tokushima 22:17–24.
Karaytug, S.G.A., and G.A. Boxshall. 1998. The Paracyclops fimbriatus-complex
(Copepoda, Cyclopoida): A revision. Zoosystema 20:563–602.
Kikuchi, T. 1992. Effects of the insect growth regulator methoprene and
organophosphorous insecticides against non-target aquatic organisms in urban
drains. Japanese Journal of Sanitary Zoology 43:65–70.
Knepper, R.G., A.D. Le Clair, J.D. Strickler, and E.D. Walker. 1992. Evaluation of
methoprene (Altosid XR) sustained-release briquets for control of Culex mosquitoes
in urban catch basins. Journal of the American Mosquito Control Association
8:228–230.
250 Northeastern Naturalist Vol. 14, No. 2
Kronenwetter-Koepel, T., J.K. Meece, C. Miller, and K.D. Reed. 2005. Surveillance
of above- and below-ground mosquito breeding habitats in a rural midwestern
community: baseline data for larvicidal control measures against West Nile Virus
vectors. Clinical Medicine and Research 31:3–12.
Marten, G., E.S. Bordes, and M. Nguyen. 1994. Use of cyclopoid copepods for
mosquito control. Hydrobiologia 292/293:491–496.
Marten, G.G., G.M. Nguyen, B.J. Mason, and G. Ngo. 2000. Natural control of Culex
quinquefasciatus larvae in residential ditches by the copepod Macrocyclops
albidus. Journal of Vector Ecology 25:7–15.
McCarry, M.J. 1996. Efficacy and persistence of Altosid pellets against Culex
species in catch basins in Michigan. Journal of the American Mosquito Control
Association 12:144–146.
Micieli, M.V., J.J. Garcia, and J.J. Becnel. 2000. Life cycle and description of
Amblyospora camposi n.sp. (Microsporidia:Amblyosporidae) in the mosquito
Culex renatoi (Diptera, Culicidae) and the copepod Paracyclops fimbriatus
fimbriatus (Copepoda, Cyclopidae). Journal of Eukaryotic Microbiology
47(6):575–580.
Munstermann, L.E., and G.B. Craig. 1977. Culex populations in the catch basins of
northern St. Joseph County, Indiana. Indiana Academy of Science Proceedings
86:246–252.
Pallant, J. 2001. SPSS Survival Manual. Open University Press, Berkshire, UK. 304 pp.
Pennak, R. 1978. Fresh-water Invertebrates of the United States. John Wiley and
Sons, New York, NY. 648 pp.
Rey, J.R., S. O’Connell, S. Suarez, Z. Menendez, L.P. Lounibos, and G. Byer. 2004.
Laboratory and field studies of Macrocyclops albidus (Crustacea:Copepoda) for
biological control of mosquitoes in artificial containers in a subtropical environment.
Journal of Vector Ecology 29:124–134.
Schoeppner, R.F. 1978. The effectiveness of Altosid briquets in controlling Culex
pipiens in catch basins. California Mosquito and Vector Control Association
46:115–117.
Schwartz, S.S. 1992. Benthic predators and zooplankton prey: Predation by
Crangonyx shoemakeri (Crustacea; Amphipoda) on Daphnia obtusa (Crustacea;
Cladocera). Hydrobiologia. 237:25–30.
Siegal, J.P., and R.J. Novak. 1997. Field trials of Vectolex CG, a Bacillus sphaericus
larvicide, in Illinois waste tires and catch basins. Journal of the American
Mosquito Control Association 13:305–310.
Southwood, T.R.E. 1966. Ecological Methods with Particular Reference to the Study
of Insect Populations. Chapman and Hall, London, UK. 391 pp.
Spencer, M., L. Blaustein, S.S. Schwartz, and J.E. Cohen. 1999. Species richness and
the proportion of predatory animal species in temporary freshwater pools: Relationships
with habitat size and permanence. Ecology Letters 2:157–166.
Tabachnick, B.G., and L.S. Fidell. 1989. Using Multivariate Statistics, 2nd Edition.
HarperCollins, New York, NY. 746 pp.
Timmermann, S.E., and H. Briegel. 1993. Water depth and larval density affect
development and accumulation of reserves in laboratory populations of mosquitoes.
Bulletin of the Society for Vector Ecology 18:174–187.
Wiggins, G.B., R.J. MacKay, and I.M. Smith. 1980. Evolutionary and ecological
strategies of animals in annual temporary pools. Archives fur Hydrobiologie/
Supplement 58.1/2:97–206.