Roof gutters on houses that have become inundated with leaf litter and cannot drain properly are an often-overlooked man-made container habitat that is suitable for mosquito larval development. In order to reduce the amount of leaf litter debris in gutters, many homeowners install debris screens, commonly referred to as “gutter guards,” on their roof gutters, but no study has examined the effect of gutter guards on mosquito production. The objective of this research was to determine the extent to which different types of gutter guards affect mosquito colonization and abundance of juvenile mosquitoes in gutter habitats. Three experimental gutters, each with 1 of 3 treatments (control with no gutter guard, a metal lock-in mesh screen gutter guard, or a foam filter gutter insert), were placed at 5 field locations to monitor mosquito colonization and production over 8 wk. Pupae were collected daily, and eclosed adults were identified to species. Mosquitoes colonized and larvae developed in all gutters regardless of the presence of a guard, although those with the foam filter guards were least likely to be colonized (P < 0.001). Once colonized, the control gutters without a gutter guard had the lowest mosquito abundance (P < 0.001), and the metal lock-in gutters had the highest abundance (P < 0.001). The results suggest that if standing water exists in a gutter, gutter guards are not an effective tool for mosquito control.

Surveillance and control of container-breeding mosquitoes within residential neighborhoods are challenging due to the high density and diversity of man-made containers found in residents' yards. One of the most cryptic container habitats occurring in residential neighborhoods is the roof gutters. Due to the difficulty in reaching roof gutters on residential homes, these potential container habitats often are not inspected for mosquito juveniles during surveillance activities (Williams et al. 2007). Roof gutters that have become inundated with debris often do not drain properly, creating very productive larval habitats that can remain waterlogged long after an initial rain event (Tinker 1974, Montgomery and Ritchie 2002, Gustave et al. 2012). Although roof gutters may only make up a small proportion of man-made container habitats, once they are colonized, they can produce large numbers of adults (Montgomery and Ritchie 2002, Pilger et al. 2011, Gustave et al. 2012). For instance, in a survey of residential yards in Australia, Montgomery and Ritchie (2002) found that less than 15% of gutters were positive for pupae, but these accounted for almost 50% of the total mosquito pupae collected. Even in areas where mosquito control is implemented, roof gutters typically do not receive abatement, making them a refuge in an otherwise hostile environment (Tinker 1974, Williams et al. 2007). Inclusion of roof gutters in management practices could enhance the efficacy of controlling container-breeding mosquitoes within residential neighborhoods.

To reduce the amount of debris entering the gutter, many residents install debris screens (hereafter “gutter guards”; Abbott et al. 2007). While there are many different types of gutter guards commercially available, 1 commonly used type is a metal lock-in gutter guard that attaches to the roof and outer edge of the gutter, creating a mesh screen that allows rainwater to enter the gutter while excluding large debris (Fig. 1A). Another commonly used gutter guard is a foam filter insert, which is inserted directly into the gutter, covering the entire opening with a porous foam while leaving an area underneath the foam for water to drain. The purpose of this study was to determine whether either of these commonly used gutter guards affects colonization and subsequent larval development of mosquitoes. Considering the ubiquity of gutter habitats in residential landscapes, the potential for gutter guards to alter the importance of these habitats is considerable, yet it has not been evaluated experimentally.

Fig. 1.

Number of mosquitoes collected from experimental gutters (A) without a gutter guard, with a foam filter insert gutter guard, and with a metal lock-in mesh gutter guard for (B) all mosquito species and (C) Ae. albopictus only.

Fig. 1.

Number of mosquitoes collected from experimental gutters (A) without a gutter guard, with a foam filter insert gutter guard, and with a metal lock-in mesh gutter guard for (B) all mosquito species and (C) Ae. albopictus only.

This study was conducted from June 6 to August 3, 2018, in Urbana, IL. An artificial gutter system was used to determine if female mosquitoes were able to oviposit and if larvae were able to develop to adulthood within gutters covered with 2 different gutter guard strategies relative to an uncovered control. Each artificial gutter was 2.5 m long, stood 1.25 m from the ground, and was capped on both ends to hold water. Three treatments were used: a control with no gutter guard, a foam filter gutter guard insert (TMJ Innovations LLC, Milwaukee, WI), and a metal lock-in mesh gutter guard (Amerimax, Norcross, GA). Three gutters (1 of each treatment) were spaced 1 m apart at each study location with 5 replicate blocks for a total of 15 gutters. Two blocks were located within Southern Arboretum Woodlands in Urbana, IL (40.05063°N, 88.12592°W), and 3 blocks were located within Trelease Woods in Urbana, IL (40.130437°N, 88.143031°W). Both sites are wooded within a mixed-use residential and agricultural landscape. Each block was placed a minimum of 100 m apart. Each gutter was filled initially with 2.0 liters of grass infusion. Additional infusion was added to the gutters when the water level was below a 1.5-liter threshold. The gutters were checked and pupae were removed daily. Pupae were placed in individual vials and were enumerated and identified to species following eclosion.

Across all treatments, 2,302 adults belonging to 5 species from 3 genera emerged from the experimental gutters. Of the collected mosquitoes, Aedes albopictus (Skuse) was the predominate species collected (64.4%), followed by Ae. triseriatus (Say) (33.2%). Three additional species were collected in low numbers: Ae. japonicus (Theobald) (2.1%), Culex pipiens Linnaeus (<0.01%), and Anopheles quadrimaculatus Say (<0.001%).

Our goal was to determine if significant differences in mosquito colonization and abundance occurred across the 3 treatments. Here, “colonization” refers to presence or absence of mosquitoes from a gutter treatment, while “abundance” refers to overall numbers of mosquitoes. Due to a large number of zeros in the data set, we conducted a zero-inflated Poisson (ZIP) model (Zuur et al. 2009, Leisnham et al. 2014) for overall mosquito colonization and abundance and another for Ae. albopictus colonization and abundance to test whether gutter guard treatment, collection site, and week were significant effects in the models (α = 0.05). We also conducted a Poisson regression model for overall mosquito and Ae. albopictus count data, and then we ran a Vuong test to determine if the ZIP model offered a significant improvement over the standard Poisson regression model. Both for overall mosquito and for Ae. albopictus data, the ZIP model was a significant improvement at α = 0.05 (P < 0.001) over the Poisson regression model. All analyses were carried out using the package “pscl” in RStudio (RStudio, Inc., Boston, MA, 2016, Zeileis et al. 2008, Jackman et al. 2017).

Results from the ZIP model showed that foam-covered gutters were significantly less likely to be colonized by mosquitoes compared to the metal lock-in covered and control gutters (P < 0.001; Table 1 and Fig. 1B). Once colonized, metal lock-in covered gutters had the greatest abundance of mosquitoes (57.8%; Table 1 and Fig. 1B), followed by foam-covered gutters (30.1%, P < 0.001; Table 1 and Fig. 1B); control gutters with no guards had the lowest numbers of collected mosquitoes (12.6%, P < 0.001; Table 1 and Fig. 1B). For Ae. albopictus, all treatment types were equally likely to be colonized (Table 1), but if colonized, foam filter gutters had the highest abundance of Ae. albopictus (P < 0.001; Table 1 and Fig. 1C), followed by metal lock-in gutters (P < 0.001; Table 1 and Fig. 1C); control gutters without a gutter guard had the lowest abundances (Table 1 and Fig. 1C). There were significant effects for both site and week for all models.

Table 1.

Zero-inflated Poisson regression model constructed for overall mosquito and Aedes albopictus presence/absence and abundance counts comparing gutters without a gutter guard, with a foam filter insert gutter guard, and with a metal lock-in mesh gutter guard.

Zero-inflated Poisson regression model constructed for overall mosquito and Aedes albopictus presence/absence and abundance counts comparing gutters without a gutter guard, with a foam filter insert gutter guard, and with a metal lock-in mesh gutter guard.
Zero-inflated Poisson regression model constructed for overall mosquito and Aedes albopictus presence/absence and abundance counts comparing gutters without a gutter guard, with a foam filter insert gutter guard, and with a metal lock-in mesh gutter guard.

This is the first experimental evaluation of the effect of gutter guards on the colonization by and abundance of mosquitoes in roof gutter habitats. The results of this study show that metal lock-in mesh gutter guards and foam filter gutter inserts do not prevent oviposition and subsequent larval development of mosquitoes in a roof gutter container habitat. In fact, metal lock-in gutter guards increase mosquito production compared to gutters without any gutter guard. While difficult to initially colonize, gutters with a foam filter insert provide a highly suitable habitat for larval development, especially for Ae. albopictus. Under conditions typical of residential environments, gutters with foam filter inserts may be more readily colonized by mosquitoes due to a greater number of access points than the experimental apparatus provided. Although not tested in this study, gutter guards may prevent gutters from becoming clogged by prohibiting debris from entering the gutter. This could potentially reduce the availability of mosquito habitats, though once clogged, the presence of gutter guards may make the gutter a more suitable habitat for larval development by reducing evaporation and keeping the water shaded, allowing for a more stable water temperature compared to the open gutter. Based on our results, we did not find evidence that gutter guards should be considered an appropriate strategy for preventing mosquitoes in roof gutters and may inhibit the application of effective control agents from entering the gutter. We recommend regular inspection and cleaning of roof gutters by homeowners, even if a gutter guard system is in place. We also recommend regular monitoring and treatment of roof gutters in residential neighborhoods in order to control mosquito production from these important, but often over-looked, container habitats.

The authors thank S. Halsey, M. Perez, L. Edwards-Blinderman, M. Velasco, W. Marshall, L. Burns, C. Dust, C. Dortch, and G. Dagher for their technical assistance. Support for this research was provided by the Department of Entomology Student Stipend Award, the Institute for Sustainability, Energy, and Environment, and the School of Integrative Biology at the University of Illinois Urbana-Champaign.

Abbott
SE,
Douwes
J,
Caughley
BP.
2007
.
A survey of the microbiological quality of roof-collected rainwater of private dwellings in New Zealand. In: Proceedings of the 13th International Conference on Rainwater Catchment Systems. 2007 August 21–23; Sydney, Australia
.
Rainwater Urban Des
9
:
1
24
.
Gustave
J,
Fouque
F,
Cassadou
S,
Leon
L,
Anicet
G,
Ramdini
C,
Sonor
F.
2012
.
Increasing role of roof gutters as Aedes aegypti (Diptera: Culicidae) breeding sites in Guadeloupe (French West Indies) and consequences on dengue transmission and vector control
.
J Trop Med
2012
:
249524
.
Jackman
S,
Tahk
A,
Zeileis
A,
Mainone
C,
Fearon
J,
Meers
Z.
2017
.
Package
pscl. ”
Stanford, CA
:
Political Science Computer Laboratory Stanford Univ
.
Leisnham
PT,
LaDeau
SL,
Juliano
SA.
2014
.
Spatial and temporal habitat segregation of mosquitoes in Urban Florida
.
PLoS One
9
:
e91655
.
Montgomery
BL,
Ritchie
SA.
2002
.
Roof gutters: a key container for Aedes aegypti and Ochlerotatus notoscriptus (Diptera: Culicidae) in Australia
.
Am J Trop Med Hyg
67
:
244
246
.
Pilger
D,
Lenhart
A,
Manrique-Saide
P,
Siqueira
JB,
da Rocha
WT,
Kroeger
A.
2011
.
Is routine dengue vector surveillance in central Brazil able to accurately monitor the Aedes aegypti population? Results from a pupal productivity survey
.
Trop Med Int Heal
16
:
1143
1150
.
Tinker
M.
1974
.
Aedes aegypti larval habitats in Surinam
.
Pan Am Heal Organ Bull
8
:
293
301
.
Williams
CR,
Long
SA,
Webb
CE,
Bitzhenner
M,
Geier
M,
Russell
RC,
Ritchie
S.
2007
.
Aedes aegypti population sampling using BG-Sentinel traps in north Queensland Australia: statistical considerations for trap deployment and sampling strategy
.
J Med Entomol
44
:
345
350
.
Zeileis
A,
Kleiber
C,
Jackman
S.
2008
.
Regression models for count data in R
.
J Stat Softw
27
:
1
25
.
Zuur
AF,
Ieno
EN,
Walker
NJ,
Saveliev
AA,
Smith
G.
2009
.
Mixed Effects Models and Extensions in Ecology with R
.
New York, NY
:
Springer
.