The response of mosquitoes to bushfire is poorly understood. During the 2019–20 summer, many regions of Australia were impacted by devastating bushfires. An area of estuarine and brackish-water wetlands alongside the Georges River, Sydney, New South Wales, was burned in January 2020. Mosquito populations within the area were monitored as part of the local authority's mosquito management program, providing a unique opportunity to record the response of key mosquitoes of pest and public health concern to bushfire. Ground pools within a tidally influenced swamp oak forest dominated by Casuarina glauca and associated wetlands dominated by Phragmites australis and Bolboschoenus spp. had been identified as suitable habitat for a range of mosquitoes, including Aedes alternans, Ae. vigilax, and Verrallina funerea. Surveys of immature stages of mosquitoes within recently burned habitats inundated by tides demonstrated that mosquito eggs survived the direct and indirect impacts of fire and immature stages successfully completed development as reflected in concomitant changes in adult mosquito populations following the bushfire. This unique observation has implications for mosquito management following bushfire in Australia and internationally.

Mosquitoes associated with coastal regions of Australia are of significant pest and public health importance (Webb et al. 2016). Notwithstanding the nuisance-biting impacts abundant mosquito populations have on coastal residential communities and recreational activities, the health risks associated with mosquitoes are of increasing concern for local authorities. The most commonly reported mosquito-borne disease in Australia is caused by Ross River virus and while dozens of mosquito species have been implicated in its transmission, one of the key vectors in coastal regions is considered Aedes vigilax (Skuse) (Ryan et al. 2000, Claflin and Webb 2015). This is an often very abundant mosquito that is a known nuisance-biting pest and disperses widely from larval habitats (Webb et al. 2016, Claflin and Webb 2017, Webb and Russell 2019). This mosquito is typically associated with tidally influenced habitats, most commonly estuarine wetlands including salt-marsh and mangrove communities (Gíslason and Russell 1997), and is a key target of mosquito control programs in coastal regions (Webb and Russell 1999, Russell and Kay 2008).

The Georges River, which runs along the southern edge of the Sydney metropolitan region, contains a combination of estuarine, brackish-water, and freshwater habitats and while a suite of mosquito species has been identified from the region that are of pest and public health concern, the dominant species is Ae. vigilax. The productive mosquito habitats are tidally influenced salt marshes (dominated by Sarcocornia quinqueflora (Bunge ex Ungen-Sternberg) A.J. Scott), mangrove forests (dominated by Avicennia marina L.), and swamp oak forests (dominated by Casuarina glauca Sieber ex Sprengel and Melaleuca spp.). There are also brackish-water and freshwater wetlands dominated by various macrophytes (e.g., Juncus spp., Bolboschoenus spp., Phragmites australis (Cavanilles) Trinius ex Steudel) in addition to ephemeral ground pools in terrestrial woodlands and water-holding containers and other water-holding infrastructure associated with residential areas. In addition to Ae. vigilax, there is a suite of other key mosquitoes of pest and public health concern in the local area (Crocker et al. 2017) and local authorities undertake routine surveillance of mosquito populations and regularly issue warnings regarding mosquito-borne disease risk (Glasgow et al. 2018).

The 2019–20 summer in Australia was marked by hot and dry conditions that led to catastrophic bushfires (often referred to as wildfire in other countries) in many parts of southeastern Australia, with >5 million ha burned and significant impacts on the local environment and wildlife (Dickman and McDonald 2020). On January 5, 2020, bushfire swept through a section of wetlands, swamp oak forest, and Eucalypt woodland alongside the Georges River in the suburbs of Hammondville and Voyager Point, New South Wales. An estimated 60 ha was burned, including wetland areas where mosquito monitoring sites had already been established in association with a mosquito control program undertaken by a local government authority (Liverpool City Council 2019).

Mosquito monitoring commenced in early December 2019 with approximately weekly adult mosquito trapping undertaken using carbon dioxide–baited encephalitis virus surveillance traps (Rohe and Fall 1979) at 2 sites in the local area (Hammondville 33°57′12″S, 150°58′05″E; Voyager Point 33°57′26″S, 150°58′07″E). In addition, immature mosquito sampling was undertaken following tidal and/or rainfall inundation of habitats using a 300-ml dipper at 20 predetermined sampling sites to determine the need for, and assess effectiveness of, larvicide applications (Webb and Russell 2001). Following the bushfire, an additional 10 sampling sites were established. All mosquito specimens were identified according to the taxonomic keys of Russell (1993) and pictorial guides of Webb et al. (2016). No mosquito control had been undertaken during these early stages of summer due to above average temperatures, below average rainfall, and a lack of extensive tidal inundation of local habitats. Due to the prevailing environmental factors and a number of operational matters associated with bushfires, no mosquito control was undertaken at the site until March 2020, outside the period discussed in this note.

Prior to the bushfire, habitat and mosquito surveys were undertaken on December 19, 2019, and December 28, 2019, following partial tidal inundation of the area. Immature stages of Ae. vigilax (mean larval density 2.13 ± 0.71 per dip; n = 20) and Ae. alternans (Westwood) (mean larval density 0.06 ± 0.05 per dip; n = 20) were collected. This difference in relative abundance was not surprising given that the larvae of Ae. alternans are predatory (Webb et al. 2016) and typically collected at much lower densities than Ae. vigilax (Webb and Russell 2001). No larval abundance data were recorded on December 28, 2019, given the extremely limited areas of habitats inundated.

Postbushfire, these habitats were again inundated by tides, and much more substantially than during the previous month, between January 10–14, 2020 (approximately 2 wk following the bushfires). Surveys of immature mosquitoes within ground pools remaining following retreat of tides were undertaken on January 14, 2020, and included a range of ground pools that were variously categorized as either burned or partly burned (Fig. 1). There were some small areas of habitat that remained only marginally impacted by fire, but these were not sufficient to allow for a formal comparison of larval densities. Habitats categorized as burned were those where all vegetation had been burned, whereas those categorized as marginally burned that had the unburned vegetation remained within the ground pools, but the surrounding vegetation and/or canopy cover had burned.

Fig. 1.

Example of ground pools containing mosquito larvae within swamp oak forest following bushfire impact at Georges River, New South Wales, Australia.

Fig. 1.

Example of ground pools containing mosquito larvae within swamp oak forest following bushfire impact at Georges River, New South Wales, Australia.

Close modal

Immature stages, mostly 3rd- to 4th-stage larvae, of a range of mosquito species were collected from ground pools within fully burned (n = 16) with Ae. vigilax (mean larval density 4.48 ± 1.49 per dip) and Ae. alternans (mean larval density 0.06 ± 0.06 per dip) and in partly burned habitats (n = 14) with Ae. vigilax (mean larval density 6.56 ± 1.66 per dip) and Ae. alternans (mean larval density 0.14 ± 0.12 per dip). Specimens of immature Verrallina funerea (Theobald) were also collected but at very low abundances (mean larval density <0.05 per dip; n = 20). This mosquito is rarely collected locally given this species is close to its known southern geographic limit (Webb et al. 2016).

Given the short time between the bushfire, tidal inundation of habitats, and hatching of immature stages, it seems highly likely that the eggs of these mosquitoes survived the fire. While it is not possible to rule out completely that eggs were laid shortly after the fire and then hatched on inundation, given adult mosquito monitoring suggested adult mosquito populations had sharply declined immediately after the fire (Table 1), it is concluded that there was unlikely to be substantial oviposition throughout the area during the short period between bushfire and tidal inundation of wetlands. There is also generally a period of at least 48–52 h of embryonic development after oviposition before hatching (Sinclair 1976) that would likely further reduce the likelihood that the larvae observed in these areas of burned wetland had hatched from eggs laid following the fire. While there is some evidence that larvae of Ae. vigilax may be redistributed by movement of tidal water (Dale et al. 1986), in this situation, it was unlikely that larvae were redistributed from unburned areas outside these study areas as there were no substantial areas of unburned habitat, apart from a few small unburned areas described above, in close enough proximity to the study site. The nature of the microtopography of the site also meant that tidal water tended to flow into and out of the site through a series of smaller tributaries connected to main creekline, as opposed to traveling across the site as a whole, and unlikely to move larvae long distances between burned and any unburned habitats. There is also little evidence that the unhatched eggs of Ae. vigilax are redistributed by tidal flows (Ritchie 1994).

Table 1.

Total adult mosquito abundance across 2 mosquito trap locations at Hammondville and Voyager Point adjacent to swamp oak forest, Georges River, New South Wales, Australia.

Total adult mosquito abundance across 2 mosquito trap locations at Hammondville and Voyager Point adjacent to swamp oak forest, Georges River, New South Wales, Australia.
Total adult mosquito abundance across 2 mosquito trap locations at Hammondville and Voyager Point adjacent to swamp oak forest, Georges River, New South Wales, Australia.

A total of 3,175 mosquitoes were collected over the 7-wk sampling period, 2 wk before bushfire and 5 wk postbushfire, representing 17 species (Table 1). Adult mosquito collections made within 48 h of the bushfire (January 6, 2020) demonstrated a sharp reduction in abundance. However, it should be noted that with the lack of substantial rainfall and tidal inundation of habitats in late December 2019, it would be expected that a decline in adult mosquito abundance is likely to have been observed irrespective of the bushfire. Therefore, it is difficult to determine the relative impact of the bushfire on the immediate decline in abundance of adult mosquitoes. There was a substantial increase in adult mosquito abundance, especially Ae. vigilax, from January 19, 2020, approximately 10 days following tidal inundation of local habitats. This was consistent with the pattern of emergence of Ae. vigilax following inundation of wetlands by tides or rainfall that trigger an egg hatch (Webb and Russell 2001, Claflin and Webb 2017). The continued abundance of Ae. vigilax over subsequent weeks strongly suggests that survivorship and emergence from immature mosquitoes recorded within burned habitats was relatively high and that conditions within the burned habitats did not adversely impact larval survivorship. It was beyond the scope of existing monitoring program to sample more frequently following the bushfire to confirm survival rates of immature stages.

There is a paucity of information on the impact of bushfire in coastal wetlands on fauna and flora in Australia (Ross et al. 2019). The observations reported here demonstrate that the eggs of some mosquitoes are tolerant to the impacts of bushfire; this tolerance may be due to eggs being laid in microhabitats that provide a buffer against the thermal stress associated with bushfire. Eggs of Ae. vigilax are known to be desiccation resistant and there is evidence that they can persist for extended periods in the environment (Lee et al. 1984). There is little information available on their susceptibility to extreme heat or indirect consequences of thermal stress. While it would not be expected that eggs would survive direct exposure to fire, it is likely that eggs, known to typically be deposited below an accumulation of vegetation and debris, at the base of dense vegetation (e.g., S. quinqueflora) or in cracks within the sediments (Kay and Jorgensen 1986, Gíslason and Russell 1997), were somewhat insulated to lethal temperatures. Similarly, other mosquito species, especially Aedes spp., may survive bushfire-prone habitats where eggs are deposited in damp soil and base of vegetation associated with depressions or rock pools within woodland or forest habitats (Webb et al. 2016). It is important to note that without information on the specific temperatures mosquito eggs were exposed to, it is difficult to draw conclusions on specific thermal tolerances.

The observations reported here are consistent with the limited published accounts of fire and impact on mosquitoes elsewhere, due either to the wildfire or intentional use of fire as a mosquito control strategy. Wildfire has been observed having an impact on the pest mosquito Culex salinarius Coq. in Texas, where mosquito populations rapidly declined and took months to recover following fire (Janousek and Olson 1994). Controlled burning of habitats where the eggs of Ae. melanimon Dyar, Ae. nigromaculis (Ludlow), and Ae. taeniorhynchus (Wied.) were present had been investigated as possible mosquito control strategies, but while it was reported that there was a decline in larval hatching from eggs exposed to fire, the decrease was not significant and nor was there evidence that this approach would be a practical mosquito management strategy (Wilder and Takahashi 1980, Wallace et al. 1990).

This is the first account of mosquitoes responding to bushfire in Australian wetlands. While it would be expected that bushfire or prescribed burns may adversely impact mosquito populations in the short term, recovery is likely to be rapid during periods of favorable environmental conditions (e.g., rainfall, tidal inundation). There is some uncertainty regarding potential long-term adverse impacts on mosquito populations associated with freshwater-dominated wetlands reliant on rainfall to stimulate and sustain abundant populations. The ecological drivers of mosquito population recovery are likely to be dependent on local conditions. The implications from the observations presented here are that it should be expected that mosquitoes, especially Ae. vigilax, associated with bushfire-impacted coastal wetlands will recover quickly and that risks associated with nuisance biting, and potentially mosquito-borne disease, must be managed by local authorities, even during the weeks immediately following bushfire (Webb 2020). As a consequence, adequate resourcing should be made available where mosquito control activities may already be in place or other responses to reduce the impact of mosquitoes are undertaken.

The observations reported here were undertaken as part of the mosquito control and monitoring program coordinated by Liverpool City Council and are acknowledged for their assistance with this study. John Clancy and John Haniotis of New South Wales Health Pathology provided assistance with mosquito identification. Al-Aabid Chowdhury of the University of Sydney provided assistance with field work and mosquito identification. We acknowledge the traditional owners of the land on which this work was conducted, the peoples of the Dhurawal and Darug Nations.

Claflin
SB,
Webb
CE.
2015
.
Ross River virus: many vectors and unusual hosts make for an unpredictable pathogen
.
PLoS Pathog
11
:
e1005070
.
Claflin
SB,
Webb
CE.
2017
.
Surrounding land use significantly influences adult mosquito abundance and species richness in urban mangroves
.
Wetlands Ecol Manag
25
:
331
344
.
Crocker
W,
Maute
K,
Webb
C,
French
K.
2017
.
Mosquito assemblages associated with urban water bodies; implications for pest and public health threats
.
Landsc Urban Plan
162
:
115
125
.
Dale
PER,
Hulsman
K,
Harrison
D,
Congdon
B.
1986
.
Distribution of the immature stages of Aedes vigilax on a coastal salt-marsh in south-east Queensland
.
Aust J Ecol
11
:
269
278
.
Dickman
C,
McDonald
T.
2020
.
Some personal reflections on the present and future of Australia's fauna in an increasingly fire-prone continent
.
Ecol Manag Restor
21
:
86
96
.
Gíslason
GM,
Russell
RC.
1997
.
Oviposition sites of the saltmarsh mosquito, Aedes vigilax (Skuse) (Diptera: Culicidae), at Homebush Bay, Sydney, NSW—a preliminary investigation
.
Aust J Entomol
36
:
97
100
.
Glasgow
KM,
Bateman-Steel
CR,
Pope
B,
Ferson
MJ.
2018
.
The right kind of viral: boosted Facebook advertisements for health protection messages
.
Med J Aust
209
:
447
.
Janousek
TE,
Olson
JK.
1994
.
Effects of a natural marsh fire on larval populations of Culex salinarius in east Texas
.
J Am Mosq Control Assoc
10
:
233
235
.
Kay
BH,
Jorgensen
WK.
1986
.
Eggs of Aedes vigilax (Skuse) and their distribution on plants and soil in south east Queensland saltmarsh
.
Aust J Entomol
25
:
267
272
.
Lee
DJ,
Hicks
MM,
Griffiths
M,
Russell
RC,
Marks
EN.
1984
.
The Culicidae of the Australian region. Volume 3
.
Canberra, Australia
:
Australian Government Publishing Service
.
Liverpool City Council.
2019
.
Stay safe from mosquitoes
[Internet].
Liverpool, Australia
:
Liverpool City Council
[accessed April 28, 2021]. Available from: https://www.liverpool.nsw.gov.au/environment/mosquitoes.
Ritchie
SA.
1994
.
Spatial stability of Aedes vigilax (Diptera: Culicidae) eggshells in southeastern Queensland salt marshes
.
J Med Entomol
31
:
920
922
.
Rohe
DL,
Fall
RP.
1979
.
A miniature battery powered CO2 baited light trap for mosquito-borne encephalitis surveillance
.
Bull Soc Vector Ecol
4
:
24
27
.
Ross
PM,
Harvey
K,
Vecchio
EM,
Beckers
D.
2019
.
Impact of fire and the recovery of molluscs in south-east Australian salt marsh
.
Ecol Manag Restor
20
:
126
135
.
Russell
RC.
1993
.
Mosquitoes and mosquito-borne disease in southeastern Australia
.
Westmead, Australia
:
Westmead Hospital
.
Russell
TL,
Kay
BH.
2008
.
Biologically based insecticides for the control of immature Australian mosquitoes: a review
.
Aust J Entomol
47
:
232
242
.
Ryan
PA,
Do
KA,
Kay
BH.
2000
.
Definition of Ross River virus vectors at Maroochy shire, Australia
.
J Med Entomol
37
:
146
152
.
Sinclair
P.
1976
.
Notes on the biology of the salt marsh mosquito, Aedes vigilax (Skuse) in southeast Queensland
.
Queensl Nat
21
:
134
139
.
Wallace
FL,
Tidwell
MA,
Williams
DC,
Jackson
KA.
1990
.
Effects of controlled burning on Aedes taeniorhynchus eggs in an abandoned rice impoundment in South Carolina
.
J Am Mosq Control Assoc
6
:
528
529
.
Webb
CE.
2020
.
Reflections on a highly unusual summer: bushfires, COVID-19 and mosquito-borne disease in NSW, Australia
.
Public Health Res Pract
30
:
3042027
.
Webb
CE,
Doggett
SL,
Russell
RC.
2016
.
A guide to mosquitoes of Australia
.
Clayton South, Australia
:
CSIRO Publishing
.
Webb
CE,
Russell
RC.
1999
.
Towards management of mosquitoes at Homebush Bay, Sydney, Australia. 1. Seasonal activity and relative abundance of adults of Aedes vigilax, Culex sitiens, and other salt-marsh species, 1993–94 through 1997–98
.
J Am Mosq Control Assoc
15
:
242
249
.
Webb
CE,
Russell
RC.
2001
.
Do we spray? Did it work? Indices for control of Aedes vigilax larvae in Homebush Bay
.
Arbovirus Res Aust
8
:
387
390
.
Webb
CE,
Russell
RC.
2019
.
Dispersal of the mosquito Aedes vigilax (Diptera: Culicidae) from urban estuarine wetlands in Sydney, Australia
.
J Med Entomol
56
:
1290
1295
.
Wilder
WH,
Takahashi
RM.
1980
.
Effect of pasture burning on survival of Aedes mosquito eggs
.
Mosq News
40
:
12
15
.

Author notes

1

Marie Bashir Institute for Infectious Diseases and Biosecurity, The University of Sydney, Sydney, NSW 2006, Australia.

2

Medical Entomology, NSW Health Pathology, Westmead Hospital, Westmead, NSW 2145, Australia.

3

Liverpool City Council, Locked Bag 7064, Liverpool BC, NSW 1871, Australia.