SEARCH

SEARCH BY CITATION

Keywords:

  • Coastal wetland;
  • management;
  • climate;
  • disease;
  • land use;
  • Aedes vigilax

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Will mangrove encroachment into saltmarshes affect saltwater mosquito habitats? To address this, we synthesized information from two perspectives: 1) at a detailed level, the immature mosquito habitat within mangroves; 2) at a more general or regional level, changes due to mangrove expansion into saltmarshes. This is a synthesis of two research projects. One showed that mosquito larval habitats in mangroves are complex, related to the detailed interactions between topography and tidal patterns and that not all parts of a mangrove forest are suitable habitat. The other, based on remote sensing and analysis of rainfall data, showed that mangrove encroachment in eastern Australia is related to both climate and human land use over several decades (1972–2004). An important question emerged: when mangroves encroach into saltmarshes will they displace saltmarsh immature mosquito habitats or will they replace them with mangrove ones? There is no simple answer: it will vary with climate change and sea level scenario and how these affect the system. We conclude that mosquito management, which is locally implemented, needs to be integrated with land use planning systems, which often operate at a more general level.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Mosquitoes are significant vectors of human diseases. Although some diseases can be fatal, such as yellow fever, dengue, and malaria, there are others that are mainly debilitating but may be fatal, such as West Nile virus, recognized in the U.S.A. since 1999. Ross River and Barmah Forest viruses in Australia are generally debilitating and not fatal (Dale and Knight 2008).

Mosquitoes are found wherever the environment provides the necessary conditions for their survival. For the larval stages this means water and that often includes wetlands. Intertidal wetlands provide suitable habitat for some mosquitoes and, for human populations living near the coast, this can lead to a nuisance and health risk.

Mosquitoes that spend their immature stages in saline intertidal wetlands (saltmarshes and mangroves) include several species. In the U.S.A., these include Aedes taeniorhynchus (Wiedemann) and Aedes sollicitans (Walker), which are nuisance mosquitoes and may transmit West Nile virus (Center for Disease Control and Prevention (CDC) 2009). In Australia, mangrove and saltmarsh wetlands provide habitats for Aedes vigilax (Skuse) and Aedes camptorhynchus (Thomson), which are significant vectors of Ross River and Barmah Forest virus diseases in the human population (Jacups et al. 2008). Between 1993 and 2012, there were averages of 4,413 cases of Ross River and 1,246 of Barmah Forest virus diseases each year in Australia, with over 50% of each in the state of Queensland (data from Department of Health and Ageing 2013). Mosquito management is therefore an important issue.

Immature mosquito (egg and larval) habitats in saltmarshes are relatively well documented. For example, Dale et al. (1986, 2008) showed relationships between eggs, larvae, and position in the saltmarsh landscape together with vegetation. These are related to elevation and tidal flooding patterns. Until recently, it has been difficult to identify and characterize immature mosquito habitats in mangroves, as ground survey is impractical, aerial survey is impeded by canopy coverage, and detailed elevation data has been lacking. Advances in the availability of high resolution LiDAR data, together with local tidal modelling, has significantly remedied this. Recent research has identified the complex tidal, elevation, and habitat structures that are associated with Ae. vigilax larval habitats in mangrove systems (Knight 2011, Knight et al. 2009, 2012).

It is of concern that mangrove systems are generally encroaching into saltmarshes and this may affect immature mosquito habitats. The shift from saltmarsh to mangrove is occurring both overseas (McKee 2004) and in eastern Australia (Eslami-Andargoli et al. 2009, McTainsh et al. 1986, Saintilan and Williams 1999, Saintilan amd Wilton 2001). Sea level changes are likely to affect intertidal ecosystems and scenarios for mangrove systems are uncertain, variable, and need monitoring (Soares 2009). What has not been explored is the likely consequence of the advance of the mangroves for vector mosquito issues.

The interaction between mosquitoes and people, and hence the risk of arbovirus disease transmission, is related to their proximity to each other. For example, residential locations close to mosquito larval habitats were shown to increase the risk of Ross River virus disease in Brisbane (Muhar et al. 2000). Thus, it is important to understand not only larval habitats within mangroves but also how those habitats may change as climate, sea level, and human land use change.

As an example of a generally applicable approach, this paper describes the detailed within-mangrove community mosquito larval habitats as characterized by Knight (2011) and outlines recent research on the general expansion of mangrove communities in southeast Queensland between 1972 and 2004. It also considers potential changes until 2031. It asks a key question: will the expansion of mangroves into saltmarshes be accompanied by an expansion of mosquito larval habitats?

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Study area

The study area (27°20′S, 153°10′E) extends over a linear distance of approximately 40 km from north to south and includes sites in northern Moreton Bay, southeast Queensland, Australia. It has a subtropical climate with maximum rainfall in the summer and a mean annual rainfall of 1,028 mm (Australian Bureau of Meteorology). Ten intertidal sites were selected for the mangrove encroachment research (Figure 1). One of these, Glass Mountain Creek, was also used as an example of the mosquito habitat assessment.

image

Figure 1. Location of study sites for mangrove encroachment. The Glass Mountain Creek site was also used for mangrove habitat assessment.

Download figure to PowerPoint

Within the intertidal zone, the vegetation towards the seaward boundary was predominantly grey mangrove (Avicennia marina (Forsk.) Vierh). At higher elevations, saltmarsh and bare ground predominated. The sites shared a similar climate, but the surrounding land cover characteristics varied, with exotic pine plantations mainly in the northern area and settlement and associated infrastructure more common to the south, closer to the state capital city of Brisbane. Southeast Queensland had a population of over three million in 2011 and is experiencing rapid population growth of 1.7 %/year (The State of Queensland (Treasury and Trade) 2013). This growth is planned to be contained within the Southeast Queensland Urban Footprint (Queensland Government 2009) which allows urban development close to the coast and hence, close to mosquito larval habitats.

The mosquitoes

The requirements of the local saltwater mosquito, Aedes vigilax, are well known and were synthesized in Knight (2011) who referred to seminal publications such Sinclair (1976). In summary, oviposition is onto a damp substrate, eggs need to be exposed and conditioned for at least three days, a hatch is stimulated by a decrease in dissolved oxygen when flooded, and water needs to persist for larval development to adult emergence (around five days in sub-tropical summers). The species is autogenous (Hugo et al. 2003) and installment hatching of eggs can occur with eggs remaining viable for up to 20 submersion – exposure cycles, as recorded by Sinclair (1976).

Data and mapping

Mosquito habitat identification within mangrove forests

To identify mosquito larval habitats within mangrove forests, a digital elevation model and detailed tidal analysis were combined to relate topography to tidal flooding (as described in Knight et al. 2009, Griffin et al. 2010, Knight 2011). The tidal analysis related tides recorded at the nearest Standard Port (Brisbane Bar, 27° 22′S, 153° 10′E) to in situ data logger measurements of water depths over tidal events as described in Knight et al. 2009. A digital elevation model (derived from LiDAR data) at a 1 m2 resolution with 0.45 m horizontal and 0.05 m vertical accuracy was used to describe the detailed micro-topography and to model tidal flooding events (also described in Knight et al. 2009). To relate these to mosquito habitats, Ae vigilax mosquito eggshells were extracted from substrate samples collected and then assessed in the laboratory. Field samples of 10 cc substrate were collected from the wetland surface and around pools and depressions as described in Dale et al. 2008. Laboratory analysis involved drying and sieving samples, then preparing them for eggshell counting, using the method of Ritchie and Jennings 1994.

Mangrove encroachment into saltmarsh

Mangrove expansion and encroachment into saltmarsh was analyzed during two periods, a relatively wet (1972–1990) and a relatively dry period (1991–2004), as established by change point analysis (described in Eslami-Andargoli et al. 2009) which identified 1990 as a significant point of change in rainfall patterns. During the wetter period (1972–1990), the mean annual rainfall was 1,321 mm (median 1,330 mm); in the drier period (1991–2004) it was 1,062 mm (median 982 mm). The mean annual rainfall in the area is 1,028 mm (median 1,074) (data from the Australian Bureau of Meteorology).

Mangrove spatial changes, land cover, and population changes were calculated as % change per year for each site, for both periods 1972–1990 and 1990–2004, using aerial photographs for mangrove mapping, Landsat satellite imagery was used for land use, and Australian Bureau of Census data for population (Eslami-Andargoli et al. 2010, 2013).

The perimeter-to-area relationship was calculated as a Mangrove-Saltmarsh Interface (MSI) index for each site. This showed the relationship between mangrove area and its boundary with saltmarsh. In theory, other things being equal, the greater the value of the index, the greater the potential for mangroves to spread into saltmarsh (Eslami-Andargoli et al. 2010).

Analysis

Mosquito larval habitats

At the local level, relevant to mosquitoes, the internal structure of the mangrove system was described using the detailed LiDAR elevation data. Tidal data from in situ loggers were used to model the patterns of flooding and hence, the likelihood of mosquito egg hatch and larval survival. To do this, thresholds of flooding across the mangrove and the frequencies of whole system flooding were established by comparing tides at the fringing edge of the mangroves with elevations within the system (berm and basin). Based on the habitat requirements of immature Aedes vigilax, this was used to identify why some habitats were particularly suitable for mosquito larval development (Knight et al. 2012).

Mangrove encroachment – Identifying relationships

A Partial Least Square Regression (PLSR) method was used to generalize and combine features from principal component analysis and multiple regression (Abdi 2003). This technique is described in more detail in Eslami-Andargoli et al. 2010 and summarized here. It is an alternative approach to the ordinary least square (OLS) regression for handling data that may exhibit high dimensionality and multi-collinearity. It is especially appropriate for ecological analysis, as ecological phenomena are frequently described by many variables that may not be independent. Also, the method can be used where there are many predictors compared to the number of observations (Carrascal et al. 2009).

Predicting mangrove spatial changes

As a hypothetical exercise, we attempted to predict potential change scenarios in mangrove distribution over 27 years until 2031. That is the time frame for the South East Queensland Regional Plan (Queensland Government 2009). There were two scenarios: one assumed a rate of change similar to that in the ‘wetter’ period; the other was based on the ‘drier’ period. For example, at the Glass Mountain Creek site, the average annual rate of change in the wetter period was 2.15% and during the drier period it was 1.02%. These figures were used to calculate the area expected in 2031.

To assess the potential effect of population and urbanization growth in the coastal zone, the Urban Footprint that constrains development (Southeast Queensland Regional Plan) was inspected to see if urban encroachment, particularly towards the coast, would be likely to have an impact on mangrove expansion. This was a qualitative assessment.

Integrating adaptation, encroachment, and internal mangrove structure

We constructed a hypothetical figure to illustrate the potential effect of two human adaptation responses to sea level rise. These were: 1) for settlement to retreat from the coast, allowing increased flooding inland; 2) to construct defenses preventing flooding. This was a qualitative assessment, constructed by modifying the schematic in Figure 2.

image

Figure 2. Schematic of intertidal wetlands showing fringing mangroves, mangrove basin, and salt marsh relevant to the immature stages of Aedes vigilax. Note that Ae vigilax is autogenous; it can produce eggs without a blood meal, thereby ensuring reproduction even if no blood meal is available.

Download figure to PowerPoint

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Mosquito larval habitats

The micro-topography analysis identified three landscape units in the mangrove system. Figure 2 shows a schematic of these units as well as indicating the upslope saltmarsh and upland human settlement. The landscape units were: 1) fringing mangroves at the seaward edge with a relatively consistent slope (i.e., no pool structure), 2) a berm or ridge that limited tidal connections into the landward side of the site to only the 10% – 15% of high tides that exceeded the berm elevation, and 3) a shallow depression or basin on the landward side that contained pools, suitable for larvae, with emergent sediment islands with some pneumatophores, suitable for oviposition. It is in the shallow basin areas with pool structures that the presence of mosquito habitats were confirmed by the eggshell densities. For example, at the Glass Mountain Creek site there were on average 0.74 eggshells/cc in the shallow basin, whereas there were no eggshells in the fringing forest samples (Knight et al. 2009). Addison et al. 1992 showed that an eggshell density of greater than 0.05/cc indicated a mosquito problem habitat. Hydrologic modelling by Knight (2011) of water loss due to infiltration and evapotranspiration from the basin (9 mm/day) indicated that pooled water would generally remain long enough for larvae to develop to adult emergence at that site.

Mangrove expansion

Overall, between 1972 and 2004, mangroves expanded, with an increase of 119 ha at the landward edge, encroaching into and replacing 117 ha of saltmarsh. The rate of encroachment was consistent during the wetter period, when all sites experienced mangrove increase and, in all but one site (Pine Rivers), this was also accompanied by saltmarsh loss. This pattern was less marked during the drier period with two sites showing a loss of mangrove and increase in saltmarsh (Hays Inlet and Little Burpengary Creek). Another site (Pine River) showing an increase in saltmarsh and no change in mangroves (Figure 3). The greatest rate of change was for Glass Mountain Creek. There, during the wetter period, mangroves increased by 2.15% per year reducing to 1.02% per year during the drier period.

image

Figure 3. Change %/yr in mangrove and saltmarsh area before and after 1990.

Download figure to PowerPoint

The increases are displayed visually in Figure 4 for the Glass Mountain Creek site. The three illustrations at the top show the situation for each of the years 1972, 1990, and 2004. These show a change from salt marsh dominance in 1972 to mangrove dominance in 2004. There appeared to be no loss of mangroves at the tidal edge and that was consistent also for all the other sites. The expansion of mangroves is accentuated in a lighter tone (yellow) in the two illustrations at the bottom of Figure 4. This shows that mangrove expansion has been at the expense of saltmarsh and that it was less during the drier period.

image

Figure 4. Example of mangrove landward expansion and saltmarsh loss at Glass Mountain Creek 1972–2004.

Download figure to PowerPoint

Predictors of mangrove expansion

The results of the PLSR analysis showed that in both the wet and dry periods there was a positive R2 relationship (P<0.05) between mangrove expansion and the other variables (Table 1). Median rainfall and the MSI index (opportunity to expand) had the greatest influence in both wetter and drier periods, but the other variables all contributed to influencing mangrove expansion (but negatively in the case of agriculture).

Table 1. Summary of the PLSR analysis for ten sites.
ResponsePredictor variables
Mangrove expansionRainMSI indexPop densityAgriculture (%)Built up (%)Plantation forest (%)R2
  1. Bold indicates significant R2 (P<0.05).

Pre 1990 -wetter0.3060.4920.217−0.4850.0540.2000.913
Post 1990 -drier0.3140.4860.189−0.1280.0150.2290.732

Extrapolating the changes

Table 2 shows that the area of mangroves by 2031 changed from 686 ha in 2004 to 912 ha if wetter (an increase of 33%) and to 736 ha if drier (7%). If the effect of future urbanization, as envisaged in the Urban Footprint of the Southeast Queensland Regional Plan, were to be added to this, then the increase in mangrove area could be greater as there was a positive relationship between mangrove expansion and both population density and built up area during both wetter and drier periods (Table 1). This remains a qualitative observation.

Table 2. Predicted change in mangrove area between 2004 and 2031 based on the established rate of change during relatively wet and dry periods (rounded to nearest integer) (% change from 2004).
SiteMangrove area 2004 (ha)Predicted mangrove area 2031 – wet period (ha) (%)Predicted mangrove area 2031 – dry period (ha) (%)
Cabbage Tree Ck5982 (39)67 (14)
Bald Hills Ck76117 (54)89 (17)
Pine Rivers7280 (11)72 (0)
Hays Inlet160189 (18)142 (−11)
Little Burpengary Ck7290 (25)66 (−8)
Burpengary Ck2031 (55)25 (25)
Southern Caboolture88123 (40)100 (14)
Lagoon Ck4449 (11)52 (18)
Ningi Ck3547 (34)46 (31)
Glass Mt Ck60104 (73)77 (28)
TOTAL686912 (33)736 (7)

Adapting to sea level rise

Figure 5 provides a qualitative visual representation of the current situation and the two extremes of potential human adaptation to sea level rise: retreat of settlement (allowing the system to move inland) and defense (constructing barriers to the sea). Figure 5a shows the schematic with indicative settlement – this is the current situation. Figure 5b shows the retreat adaptation option whereby the intertidal mangrove systems move inland, and so do the saltmarshes. With this scenario there would be a risk of increased intertidal and mosquito habitat, most likely moving closer to human settlement (depending on how retreat would be implemented). With the defense option, the scenario, shown in Figure 5c illustrates building of, for example, a sea wall. The seaward side would be under tidal water where the intertidal (and mosquito) habitat would be lost, replaced by a marine system.

image

Figure 5. Adaptation options and scenario for mosquito habitats: (a) current situation; (b) retreat option; (c) defense option.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Mosquito habitats

Immature mosquito distribution in mangroves is related to micro-habitat characteristics including tidal patterns. This is widely applicable to other mangrove systems as shown by Knight et al. 2012 who found similar relationships in southern Moreton Bay and this was subsequently confirmed by similar findings, including larval and eggshell surveys, in other sites (Knight, unpublished data). This supports the view that not all parts of mangrove systems are suitable for all the immature stages of Ae. vigilax. In general, if around 10% to 15% of tides flood basin areas within mangrove forests, there will be habitats suitable for oviposition, hatch, and larval survival. This is also consistent with Ritchie and Addison 1992. A greater percentage of flooding tides, as with sea level rise, creates a longer lasting impounding effect and would be detrimental for mosquitoes as shown in Knight (2011). It would restrict oviposition and hatch as well as allowing increased access to fish as potential predators.

Mangrove encroachment

At a regional level, mangrove systems are encroaching into saltmarshes. Saltmarshes also contain mosquito larval habitats in the pools and depressions (Dale et al. 1986). Whether the saltmarsh oviposition and larval habitats will become mangrove ones remains speculative. Saltmarsh topography (with local pools and relatively small depressions) may be transformed into problem mosquito habitats as mangroves encroach. Conversely saltmarsh pools may become deeper and act as refuges for resident fish and hence sustain fewer mosquito larvae. The likely situation is uncertain.

Climate and sea level change

Another factor is the uncertain impact of climate change. In the southeast Queensland area of the research, climate change is anticipated to result in less rainfall with more dry days (Burton et al. 2009). As shown in Figure 3 and Table 2 this would be likely to result in a reduced rate of expansion and sometimes in a loss of mangroves. In that scenario, although mosquito larval habitats may not change a great deal, mosquitoes might still pose an increased health risk. This is because predicted increases in temperatures can speed up development of the immature mosquito stages, virus replication, and also potentially extend the problem season (during the warmer months). This aspect of arboviral disease is discussed in more detail in Weaver and Reisen 2010.

Sea level changes are also likely to impact the intertidal systems, especially in the longer term. For example, Traill et al. 2011 modelled the effect of sea level rise on coastal vegetation in the southern part of Moreton Bay, Queensland, until 2100. They used two scenarios: a conservative rise of 0.64 m and a worst-case scenario of a rise of 1.80 m. For the conservative prediction they estimated a loss of 33% of saltmarsh and a gain of 35% of mangroves. This may represent the scenario whereby mangrove habitats replace saltmarsh ones. For the extreme scenario there would be a 19% loss of mangroves and a 98% loss of saltmarsh. In that scenario mosquito habitats would be reduced. In our research we found minimal change (0.14 to 0.17%) in the extent of mangroves at the seaward edge over the 32-year period (1972 to 2004), indicating that mangroves in the area were not being lost to the sea, at least in decadal time frames. This may reflect a process of sediment accumulation and there are data that suggest an accumulation rate of ∼5 mm/year over the last 100 years in two mangrove sites in the general area (Knight, unpublished data). Sea level is generally rising in the area at a rate of 3.1 mm/year (Burton et al. 2009). Nevertheless, there is considerable uncertainty, with outcomes dependent on the various models used. For example, Oliver et al. 2012, for the southeast coast of New South Wales, Australia, predicted virtually 100% mangrove cover by 2100 under one sea level rise model but most of the area under water in another.

In summary, sea level rise scenarios may result in a range of mosquito-related responses. They include: increased mosquito problems, for example, if mangrove basins replace saltmarsh larval habitats; reduced mosquito problems, for example, if the intertidal areas are inundated more frequently and become wetter, inhibiting mosquito egg conditioning; or eradicating the mosquito problem if inundation is more or less continuous (one of the scenarios in Oliver et al. 2012).

Management

Management needs to regard the uncertainty of ecological responses to climate and sea level changes. Planning time-frames for human adaptation to the changing intertidal environments is also important. Rather than attempting to adapt to end-of-century scenarios, the planning time-frame realistically needs to look forward for a few decades, for example within the 2031 planning horizon of the Southeast Queensland Regional Plan (at least for southeast Queensland, as we have used here). Models by Oliver et al. 2012, for eastern Australia showed large variations by 2100, but, whichever model they applied, the changes by 2030 were relatively minor.

From a human health perspective, the key questions that remain are: will encroaching mangroves around areas of increasingly dense human settlement result in an increased or decreased mosquito problem? How can mosquito control and urban planning help reduce the risk of arbovirus disease transmission?

It is beyond the scope of this paper to address the human land use planning issue, but population and land use are also drivers of mangrove encroachment into saltmarshes, especially when close to the wetland (e.g., Eslami-Andergoli et al. 2013). Land use planning in Queensland under the Sustainable Planning Act (2009) is intended to manage the ‘effects of development on the environment’ (Section 3b). Environment is broadly defined and so includes issues of both ecosystems and human health. Effective management to conserve ecosystem and human health needs collaboration between planning and mosquito agencies (Muhar and Dale 2001) and needs to be developed as a priority action.

The question raised at the start was: will mangrove encroachment into saltmarshes affect saltwater mosquito habitats? To address this we have synthesized research findings at two levels of resolution. At the detailed level, the internal structure of mangroves relates to mosquito larval habitats and, at the general level, mangrove encroachment into saltmarshes has the potential to replace saltmarsh mosquito habitats with mangrove ones. This is a general pattern in Australia and overseas. The simple conclusion is that mangrove encroachment into saltmarshes will have implications for mosquito populations. There is uncertainty as to the detailed effects in specific locations and each will, at least for the present, need to be monitored and addressed on a case-by-case basis, supported by resources and capacity building. Mosquito management, which is locally implemented, needs to be integrated with land use planning systems, which often operate at a more general level.

Acknowledgments

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

This research was supported by data provided by State and National agencies in Australia, and by in-kind provision of facilities by Griffith University. Eslami-Andergoli was supported by Tehran University.

REFERENCES CITED

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  • Abdi, H. 2003. Partial Least Squares (PLS) Regression. In: M. LewisBeck, A. Bryman, and T. Futing (eds.) Encyclopedia of Social Sciences Research Methods. Thousand Oaks (CA), Sage.
  • Addison, D., S. Ritchie, L. Webber, and F.V. Essen. 1992. Eggshells as an index of aedine mosquito production 2: relationship of Aedes taeniorhynchus eggshell density to larval production. J. Am. Mosq. Contr. Assoc. 8: 3843.
  • Burton, D., K. Mallon, P. Taygfeld, E. Laurie, and L. Bowra. 2009. Climate risk report Scoping Climate Change Risk for Moreton Bay Regional Council (MBRC). 82 pp. avilable online: http://www.moretonbay.qld.gov.au/uploadedFiles/moretonbay/environment/ScopingClimateChangeRisk.pdf. Accessed 3 July 2013.
  • Carrascal, L. M., I. Galván, and O. Gordo. 2009. Partial least squares regression as an alternative to current regression methods used in ecology. Oikos 18: 681690.
  • Center for Disease Control and Prevention (CDC) Division of VectorBorne Diseases (2009) West Nile Virus. http://www.cdc.gov/ncidod/dvbid/westnile/mosquitoSpecies.htm#99. Accessed 29 April 2013.
  • Dale, P.E.R., K. Hulsman, D. Harrison, and B. Congdon. 1986. Distribution of the immature stages of Aedes vigilax on a coastal salt marsh in South East Queensland. Aust. J. Ecol. 11: 269278.
  • Dale, P.E.R. and J.M. Knight. 2008. Wetlands and mosquitoes: a review. Wetl. Ecol. Mgt. 16: 255276.
  • Dale, P.E.R., J.M. Knight, B.H. Kay, H. Chapman, S.A. Ritchie, and M.D Brown. 2008. Habitat characteristics and eggshell distribution of the salt marsh mosquito, Aedes vigilax, in marshes in subtropical eastern Australia. 8 pp. J. Insect Sci. 8:25, available online: insectscience.org/8.25
  • Department of Health and Ageing. 2013. National Notifiable Diseases Surveillance System. http://www9.health.gov.au/cda/Source/CDAindex.cfm Accessed April 30 2013.
  • EslamiAndargoli, L., P.E.R. Dale, and N. Sipe. 2013. Does spatial scale affect the pattern of mangrove change under different rainfall regimes? An example in southeast Queensland, Australia. Austral. Ecol. 38: 208218.
  • EslamiAndargoli, L., P.E.R. Dale, N. Sipe, and J. Chaseling. 2010. Local and landscape effects on spatial patterns of mangrove forest during wetter and drier periods: Moreton Bay, Southeast Queensland. Estuar. Coast. Shelf Sci. 89: 5361.
  • EslamiAndargoli, L., P.E.R. Dale, N. Sipe, and J. Chaseling. 2009. Mangrove expansion and rainfall patterns in Moreton Bay, southeast Queensland, Australia, Australia. Estuar. Coast. Shelf Sci. 85: 292298.
  • Griffin, L.F., J.M. Knight, and P.E.R. Dale. 2010. Identifying mosquito habitat microtopography in an Australian mangrove forest using LiDARderived elevation data. Wetlands 30: 929937.
  • Hugo, L.E., B.H. Kay, and P.A. Ryan. 2003. Autogeny in Ochlerotatus vigilax (Diptera: Culicidae) from Southeast Queensland, Australia. J. Med. Entomol. 40: 897902.
  • Jacups, S.P., P.I. Whelan, and B.J. Currie. 2008. Ross River virus and Barmah Forest virus infections: A review of history, ecology, and predictive models, with implications for tropical northern Australia. Vector Borne Zoonot. Dis. 8: 283297.
  • Knight, J.M. 2011. A model of mosquitomangrove basin ecosystems with implications for management. Ecosystems 14: 13821395.
  • Knight, J.M., P.E.R. Dale, J. Spencer, and L. Griffin. 2009. Exploring LiDAR data for mapping the microtopography and tidal hydrodynamics of mangrove systems: An example from southeast Queensland, Australia. Estuar. Coast. Shelf Sci. 85: 593600.
  • Knight, J.M., L. Griffin, P.E.R. Dale, and S. Phinn. 2012. Oviposition and larval habitat preferences of the saltwatermosquito, Aedes vigilax, in a subtropical mangrove forest in Queensland, Australia. J. Insect Sci. 12: On line: insectscience.org/12.16.
  • McKee, K.L. 2004. Global Change Impacts on Mangrove Ecosystems. U.S. Department of the Interior, U.S. Geological Survey.
  • McTainsh, G., B. Iles, and P. Saffigna. 1986. Spatial and temporal patterns of mangroves at Oyster Point Bay South East Queensland, 1944–1983. Proc. R. Soc. Qld. 99: 8391.
  • Muhar, A. and P.E.R. Dale. 2001. Minimizing impacts of mosquito managemnt in ecologically sensitive urban areas: a mulitscale approach. In: Nature Conservation 5. J.L. Craig, M. Mitchell and D. Saunders (eds.) pp. 448455. Surrey Beatty and Sons, Chipping Norton, NSW, Australia.
  • Muhar, A., P. Dale L. Thalib, and E. Arito. 2000. The spatial distribution of Ross River virus infections in Brisbane: significance of residential location and relationships with vegetation types. Environ. Hlth. Prev. Med. 4: 184189.
  • Murray, A. B., M.A.F. Knaapen, M. Tal, and M.L. Kirwan. 2008. Biomorphodynamics: Physicalbiological feedbacks that shape landscapes. Water Resour. Res. 44: W11301, doi:10.1029/2007WR006410.
  • Oliver, T. S. N., K. Rogers, C.J. Chafer, and C.D. Woodroffe. 2012. Measuring, mapping and modelling: an integrated approach to the management of mangrove and saltmarsh in the Minnamurra River estuary, southeast Australia. Wetl. Ecol. Mgt. 20: 352371.
  • Queensland Government. 2009. South East Queensland Regional Plan 2009–2031. Queensland Department of Infrastructure and Planning, Brisbane, Queensland.
  • Queensland Government. 2012. Queensland Coastal Plan State Planning Policy 3/11 Guideline: Coastal Protection. Department of Environment and Resource Management, Brisbane, Queensland.
  • Ritchie, S.A. and D.S. Addison. 1992. Oviposition preferences of Aedes taeniorhynchus (Diptera: Culicidae) in Florida mangrove forests. Environ. Entomol. 21: 737744.
  • Ritchie, S.A. and C.D. Jennings. 1994. Dispersion and sampling of Aedes vigilax eggshells in Southeast Queensland, Australia. J. Am. Mosq. Contr. Assoc. 10: 181185.
  • Saintilan N. and R.J. Williams. 1999. Mangrove transgression into saltmarsh environments in southeast Australia. Global Ecol. Biogeogr. 8: 117124.
  • Saintilan N. and K. Wilton. 2001. Changes in the distribution of mangroves and saltmarshes in Jervis Bay, Australia. Wetl. Ecol. Mgt. 9: 409420.
  • Sinclair, P. 1976. Notes on the biology of the saltmarsh mosquito, Aedes vigilax (Skuse) in south east Queensland. Qd. Nat. 21: 134139.
  • Soares, M. L. G. 2009. A conceptual model for the responses of mangrove forests to sea level rise. J. Coast. Res. Special Issue 56: 267271.
  • The State of Queensland (Office of Treasury and Trade). 2013. Population and Dwellings, South East Queensland. http://www.oesr.qld.gov.au/products/profiles/.../pophousingprofilesseq.pdf. viewed 7th May 2013.
  • Traill, L.W., K. Perhans, C.E. Lovelock, A. Prohaska, S. McFallan, J.R. Rhodes, and K.A. Wilson. 2011. Managing for change: wetland transitions under sealevel rise and outcomes for threatened species. Divers. Distrib. 17: 12251233.
  • Weaver, S.C. and W.K. Reisen. 2010. Present and future arboviral threats. Antivir. Res. 85: 328345.