An Operational Timeline for Initiating EEE Bridge Vector Control

Weekly Data (By Location)

Column 1 CTA

Control Programs

Column 2 CTA

Service Request

Column 3 CTA

Mosquito Species with Monthly and Yearly Distribution

Column 4 CTA

By Jeffrey S. Brown and Rick Hickman


The Centers for Disease Control (CDC) and Prevention has published Guidelines for Arbovirus Surveillance in the Unites States (April 1993) to standardize the surveillance for mosquito borne viral encephalitis. These guidelines emphasize predictive, proactive and efficient methods of surveillance whenever possible (Moore et al. 1993). The national trend in mosquito control has been to decentralize arbovirus surveillance programs from federally funded state programs to local county programs. The operational emphasis at the local level has been placed on incorporating Integrated Pest Management (IPM) practices into mosquito control operations. This can be difficult to achieve because of the technical, financial and political constraints placed on mosquito control activities. Operational mosquito control programs are tasked with managing mosquito populations to minimize the risk of both nuisance and public health mosquito pests.

The CDC defines surveillance as “the organized monitoring of levels of virus activity, vector populations, infections in vertebrate hosts, human cases, weather and other factors to detect or predict changes in the transmission dynamics of arboviruses.” They further suggest “a sound (arbovirus) surveillance program requires a thorough understanding of the biology, ecology and interactions of the vertebrate and mosquito hosts (Moore et al. 1993). Very few local mosquito programs have the financial resources, laboratory facilities or technical expertise to address the dynamics of arboviruses. Most focus their energies on managing mosquito vector populations. Local programs are often a key source for information regarding historical mosquito populations. Control measures are typically initiated when a particular predictor exceeds the action threshold (usually determined from historical data and experience) (Moore et al. 1993). The CDC suggests that different measures or predictors for epidemic arboviral transmission are effective at different times of the year. The earliest useful predictors are climatological factors that influence the size of the early mosquito population. Midseason predictors usually consist of population estimates of vectors and vertebrate hosts (especially the young of the year), and evidence of early virus transmission in the natural cycle. The likelihood of an outbreak is estimated by comparing current vector and vertebrate host population densities and the age structures with long-term averages. Late season predictors consist of evidence of virus spillover to sentinel bird/chicken flocks, epidemic/epizootic vectors and domestic animals. The likelihood of transmission to humans or domestic animals becomes more accurate as the virus begins to circulate in vector and vertebrate host populations (Moore et al. 1993). Local programs without arbovirus testing capabilities are forced to rely solely on climatological and mosquito surveillance data.

This project involves using weekly mosquito surveillance and monitoring information to create a timeline that can be used to project the potential for increased arbovirus activity. The timeline will help to identify the best opportunities for collecting, pooling and testing mosquitoes that may be carrying Eastern Equine Encephalitis (EEE). Because mosquitoes reside in a constantly changing environment, recognizing the environmental trends that limit or enhance mosquito production is essential for successful control operations. Mosquito production is regulated seasonally by climactic variables, temperature, day length, groundwater levels and rainfall and other variables such as predators, suitable oviposition habitat, effects of ongoing control efforts and the reproductive capacity of the mosquito species in question. Mosquito surveillance and monitoring are essential for successful control activities.

The timeline (Diagram 1)anticipates arbovirus activity for EEE based on maintenance vector activity of Culiseta melanura and associated bridge vector activity. This is not a scientific model per say, but a surveillance strategy that encourages weekly surveillance and monitoring of mosquito species and their associated population trends to assess both nuisance mosquito populations and to project the potential for arbovirus activity. This can ultimately enhance nuisance mosquito control operations and potentially minimize the risk of arbovirus transmission.

The timeline promotes mosquito and environmental monitoring to identify potential EEE virus-scenarios. For example, by using larval dipping and light trapping to document the adult emergence of Cs. melanura, programs can track the sequence of events that must occur for EEE transmission. The timeline considers 6 variables that must be in place:

  • EEE virus activity
  • Presence of reservoirs in the environment
  • Maintenance vector (abundance and seasonal occurrence)
  • Bridge vector (abundance and seasonal occurrence)
  • Susceptible mammal population
  • Suitable environmental conditions

The timeline chronology begins when the virus activates and infects a wild bird and progresses until there is a laboratory confirmation of a positive virus isolate in a mammal. It basically follows one virus through all the possible scenarios that can occur in southeastern North Carolina. The progression includes virus activation, wild bird infection, maintenance vector activity, sentinel flock seroconversion, laboratory analysis of serum from the sentinel flock, potential bridge vector activity, mammal infection and laboratory results of the mammal infection. Given that the virus is endemic in the area there are sufficient reservoirs, a chronological description of the timeline progresses as follows:

  • Day 1 EEE virus activates
  • Day 1-4 Wild bird infected, virus amplifies in bird
  • Day 5-7 Wild bird infectious
  • Day 5-7 Cs. melanura bites infected bird
  • Day 5-17 EEE virus amplifies in Cs. melanura
  • Day 18-25 Cs. melanura capable of vectoring virus
  • Day 18-20 Cs. melanura infects a wild bird or sentinel chicken
  • Day 20-23 Bird host becomes infectious
  • Day 20-25 Bridge vector bites infected bird
  • Day 21-32 Virus amplifies in competent bridge vector
  • Day 23-36 Local mosquito control program could be notified of seroconversion
  • Day 32-42 Bridge vector capable of virus transmission to mammals
  • Day 34-42 Mammal infection possible
  • Day 34-42 Virus incubation period in mammal
  • Unknown Virus diagnosed by physician or veterinarian
  • Unknown Laboratory confirmation
  • Unknown Local mosquito control program is notified

In southeastern North Carolina, the typical Cs. melanura adult female lives about 25 days (based on weekly light trap data). The EEE virus is only infectious in the mosquito for the last week to ten days of its life because of virus amplification considerations. Some of the questions to consider when evaluating the EEE potential are listed below.

  • Is there a new emergence of competent bridge vectors within the projected 7-day window that Cs. melanura is infectious?
  • What is the overlap in habitats of the bridge vector population and the infectious Cs. melanura?
  • Is it early in the arbovirus transmission season?
  • How significant are the maintenance and bridge vector populations?
  • Are environmental conditions favorable for future populations of Cs. melanura and bridge vector species?

Interpreting this information can help to identify the risk to the human population and weekly monitoring allows this risk to be assessed in real time. Therefore, the timeline is proactive and not of reactive. In order to apply the timeline in an operational setting, some fundamental information about EEE and how it interacts with maintenance vectors, bridge vectors, wild birds, sentinel chickens, horses, humans and the environment should be reviewed.

Eastern Equine Encephalomyelitis (EEE)

Eastern Equine Encephalitis (EEE) virus is an arthropod borne virus belonging to the virus family Togaviridae, genus Alphavirus. The initial isolate of EEE was recovered in 1933 from the brain of a dead horse in New Jersey by Giltner and Shahan (Eldridge and Edman 2000). The first mosquito isolate of EEE virus was obtained from Coquillettidia perturbans in 1948. The first isolation of EEE in the enzootic mosquito vector, Culiseta melanura, was reported by Chamberlain et al. in 1951 (Scott and Weaver 1988).

In North America, the principal area of transmission is the Atlantic Coast from New Hampshire to Florida and along the Gulf States to Texas (Scott and Weaver 1988). In the northern part of this virus range, both human and equine cases occur between July and October. In Florida, human cases occur throughout the year, but are concentrated between May and August. The virus or disease is often recognized in wild birds, penned pheasants, or equines prior to human involvement (Morris 1988). Epidemics typically occur during August and September and are often preceded (from 2 weeks to 2 months) by epizootics among game birds or horses (Scott and Weaver 1988). The seasonal changes in Cs. melanura biology and its relationship to EEE virus transmission vary with geographic location and its associated climate. In Florida, for example, transmission occurs throughout the year but peaks during the summer months (Bigler et al. 1976; Wilson et al. 1986)*. Conversely, in temperate areas like Maryland, there is a distinct transmission season. Virus is not detected until July and can remain active in the mosquito-bird cycle until the first heavy frost in November or December. Most virus isolations in Maryland have been recovered in late August through October (Williams et al. 1972, 1974; Watt et al. 1987; Scott et al. 1987)*.

There seems to be no clear-cut relationship between epidemics and any known environmental factors. It is likely that a complex of environmental conditions must simultaneously impact several parameters, such as vertebrate-host population density, brood size and nutritional status, vector population density and longevity, and winter survival of both vectors and vertebrate hosts (Centers for Disease Control 1996). The isolation of EEE from mosquitoes is very common during epizootics but much less common during inter-epizootic periods (Morris 1992).

EEE and Culiseta melanura

It is now generally accepted that in North America EEE is transmitted among passerine birds primarily, if not solely, by Culiseta melanura in freshwater, forested swamp habitats. (Scott and Weaver 1988). The disease is rural in distribution and most cases are associated with wooded areas adjacent to swamps and marshes (Morris 1988).

EEE epidemiology typically concludes that

  • Cs. melanura is the principal endemic vector.
  • Wild birds, primarily passerines, serve as the amplifying hosts.
  • Epizootics begin in swamps and move locally outward through viremic birds.
  • In some cases, migrating birds introduce the virus into distant “clean” ecosystems.
  • Humans and equines are dead-end hosts.
  • There is a single most important epidemic or epizootic vector, non-avian vertebrates are potential but unproven maintenance hosts.
  • The over-wintering mechanism is unknown or an over-wintering mechanism is not required (Morris 1988).

Cs. melanura is a swamp mosquito, probably limited in its flight range and nocturnal or crepuscular habitat. Apparently increased humidity and temperature are critical factors in determining its range of distribution, population density, and seasonal prevalence. Developmental rates in nature are unpredictable and are subject to such variations imposed by weather factors and fluctuations in water level. In the southern states, breeding may be continual, although retarded during the winter months. In the northern states, development is interrupted during the winter; larvae may sustain themselves in protected habitats and complete their development the following spring. The number of annual generations ranges from a few in the northern states to several in the south (Siverly and Schoof 1962).

Population densities of this mosquito are highest deep in the interior of the swamp habitat, which is also where most enzootic EEE virus transmission occurs (Williams et al. 1972, 1974). However, mosquitoes will leave the swamp breeding sites and move to drier upland, forested habitats, especially during the late summer and fall (Joseph and Bickley 1969; Saugstad et al. 1972; Muul et al. 1975; Morris et al. 1976, 1980; Nasci and Edman 1984). It is believed that in these upland locations, blood feeding occurs after engorging female mosquitoes return to the swamp to oviposit (Morris et al., 1980; Nasci and Edman, 1984).*

The complexity of the swamp ecosystem is increased by the uneven upbuilding of the swamp floor. Because of the high water table, the substrate is habitually saturated so that root systems do not penetrate deeply into the organic soil. Instead, they grow near the surface, spreading widely. The resulting root buttress brings about a rapid and unequal elevation of the land, with levels around trees as much as 1 m above the general swamp level. Thus, the built-up areas associated with tree bases are also well above water level, while the intervening spaces are be below it. As the water level recedes, the exposed muck becomes subject to invasion by mosses, liverworts, and swamp annuals. Tussocks of ferns, uprooted trees and slowly decomposing fallen trunks provide a second series of rises and falls between the trees, making foot travel hazardous and slow. This uneven topography, when coupled with differences in moisture, temperature, illumination and wind exposure, creates a wide diversity of niches in the swamp forest environment. Which microhabitats favor the development of Cs. melanura and influence the epizootiology of EEE are yet unidentified. The relationship between soil types and land use has significant impact on the present and future determinations of the human population at risk to EEE (Morris et al. 1980).

Various observations have been made on the development of Culiseta melanura. Average periods of development are as follows: eggs-2 days, first instar larvae-3 days, second instar larvae-2 days, third instar larvae-2 days, fourth instar larvae-3 days, and pupa-2 days (Joseph and Bickley 1969). The life span of the adult female includes emergence, feeding (survival), resting, mating, and oviposition (Siverly and Schoof 1962). These mosquitoes breed in swarms above larval sites (Hayes 1958)*. Larvae develop in cool, acidic water that is shaded and often retained in depressions at the base of tree trunks, stumps or uprooted trees (Scott and Weaver 1988). Cs. melanura breeding and oviposition is usually concentrated at the edge of the swamps but also occurs in the swamp interior (Morris 1988). Larvae are most often found in sphagnum bogs, cedar swamps, or maple swamps (Scott and Weaver 1988).

In temperate regions, Cs. melanura populations have two or three major peaks of adult abundance, typically one each during May-June, July-August, and August-October (Morris et al. 1976, 1980; Pierson and Morris 1982; Nasci and Edman 1984; Watts et al. 1987; Scott et al. 1987)*. Host-seeking activity, and presumably blood-feeding, is greatest during the first 2 hours after sunset and then continues at a low but constant level until sunrise (Hayes 1962b, Nasci and Edman 1981a)*. During sunrise, there is an increase in activity that is believed to be more closely associated with location of suitable daytime resting sites than with location of hosts on which to blood-feed (Nasci and Edman 1981a)*. This mosquito does not usually feed diurnally. Results from analyses of Cs. melanura distribution within the forest stratum vary from equal activity at all heights tested to a flight preference to the canopy (Love and Smith 1958; Bast and Rhen 1963; Main et al. 1966; Joseph and Bickley 1969; Nasci and Edman 1981b)*

(Scott and Weaver 1988).

Finally, the longevity of mosquito encephalitis vectors is of special interest since it has been demonstrated that once mosquitoes became infective, EEE virus persisted as long as they remained alive (Davis 1940)*. Cs. melanura appears to have a survival of 3.5 months under insectary conditions (Love and Goodwin 1961)*. Studies of eastern (EEE) and western (WEE) equine encephalitis in and around swamplands have demonstrated that densities of Culiseta melanura are higher within swamps than the densities found at river and upland boarders of swamps (Williams et al. 1971a, Saugstad et al. 1972)**, that the largest numbers of mosquitoes infected with these viruses also occur inside swamps (Williams et al. 1972)**, and that infection of sentinel birds may occur at twice the rate within swamps compared to sentinel infection at localities immediately outside the swamps (Hayes et al. 1960, Dalrymple et al. 1972)**.

The greater number of mosquitoes at the perimeter reflects the true distribution resulting from an edge effect (Morris et al. 1980). Parity data indicates that Cs. melanura remains in the swamp when populations are low or during adverse weather conditions but, during periods of high populations or warm, wet weather, older females, and presumably older males, spread at least 2 km from the nearest known Cs. melanura breeding area (Morris et al. 1980). During EEE transmission seasons, Cs. melanura populations at the perimeter have the greatest vector potential, but as late season populations increases, vector potential is relatively greater more distant from the swamp (Morris et al. 1980).

Dalrymple et al. (1972) presented evidence indicating that the fresh water swamp habitat is the enzootic focus for WEE and EEE in the eastern U.S. and that under normal weather conditions these viruses are largely restricted to this habitat. Adult mosquito population densities are closely linked to environmental conditions and densities are highest during years with high rainfall. Associations between EEE outbreaks and excessive rainfall have been observed when there is unusually heavy rainfall during the summer of the outbreak and during the autumn months of the preceding year (Hays and Hess 1964). The relative infrequency of EEE or WEE infection in man in the coastal states seems attributable to the infrequency of the endemic vector, Cs. melanura, feeding on man, and its greatest densities occurring in swampy habitats away from man. Cs. melanura seems to be important in endemic transmission because it is ornithophilic, multivoltine and abundant throughout the summer. Its abundance is attributable to its unusual breeding habits and because its larvae are able to survive in the saturated root mat when the water table declines, leaving other breeding sites dry (Muul et al. 1975). Late season dispersal of Cs. melanura could carry virus from the swamp to outlying areas and thereby initiate outbreaks in non-avian vertebrates (Morris et al. 1980). Epidemics of disease produced by EEE and WEE generally occur during late summer and fall when wild birds are at their maximum population density and are most mobile. It would appear logical that these vertebrates could largely be responsible for the maintenance of annual virus enzootics (Dalrymple et al. 1972).

EEE and Bridge Vectors

A major question in the ecology of EEE is the identity of the bridging vectors that transfer the virus from the enzootic cycle to humans and equines. A variety of species serve as vectors, depending on time of year, environmental conditions, geographic location and population dynamics (Moore et al. 1993). Among the factors that determine the vector potential of various mosquito species is their distribution in time and space. Mosquito larvae are confined to aquatic habitats which can be roughly correlated with patterns of vegetation, especially those patterns which are produced as a result of edaphic factors (influenced by the soil rather than the climate). Adult mosquitoes are more mobile and their distribution in space may or may not conform to the distribution of the larval stages. Furthermore, adult distribution varies with time and adult activity cycles. If suitable larval habitats are present in one type of area, but suitable hosts for adults are scarce or absent, adult females must migrate from their breeding areas to other areas in order to obtain the blood necessary for reproduction (Saugstad et al. 1972).

The endemicity of EEE may vary in intensity from year to year depending on the size of the population of Cs. melanura, the timing of virus introduction during the population growth phase of the endemic vector (late introductions may not allow sufficient time for massive replication and proliferation), and the availability of susceptible hosts, especially non-immunes. An epidemic may occur depending on the population size of the endemic vector, and the intensity of the endemic transmission cycle, particularly in the vicinity of human habitations (Muul et al.1975). A mechanism for “spillover” of virus from the swamp to areas inhabited by man and domestic animals is probably dependant upon unusually large populations of “invader” species entering and leaving the swamp during periods of abnormally high precipitation and temperature (Saugstad et al. 1972). It is important to emphasize that there is no single epizootic vector species for all foci or for all hosts within a focus (Morris 1988).

Mosquito survival is one of the most important factors used to determine the extent in which a mosquito participates in arbovirus transmission. Mosquito survival affects the vectorial capacity of the mosquito population, the rate of mosquitoes biting on man and the duration of the virus extrinsic incubation (EI) period in the mosquito. Mosquito biting rate, mosquito population size, and the EI period of viruses are also basic considerations in any analysis of arbovirus transmission (Scott and Weaver 1988). Extrinsic incubation in mosquitoes requires that susceptible mosquitoes must locate and blood feed on a viremic vertebrate host. These mosquitoes must then survive an extrinsic incubation period. During which time the virus replicates in posterior midget epithelial cells, disseminates through hemolymph to salivary glands, replicates in salivary acinar cells, and is released into the salivary matrix. During re-feeding, infected mosquitoes transmit virus via their saliva to susceptible hosts, which may become viremic and infect additional mosquitoes that imbibe their blood (Scott and Weaver, 1988).

It is important to emphasize that mosquito population density is not, in itself an indicator of either detectable endemic or epidemic EEE activity (Morris 1988). For example, Ochlerotatus sollicitans is a brooded mosquito and in brooded mosquitoes, high populations are usually an indication of a fresh emergence. Annoyance can be considerable but the adults are mainly nulliparous, thus, the vector potential is low.

Over time, annoyance declines but the adults become parous and the vector potential of the biting population increases. The technique of evaluating vector potential by measuring the number of parous mosquitoes coming to bite yields the index that New Jersey uses to institute control for the prevention of human disease. When EEE is known to be circulating in Cs. melanura, chemical control is directed toward “old” populations of Oc. sollicitans. This results in spot treatments in designated areas of the state rather than broad scale chemical coverage during emergency periods (Crans and McCuiston, 1993).

In some foci, endemic passerine birds and mosquitoes are monitored for endemic EEE activity (Crans 1985, Howard 1985)*. Virus isolations from mosquitoes other than Cs. melanura signal possible epidemic activity and elude appropriate warnings and recommendations for increased mosquito control activity (Crans 1985, Maxfield 1982, Howard 1985)*. Presently these techniques are time consuming and costly. Often the turn around time from the field to laboratory results is too long for implementing effective EEE preventative mosquito control.

Komar et al. (1999) permitted Oc. triseriatus, Ae. aegypti and Ae. albopictus to feed on starlings 2 days after they were inoculated with virus. Engorged females were incubated for 10 days and then individually tested for virus by plaque assay. Specimens of all three species became infected. Although some mosquito species might be involved in epidemic transmission more often than others, the identification of the epidemic vectors of EEE virus has been difficult because no single species is consistently associated with the transmission of virus to horses or humans. Depending on time of year environmental conditions, geographic location, or dynamics of respective mosquito populations, various species probably serve as epidemic vectors (Scott and Weaver 1988).

Mosquito arbovirus collection and pooling for EEE in North Carolina is conducted by the Department of Environment and Natural Resources’ Public Health Pest Management Section. This section works closely with county health officials and operational mosquito programs to monitor and respond to EEE activity across the state. They provide technical assistance and facilitate laboratory testing for arthropod vectors and the diseases they transmit.

The potential EEE bridge vector species in North Carolina include Aedes albopictus, Aedes vexans, Anopheles crucians, Anopheles punctipennis, Anopheles quadrimaculatus s.l., Coquillettidia perturbans, Culex erraticus, Culex pipiens complex, Culex restuans, Culex salinarius, Ochlerotatus atlanticus/tormentor, Ochlerotatus c. canadensis, Ochlerotatus infirmatus, Ochlerotatus sollicitans, Ochlerotatus triseriatus, and Psorophora columbiae. Aedes vexans, Culex restuans, Culex salinarius, Culiseta melanura, Ochlerotatus atlanticus/tormentor, Ochlerotatus triseriatus and Psorophora columbiae have tested positive for EEE in North Carolina (Harrison 2001).

EEE and Wild Birds

As stated earlier, in North America, EEE is transmitted among passerine birds primarily by Cs. melanura. Blood meal analysis of engorged Cs. melanura show that this ornithophilic mosquito feeds almost exclusively on passerines. Nestlings or young birds that have just left the nest are considered the most important vertebrate hosts for EEE because they are the most susceptible to infection. Their viremia has greater magnitude and duration, and they are behaviorally less defensive than adult birds toward host seeking mosquitoes (Dalrymple et al. 1972; Edman and Scott 1987; Scott et al. 1988a)*

Crans et al. (1994) discussed the cryptic cycle of EEE that begins with recrudescence of latent virus in previously infected AHY (after hatch year) birds. In the early spring, reactivation of latent virus may result from changes in the physiological state of local birds (i.e., summer residents, residents) related to factors such as stress of migration, establishment of territory, or other breeding activities. Epiornitic (bird to bird) cycling requires an influx of newly emerged Cs. melanura. If recrudescing virus is acquired by Cs. melanura females that have already completed one gonotrophic cycle, it is unlikely these females will live long enough for the virus to replicate and be transmitted to other hosts.

In some years, the virus can become epiornitic by early July, when conditions are favorable for Cs. melanura. If vector populations are limited by unfavorable conditions during that critical period, the virus cycle is either postponed until the August emergence or eliminated entirely (Crans et al. 1994). Birds that frequent the deep swamp are at greater risk of infection with EEE virus than wild birds at the river and upland boarders of the swamp. Similarly, sentinel birds located deep in the swamp have been found to acquire infections more readily than sentinels located on its borders (Williams et al. 1972, Dalrymple et al. 1972)*. Thus, the innermost swamp appears to be a central focus for the transmission of EEE virus. It is the optimum larval habitat for Cs. melanura, and the highest densities of adult female Cs. melanura mosquitoes occur there. Significant movements of Cs. melanura occur between the swamp and outlying areas (Saugstad et al. 1972)*, perhaps introducing viruses to areas outside of the swamp. Also, viremic birds may depart the deep swamp during migration or at other times and convey viruses to new areas.

Characteristic symptoms of EEE infection in birds begins with fever, weakness, inactivity, ruffed plumage, or stupor. Later, ataxia, trembling and paralysis may ensue (Scott and Weaver 1988). Komar et al (1999) suggest EEE viremia in starlings remains sufficiently intense to infect mosquitoes for three days, compared to just one day for robins. Although starlings sustained a viremia intense enough to infect mosquitoes for at least two days, viremias were sustained at this level in other birds for only one day.

County mosquito programs in North Carolina do not have the resources to conduct wild bird sampling as a surveillance tool for monitoring EEE. State personnel conduct wild bird sampling (using mist nets) on a very limited scale to address specific arbovirus questions. Historically, North Carolina has relied heavily on their sentinel chicken program as the primary surveillance tool for monitoring EEE activity.

EEE and Sentinel Chickens

The CDC guidelines suggest that captive sentinel animals be used to establish the presence of arboviruses and to monitor temporal and spatial changes in virus activity in an area. Sentinels are sometimes used to attract mosquitoes for virus isolation. The primary advantage of using captive sentinels is that the time and place of exposure are known. The use of sentinel animals also assures uniformity in selection of location, habitat, number, breed, age and source of the animals and the sampling schedule. Seroconversion and field infection rates are reliably determined when the foregoing factors are controlled. The disadvantages of sentinel animals include the expense of buying the animals, building shelters or cages and maintaining the animals in the field. The lack of mobility of sentinel animals affects their exposure to mosquitoes and limits the geographic area represented (Moore et al. 1993).

Probably the most widely used sentinel animal for arbovirus surveillance is the domestic chicken. Chickens are attractive hosts for Culex mosquito vectors and are susceptible to arboviral infections. Additionally, they can tolerate these infections and produce readily identifiable antibodies. Chickens are hardy and are easily handled and bled. They are relatively inexpensive to maintain on farms or in urban-suburban locations by residents or health officials. Eggs laid by the birds may provide an added incentive and help defray any costs of maintaining the birds (Moore et al. 1993).

In the spring, monitoring sites are stocked with 10 to 30 pre-tested, non-immune, individually banded birds approximately 6 to 8 weeks old. Dispersing smaller groups throughout an “at risk” area yields a more representative estimate of arbovirus activity. The choice of monitoring sites should be based on historical records of virus activity, vector resting sites or flight corridors, and the likelihood of virus transmission rather than convenience. The chickens are kept in standard sentinel sheds or similar structures (Moore et al. 1993).

Sentinel chickens are bled biweekly or monthly from the wing, the jugular vein, or from the heart throughout the transmission season. Seroconversions may occur 2-3 weeks before the detection of equine or human cases of EEE. If the intent of surveillance is to monitor season-long transmission, birds that seroconvert are replaced by non-immune birds, preferably of the same age (Moore et al. 1993).

North Carolina uses sentinel flocks consisting of 5 white leghorn laying hens per flock to detect virus EEE activity in specific endemic areas. Although the flocks do not indicate the degree or distribution of the EEE virus, they do provide valuable data about current EEE activity. Because the flocks are bled on rigid schedules, the approximate date of infection can be projected when a bird serocoverts. Comparing this result against weekly mosquito light trap data provides valuable information about the maintenance and bridge vector populations and their potential for disease transmission. Consecutive seroconversions from a sentinel flock location and elevated mosquito light trap counts generally indicates a need for increased mosquito control measures.

North Carolina has maintained sentinel chickens for EEE since 1984. Their operational response to a seropositive bird in a sentinel flock is considered to be a focal in nature. Typically spot spraying around the flock and its associated mosquito habitat is appropriate. Additionally, local health departments may issue press releases recommending mosquito avoidance practices and the use of personal protection measures such as mosquito repellants.

EEE and Horses

Surveillance for equine cases in areas with susceptible horse populations may provide the most practical and sensitive tool for the recognition of a potential public health problem caused by EEE viruses. This is especially true in areas that lack the resources to monitor virus activity in birds and mosquitoes. As a result of their high field exposure, horses are subject to high vector attack rates (Moore et al. 1993). A survey of 67 equine deaths due to EEE in the US in 1971 showed that the most common signs of illness were depression, fever above 103o F, ataxia, paralysis, anorexia and stupor. Other symptoms of note include irregular gate, grinding of teeth, lack of coordination, circling, staggering, recumbency and hyper-excitability. Death typically occurs in 1 to 3 days (Morris 1988) and the mortality rate is 80 to 90% (Eldridge and Edman 2000). Most EEE endemic states have a diagnostic service for veterinarians and physicians to submit blood and organ samples from suspected EEE cases. Reporting is required in some states and voluntary in others. These systems are inappropriate for early warning since the turnaround time for most is too long (Morris 1988). Limitations to using equines include their vaccination status, movement into and out of the surveillance area, and lack of prompt reporting of morbidity by attending veterinarians (Moore et. al 1993).

Franklin et al. (2002) reported a 14 month-old gelding quarter horse receiving routine vaccination for EEE and becoming symptomatic with EEE one week later. This horse case occurred in California, which is outside the known geographic range of EEE. Although the source of the EEE infection cannot be conclusively determined, it may suggest a possible incubation period for the EEE virus before clinical symptoms develop.

Nationally, EEE activity is monitored using passive surveillance for humans and horses. North Carolina averages 1 to 2 human cases and 10 to15 horse cases of EEE a year. Passive veterinary surveillance provides county programs in North Carolina with information about positive horse cases of EEE within their jurisdictions. Usually, by the time the county receives the laboratory test results it is too late to implement effective control measures with respect to the mosquito population causing the initial infection. Responding to a confirmed horse case of EEE does allow the program to determine if the environmental conditions are favorable for continued virus transmission. Documenting and mapping the mosquito production habitats in the area around a verified EEE horse case may provide information about the potential bridge vector species involved in future EEE transmission in that area.

EEE and Humans

Human cases of EEE can offer important insight into the dynamics of EEE and mosquito populations. An understanding of the etiology of EEE in humans can assist in determining the initial time of the EEE infection. It takes 4 to 10 days after a bite for symptoms to develop (Centers for Disease Control 1996). The illness can be systemic or encephalitic. A systemic infection is abrupt and characterized by malaise, arthralgia (joint pain) and myalgia (muscle pain). The patient experiences chills and severe muscular shaking which may last for a few days. Temperatures reach 100oF to 104oF and the illness can last 1 to 2 weeks. There is no central nervous system involvement and recovery is complete (Morris 1988).

The encephalitic form of the disease is characterized by an abrupt onset in infants, whereas in older children and adults onset of active encephalitis typically occurs after a few days of indisposition. Symptoms include fever from 102oF to 106oF, irritability, restlessness, drowsiness, anorexia, vomiting, diarrhea, headache, cyanosis (blue skin color due to lack of oxygen), convulsions and coma. Death occurs 2 to 10 days after being bitten (Morris 1988).

During epidemics, the initial mortality approximates 70%. Nearly all that recover have some degree of progressive mental and/or physical disabilities. In one study, the initial mortality was 74%, but 9 years post-infection total mortality reached 90%, with only a 3% complete recovery rate. People with the greatest risk of developing EEE are children under 15 and adults over 55. These age groups make up 70 to 90% of the cases in any given outbreak (Morris 1988).

The public health response for mosquito transmitted disease is initiated when there is confirmation of increased arbovirus activity in an area. At the local level, confirmation is received after a human or veterinary case is identified and reported to the State Health Department. This type of surveillance is often equated to closing the barn door after the horse is already out. The impact of prevention or control measures on the course of a potential epidemic is diminished by even the smallest delay. By the time human cases are confirmed (a very accurate predictor), the epidemic may be waning of its own accord and control measures may have little impact (Moore et al. 1993).

Most of the human cases of EEE in North Carolina occur in areas of the state that do not have strong county operational programs. Incorporating IPM strategies with mosquito surveillance and monitoring is paramount to protecting the health of the citizens of North Carolina.

Environmental Considerations

Vectorborne disease surveillance systems seldom incorporate weather data, partly because of the lack of understanding of the relationship between weather patterns and disease outbreaks (Eldridge and Edman 2000). The effects of precipitation with regard to mosquito production cannot be understated. Letson evaluated rainfall patterns in states and locales where human EEE occurred between 1983 and 1989. He found a significant association between the occurrence of human cases and excess rainfall in the years when the cases occurred. The association was stronger with data from local weather stations than those from statewide rainfall averages (Centers for Disease Control 1996).

The National Weather Service maintains stations that collect environmental variables such as daily precipitation, minimum daily temperature, maximum daily temperature, average daily temperature and wind speed and direction. The average annual precipitation for southeast North Carolina is 54.3 inches. The average high temperature is 73.8oF, the average low temperature is 53.0oF and the average mean is 63.4oF. The weather data for this project was collected from the Wilmington Weather Forecast office (National Weather Service). The Wilmington office monitors Bladen, Brunswick, Columbus, New Hanover, Pender and Robeson Counties in southeastern North Carolina.

Documenting daily precipitation and temperature within a defined control area using a standardized weekly format that can be compared to weekly light trap data is essential to reconciling mosquito production over time, within and across habitat types. Creating and maintaining a long-term weather monitoring database at the local level can help to determine the relationship between weather patterns, Cs. melanura population densities and potential EEE virus amplification (Centers for Disease Control 1996).

The environmental variables that can influence mosquito production within the system throughout the year should be identified. Each habitat type has very specific environmental influences that affect mosquito production over time. Each new site should be monitored and compared against existing sites with similar designations throughout the year. The goal is to identify similar seasonal trends for mosquito production within each habitat type and to monitor the effects of fluctuating environmental variables affecting each habitat type.

Timeline Pooling Projections for 2000, 2001 and 2002

In 1999, this timeline was developed to address EEE potential at the local mosquito program level. The purpose was to combine and extrapolate weekly mosquito surveillance and monitoring data across multiple counties (Map 1) to project the potential for EEE activity across local and regional mosquito populations. Brunswick County Mosquito Control has been monitoring Cs. melanura emergence since 1990 using a New Jersey light trap. Using their weekly surveillance data as the foundation (Graphs 1, 2, and 3), the EEE timeline was used to project mosquito pooling dates in 2000, 2001 and 2002. When a pooling date was selected, each participating county placed CDC miniature light traps baited with CO2 near their sentinel flocks. These sites were selected because of their proven history of EEE activity. The light traps were placed at least ten feet into the tree line edge between the sentinel flock and the mosquito habitat the flock monitors. Counties with extra resources placed additional traps at other potential EEE sites. The additional trap sites were chosen based on a history of Cs. melanura activity from previous light trap collections.

2000 – September 26 Pooling (Week 39)

Using the EEE timeline, the 2000 pooling date was selected based on Brunswick County’s weekly light trap data collected during Week 35 (August 28-September 3, 2000) (Graph 1). The goal of this pooling effort was to project a pooling date when the bridge vector species could transmit the EEE virus to humans. Cs. melanura populations were peaking during the week of August 28. Based on the previous week’s trap data, it was evident that this brood of Cs. melanura emerged the week of August 21 (Week 34). Using the timeline and counting August 27 (Week 34) as day 1, 29 days were added and a pooling date was projected for September 24 (Week 38) in hopes of collecting bridge vector mosquito species. Note that any date between timeline (TL) day 29 and day 35 could be used as the pooling date. A pooling date of September 26 (Week 39) was chosen and the participating counties were contacted. The county mosquito programs participating in this multi-county effort included Brunswick, Bladen, New Hanover, and Pender Counties.

2001 – October 3 Pooling (Week 40)

Following the strategy developed from the 2000 season, Cs. melanura populations were closely monitored at the Brunswick County trap site (Graph 2 ). Week 35 (Aug 27-Sept 2) indicated an increase in the Cs. melanura population from 7 in Week 34 to 26 in Week 35. A pooling date of Sept 20 (Week 38, TL day 25) was projected. The goal of this pooling was to assess the Cs. melanura populations between the woodland pool habitat in Brunswick County and the flood plain habitat in New Hanover County. It was noted that another brood of Cs. melanura emerged during Week 38 (Sept 17-23). On September 20, a multi-county pooling date of October 3 (Week 40) was projected. This pooling would have two objectives, to collect EEE positive bridge vectors from the initial brood of Cs. melanura (TL day 38) and to collect EEE positive Cs. melanura from the second brood of Cs. melanura (TL day 17). The mosquito programs participating in this multi-county effort included Brunswick, Bladen, Columbus Onslow, New Hanover, and Pender Counties.

2002 – October 9 Pooling (Week 41)

Following the previous strategy, Cs. melanura populations were closely monitored at the Brunswick County trap site (Graph 3). Week 37 (Sept 11) indicated an increase in the Cs. melanura population from 3 in Week 36 to 65 in Week 37. Daily collections of the New Jersey trap identified an increase in Cs. melanura populations on Sept 11 (Week 37, TL day 6). A pooling date of Oct 8 (Week 41, TL day 33) was projected. The programs participating in this multi-county effort included Brunswick, Bladen, Columbus, Onslow, New Hanover, and Pender Counties.


Multi-County Mosquito Arbovirus Pooling Results 2000, 2001 and 2002

2000 (Week 39)

Five mosquito pools tested positive for an arbovirus from this pooling effort. Bladen County had 2 pools of An. crucians testing positive for a California group virus. Brunswick County had 1 pool of Ochlerotatus taeniorhynchus test positive for a California group virus. New Hanover County had 1 pool of Ochlerotatus infirmatus test positive for Highlands J virus. Finally, Pender County had 1 pool of Culex salinarius test positive for EEE. Table 1 provides an overview of the 2000 multi-county pooling results. Graph 4 shows the distribution of mosquito pools by genus.

Table 1. September 26, 2000 Multi-County Pooling Results

Total Mosquitoes Collected
Bridge Vectors Collected
Total Cs. melanura Tested
% Cs. melanura of Total
Number of Mosquito Species Collected
Number of Mosquito Pools Collected
Number of Trap Nights
Mosquito Pools Testing Positive EEE
Mosquito Pools Testing Positive an Arbovirus


2001 (Week 40)

One mosquito pool tested positive for an arbovirus from this pooling effort. Pender County collected 1 pool of Culiseta melanura that tested positive for Highlands J virus. Table 2 provides an overview of the 2001 multi-county pooling results. Graph 5 shows the distribution of mosquito pools by genus.

Table 2. October 3, 2001 Pooling Results (Week 40)

Total Mosquitoes Collected
Bridge Vectors Collected
Total Cs. melanura Tested
% Cs. melanura of Total
Number of Mosquito Species Collected
Number of Mosquito Pools Collected
Number of Trap Nights
Mosquito Pools Testing Positive EEE
Mosquito Pools Testing Positive an Arbovirus


2002 (Week 41)

Two mosquito pools tested positive for an arbovirus from this pooling effort. Pender County collected 1 pool of Culiseta melanura testing positive for Highlands J virus. New Hanover County collected 1 pool of Culiseta melanura testing positive for Highlands J virus. Table 3 provides an overview of the 2002 multi-county pooling results. Graph 6 shows the distribution of mosquito pools by genus.

Table 3. October 9, 2002 Pooling Results (Week 41)

Total Mosquitoes Collected
Bridge Vectors Collected
Total Cs. melanura Tested
% Cs. melanura of Total
Number of Mosquito Species Collected
Number of Mosquito Pools Collected
Number of Trap Nights
Mosquito Pools Testing Positive EEE
Mosquito Pools Testing Positive an Arbovirus


Sentinel Chicken Flock Results from Project Counties 2000-2002

Table 4 shows that the counties in the project area had at least one bird seroconvert for EEE during the 3-year study period. The 2000 season identified 10 sentinel birds serocoverting for EEE. Two birds seroconverted in both 2001 and 2002. The table also shows which week each year the birds seroconverted and which week each year the multi-county pooling effort was projected and conducted. For example, during the 2000 season, the pooling date of September 26 (Week 39) was selected in Week 34 based on EEE activity in the sentinel flock in Bladen County and the emergence of a new brood of Cs. melanura during the same week in Brunswick County (Graph 2). In order to properly interpret the data it should be noted that a seroconversion in a sentinel bird provides information about an earlier brood of Cs. melanura. In the 2000 example, the brood of Cs. melanura that caused the seroconversion in Week 34 emerged in Week 31. The operational information needed when a bird seroconverts are the flocks bleeding schedule and the sample collection date. By overlaying this information onto weekly light trap data, the brood of Cs. melanura most likely involved in EEE transmission to the sentinel bird can be identified.

Table 4. Sentinel Flock Surveillance Supporting the Projected Pooling Dates 2000-2002



Week that Pooling was Projected

Actual Pooling Week


Sentinel Flock EEE Positives by County
















New Hanover, Onslow










2001, 2002














Brunswick, New Hanover, Onslow












Brunswick, Onslow


New Hanover

Statewide Passive Human and Veterinary Surveillance 2000-2002

During the 2000 season, 2 human and 17 horse EEE cases were identified. The 2001 season produced no human cases and 13 horse cases, in 2002, there were 5 horse cases and 1 human case (Graph 7). The summary of reported human and veterinary surveillance for the project counties includes 1 EEE positive horse in 2000 (Week 30) from New Hanover County, 1 EEE positive horse in 2001 (Week 38) from Bladen County and 1 EEE horse positive from Onslow County in 2002 (Week 52). There were no positive human EEE cases reported from any of the project counties during 2000-2002.

Environmental Results 2000-2002

Graph 8 shows the weekly precipitation amounts for 2000, 2001, 2002 and the average rainfall for southeastern North Carolina for the last 8 years. Rainfall for 2000 was 0.42 inches below the norm, 2001 was 17.36 inches below the norm and 2002 season was 4.92 inches below the norm. Operationally speaking, annual rainfall amounts are not nearly as important as which week during the year has significant precipitation.

During this project, the affects of temperature on Cs. melanura populations were observed. While this was difficult to measure, it was apparent that heat indices above 100oF had an impact on the Cs. melanura populations. The possible explanations for this may be that temperatures over 100oF caused significant mortality to the Cs. melanura populations, the high temperatures drove the Cs. melanura population deeper into the swamp habitat reducing the likelihood of collection, and finally, exceedingly high temperatures could be initiating summer aestivation in the existing Cs. melanura population.


Benefits of the Timeline

The EEE timeline gives county programs responsible for operational mosquito control a virus transmission scenario to look for within the data they are collecting. Combining mosquito light trap data with sentinel flock and environmental data to anticipate a potential increase in EEE activity is a proactive initiative and not a reactive response, which is so often the case. The timeline also encourages data sharing and inter-agency communication among city, county, state and federal officials.

Operationally, the timeline shows local programs the mosquito population situation. This can reduce unneeded mosquito spraying and increase the effectiveness of their applications by targeting the correct species at the correct locations at the correct time.

One of the toughest things for a local program to assess is the need for adulticide control measures based on passive arbovirus surveillance results. Typical limitations include inadequate reporting by the medical and veterinary community and the laboratory confirmation process. Utilizing the timeline, local programs can determine which part of the timeline they are dealing with when notified of a EEE positive horse, for example. Additionally, the timeline can be used to support operational control decisions. The timeline will indicate that the mosquito responsible for infecting a horse between days 19 and 41 is already dead by the time the lab results come in, sometimes weeks or even months after the initial infection.

The Centers for Disease Control has suggested that one of the limitations to in an arbovirus situation is that long-term surveillance information about local mosquito populations is rarely available. Encouraging long-term surveillance using the EEE timeline as a regular component of nuisance mosquito operations will streamline control efforts and facilitate a public health response to increased arbovirus activity.

References Cited

Centers for Disease Control 1996. Eastern Equine Encephalities Prevention and Control in the United States Arbovirus Disease Branch DVBID, NCID Fort Collins CO.

Crans WJ and McCuiston 1993. New Jersey’s Approach to Encephalitis Prevention. Bull. Soc. Vector Ecol., 18(1): p45-48.

Crans WJ, Caccamise DF and McNelly JR 1994. Eastern Equine Encephalomyelitis Virus in Relation to the Avian Community of a Coastal Cedar Swamp. J. Med.Ent. Vol. 31, no 5, p711-728.

Dalrymple JM, Young OP, Eldridge BF and Russell PK 1972. Ecology of Arboviruses in a Maryland Freshwater Swamp. J. of Epidemiology Vol. 96, no. 2: p129-140.

Eldridge BF and Edman JD 2000. Medical Entomology: A textbook on Public Health and Veterinary Problems caused by Arthropods Dordrecht, The Netherlands:Kluwer Academic Publishers.

Franklin RP, Kinde H, Jay MT, Kramer LD, Green EN, Chiles RE, Ostlund E, Husted S, Smith J and Parker MD 2002. Eastern Equine Encephalomyelitis Virus Infection in a Horse from California. Emerging Infectious Diseases. Vol. 8 no.3 p283-288.

Harrison BA 2001. Personal Communication. North Carolina Dept. of Environment and Natural Resources. 585 Waughtown Street. Winston-Salem N.C. 27107-2241.

Hays RO and Hess AD 1964. Climatological Conditions Associated with Outbreaks of Eastern Encephalitis. Am. J. Trop. Med. Hyg. Vol. 13: p851-858.

Joseph SR and Bickley WE (University of Maryland Agricultural Experiment Station College Park, Maryland) 1969. Culiseta melanura (Coquillett) on the Eastern Shore of Maryland (Diptera Culicidae). Bulletin A-161, 84 pages.

Komar N, Dohm DJ, Turrell MJ and Spielman A 1999. Eastern Equine Encephalitis in Birds: Relative Competence of European Starlings (Sturnus vulgaris). Am. J. Trop. Med. Hyg., 60(3), p387-391.

Moore CG, McLean RG, Mitchell CJ, Nasci RS, Tsai TF, Calisher CH, Marfin AA, Moore PS, and Gubler DJ 1993. Guidelines for Arbovirus Surveillance Programs in the United States. U.S. Department of Health and Human Services

Morris CD, Zimmerman RH and Edman JD 1980. Epizootiology of Eastern Equine Encephalomyelitis Virus in Upstate New York USA. J. Med. Ent. Vol. 17, no. 5: p 453-465.

Morris CD, Corey ME, Emord DE, Howard JJ 1980. Epizootiology of Eastern Equine Encephalomyelitis Virus in Upstate New York USA. J. Med. Ent. Vol. 17, no. 5: p 442-452.

Morris CD 1988. The Arboviruses: Epidemiology and Ecology Vol. III. Monath T P ed., Boca Raton Florida: CRC Press. p 1-20.

Morris CD 1992. Eastern Equine Encephalitis. J. Florida Mosquito Control Ass. 63(1): p 23-34.

Muul I, Johnson BK and Harrison BA 1975. Ecological Studies of Culiseta melanura (Diptera: Culicidae) in Relation to Eastern and Western Equine Encephalomyelitis Viruses on the Eastern Shore of Maryland. J. Med. Ent. Vol. 11, no. 6: p739-748.

National Weather Service. Climactic Data. Wilmington Weather Forecast Office. 2015 Gardner Drive, Wilmington NC 28405.

Saugstad ES, Dalrymple JM and Eldridge BF 1972. Ecology of Arboviruses in a Maryland Freshwater Swamp. J. of Epidemiology Vol. 96, no. 2: p114-122.

Scott TW and Weaver SC 1988. Eastern Equine Encephalomyelitis Virus: Epidemiology and Evolution of Mosquito Transmission. In: Advances in Virus Research, Vol. 37. Department of Entomology and Agricultural Biotechnology Center, University of Maryland College Park, Maryland 20742. P277-328.

Silverly RE and Shoof HF 1962. Biology of Culiseta melanura (Coquillett) in SouthEast Georgia. Mosquito NewsVol. 22, No. 3 p274-282.

Williams JE, Young OP and Watts DM 1974. Relationship of Density of Culiseta melanura Mosquitoes to infection of Wild Birds with Eastern Equine Encephalitis Viruses. J. Med. Ent. Vol. 11, no. 3: p.352-354.



179 MARCH 9, 1764 DRIVE

(910) 253-2515
open mon-fri:
8:00 a.m. – 4:30 p.m.

News & Events

Get the latest updates and happenings on events throughout the county.

Latest News

Read More Mosquito Control News

Latest Events

See Event Calendar