Twenty Five and a Hundred: What Manitoba's Mosquito Programme Shows About Deciding to Treat

1. Introduction: a threshold written down

The economics article published elsewhere in this journal reported a finding from nine elementary schools: monitoring revealed that most of the conventional, calendar-based pesticide treatments being delivered were unnecessary. The problem was not the product. It was that nobody had defined a condition under which treatment was warranted.

Manitoba's mosquito programme is the local counter-example, and it is unusually explicit.

1.1 Why this is worth an article

Because almost nothing else in pest management states its action threshold in public, and the contrast with how structural contracts are written is instructive.

2. Two mosquitoes

Public understanding of mosquito risk is hampered by a biological fact that is rarely explained clearly.

While there are dozens of mosquito species in southern Manitoba, the main variety the city tries to control is Aedes vexans, unofficially known as the summer nuisance mosquito. This is the one most likely to give a nasty, itchy bite. Aedes vexans is relatively harmless, as it is not known to transmit illness to people very often. Another species, Culex tarsalis, is a stealthier biter, known to transmit West Nile virus to people and horses.⁵

2.1 The communication problem this creates

Residents assess mosquito risk by how much they are being bitten. That is a reasonable heuristic and it tracks the wrong species.

A summer with heavy Aedes pressure feels dangerous and is mostly uncomfortable. A dry summer with few nuisance bites can coincide with elevated Culex tarsalis activity, which is the condition that matters for West Nile virus.

The two programmes are correspondingly separate. The City works with the Province on a targeted larviciding programme focused on Culex tarsalis and Culex restuans, while a West Nile adulticiding programme would only occur under a provincial health order.²

3. The nuisance programme and its numbers

The thresholds quoted in §1 deserve unpacking because their structure is deliberate.

The city-wide count of 25 establishes that the problem is general rather than local. The quadrant count of 100 establishes that somewhere it is severe. Requiring both prevents city-wide adulticiding in response to one bad neighbourhood, and prevents it in response to a mild but uniform elevation.⁵

The two-consecutive-day requirement filters out single-night anomalies from weather or trap placement.

3.1 The scale of the alternative

Before adulticiding is considered, the city runs a larval programme. Winnipeg's insect control branch operates four larviciding helicopters.⁵ Crews work to avoid a fogging programme rather than to trigger one.⁵

In one reported instance, relatively cool spring temperatures allowed the branch to limit development of adult mosquitoes to the point where the city-wide average trap count for adult female nuisance mosquitoes was only nine.⁵ That is well below the threshold of 25, and the intervention that achieved it was larval rather than adult.

3.2 The product caution

The reticence about adulticiding is explained in the reporting. Deltamethrin has broad-insecticidal-spectrum properties, meaning it is not just a mosquito killer and will kill other insects, which is why the city applies it in the precise dose and concentration prescribed by its label.⁵

An agency that states publicly that its adulticide kills non-target insects, and treats that as a reason for restraint, is being more candid about trade-offs than most of this industry manages.

4. Why a threshold matters

The general principle here is the one this journal has returned to repeatedly.

A threshold converts the question from how often should we treat into under what measured condition is treatment warranted. The first question has no correct answer and defaults to a schedule. The second has an answer that can be checked.

4.1 The three things a threshold requires

A measurement that is consistent enough to compare across time and place. The trap catch definition in §6 exists for this reason.

A stated number agreed before the situation arises, which removes the pressure to adjust the standard to the circumstances.

A willingness to do nothing when the number is not met, which is the part that is difficult when the public is uncomfortable and the equipment is available.

4.2 What it protects against

Thresholds protect in both directions. They prevent treatment that is not warranted, which is the unnecessary-treatment finding from the schools study. They also commit the agency to act when the condition is met, which removes discretion in the other direction.

5. The provincial surveillance network

Underneath the thresholds sits a measurement system.

Permanent mosquito traps are set up in 21 community areas in southern Manitoba to monitor Culex tarsalis, the mosquitoes most likely to transmit West Nile virus to humans in the province.¹

5.1 Method

Normally two Centers for Disease Control mosquito traps baited with carbon dioxide are run for two nights each week in each Manitoba community, except Winnipeg and Brandon where proportionately more traps are used because of larger community size.¹

Carbon dioxide baiting matters because it approximates the cue a host provides, which is what the insect is actually responding to.

5.2 The network has contracted

An earlier season reported permanent traps in 37 community areas, against 21 now.⁶¹

We note the change without knowing the reason, which is not given in the sources we located. Whether it reflects programme refinement, resource constraint or both is a reasonable question for anyone assessing the system's sensitivity.

6. What the metric measures

The definition is published and precise, which is what makes the data usable.

A trap catch, also called a collection, represents the mosquitoes collected by one trap during one night from approximately one hour before sunset until approximately one hour after sunrise. The average number of Culex tarsalis per trap catch is calculated by dividing the total number collected for the whole week in a given health region by the number of collections.¹

6.1 What it is for

The published framing is careful. The measure provides a relative indication of activity levels between regions and over time, and indicates whether populations are rising, falling or staying steady.¹

It is explicitly not a population estimate. Counts in specific locations within a health region will vary depending on local conditions, temperature and precipitation.¹

6.2 Why that restraint is the right call

This is the same distinction the urban rodent article published in this journal argued for: activity indices measured consistently are defensible, while converting them into absolute population figures is not.

The province publishes an index and describes it as an index. That is a more honest treatment of surveillance data than most of the sector manages.

6.3 Current season figures

In week 30 of 2026, adult mosquito surveillance detected C. tarsalis in all 21 trapping communities. Mosquitoes per trap decreased compared with the previous week for the Interlake, Southern and Winnipeg health regions but increased in Prairie Mountain, and the provincial average was significantly higher than the historical average for that week. Ninety five West Nile virus positive C. tarsalis pools were detected in that week alone.⁷

7. The West Nile trigger

Nuisance fogging and West Nile fogging are separate decisions with separate authority.

The Manitoba government issues fogging orders under the Environment Act, which requires municipalities to fog when the risk of West Nile virus transmission via Culex tarsalis mosquitoes is high.⁴

Surveillance data showing high numbers of adult Culex tarsalis and evidence of infection with West Nile virus is the stated basis for such an order.⁴

7.1 The notification requirement

Areas to be fogged on any given night are announced to the public at least eight hours in advance, and Winnipeggers can register online for notifications and receive an automated phone or email alert.⁴

This is a small detail with a large implication. Residents with chemical sensitivity, beekeepers, and anyone wishing to close windows are given a mechanism to act, which treats the application as something the public has a legitimate interest in rather than as a municipal operation conducted around them.

7.2 The stakes

A single bite from an infected mosquito can transmit West Nile virus to humans, and infection can cause severe illness resulting in encephalitis, long-term complications and death.⁴

8. Temperature and the compression of risk

One mechanism explains why the risk window is narrow and why hot summers matter disproportionately.

High temperatures across southern Manitoba shorten the time required for West Nile virus to develop within infected Culex tarsalis mosquitoes and be transmitted to humans.⁴

8.1 Why this is not simply more mosquitoes

An infected mosquito is not immediately infectious. The virus must replicate and reach the salivary glands before transmission is possible, and the duration of that process depends on temperature.

Warm conditions therefore raise risk through two independent routes: more mosquitoes, and a larger proportion of infected ones becoming infectious within their lifespan. The second effect is nonlinear and is the reason a hot spell produces a disproportionate response.

8.2 The risk window

Manitoba identifies the highest-risk period for human exposure as occurring between 15 July and 15 August each year.⁴

A defined month is a useful thing for the public to hold, and it corresponds to the convergence of accumulated seasonal population growth with the temperatures that compress the incubation period.

9. Larviciding and the preference for the larval stage

The programme's structure reflects a consistent preference for intervening before adults emerge.

The urgency is straightforward: mosquito larvae, which do not bother people at all while they remain in stagnant water, develop into biting adults far more quickly when weather gets very warm.⁵

The province offers a cost-shared larviciding programme for municipalities and communities to help control Culex tarsalis, with 53 communities participating in one reported year.³

9.1 Adaptive targeting

One example demonstrates surveillance feeding back into where control is applied. In a drought year, many sites where mosquitoes would normally lay eggs dried up, leaving mosquitoes searching for water. Some found this scarce resource in unusual locations such as smaller rivers and creeks which, though normally unsuitable, had stopped flowing and thereby created ideal sites for Culex tarsalis. Larval control was consequently conducted along these unusual development sites.⁸

A programme that relocates its larval control in response to hydrological conditions is doing something structural pest management rarely does, which is changing where it works rather than only how often.

10. The private property share

One figure reframes the whole programme.

Fifty per cent of Winnipeg's mosquito population comes from standing water on private property.²

The city launched an awareness campaign in 2016 encouraging homeowners to fight mosquitoes before they hatch by draining, dumping or covering standing water in their yards.²

Provincial guidance is specific: clean eavestroughs, regularly empty bird baths, old tyres and other items that collect water, ensure rain barrels are covered with mosquito screening or tightly sealed around the downspout, and improve landscaping to prevent standing water around the home.³

10.1 The parallel to structural work

This is the same structure as the exclusion and sanitation argument running through this journal. The decisive intervention is habitat removal by the occupant, and the professional programme operates on the remainder.

It is also the same problem: the measure with the highest return is the one nobody is paid to perform.

11. The corvid programme and its retirement

A piece of programme history worth recording because retirement of a surveillance method is unusual.

Birds from the crow family are very susceptible to West Nile virus and many die soon after becoming infected, so testing dead crows and other corvids provided an early indication of the presence of the virus.⁹

By 2008 the dead corvid pick up programme was no longer in effect. Dead corvids were no longer needed as an early indicator of West Nile virus in Manitoba or to determine geographic distribution. Larval sampling, mosquito sampling and testing, and other factors including temperature, would be used to guide the assessment of risk.⁶

11.1 Why this is a good sign

The corvid programme was the early-warning instrument when the virus was newly arrived and its distribution unknown. Once the mosquito trapping network could answer those questions directly, the bird programme was redundant and was stopped.

Retiring a surveillance method once a better one exists is harder institutionally than it sounds, because programmes acquire constituencies. Doing it is evidence the system is oriented toward the question rather than toward its own activity.

12. What structural pest control could take from this

Four transferable points.

Publish the threshold. A contract stating the monitoring result at which treatment occurs is a different product from one stating a visit frequency, and the mosquito programme demonstrates the form.⁵

Call an index an index. The province is explicit that trap averages indicate direction and relative level rather than population size.¹ Structural monitoring data deserves the same restraint.

State non-target effects plainly. The city says publicly that its adulticide kills insects other than mosquitoes and treats that as a reason for caution.⁵

Treat the occupant as part of the system. Half the larval habitat is on private property.² The equivalent in buildings is harbourage, moisture and food, and it is equally decisive and equally outside the contractor's control.

13. Limitations and open questions

We have not evaluated whether the programme works. This paper describes an architecture. Whether Manitoba's thresholds are set at the right levels, and whether the interventions reduce human cases, are epidemiological questions we have not attempted and could not answer from these sources.

Several figures come from news reporting. The fogging thresholds, helicopter count and deltamethrin characterisation are from media coverage quoting city officials.⁵ We have not located them in a municipal technical document.

The trap network contraction is unexplained. We report 37 community areas previously against 21 now¹⁶ without knowing why.

Figures are season-specific. The week 30 data and the 53-community larviciding participation are point-in-time values³⁷ and readers should consult the provincial site for current status.

We provide mosquito control, but not as part of this programme. This is an outside examination of a public programme, which limits our operational insight into it. It does not remove our commercial interest: we sell mosquito control to private clients, and public surveillance findings and fogging decisions shape demand for it. An earlier version of this paragraph stated that we did not perform mosquito control, which was wrong.

14. Conclusion

Winnipeg will not fog for nuisance mosquitoes until two consecutive days show a city-wide trap count of 25 or more and at least one quadrant at 100 or more.⁵ Underneath that sits a provincial network of permanent traps in 21 community areas running carbon dioxide baited CDC traps two nights weekly, producing a published weekly average that the province describes accurately as a relative index rather than a population estimate.¹ West Nile fogging is a separate decision, ordered under the Environment Act when risk is high, with eight hours of public notice and an alert system.⁴

Two elements deserve wider attention. The species that bites people is not the species that carries the virus,⁵ so public intuition about risk tracks the wrong insect. And half of Winnipeg's mosquito population originates on private property,² which means the municipal programme can address at most half the habitat however well it is run.

For the rest of pest management the lesson is narrower and, we think, important. This is a programme that measures first, publishes the numbers, states in advance what condition justifies intervention, admits that its adulticide harms non-target insects, redirects its larval control when hydrology changes, and retired a surveillance method once a better one existed.⁵⁶⁸

Structural pest control operates almost entirely on calendars. The apparatus for doing it otherwise is running in this province every summer, in public, with the numbers posted weekly.

References

1. Surveillance Data for West Nile virus in Manitoba. Province of Manitoba, Health. Source for permanent mosquito traps in 21 community areas in southern Manitoba monitoring Culex tarsalis, the use of two CDC traps baited with carbon dioxide run two nights weekly per community with proportionately more in Winnipeg and Brandon, the definition of a trap catch as one trap for one night from approximately one hour before sunset to one hour after sunrise, the calculation of the weekly regional average, and the explicit framing of the measure as a relative indication of activity levels rather than an absolute population estimate. https://www.gov.mb.ca/health/wnv/stats.html
2. Insect Control frequently asked questions. City of Winnipeg, Public Works. Source for the statement that 50 per cent of Winnipeg's mosquito population comes from standing water on private property, the 2016 Agents of SWAT awareness campaign, the collaborative surveillance and control arrangement with Manitoba Health, the targeted larviciding programme focused on Culex tarsalis and Culex restuans, and the condition that a West Nile adulticiding programme would only occur under a provincial health order. https://legacy.winnipeg.ca/publicworks/insectcontrol/insect/faq.stm
3. West Nile Virus Update. Province of Manitoba news release. Source for the cost-shared provincial larviciding programme with 53 participating communities in the reported year, the specific homeowner guidance on eavestroughs, bird baths, tyres, rain barrel screening and landscaping, and the reporting of the season's first West Nile virus positive Culex tarsalis samples. https://news.gov.mb.ca/news/print,index.html?item=38831&posted=2016-07-29
4. Manitoba orders fogging in several communities as West Nile-carrying mosquito count soars. CBC News. Source for the issuance of fogging orders under the Environment Act requiring municipalities to fog when West Nile transmission risk via Culex tarsalis is high, the role of high temperatures in shortening the time required for the virus to develop within infected mosquitoes and be transmitted, the identification of 15 July to 15 August as the highest-risk period for human exposure, the eight hour advance public notification with online registration for automated alerts, and the statement that a single bite from an infected mosquito can transmit the virus with infection capable of causing encephalitis, long-term complications and death. https://www.cbc.ca/news/canada/manitoba/west-nile-mosquito-fogging-9.7291929
5. Winnipeg larviciding and mosquito control reporting. CBC News. Source for the fogging thresholds of two straight days at a city-wide trap count of 25 or more with at least one quadrant at 100 or more, the four larviciding helicopters operated by the Insect Control Branch, the reported city-wide average count of nine adult female nuisance mosquitoes in a cool spring, the characterisation of deltamethrin as having broad-insecticidal-spectrum properties requiring precise label dosing, the distinction between Aedes vexans as the summer nuisance mosquito and Culex tarsalis as the stealthier West Nile vector affecting people and horses, and the observation that larvae develop into biting adults far more quickly in warm weather. https://amp.cbc.ca/news/canada/manitoba/winnipeg-larvicding-mosquitoes-1.6488972
6. 2008 Surveillance for West Nile virus in Manitoba. Province of Manitoba, Health. Source for the earlier network of permanent mosquito traps in 37 community areas, and for the discontinuation of the dead corvid pick up programme on the basis that dead corvids were no longer needed as an early indicator or to determine geographic distribution, with larval sampling, mosquito sampling and testing and other factors including temperature used instead to guide risk assessment. https://www.gov.mb.ca/health/wnv/stats2008.html
7. West Nile virus Weekly Activity Reports. Province of Manitoba, Health. Source for week 30 of 2026 surveillance detecting C. tarsalis in all 21 trapping communities, the regional direction of change against the previous week, the provincial average being significantly higher than the historical average for that week, and the detection of 95 West Nile virus positive C. tarsalis mosquito pools. https://www.gov.mb.ca/health/wnv/update.html
8. West Nile Virus Bulletin. Province of Manitoba news release. Source for the drought season account in which normal oviposition sites dried up, mosquitoes sought scarce water in unusual locations including smaller rivers and creeks that had stopped flowing, and larval control was consequently redirected to those unusual development sites. https://news.gov.mb.ca/news/index.html?item=51741
9. 2004 Surveillance for West Nile virus in Manitoba. Province of Manitoba, Health. Source for the susceptibility of corvids to West Nile virus, their death soon after infection, and the use of dead corvid testing as an early indication of viral presence, together with the earlier trapping protocol of two or three carbon dioxide baited CDC traps run one or two nights weekly per community. https://www.manitoba.ca/health/wnv/stats2004.html