Four Methods, One That Works: What Wasp Control Actually Delivers

1. Introduction: four things a contractor can do

Trap queens in spring, trap workers through the season, put out toxic protein bait, or find the nest and destroy it. Those are the options, and they have been assessed.

1.1 That is extension guidance about all three at once

Written for beekeepers rather than for our trade, which may be why it is blunter than trade material on the same question.¹

1.2 What this article argues

That one of the four methods does reduce populations, that it is unavailable when people want it, and that the other three deliver something real which is not what is usually claimed for them. Sections 10, 16 and 24 are the case.

2. What this article assumes

And does not re-argue.

Our article on the late summer shift set out the colony's internal food economy: workers hunt protein they cannot digest, larvae convert it and return a sugary secretion, and when the queen stops laying and the last larvae pupate that supply ends and the workers go looking for carbohydrate where people keep it.

2.1 The conclusion there was about explanation

That the nuisance peak is a symptom of the colony's decline rather than its growth.

2.2 This article takes that as given and asks what follows for method

Which turns out to be more awkward than the phenology itself.

3. The seasonal constraint in one paragraph

Because every method below is governed by it.

A bait is only a bait if it is taken. An insect provisioning larvae takes protein. An insect feeding itself takes sugar. The same nest therefore requires different bait in July and in September, and §15 explains why only one of those two is usable.

4. Queen trapping

The intervention that should work best on any intuitive account.

One queen founds one colony, and a colony reaches thousands of workers. Removing a queen in May should be worth thousands of wasps in September, at a leverage no other method approaches.

4.1 And queens are catchable

In spring, queens and low numbers of workers were captured in lure-baited traps until early July, when the number of foragers increased dramatically.⁴ Queens and workers alike are attracted to the standard synthetic lure.³

4.2 So the method is not failing for want of catching anything

Which is what makes the result in §5 informative.

5. Which does not work

Reported flatly by extension guidance.

A study in California found that traps containing heptyl butyrate effectively attract queen and worker yellowjackets. However, in this study, the number of trapped queens did not reduce the number of yellowjacket workers trapped in late summer.¹

5.1 And the finding has been restated since

More recent literature suggests that early trapping, specifically of queens, does not reduce overall numbers of foraging workers, but that trapping throughout the season does reduce interactions between yellowjackets and people.³

5.2 With a residual use

Early season queen trapping may be used as a monitoring tool by residents and pest control operators, and traps are effective tools for monitoring activity in an area.³

6. Why it does not work

Our explanation, offered because the sources report the result without one.

Founding queens die at very high rates for reasons unrelated to trapping. The number of colonies that establish is set by how many queens survive the founding period rather than by how many emerge from overwintering, and removing a fraction of a pool in which most individuals were going to fail changes the total very little.

6.1 Which is compensatory mortality

The same concept our wildlife article used to explain why removing animals from a site does not reduce the number of animals at that site. Added mortality substitutes for existing mortality rather than adding to it, wherever the existing mortality is high.

6.2 And it is a limit on attempts rather than on success

Trapping reduces the number of founding attempts. Weather acts on the probability that an attempt succeeds. A method that reduces attempts has little effect when the outcome is governed by a success rate rather than by the number of tries.

That formulation is ours and is the same structure as §6.1 seen from the other side.

7. What sets the number instead

Which both sources identify.

Weather in the spring is the most important factor in colony establishment.⁷ If spring weather is rainy or cold, queens may not be able to establish a nest or forage for food for larvae; if spring is warm and dry, queens may be successful and produce many offspring that grow into large populations in late summer.¹

7.0b Which is a weather effect acting in May on an outcome visible in September

The gap between cause and symptom is four months, and nothing observable in the interval indicates which kind of year is underway until the foragers appear. A contractor asked in August why this year is bad is being asked about a process that finished before the season's first service visit.

That framing is ours, and it is the reason §27 treats the weather term as dominating rather than merely contributing.

7.1 So the limiting factor is not queen supply

Which means an intervention acting on queen supply has nothing to act on.

8. Worker trapping

The method most householders actually buy.

Traps are readily available in garden centres and hardware stores, and many designs are containers with small entrances to minimise exit, some using an attractant suspended over soapy water to drown the wasps.¹

9. Which does something real

And we want to say so before saying what it does not do.

Trapping throughout the season does reduce interactions between yellowjackets and people.³

9.0b And the benefit is immediate rather than cumulative

Unlike a population effect, which would take a season to appear and a season to verify, a diversion effect operates on the afternoon the trap is hung. Whatever else is true of the method, the customer can tell within a day whether it is doing the thing it does.

9.1 A trap near a deck removes the wasps that arrive at the deck

Which is a genuine and immediate benefit at the place a person occupies.

10. But not the thing implied

It does not reduce the number of wasps in the neighbourhood.

Extension guidance places worker trapping in the same list as poison baiting and nest destruction as unlikely to reduce the overall population.¹

10.1 And the two are not the same purchase

Our position, and it applies to traps sold in hardware stores as much as to anything a contractor places.

11. A distinction this journal keeps finding

Between reducing a population and reducing contact.

Our articles on rodent monitoring, on exclusion and on wildlife removal each arrived at a version of it. Here it is unusually clean, because one source reports both halves in consecutive clauses: queen trapping does not reduce workers, season-long trapping reduces interactions.³

11.0b And the two can be measured with the same instrument

A trap catch is evidence of both and distinguishes neither. Wasps in the trap near the deck are simultaneously wasps removed from the deck and wasps that were present in the area, and a falling catch is consistent with a shrinking population, a successful diversion, or a change in what the animals are foraging for.

Which is why the California result matters as much as it does: it compared queens caught in one period against workers caught in another, and found the first did not predict the second.¹

11.1 And contact reduction is usually what the customer wanted

Nobody experiences a neighbourhood wasp population. They experience wasps at the place they are sitting.

11.2 So the method is sound and the claim attached to it often is not

Which is a different criticism from saying it does not work, and a fairer one.

12. Protein baiting

The method that does reduce populations.

A university urban entomology text sets out the window: when the workers appear, they search for proteinaceous food to feed the developing larvae in the nest. There follows a long period when a protein bait can be used that attracts yellowjackets and practically no other insects.²

12.1 And the field record supports it

Trials used canned chicken mixed with fipronil to a final concentration of 0.025 per cent in an urban recreational setting.⁴ Earlier work used a different toxicant in canned chicken with acceptance enhanced by the synthetic lure, and was in many instances effective in drastically reducing populations.³

13. Its unusual selectivity

Which deserves noting in a journal that has spent several articles on nontarget exposure.

The same text states that when protein foods are being scavenged, the baiting method is safe, convenient, adds practically no insecticide to the environment, and is harmless to all species but the two target ones.²

13.1 That is a strong claim and we report it as the source's

Harmless to all species but two is a categorical statement, and the text does not present the testing behind it.²

14. Which is the same fact as its seasonality

Our observation, and it is the neatest thing in this material.

Protein bait is selective because no other insect in the vicinity is scavenging protein. That is true only while these wasps are provisioning larvae.

14.1 So the property that makes it safe gives it a season

The two are not separate features of the method to be traded off. They are one feature described twice.

14.1b Which is unusual among selective methods

Selectivity in pesticides normally comes from the chemistry: a compound that binds an invertebrate target and not a vertebrate one, or a formulation placed where only the target goes. Here it comes from the behaviour of the target at a particular moment in its life cycle.²

A chemical basis for selectivity travels with the product. A behavioural basis does not, because the behaviour is a property of the season. That comparison is ours and it explains why this method cannot simply be improved into a year-round one.

14.2 And nothing can be done about that

A more persistent formulation or a better lure does not extend the window, because the window is a property of the insect rather than of the product.

15. And why carbohydrate bait was not pursued

The obvious response to §14, ruled out for a reason that has not gone away.

Carbohydrate baits were not tested, since they might have the disadvantage of attracting honey bees and other nontarget insects.²

15.1 Repeated in extension guidance on trap construction

Advising against sugary attractants because they could inadvertently attract bees.¹

15.2 So the method that would work in September is the one that cannot be used

The wasps are taking sugar, and sugar is what a pollinator takes too.

16. So the effective method expires as demand opens

Which is the central practical problem in this species' management.

The protein foraging period is the long quiet stretch when nobody is complaining. The sugar foraging period is when the calls come, and it is precisely when protein bait stops being taken.

16.1 The field trials show the same timing from the other side

In two consecutive years, sufficient numbers of foragers were not captured until early August when baiting trials were initiated.⁴

16.2 And it explains a commercial pattern

The method with the best evidence behind it is not a product anybody sells to householders, because a householder in June has no complaint and will not buy a treatment for a problem they have not yet noticed. That inference is ours.

17. Species differ in what they take

Which complicates the bait choice more than the literature's confidence suggests.

In British Columbia one species responded preferentially to rotisserie chicken over canned chicken or canned sardines, while the synthetic lure was more attractive than rotisserie chicken; and counterintuitively, combining rotisserie chicken with the synthetic lure reduced catches. A second species showed a different preference.⁴

17.0b Which matters more for a contractor than for a park district

A programme covering a large public space can afford to deploy several bait types and let the wasps sort out which they prefer. A single property visit cannot, so the species composition of the local population determines whether one bait choice is right or wrong, and §30.1 is the reason that choice cannot currently be made on evidence here.

17.1 The combination result is the interesting one

Two attractants that each work, used together, performing worse than either. Our sources report it without explaining it and we will not speculate beyond noting that it rules out treating lures as additive.⁴

18. The transfer demonstration

A piece of method worth recording, because it shows the mechanism our horizontal transfer article described operating in a vespid colony.

This transfer was substantiated by placing a wire-mesh cage soaked with 0.05% fipronil over a nest entrance for 15 min. The few caged exiting wasps were released and allowed to return to the colony, whereupon the colony was eliminated within 1 day.³

18.1 A small number of contaminated individuals

Enough to eliminate a colony of thousands within a day.³

18.1b And it clarifies what the bait is actually for

A bait is often described as a delivery method that gets the compound to animals the operator cannot reach. In a colony that provisions larvae, it is more than that: the forager is not carrying the material for itself, so the quantity that moves into the nest is set by the colony's demand rather than by the individual's appetite.

Which is why the caged-wasp demonstration worked from contact alone, without any bait at all, and why §16's timing constraint is about the direction of food flow rather than about palatability.³

18.2 Which is why baiting outperforms spraying here

Delivery to the colony by its own members rather than by the operator, and it works precisely because the foragers are provisioning others rather than feeding themselves.

19. Nest destruction

The method everyone assumes is primary.

Identified colonies may be eliminated by directing toxicant into the nest opening.³ Nest sites are underground cavities, holes or openings in structures, or aerial locations.¹

20. Most nests cannot be found

Which is the constraint that governs the whole method.

However, because they are well concealed it is nearly impossible to locate most yellowjacket nests. Except for incidental nests, individual nest treatment is not an option.³

20.1 Incidental is the operative word

The nests that get treated are the ones that happened to be noticed.³

21. Which biases what gets treated

Our observation, following directly from §20.

Findability is a property of location rather than of size, activity or contribution to the local forager population.

21.1 So the treated sample is not the important sample

A large ground nest in a neighbour's hedge contributes foragers to the same deck as a small aerial nest under the customer's soffit, and only the second one gets destroyed.

21.1b And the selection is not random with respect to risk either

The nests people find are the ones near the places people are, which is also where stings happen. So the findable sample is biased toward the nests that matter for immediate risk and away from the nests that matter for abundance.

Which means nest destruction is better targeted for the purpose §25 describes than for the purpose §24 rules out, and that is a defence of the method rather than a criticism. That reading is ours.

21.2 Which is a sampling problem disguised as a treatment

And it explains part of why nest destruction appears in the list of methods unlikely to reduce the overall population.¹

22. The catchment problem

With numbers attached.

Microsatellite work identified eighteen colonies actively foraging within one park, with foragers from eleven different colonies collected at a single trap site.⁴

22.1 A deck is a trap site in every respect but intent

The wasps arriving at it are drawn from a catchment containing many nests, most of them on land the contractor has no access to and could not search if he did.

22.2 Which sets a ceiling on what nest destruction can achieve

Removing one nest of eleven contributing to a location removes a fraction of the foragers at that location, and the other ten are unaffected. That arithmetic is ours.

23. Late season removal

Which is what a contractor is most often asked to do.

The timing of the call is fixed by the seasonal shift: people notice wasps when the wasps come to them, which is after the colony has stopped growing.

23.1 And by then the reproductive output has left

Males and new queens are produced in August and September, mate, and the inseminated queens enter diapause to overwinter.⁵ Virgin queens search for overwintering sites at the time the colonies are rapidly declining, and only the newly mated queens survive the winter.²

24. What it cannot do

Stated as directly as we can, given that it is our own service.

Destroying a nest in September cannot reduce next year's wasp population at that property, because the individuals that would found next year's colonies have already been produced and have already dispersed.

24.1 And it cannot much shorten the colony's life

Every member except the mated new queens dies at the first hard frost.⁶ Removal in mid-September anticipates by a few weeks what the season does for nothing.

24.2 Nor does it prevent reoccupation of the site

Because the nest would not have been reused anyway.⁷

24.2b And the removal is not free of risk either

A concealed nest disturbed in the most defensive phase of its cycle exposes the operator to the hazard the customer is paying to have removed. Our occupational article treated physical and exposure hazards in this trade generally; this is a case where the hazard falls almost entirely on the person doing the work rather than on the occupant.

24.3 So three of the four things a customer may believe they are buying are unavailable

Fewer wasps next year, a shorter remaining season, and a site that will not be reoccupied. That enumeration is ours.

25. What it can do

Which is not nothing, and deserves the same directness.

It removes, immediately, a concentrated source of stinging insects at a specific location where people are. For a nest in a doorframe, a play structure, or a wall void beside a bedroom window, that is the whole of the value and it is a real value.

25.1 Especially where somebody is sensitised

Our sting risk article concluded that the population at genuine risk is smaller than public anxiety suggests and that for those people the risk is serious.

25.2 And weeks matter when the exposure is daily

Three or four weeks of a nest beside a back door is a large number of encounters.

25.3 So the honest sale is immediate risk removal at one place

Not population control, and we think the trade generally allows the second to be inferred without saying it.

26. Spring weather

The variable that turns out to matter most.

Trapping across sites found a nearly fortyfold annual variation in the total number captured at some sites, with the ultimate number of foragers apparently related to spring weather episodes.³

26.1 Fortyfold

Between a bad year and a good one, at the same site, with the same method.³

27. Which dwarfs the control decision

Our reading, and it should temper any claim made about a season's outcome.

No structural intervention available to a contractor produces a fortyfold change in local abundance. How bad a wasp year is was settled by weather before anybody was asked to do anything.

27.1 With the usual attribution consequence

A programme run in a declining year will appear to work and a programme run in a rising year will appear to fail, and neither appearance is informative.

27.2 Which is stronger here than in most of our attribution articles

Because the background variation is larger than in any other system this journal has covered.

28. Winter as a control measure

An observation specific to our climate.

The annual cycle is not universal. In warmer climates nests overwinter, become polygynous and persist for more than one season, and one study found three perennial colonies, 2.4 per cent of all colonies, with tenfold higher peak traffic rates than annual ones.⁴ Polygyny is described as a likely prerequisite for successful overwintering.³

28.1 Which makes the perennial colonies the informative case

They show what the species does when the annual reset is removed, and the constraint on colony size is therefore the calendar rather than anything intrinsic to the animal.

28.1b Which also bounds what our climate guarantees

The annual reset is not a control measure anybody applies and cannot be credited to any programme, but it does mean that every failure here is a failure within one season. A method that achieves nothing still leaves the site clear by November, and any winter comparison between a treated and an untreated property will show both at zero.

That is a measurement warning as much as a climate observation, and it is ours.

28.2 None of which is available here

Our winters end every colony annually and reset the population every year regardless of what anyone does.

29. Our own position

The disclosure, and it is a substantial one.

We remove wasp nests, it is a seasonal revenue line, and §24 says that three of the four benefits a customer might reasonably assume are not delivered by it.

29.1 We think §25 is the defensible sale

Immediate removal of a concentrated hazard at a specific place, priced and described as that.

29.2 And it argues for a service we do not currently offer

Protein baiting during the summer window at properties with a recurring problem, which is the selective and effective method and which nobody asks for because nobody is being bothered yet.²

29.3 With the commercial problem stated honestly

It would have to be sold in June against a complaint that arrives in September, which is a hard sale and is probably why it is not offered anywhere we could find.

30. The Manitoba position

Where the gaps are large.

30.1 What we could not find

Any Manitoba or Winnipeg survey of which vespid species are present and in what proportions, any local phenology data giving the dates of the foraging switch in this climate, and any provincial guidance on timing of wasp management.

30.2 Which matters because the bait choice depends on species

Section 17 shows species differing in which protein they take.⁴

30.2b And one thing a contractor here could establish cheaply

Which species are being removed. Every nest removal is an opportunity to identify what was in it, and a few seasons of that would produce the composition data §30.1 says does not exist, at no cost beyond recording it.

30.3 And because the dates are climate-dependent

The field evidence throughout comes from regions with a very different frost calendar, and we would not transfer the timing without checking.

31. Limitations and open questions

Most field data comes from one region and one dominant species. California, Hawaii and the Pacific Northwest, with the trapping study reporting one species at more than 85 to 90 per cent of the total at every test site.³ Whether the same species dominates here is what §30.1 says we could not establish.

The primary papers were read as abstracts or as a conference paper. The trapping and protein bait work and the fipronil bait trials reach us through abstracts, extracts and one full conference text.³⁴

Two sources are weak and are flagged in place. One encyclopedia article and one trade marketing page, used for chronology and for the frost statement rather than for any finding.⁶⁷

We have no cost data on protein baiting. Whether §29.2 is commercially viable at a single property rather than across a park is not something our sources address, and §29.3 is speculation about why it is not offered.

The selectivity claim in §13 is reported, not verified. Harmless to all species but two is categorical, the text does not present the testing behind it, and we would want to see that before repeating it to a client.²

That is the most important gap here because §§14 and 29.2 recommend a method on the strength of a selectivity claim we have not been able to check.

Sections 6, 10.1, 11, 14, 16.2, 17.1, 21, 22.1, 22.2, 24, 25.3, 27 and 28.1 are our reasoning. The compensatory mortality explanation, the selectivity-equals-seasonality observation, the sampling argument about findable nests, the catchment arithmetic and the enumeration of what late-season removal cannot deliver are ours rather than sourced positions.

32. Conclusion

Of the four methods available against this species, extension guidance places three of them in one sentence as unlikely to reduce the overall population: trapping workers, poison baiting to kill a colony, and destroying nests with insecticide.¹ Trapping queens in spring, which on the arithmetic should be the highest-leverage intervention available, did not reduce the number of workers trapped in late summer, and the reason is that queen supply is not what limits the number of colonies. Spring weather is, and the variation it produces approaches fortyfold at the same site between years.¹³ What season-long trapping does do is reduce interactions between wasps and people.³ That is a real benefit and it is not the benefit usually claimed.

The method that does suppress populations is toxic protein bait, which is described as attracting practically no other insect and as harmless to all species but the two targets.² Its selectivity and its seasonality are the same fact: nothing else is scavenging protein, and that holds only while the workers are provisioning larvae. The obvious extension to carbohydrate bait was ruled out because it would attract bees.² So the method with the best evidence behind it is available during the months when nobody is complaining and gone by the time they are.

Nest destruction is limited before it starts, because most nests are concealed well enough that individual nest treatment is not an option and the ones treated are simply the ones that were noticed.³ Foragers from eleven separate colonies were collected at one trap site, which is roughly what a deck is, so removing the findable nest removes a fraction of the wasps arriving there.⁴ And by September the colony has released the queens that will found next year's colonies and will die at the first frost regardless. Late-season removal therefore cannot reduce next year's population, cannot much shorten the colony's life, and does not prevent reoccupation of a nest that would never have been reused. What it does is take a concentrated stinging hazard out of one specific place where people are, for the weeks that remain. That is worth buying. It is simply not the thing most customers think they are buying.

References

1. Protecting honey bees from yellowjacket wasps. University extension service publication. Source for the assessment that if yellowjacket pressure is high and negatively impacting honey bee colonies control methods may be attempted, but that research indicates these attempts are unlikely to reduce the overall population, with the methods listed as trapping workers, using poison bait to kill a colony, or destroying nests with insecticide; for the report that a California study found traps containing a named synthetic lure effectively attracted queen and worker yellowjackets but that the number of trapped queens did not reduce the number of workers trapped in late summer; for the statement that if spring weather is rainy or cold queens may not be able to establish a nest or forage for food for larvae while if spring is warm and dry queens may be successful and produce many offspring growing into large populations in late summer; for the description of trap designs as containers with small entrances to minimise exit, some using an attractant suspended over soapy water, available in garden centres and hardware stores; for the construction advice that sugary attractants should be avoided because they could inadvertently attract bees; and for the description of nest sites as underground cavities, holes or openings in structures, or aerial locations. https://extension.oregonstate.edu/catalog/em-9211-protecting-honey-bees-yellowjacket-wasps
2. Urban entomology text, chapter on pests attacking man and his pets, hosted by a university entomology department. Source for the statement that when the workers appear they search for proteinaceous food to feed the developing larvae in the nest and that there follows a long period when a protein bait can be used that attracts yellowjackets and practically no other insects; for the statement that carbohydrate baits were not tested since they might have the disadvantage of attracting honey bees and other nontarget insects; for the account that virgin queens appearing in the fall search for overwintering sites at the time colonies are rapidly declining and that only newly mated queens survive the winter; and for the assessment that when protein foods are being scavenged the baiting method is safe, convenient, adds practically no insecticide to the environment, and is harmless to all species but the two named target ones, a categorical claim the text does not support with presented testing. https://entomology.ucr.edu/we-ch-9-2
3. Traps and protein bait to suppress populations of yellowjackets. Conference paper, read as full text. Source for the finding of a nearly fortyfold annual variation in total numbers captured at some sites with the ultimate number of foragers apparently related to spring weather episodes; for the report that one species comprised more than 85 to 90 per cent of the total at every test site and that queens and workers were both attracted to the synthetic lure; for the statement that identified colonies may be eliminated by directing toxicant into the nest opening but that because they are well concealed it is nearly impossible to locate most nests, so that except for incidental nests individual nest treatment is not an option; for the demonstration in which a wire-mesh cage soaked with 0.05 per cent fipronil was placed over a nest entrance for fifteen minutes, the few caged exiting wasps released and allowed to return, whereupon the colony was eliminated within one day; for the summary that early trapping specifically of queens does not reduce overall numbers of foraging workers while trapping throughout the season does reduce interactions between yellowjackets and people, and that early season queen trapping may serve as a monitoring tool; for the note that traps are effective tools for monitoring activity in an area; for the earlier trial in which a named toxicant in canned chicken with acceptance enhanced by the synthetic lure was in many instances effective in drastically reducing populations; and for the statement that polygyny achieved by adding queens to an established colony is a likely prerequisite for successful overwintering. https://ncbees.org/library/TRAPS_AND_PROTEIN_BAIT_TO_SUPPRESS_POPULATIONS_OF_YELLOWJACKETS.pdf
4. Controlling yellow jackets with fipronil-based protein baits in urban recreational areas, together with the linked record for a study of annual and multi-year nests of the same species. Both read as abstracts on a research-sharing platform rather than at the publisher. Source for the statement that in spring queens and low numbers of workers were captured in lure-baited traps until early July when the number of foragers increased dramatically; for the microsatellite finding that eighteen colonies were actively foraging within one park with foragers from eleven different colonies collected at a single trap site; for the statement that in two consecutive years sufficient numbers of foragers were not captured until early August when baiting trials were initiated; for the bait preparation of canned chicken mixed with fipronil to a final concentration of 0.025 per cent; for the finding that in one Canadian province one species responded preferentially to rotisserie chicken over canned chicken or canned sardines while a synthetic chemical lure was more attractive than rotisserie chicken, that combining the two counterintuitively reduced catches, and that a second species showed a different preference; for the statement that in warmer climates nests overwinter, become polygynous and persist for more than one season; and for the finding of three perennial colonies, 2.4 per cent of all colonies, with tenfold higher peak traffic rates than annual ones. https://www.researchgate.net/publication/308010690_Controlling_yellow_jackets_with_fipronil-based_protein_baits_in_urban_recreational_areas
5. Species profile for the western yellowjacket on a hobbyist natural history reference site. A compiled secondary source of uncertain editorial standard, flagged as such and used only for life-cycle chronology. Source for the account that males and new queens are produced in August and September, mate, after which males die and inseminated queens enter diapause to overwinter. https://bugswithmike.com/guide/arthropoda/hexapoda/insecta/hymenoptera/vespoidea/vespidae/vespinae/vespula/pensylvanica
6. Pest control company article on yellowjacket diet through the season, anchored to one American state's frost calendar. Trade marketing material, cited as attributed material and used for a single chronological statement which we explicitly decline to transfer to this climate. Source for the statement that at the first hard frost every colony member except mated new queens dies within a week. https://www.anchorpestservices.com/yellowjackets/what-do-yellow-jackets-eat
7. Yellowjacket. General encyclopedia article, a tertiary and openly editable source, used for two uncontroversial points and flagged accordingly. Source for the statement that nests can persist as long as they are kept dry but are rarely used again and that the cycle is repeated in spring, and for the statement that weather in the spring is the most important factor in colony establishment, this last carrying the source's own flag indicating it is uncited. https://en.wikipedia.org/wiki/Yellowjacket