Two Beetles Per Kilogram, or One Cockroach in the Building: Why the Threshold Concept That Transformed Agriculture Inverts Indoors

1. Introduction: when is enough enough

Agriculture can answer a question this trade cannot. At what point is a pest population worth treating, and at what point has treatment done enough?

There is a framework for that, it is sixty-odd years old, and it does not work in buildings. What is interesting is not that it fails but how: the arithmetic does not merely give a small answer, it gives an impossible one.

1.1 The question underneath

Not how do you kill them, which this journal has covered at length, but how do you decide whether to, and how do you know when to stop. Those are decision problems rather than biological ones.

1.2 Why this is a methods article

Because a threshold is a decision rule, and the absence of one determines how every other decision in this trade gets made. Section 19 is what follows.

2. The paper everything rests on

The origin, which is unusually specific for a whole discipline.

In the 1960s, agricultural entomologists formulated a concept of Integrated Pest Management based on the economic injury level,¹ set out in a 1959 paper titled The integrated control concept.⁴

Of decision rules put forward, none has met with more success than those in the economic injury level concept of that paper.⁴ Its elements are essentially the same as those used today.⁴

2.1 What integrated meant originally

The integrated control concept, as the title has it.⁴ The integration was of chemical control with biological control, which is why §5 concerns natural enemies and why the threshold existed to protect them.

2.2 Sixty-seven years unchanged

A framework proposed in 1959 and still in use in the same form is either correct or unexamined.

The weight of application suggests the first, and this journal would ordinarily test that. Here the interesting question is different: it is what happens at the edge of its domain, which is §12.

3. The three elements

The definitions, which are worth having exactly.

Economic damage is the amount of injury which will justify the cost of artificial control measures.⁴

Economic injury level is the lowest population density of a pest that will cause economic damage; or the amount of pest injury which will justify the cost of control,² understood as the lowest population density above which it is cost-effective to control the pests.¹

Economic threshold, also called the action threshold, is the pest density at which control measures should be implemented to prevent it from reaching the economic injury level.²

3.1 A fourth phrasing of the same idea

The economic injury level has also been defined as the lowest pest density at which the cost of reducing the pest population equals the economic loss prevented by implementing control measures.⁷

That version makes the break-even explicit: the point at which spending on control exactly repays itself. Below it you are spending more than you save.

3.2 The damage curve underneath

One of the fundamental concepts of the approach is that each pest species has a definable relationship in terms of damage to the plant or animal host that it attacks, often referred to as the damage curve.⁶

Everything else is built on that curve. Without a relationship between how many there are and how much harm results, none of the three elements can be calculated.

Which is the first place the framework fails indoors, before any of the arithmetic in §15. Nobody has published a curve relating cockroaches per dwelling to dollars, and it is not obvious what the vertical axis would even be.

4. The conceptual breakthrough

What was actually new, stated well in a later account.

In the 60s, researchers came up with the idea of tolerating some certain level of pest damage, reasoning that most biological species are not pests and most pest species do not cause significant harm at all times and in all locations.⁵

4.1 Tolerating damage was the radical part

Not the arithmetic. The proposition that a pest could be present, doing harm, and correctly left alone.

Everything before that had treated control as a matter of killing what was found, which is still how most people think about pest management and is the position this trade is largely obliged to hold.

4.2 What it replaced

Calendar spraying. A schedule of applications made whether or not a population was present, which this journal's article on degree-day models found structural pest control still using.

So the threshold concept did two things at once: it stopped unnecessary treatments and it tied the necessary ones to an observation.

4.3 The reasoning behind it

Most species are not pests, and most pest species are not harmful everywhere and always.⁵ Pest status is contextual rather than intrinsic, which this journal has argued in several articles about identification. The same beetle is a pest in a granary and an ordinary decomposer in a woodland, and only one of those settings has a threshold.

5. Why that mattered beyond economics

The consequence the authors were reaching for.

This conceptual breakthrough implied that limiting pesticide use could lead to conserving natural enemies.⁵

5.1 The logic of the connection

Spraying below the threshold kills the pest and also the things eating it. Since the natural enemies are usually rarer and slower to recover, an unnecessary treatment can leave the population higher afterwards than before.

That is the mechanism behind secondary pest outbreak and resurgence, and this journal's article on cockroach rebound described the same shape of result from a different cause.

5.2 The thing being conserved

Virtually all pest populations are affected by natural enemies to some extent. In many cases, natural enemies are the primary regulating force of the pest populations.²

So a threshold is not only a cost calculation. It is a way of leaving the free control in place, which this journal's article on sequence-based pesticides found modern non-target work still pursuing by other means.

6. The ordering of threshold and injury level

The structural relationship, which is what §15 breaks.

The threshold is the density at which control measures should be used to prevent an increasing pest population from reaching economic injury level.⁵

6.1 Below, not at

The threshold must sit below the injury level, because its whole function is to trigger action early enough that the injury level is never reached.

That gap is where the decision lives. A population between the two is a population you are watching and have not yet treated, which is a state this trade has no name for and no procedure covering.

6.2 Why two numbers rather than one

Because populations grow. By the time a population reaches the level at which it causes economic damage, it is still growing, and control takes time to act.

The gap between threshold and injury level is therefore a buffer sized by the growth rate and the response time, which this journal's degree-day article suggests is calculable from accumulated heat for any outdoor species.

6.3 Intuitively

If the pest population and the resulting damage are low enough, it does not pay to take control measures.⁵

7. The criticism of the original definition

A weakness in the 1959 formulation, acknowledged in the literature.

The definition of economic damage has been criticized by several workers because a quantitative expression of economic justification was not given.⁴

A practical mathematical expression was supplied later, relating yield, price per unit of yield, level of pest injury, the control action and the cost of that action, stating essentially that cost of the control tactic equals yield times price when the tactic is applied minus yield times price without it.⁴

7.1 Why a quantitative expression was needed

Because justify the cost is a judgement unless somebody writes down what is being compared with what. The 1959 formulation named the right quantities and left the arithmetic implicit.⁴

7.2 What that formula needs

A yield, a price, a measurable level of injury, and a known effect of the treatment on that injury.

Three of those four do not exist for a cockroach in a kitchen, which is the arithmetic version of the whole problem and is our observation.

8. What determines where the threshold sits

The variables, and the last one is the interesting one.

The rise and fall of the control action threshold is determined by the importance of the ecosystem, value of the crop, the pest status, and consumer standards.³

8.1 Density in context

The threshold rests on the fact that pest density must be considered in the economic context in which it occurs, with both concepts helping in decision making to determine the moment of action.⁸

8.2 Consumer standards

An entomological threshold that depends on what buyers will accept. The insect's biology is constant and the number is not.

That is the hinge of this article, and §9 is the demonstration.

9. The same insect, two crops

The example that shows the threshold is not about the organism.

One pest species feeding on cotton boll has an economic threshold of four larvae per plant, and generally requires insecticidal treatment several times yearly. The same species feeding on sweet corn has an economic threshold approaching zero population, since the consumer will reject sweet corn with any damage or one larva on it.³

9.1 And the treatment frequency follows

The cotton crop generally requires insecticidal treatment several times yearly,³ which is what a working threshold produces: repeated decisions, some of which are to act and some of which are not.

Sweet corn produces no such sequence, because there is no interval during which waiting is correct.

9.2 Four against nothing

Same genus, same feeding behaviour, two commodities.³ On one crop four larvae per plant is acceptable; on the other a single larva makes the product unsellable.

10. What the sweet corn case establishes

The general principle, which carries directly into buildings.

The threshold collapses to zero when the product is inspected by the person who will consume it and any evidence is disqualifying.³

10.1 The general rule we would extract

The closer the customer stands to the evidence, the lower the threshold. Cotton is spun, woven, dyed and sewn before anybody sees it; sweet corn is looked at in a shop.

That rule orders every setting this trade works in, from a warehouse holding sacks to a restaurant dining room, and it predicts the tolerance without knowing anything about the insect.

10.2 Cotton is processed

Nobody examines a cotton boll before buying a shirt. The damage is absorbed upstream and shows up as a yield number, which is exactly what the formula in §7 requires.

10.3 Sweet corn is looked at

By the purchaser, at the point of sale, with the evidence visible. The threshold is therefore set by perception rather than by loss.

A hotel room is sweet corn. This journal's article on hygiene grade disclosure described exactly what happens when the customer can see the evidence, and the answer was that revenue moves. Our framing.

11. The vector exception

A second route to zero, on different grounds.

In special cases, where pests serve as vectors of plant, animal and human diseases, the economic threshold is zero.³

11.1 Why a vector cannot have a tolerance

Because the harm is not proportional to the number. One infected individual can transmit, so the damage curve in §3.2 is not a curve at all; it is a step at one.

11.2 Two different zeros

One arrives because a single individual makes the product unsellable. The other arrives because a single individual can transmit something.

They produce the same number for entirely different reasons, and this journal's articles on West Nile surveillance and on tick-borne disease concern the second kind.

12. The comparison nobody had made

The paper at the centre of this article, and its starting point.

The author notes that as no comparative study has been published so far, I have quantified the tolerable pest population size for both environments, and that the results reveal huge numerical differences.¹

12.1 The gap the paper identifies

No comparative study had been published.¹ A framework in universal use, applied to two very different settings, with nobody having checked whether the numbers behaved similarly in both.

That is the same structure this journal found behind rotation strategies and behind applicator training: a practice adopted widely and examined narrowly.

12.2 Two environments

Agricultural, where the framework was built, and urban and quarantine, where it is applied without anyone having checked whether the numbers permit it.

The stated goal is to report that very low tolerance to urban and quarantine pests can lead to situations requiring a zero injury level, and to explore the implication of extremely low pest tolerance for the threshold-based pest control theory.¹

13. The grain store figure

The agricultural end of the comparison.

For a small-sized grain silo of 50 tonnes, the total tolerated pest load is given as 0 to 100,000 beetles, at a threshold of 0 to 2 beetles per kilogram of grain.¹

13.1 Note the range starts at zero

Zero to two beetles per kilogram, and zero to a hundred thousand in the silo.¹ The lower bound is zero because some buyers and some end uses tolerate nothing, which is §8.2 operating inside a single commodity.

13.2 A hundred thousand beetles

In a silo that would be considered small.¹ That is a working tolerance, expressed per unit of commodity, and it is the kind of number the framework was designed to produce.

13.3 The bibliographic company it keeps

The same author's related work includes a paper on distinguishing injury from damage in post-harvest protection.⁷

That distinction matters here: injury is what the insect does to the commodity and damage is what it costs. A building can suffer injury with no measurable damage, and the framework needs the second.

13.4 Why grain can tolerate that

Because the loss is proportional and measurable. Two beetles per kilogram consume a calculable quantity, and this journal's article on phosphine in stored grain dealt with the economics of treating that.

14. The building figure

The other end.

For a large-sized building, given as a hotel, the tolerated load is 0, possibly 1, of the German cockroach, at a threshold of 0, possibly 1 cockroach.¹

14.1 The question mark is doing work

The author writes zero with a parenthetical one and a question mark.¹ That is an honest way of saying that the tolerable number is either nothing or as close to nothing as makes no difference, and that nobody has established which.

14.2 For an entire building

Not per room, per kilogram or per square metre. One insect in a hotel, which may contain hundreds of rooms.¹ Expressed per unit of anything, that is a density so low it is hard to write down.

14.3 Why a hotel and not a house

Because a hotel sells the room. A guest finding a cockroach does not absorb a small loss; they leave, complain publicly, and do not return, which this journal's article on hygiene disclosure quantified in revenue terms.

A householder has no such transaction and might tolerate a good deal more, which suggests the number is a property of the commercial relationship rather than of the building. That is our reading.

14.4 And quarantine is stricter still

For transport, meaning boxes, bags, railway wagons and ship chambers, the tolerance is 0 of any quarantine organism per unit of exported commodity.¹

15. The inversion

The finding in the paper's title, and the conceptual heart of the article.

The paper is titled as concerning an inversion relationship between action threshold and economic or aesthetic injury level for the control of urban and quarantine pests.¹

15.1 Why the word inversion is precise

It is not that the threshold is low. It is that the ordering of two quantities reverses, so a rule written for one ordering produces nonsense under the other.

A rule that says act when density reaches a value below zero cannot be followed.

15.2 What inverts

In agriculture the action threshold sits below the injury level, as §6 sets out. Where the injury level is zero, no threshold can sit below it.

The ordering that makes the framework work is therefore unavailable, and the relationship between the two quantities reverses.

15.3 Our reading of the consequence

If the threshold cannot precede the injury level in population terms, it has to precede it in time. Action must happen before the pest arrives rather than after it reaches a density.

That is the mathematical justification for preventive pest control, and we had not previously seen one. It is our inference from the paper's framing rather than a sentence we can quote.

16. Why there is nothing below zero

Labouring the point slightly, because it determines everything after it.

A decision rule of the form treat when density exceeds X requires an X greater than zero. Otherwise the rule fires on detection, which is not a threshold but an alarm.

16.1 Where the difference bites

A threshold can be crossed downward as well as upward, which gives a treatment programme an endpoint: you stop when the population is back below it.

An alarm has no downward crossing. Nothing tells you that the last one is gone, which is §19's missing stopping rule and is the single most consequential absence in this trade.

16.2 An alarm is a different instrument

A threshold answers is it bad enough yet. An alarm answers is it here. The first requires a damage curve and the second requires only detection.

This journal's detection probability article spent five thousand words on why the second question is harder than it looks, and §20 is what that implies.

16.3 The author's own caution

The paper notes that it is not only the notorious toxicological reasons that do not enable permanent pest control and eradication to be carried out.¹

Which we read as acknowledging that even where the target is zero, zero is frequently not achievable, so the framework fails in both directions at once.

17. What that justifies

The practical conclusion, which reframes something this trade does by habit.

Preventive service is usually defended commercially, as insurance or as peace of mind. On the argument above it has a stronger defence: where the tolerable population is zero, prevention is the only strategy the decision framework permits.

17.1 The uncomfortable symmetry

This article argues that a trade selling preventive contracts is entitled to a theoretical justification for them, which is an argument in that trade's favour, reached from a source with no commercial interest at all.

We would rather the argument came from somewhere else, and it does: a 2002 note in a pest science journal by an author quantifying tolerances for entirely academic reasons.¹

17.2 Responsive control is not available

Waiting until a threshold is crossed means waiting until the injury level is crossed, because they are the same number.¹

By then the damage the threshold existed to prevent has already occurred.

17.3 The paper's own phrasing

It refers to a responsive strategy of the control of urban pests,¹ which is the thing being contrasted with the agricultural model.

18. The aesthetic injury level

The adaptation the field made for non-agricultural settings.

As landscape and urban pest management evolved, so did attempts to consider not only economic profits, but the aesthetic value of pest control as well, and recent efforts have focused on incorporating costs to the environment and society from pest control practices.²

18.1 The third cost being counted

Recent work incorporates costs to the environment and society from pest control practices.² Which means the calculation now weighs the harm of treating against the harm of not treating.

That is a genuine improvement and it pushes thresholds upward, since treatment carries a cost the original formula ignored. This journal's articles on pollinator exposure and on secondary poisoning are that cost being made explicit.

18.2 What an aesthetic injury level is

The same structure with tolerance set by appearance rather than by yield. It is the honest acknowledgement that in a garden or a building, the loss being prevented is not a crop.

18.3 Why it does not rescue the framework indoors

Because it changes what is being measured and not the number. If the aesthetic tolerance is one cockroach in a hotel, the arithmetic problem in §15 is unchanged.

Renaming a zero does not give you something to put below it. Our argument.

19. What this trade lacks as a result

The consequences, gathered.

No damage curve. Section 3.1 says the whole framework is built on one,⁶ and we are not aware of a published curve relating cockroach density to a cost in a dwelling.

No stopping rule. Nothing that states when a treatment programme has demonstrably done enough, which this journal's articles on clearance and on detection have circled repeatedly.

No way to decline work. An agricultural adviser can tell a grower that treatment is not economically justified. Nobody in this trade can tell a client that one cockroach does not warrant a visit.

And no basis for a warranty term. This journal's article on pest damage and insurance found warranties functioning as the effective liability instrument, and a warranty needs a defined standard of performance that nothing here supplies.

19.1 No way to compare two programmes

With a threshold, two approaches can be compared on whether they keep the population below it and at what cost. Without one, the only available comparison is whether anybody complained.

Which is why this journal's article comparing integrated management with conventional spraying had to reach for complaint counts and product volumes rather than for a population figure.

19.2 The commercial shape of that

Every problem is worth treating, because the tolerable number is zero. That is convenient and it is also the correct answer, which is an uncomfortable combination and is why we have set the reasoning out in full rather than asserting it.

20. What monitoring is for when the threshold is zero

A reframing that follows from §16.

If there is no density at which action becomes justified, then counting is not the purpose. Detecting presence is.

20.1 What gets lost

Counting still has a use, because a trend is information even where a level is not actionable. But the count is no longer being compared against anything; it is being compared against last month.

That is a weaker instrument and it is the one this trade actually has.

20.2 Which changes what a monitor should be optimised for

Sensitivity rather than calibration. A device that reliably finds the first individual is more useful than one that accurately counts thirty.

This journal's articles on interceptors, on light traps and on canine detection all concerned sensitivity at low density, and this is why that is the relevant property.

20.3 And it raises the cost of a negative

Where the threshold is one, a false negative is not a small error. This is the argument the detection probability article made about effort, arriving here from the decision-theory side.

21. Where a threshold could still be built

The constructive part, because the article should not end in a dead end.

Stored product in bulk. The grain figure exists and applies to commercial food premises holding commodity.¹

Nuisance invaders. Cluster flies and boxelder bugs cause no damage and no disease, so tolerance is genuinely a matter of preference rather than of zero.

Outdoor populations. Wasp nests at a distance, ants in a lawn, and anything whose presence is not inside the envelope.

Trend rather than level. Where an absolute threshold is unavailable, a rising count is still actionable information, which is what this journal's light trap article recommended.

And an aesthetic threshold can be negotiated. A client who states what they will tolerate has supplied the number the literature cannot.

Which is worth asking for explicitly. A contract that names the tolerance turns an unbounded obligation into a defined one, for both parties.

22. Limitations and open questions

The central source is a single short paper. A note in a pest science journal, read through a repository record, and we have not seen the full table or the reasoning behind the figures.¹

The building figure is one author's estimate. Zero with a parenthetical one and a question mark is explicitly uncertain, and it is not a measured quantity.¹

The inversion argument in §15.2 is ours. The paper's title states an inversion relationship; the reading that this constitutes a mathematical justification for preventive treatment is our inference.

We have not read the 1959 paper. Everything about it comes from later accounts and extension summaries.²⁴⁵

The crop example is old. The cotton and sweet corn thresholds are cited to work from the 1960s and current figures will differ.³

No structural pest thresholds exist to test this against. Which is the article's point and is also its main weakness, since we cannot demonstrate the absence of something by failing to find it.

We have not considered health-based thresholds. Cockroach allergen has measurable dose-response relationships with asthma outcomes, which this journal covered, and that is arguably a damage curve for a structural pest. We have not worked out whether it could support a threshold and it may be the most promising route.

Sections 2.1, 4.1, 7.1, 10.1, 10.2, 11.1, 13.2, 14.1, 15.2, 16, 18.2, 19, 20 and 21 are our reasoning. The longevity observation, the reading of what was radical, the formula requirements, the processed against inspected distinction, the hotel comparison, the two zeros, the grain economics, the reading of the question mark, the inversion consequence, the alarm distinction, the aesthetic argument, the list of consequences, the monitoring reframing and the constructive list are ours rather than sourced positions.

23. Conclusion

The conceptual advance of 1959 was not the arithmetic. It was the proposition that a pest could be present, doing measurable harm, and correctly left alone, on the reasoning that most species are not pests and most pest species do not cause significant harm at all times and in all locations.⁵ Everything since has been built on a damage curve relating how many there are to how much it costs.⁶

Thresholds turn out to be set by the buyer rather than by the insect. The same species is worth treating at four larvae per plant on cotton and at nothing at all on sweet corn, because the consumer rejects sweet corn with any damage or one larva on it.³ Extend that and a small grain silo tolerates up to two beetles per kilogram while a large hotel tolerates zero, possibly one, cockroach.¹

At that point the framework stops working, because an action threshold has to sit below an injury level and there is nothing below zero. What replaces it is not a better number but a different discipline: since you cannot act when a density is reached, you act before anything arrives.¹ This trade has been doing that for a century and calling it prevention. It turns out to be the only thing the decision theory permits, which is a better reason than the one usually given, and it comes with a cost the theory also predicts: there is no density at which anybody can say the work is finished.

References

1. Stejskal, V. (2002). Inversion relationship between action threshold and economic or aesthetic injury level for the control of urban and quarantine pests. Journal of Pest Science, 75, 158 to 16x. Principal source, read through a repository record. Used for the account that in the 1960s agricultural entomologists formulated a concept of integrated pest management based on the economic injury level, understood as the lowest population density above which it is cost-effective to control the pests; for the author's statement that as no comparative study had been published, the tolerable pest population size was quantified for both environments and the results reveal huge numerical differences; for the stated goal of reporting that very low tolerance to urban and quarantine pests can lead to situations requiring a zero injury level and of exploring the implication of extremely low pest tolerance for threshold-based pest control theory; for the tabulated figures giving a small-sized grain silo of 50 tonnes a total tolerated load of 0 to 100,000 beetles at a threshold of 0 to 2 beetles per kilogram of grain, giving transport means such as boxes, bags, railway wagons and ship chambers a tolerance of zero of any quarantine organism per unit of exported commodity, and giving a large-sized building such as a hotel a tolerance of zero, possibly one, German cockroach; for the reference to a responsive strategy of the control of urban pests; and for the observation that it is not only toxicological reasons that prevent permanent pest control and eradication from being carried out. https://www.researchgate.net/publication/230001202_Inversion_relationship_between_action_threshold_and_economicaesthetic_injury_level_for_the_control_of_urban_and_quarantine_pests
2. Pest Management Decision-Making: The Economic-Injury Level Concept. University extension plant health programme. Used for the definition of the economic injury level as the lowest population density of a pest that will cause economic damage, or the amount of pest injury which will justify the cost of control; for the definition of the action threshold as the pest density at which control measures should be implemented to prevent it reaching the economic injury level; for the statement that the concept was developed hand in hand with the integrated pest management concept; for the account that as landscape and urban pest management evolved so did attempts to consider not only economic profits but the aesthetic value of pest control as well, with recent efforts focused on incorporating costs to the environment and society from pest control practices; and for the statement that virtually all pest populations are affected by natural enemies to some extent and that in many cases natural enemies are the primary regulating force of pest populations. https://extension.usu.edu/planthealth/research/eil-concept
3. Principles and theory of integrated pest management. United Nations agricultural organisation technical document. Used for the definitions of economic injury level as the lowest pest population that will cause economic damage and of the control action threshold as the density at which control measures should be applied to prevent an increasing pest population reaching it; for the statement that the rise and fall of the control action threshold is determined by the importance of the ecosystem, the value of the crop, the pest status and consumer standards; for the example that one pest species feeding on cotton boll has an economic threshold of four larvae per plant and generally requires insecticidal treatment several times yearly, while the same species feeding on sweet corn has an economic threshold approaching zero population since the consumer will reject sweet corn with any damage or one larva on it; and for the statement that in special cases, where pests serve as vectors of plant, animal and human diseases, the economic threshold is zero. https://www.fao.org/4/x5048e/x5048e07.htm
4. Economic Thresholds and Economic Injury Levels. University integrated pest management textbook chapter. Used for the statement that of decision rules put forward none has met with more success than those in the economic injury level concept of the 1959 paper; that the elements proposed formally in 1959 are essentially the same as those used today, being economic damage, economic injury level and economic threshold, collectively forming the concept; for the definition of economic damage as the amount of injury which will justify the cost of artificial control measures; for the note that this definition has been criticised by several workers because a quantitative expression of economic justification was not given; and for the later mathematical expression relating yield, price per unit of yield, level of pest injury, the control action and the cost of that action, stating that the cost of the control tactic equals yield times price when the tactic is applied minus yield times price without it. https://ipmworld.umn.edu/pedigo
5. Economics of Decision Making in Pest Management. University integrated pest management research paper. Used for the account that in the 1960s researchers came up with the idea of tolerating some certain level of pest damage, reasoning that most biological species are not pests and most pest species do not cause significant harm at all times and in all locations; that this conceptual breakthrough implied that limiting pesticide use could lead to conserving natural enemies, so that the concepts of economic damage, economic injury level and economic thresholds were developed; that these three constitute the basic elements of the economic threshold model; that intuitively the concept of economic threshold implies that if the pest population and the resulting damage are low enough it does not pay to take control measures; that in practice the term economic threshold has been used both to denote the population level at which economic loss begins and to indicate the level at which control should be initiated given the cost of control; and for the 1973 definition of the economic threshold as the pest density at which control measures should be used to prevent an increasing pest population from reaching the economic injury level. https://www.virginiafruit.ento.vt.edu/OliveProj/IPMPaper.html
6. Riley, D. G. (2008). Economic Injury Level and Economic Threshold Concepts in Pest Management. In Encyclopedia of Entomology. Springer. doi:10.1007/978-1-4020-6359-6_3497. Used for the statement that one of the fundamental concepts of integrated pest management is that each pest species has a definable relationship in terms of damage to the plant or animal host that it attacks, and that this relationship is often referred to as the damage curve. https://link.springer.com/rwe/10.1007/978-1-4020-6359-6_3497
7. Stejskal, V. Economic Injury Level and preventive pest control. Research publication record, with associated reference list. Used for the definition of the economic injury level as the lowest pest density at which the cost of reducing the pest population equals the economic loss prevented by implementing control measures; and for the bibliographic context placing the inversion paper alongside related work on distinguishing injury from damage in post-harvest protection and on alternatives to pesticides in stored-product integrated pest management. https://www.researchgate.net/publication/225454727_'Economic_Injury_Level'_and_preventive_pest_control
8. Pattern-Based Prediction of Population Outbreaks. Preprint. Used for the statement that the economic threshold is based on the fact that pest density must be considered in the economic context in which it occurs; for the restatement of the 1959 definitions of the economic threshold as the density at which control measures should be determined to prevent an increasing pest population from reaching the economic injury level, and of the economic injury level as the lowest population density that will cause economic damage; and for the observation that both concepts help in decision making to determine the moment of action. https://arxiv.org/pdf/2209.02346