Plant Bug Thresholds in Arizona Cotton Cotton Conferences, Plant Bugs January 10, 2007 Ellsworth, Ly gus thresholds in Arizona 1 JanuaryT 10, 2007 Ellsworth/UA Plant Bug Thresholds

Beltwide Cotton Conferences, Plant Bugs January 10, 2007

Ellsworth, Lygus thresholds in Arizona 1

Ellsworth/UAJanuary 10, 2007

Plant Bug Thresholds

in Arizona Cotton

Peter C. Ellsworth

Arizona Pest Management Center /Department of Entomology

University of Arizona

Beltwide Cotton Conferences

Plant Bug and Stink Bug Management Workshop

This presentation was invited by organizers of thePlant Bug Management Workshop as part of theBeltwide Cotton Conferences. The subject of thispresentation is Lygus hesperus thresholds forArizona cotton.

20 min. / 150



Ellsworth/UA

Plant / Pest /PesticideInteractions

• Historical Perspective

• Key Questions

• Experimental Progress

The development of a threshold for the control ofan insect pest is a fundamental aspect to integratedpest management. As straightforward as thepractical goal is, however, this development isbased on a 3-way interaction among plant - pestand pesticide.

In today’s talk I hope to start and end with a bit ofa historical perspective, bring out some keyquestions that need to be asked and answered inthe development of a Lygus threshold and thenreview some of the experimental progress we havebeen making in Arizona over the last 10 years.



Ellsworth/UA

Historical Trends in LygusControl in Arizona Cotton

IGRs, Bt cotton,& AZ IPM Planintroduced

Lygus is not a new pest to Arizona cotton growers.It is one that has been present and at play for along time. This chart shows the statewide foliarspray intensity for Lygus bugs since 1990. Ingeneral, you can see that we have been sprayingLygus ca. 1-3 times per season. This trend appearsconsistent even after dramatic and major changesto our system. The introduction of Bt cotton andselective whitefly IGRs has not changed the basicneed to control Lygus in our system.



Ellsworth/UA

Lygus No. 1 Since 1998

$71 / A in foliar insectcontrol since 1998

($26 / A to Lygus)

What has changed, however, is the relative statusof Lygus, which has been our number one insectpest of cotton since 1998. Lygus has gained inimportance simply because it occupies a greaterproportion of our spray requirements and budgets.In fact, it is the largest yield threat to AZ cotton.



Ellsworth/UA

• Key Questions in Developing Thresholds– Which control agents will be used?

– What will the action level be based on?• Plant or plant damage, insect, or both?

– What is the action level that maximizes grower revenue?

– When to terminate action?• Based on plant factors, insect factors, both?

Constraints: Weather, Varieties, and Length of Season

There are several key questions we must ask andanswer if we wish to develop meaningful andpractical thresholds for Lygus control…1)… 2)… 3)While we have to build yield relationships towhatever threshold is tested, ultimately, it isnecessary to maximize the relationship to growerrevenue and not really yield itself, 4)…



Ellsworth/UA

Lygus IPM

1

2

3

The 3 basic keys to any IPM plan are sampling,effective chemical use, and “avoidance”. And,thresholds are a major building block of any IPMprogram. However, as revealed in the questions inthe previous slide…



Ellsworth/UA

Lygus IPM

1

2

3

…one can hardly examine Lygus action thresholdswithout giving due consideration of the chemistrythat will be used and the sampling plans thatsupport the system.



Ellsworth/UA

Exit hole

Pink Bollworm

Live 4th Instar

Bemisia tabaci

But before I go further, I should point out thatmajor advances in the selective controls of pinkbollworm with Bt cotton, and whiteflies with insectgrowth regulators have enabled our ability toisolate the system on Lygus and develop veryspecific information. Prior to 1996, it simply wasnot possible to develop a research plan fordevelopment of Lygus action thresholds because ofthe confounding effects of these other two keypests on our Arizona system. However, now wehave essential and nearly complete and selectivecontrol of these other two pests.



Ellsworth/UA

Whatcompoundsare effective?

The first question centers around what compoundsare effective against the target pest, in this caseLygus. Without a clear understanding of the controlsystem, it is not possible to develop a threshold forthe triggering of remedial action.



Ellsworth/UA

> 10-fold Increase in Yields (02F4L)

C V NeonicotinoidsO x r2Txx C r4

Yield in bales per acre

Control(0.11 bales)

Orthene(1.28

bales)

I will not review the more than 10-yrs worth ofefficacy information that we have developed inArizona. But I do want to let you know that allthese evaluations have been done under extremeLygus pressure where once again we have isolatedthe impacts on yield to this pest through the use ofBt cotton and selective controls of other insects. Inthis case, you can see a 10-fold increase in yieldwhere plants were adequately protected againstLygus damage. It is not at all unusual for us to see3, 4, or 5-fold differences in our efficacyevaluations of this pest.



Ellsworth/UA

Broad SpectrumOptions

Acephate (e.g., Orthene)

Methamidophos (Monitor)

Oxamyl (Vydate C-LV)

Endosulfan (e.g., Thionex)

Fipronil (e.g., Regent)

PotentiallySelective Options

Flonicamid (e.g., Carbine)

Metaflumizone (BAS320)

For many years now, we have depended on a set ofbroad spectrum options. That is, compounds thatare effective but are rather broad spectrum in theirimpact on the arthropods present in the system.These include acephate, methamidophos (whichsees little use today), oxamyl, endosulfan, andfipronil or Regent, which does not have a U.S.registration in cotton but is a standard for miridcontrol in Australia and in Mexico. More recently,we have had exciting new advances withpotentially selective options: flonicamid wasrecently registered as Carbine and is very effectiveagainst Lygus, and metaflumizone is on track forregistration very soon. These two compounds bringus new chemistry that might in fact be moreselective than our traditional, broad spectrumoptions.



Ellsworth/UA

Product Comparisons05F2L

Acephate

Metaflumizone UTC

Flonicamid

05F2L-T18 Carbine WG (2.8 oz) in NE, Orthene (1.0lb ai/A) in NW, 320WVI (0.25 lbs ai/A) in SW, UTCin SE; 5-fold yield increase over the UntreatedCheck.

While flonicamid and metaflumizone haveperformed admirably against Lygus and helped topreserve the yield component in dramatic fashion, amajor potential attribute of their use in our systemwill be to drive more and more to a selectivesystem that helps preserve our natural enemycomplex…



Ellsworth/UA

Principal Response CurveNatural Enemies to Broad Spectrum Insecticide

P < 0.01

Naranjo, Ellsworth, Hagler (2004)

1997

Community response of ca.20 natural enemies

As an example, I would like to briefly explain amultivariate analytic approach that is representedgraphically in Principal Response Curves. In thisexample we can see the green ‘U’ line representingthe UTC as a baseline from which we compare othertreatments. In this analysis, we examine theresponse of an entire community of naturalenemies, here about 20 species. Departures fromthe baseline may be interpreted as density changesin this natural enemy community. The red arrowindicates the timing of a single, very broadspectrum insecticide sprayed to control Lygus in astudy that I did several years ago with SteveNaranjo and James Hagler…



Ellsworth/UA

Principal Response CurveNatural Enemies to Broad Spectrum Insecticide

P < 0.01

Naranjo, Ellsworth, Hagler (2004)

1997

Community response of ca.20 natural enemies

Season-longeffects

…What we see is a dramatic and immediatelowering of the density of these natural enemies incomparison to the UTC. What is more sobering isthe duration and significance of this effect, all theway out to 7 weeks post-treatment. These season-long effects have grave consequences in the controlof many other primary and secondary pests, as wellas Lygus. So having potentially selective options toreduce the risks of natural enemy destruction isquite important to us.



Ellsworth/UA

Basis ofThreshold?

• Yield response

• Bug susceptibility

• Visual evidence

The next question we need to ask is what will bethe basis of our threshold? For our Arizona cottonsystem, we have evidence from efficacy and otherstudies of yield response and bug susceptibility bylife stage; however, our visual evidence providesthe most compelling case for the choices we havemade in our threshold system…



Ellsworth/UA

Adults move; Nymphs don’t

00F3threshold

First, we all know that only adults have wings andtherefore only they can move any significantdistances. In fact it is unlikely that nymphs moveacross or down rows very much.

This is an aerial photograph of my 2000 Thresholdstudy. In the outlined area you can see severalborders of cotton each with 3 harvested stripstaken from them. However, in addition to the 3dark stripes down each border, we can also seesome darker areas of growth.

This is photographic evidence that adults are notmajor damagers of cotton. If they were…



Ellsworth/UA

UTC

Adults move; Nymphs eat!

We would expect a halo of damage (as drawn in onplot in center) to develop around these UTC plotsdue to the frequent movements of Lygus adultsfrom those plots. However, the demarcationbetween unprotected v. adjacent protected areas isdistinct. This indicates damage by a plant-boundlife form, nymphs. Lygus were well-managed in allareas around these untreated plots. Yet, no patternof damage occurs around these UTC plots.

Indeed, adults do move and probably do eat aswell, but comparatively they are in this world tomove and reproduce, whereas nymphs have oneobjective in life, to eat and grow.

So nymphs have become the basis for developmentof our threshold system.



Ellsworth/UA

0 Sprays3 Sprays

Note height difference

As further evidence, these effects occur over theshortest spatial scale. That is in adjacent rows,shown here in a commercial trial where cotton wassprayed 3 times on the left for Lygus and not at allon the right. The height and eventual yielddifferences we see are as a result of Lygus feedingand damage, as these plots were planted to Btcotton and all other pests were selectivelycontrolled.



Ellsworth/UA

Basis ofThreshold?

• Nymphs

– Primary feeders

• Adults– Reproduction &

movement in field

• Plant?

– Fruiting potential

So in our system, the basis of our threshold isnymphs as the primary feeders in the system;however, adults are also considered and are ofcourse key to the reproduction and movement inthe system. What we have not done as yet isincorporate any plant-based factors such as fruitingpotential.



Ellsworth/UA

Action Threshold Studies

• Obj: Test nominal thresholds & identify mosteconomically effective approach

In some range-finding experiments in 1997, wefound that 15 total lygus per 100 sweeps wasassociated with our yield protection. So starting in1999, we initiated trials where we tested nominalthresholds that had as a criterion that there be atleast 15 total lygus per 100 sweeps with varyingnumbers of nymphs contained within thesesamples. Specifically, we looked at 15 total Lyguswith 0 nymphs, 1 nymph, 4 nymphs, 8 nymphs, andthe nonsensical 16 nymphs per 100 sweeps. Wethen triggered applications of Lygus-effectiveinsecticides in a set rotation of chemistry.



Ellsworth/UA

Yield : Density

• Regression ofstandardized yieldsfrom two years

• Excellent fit

After two years of study, we conducted regressionsof standardized yields using a 2nd degree fittedpolynomial (quadratic). The fit was excellent…



Ellsworth/UA

Yield : Density

• Regression ofstandardized yieldsfrom two years

• Excellent fit

• Maximum Yield @1.7 nymphs / 100

From this relationship, we can examine the point ofmaximum yield, which occurred at 15 total Lyguswith ca. 1.7 nymphs per 100 sweeps. Of course,yield is only partly the answer,…



Ellsworth/UA

Revenue : Density

• Regression ofstandardizedrevenues from twoyears


To understand the point of diminishing return, wedeveloped this regression that shows therelationship between standardized revenues andour tested thresholds.



Ellsworth/UA

Revenue : Density

• Regression ofstandardizedrevenues from twoyears


• Maximum Revenue@ 5.2 nymphs / 100

Following this curve to its maximum, we see thatmore money is made when a threshold of 15 totalLygus with 5.2 nymphs per 100 sweeps is observed.Furthermore, this basic relationship held up under ahuge variety of cotton economic conditions ($0.20-1.20 / lb). So these studies have given rise to ourcurrent recommendation which is intentionally setto be somewhat conservative to guard againstexcessive yield loss…



Ellsworth/UA

‘15:4’ Threshold Recommendation100 sweeps

This so-called ‘15:4’ threshold represents 15 totalLygus per 100 sweeps with at least 4 nymphs per100 sweeps. I should add here that a sweep-net isa standard method used by our consultingcommunity in Arizona. It is widely used and wedon’t have limitations of early morning dews thatcause the sweep bag to get too wet to efficientlysample.



Ellsworth/UA

‘15:4’ Threshold Recommendation

13 Adults+

4 Nymphs(17:4)

is over

‘15:4’

Spray

100 Sweeps

Note that the 15 in the 15:4 represents the totalcount of Lygus (and not just the adults), and thatboth components must be satisfied in order toproperly interpret the threshold. That is, there mustbe at least 15 total Lygus, but also, there must beat least 4 nymphs per 100 sweeps. When theseconditions are met, a grower is advised to spray;short of this, he/she should continue to wait andmonitor the situation. Even at 15:4, I advisegrowers to consider the abundance of naturalenemies, other pests, and fruiting potentials. Thesefactors could serve to further delay or hasten anapplication. I do work with some growers that arevery comfortable with 15:8 and routinely averageover 3-3.5 bales / A on their acreage.



Ellsworth/UA

When to stopspraying?

• Plant factors?

• Bug factors?

• Season?

Threshold work is not done, however. Once athreshold is established, at a minimum, one has toconsider under what conditions should spraying bestopped. Decision-making could potentially bebased on plant factors, bug factors, or elementsrelated to length of season remaining (environmentx variety).



Ellsworth/UA

Cotton’s Fruiting Curve

This is a graphical depiction of Cotton’s FruitingCurve expressed as blooms per unit area over time(HUs). In AZ cotton, we do have the capacity insome varieties to grow a second or true top cropafter crop cut-out. Some growers manageirrigations to mature out the primary fruiting cycleonly, while others elect to continue irrigations tomature out the top crop.

So experimentally, we set-out to examine 4different Lygus chemical control terminationtimings centered around initiation of cut-out, wherecut-out is defined as NAWF ! 5.



Ellsworth/UA

Lygus Termination Studies

• Obj: Identify besttiming (of 4 tested) fordiscontinuing Lyguschemical controls

• Compare 12 differentproduction scenarios(3 x 2 x 2)

– Variety (S, M, L)

– Planting (Opt., Late)

– Irrigation termination(Opt., Late)

However, given this 3-way interaction, it seemedimportant to consider different productionscenarios that influenced season length and fruitingpotential. So we looked at 12 production scenariosin a factorial arrangement of varieties (short-,medium-, or long-season), planting dates (optimumor “early”, and late), and irrigation termination(optimum or “early”, and late).

This was an ambitious plan and results in someextensive data that I wish to share with you. Mindyou, my goal will not be to examine each pointtoday, but to look at broader trends. So don’t panicwhen the screen fills with data.



Ellsworth/UA

5-Aug 16-Aug 2 wk < c.o.

23-Aug 1 wk < c.o.

6-Sep 1 wk > c.o.

20-Sep 3 wk > c.o.

LT4

LT3

LT2

LT1

Spray DatesLygus Termination (LT)

Timing Late Season Controls(when should you stop spraying?)

c.o. = cut-out or nodes above white flower = 5

In a range-finding experiment that we conducted in2002, we looked at 4 chemical termination timingsthat resulted in a varying number of sprays asshown here, 2–5 sprays. After this initial pilot test,we settled on 3 timings to test and included an UTCfor comparative purposes.



Ellsworth/UA

DP444BR

14 May

Last Irrigation

1 September

Last Irrigation

16 September

H a r v e s t e d H a r v e s t e d

And before I share the data with you, I’d like toshow you how these production scenarios wereimplemented in the field. In this case (2005), weare looking at a 12-row whole plot of a short-season variety, DP444B/R, that was planted late(14 May). This shot is after our machine harvest ofthe two subplots. So 4 of 6 rows were harvestedout in each case where irrigations were stoppedearlier (on the right) or later (on the left). Thenumbers you see represent the total number ofLygus sprays made to these plots. Yields were goodin the 2-spray plots, not statistically better in the 4-spray plots, and off 1 whole bale in the UTC. Youcan even see the darkened area where the UTC wasmuch more rank than the surrounding cotton. Note,too, that this short-season variety did not respondto the added water.



Ellsworth/UA

a a a a a a

b b

a

b

a a a

b

a a a a a

b

c c

a

b

c c

a a

b b

c c

a a

b

a

b b

a a

b b

c c

a a

b b

c

a

b b

c c

a

b b b

0

1

2

3

0

1

2

3

0

1

3

4

0

1

3

4

0

1

3

4

0

1

3

4

0

1

2

3

0

1

2

3

0

1

2

4

0

1

2

4

0

2

3

4

0

2

3

4

0 1 4 03 3 2 1

2 4

3 2

03F4LT

12 Production Scenarios

So let’s look at the yield data (in bales / A) forthese 12 production scenarios, early planted for the6 cells on the left and late planted for the six on theright. Yields are plotted for each lygus terminationtiming and the number of sprays are shown above;for the two irrigation termination timings, eitherwith more water (on left) or less water (on right)sides of each cell.

As we reveal each cell, you can see how yieldpotential was consistently greater as we adopted amore normal planting date and as we moved tomore determinant varieties. The green arrows andnumbers indicate the statistical break-off pointswhere additional sprays failed to returnsignificantly more yield.



Ellsworth/UA

Yield : NymphsRelationship

03F4LT

2003

ns ns ns 0.56 0.47 0.33 0.73 0.40 0.48 0.75 0.85 0.51R2

!

Make note here that where yields are highestamong the production scenarios, that’s whererevenue is the highest. Later on, I will show yourelative, marginal revenues, that focus onincremental improvements over the poorestoutcome for each variety x irrigation scenario.

But before we do, let’s look at the relationshipbetween nymphs per 100 sweeps (blues on rightscale) and yield (bales/A on left scale, orange).There appears to be a very good correspondencebetween the two measures with yields going up asnymphs go down. In fact, the R2 for the simplelinear regressions are quite good in most cases andhighly significant. This is the 2003 data.



Ellsworth/UA


04F3LT

2004

ns ns 0.40 ns 0.33 ns 0.66 0.57 0.91 0.70 0.66 0.37R2

!

A different year (2004) with some different yieldoutcomes — in this case, the late planted varietiesresponded much more to the added irrigations thanin 2003. However, the overall relationshipsbetween nymphs and yields is once again veryimpressive for very simple linear regressions. Note,too, that nymphs were usually the best predictor ofyield, though occasionally the regressions wereimproved somewhat when adults were added in aswell. Adults by themselves were very poorpredictors of yield.



Ellsworth/UA


05F4LT

2005

ns ns 0.42 ns 0.74 0.74 0.82 0.70 0.79 0.77 0.88 0.62R2

!

In 2005, we hit yields as high as 4 bales and as lowas 0.5 bales. And the yield : nymph relationshipsheld up with very high degrees of fit.

Of course, as stated before, yields are only part ofthe answer. We need to look at revenues as well.The data I am about to show is still being analyzed.



Ellsworth/UA

50¢ Cotton Marginal Returns

03F4LT

a a

b b

a

b

a a a a a

b

a

b

a

b

a a a a a a

b b

a

b

c c

a a a

b b

a a a a a a

b

a

b b

a a

b

a

b b

a

b b b

a a a a

Ave Spray Cost = $16.71

0

1

2

3

0

1

2

3

0

1

3

4

0

1

3

4

0

1

3

4

0

1

3

4

0

1

2

3

0

1

2

3

0

1

2

4

0

1

2

4

0

2

3

4

0

2

3

4

Looking at 50¢ cotton, and comparing each pair ofcells (variety x planting), we get this relationship.My goal was to look at what happens after a growerplants. This fixes his variety and his planting date.His decisions after this relate to when to terminateirrigations and when to terminate Lygus controls.So each pair of cells shows the amount of revenueabove the minimum baseline for each pair of cells.

Clearly, sometimes no action is the best, mostprofitable course (late irrigated, short-seasoncotton planted early); other times, more aggressiveaction is called for as in the late irrigated, long-season early planted scenario.

Lygus spray costs were held to $16.71 per spray (5-yr ave.; and includes application costs). One extrawater was priced at $12, but this varies widelyacross the state.



Ellsworth/UA


04F3LT

a a

b b

a

b

a a a a a a

b b b

a a a a a a

b b

a a

b b

a

b

c c

a

b b b

a

b

c c

a

b b b

a

b

c c

a

b b b

0

1

2

4

0

1

2

4

0

1

2

3

0

1

2

3

0

1

2

3

0

1

2

3

0

1

2

3

0

1

2

3

0

1

2

3

0

1

2

3

0

1

2

3

0

1

2

3

Again in 2004, the additional irrigations in the lateplanted scenarios really paid off and especially so, ifprotection against Lygus was also extended.However, the short season system (early plantedand early terminated) failed to benefit by anyadditional Lygus control.



Ellsworth/UA


05F4LT

66

00

a a a a a a a a a a a a a a a a a a

b b

c

a a a

b b

a

b b b

a

b b b

a

b

c c

a

b b b

a

b b

c c

a

b b b

011

3

0112

0123

0123

012

6

012

5

0

2

44

0

2

44

0

2

44

0

2

44

0

2

45

0

2

45

In 2005, we got our greatest contrast where as fewas 0 sprays were indicated and as many as 6 sprayswere warranted depending on the productionscenario.

As I said, we are not done analyzing this data asyet. But what is clear is that this 4-way interaction(planting date, termination date, variety, and Lyguscontrol termination) late in the season is dynamicand complex. Yet there are huge pay-outs possibleif the “right” decisions are made.



Ellsworth/UA

Arizona Lygus Guidelines

• Before 1997– Ca. 15–20 Lygus / 100 sweeps

• 1998– 15–20 total Lygus / 100 with nymphs

present and/or 25% square damage

• 1999– 15–20 total Lygus / 100 with ca. 1/3 of total

being nymphs

• After 2000– ‘15:4’ or 15 total Lygus with at least 4

nymphs per 100 sweeps

• Consider natural enemies, fruit retentionand damage levels

• Termination rules?

To summarize a decade’s worth of thresholdresearch with this insect, let me review our cottonguidelines. Prior to 1997, we adopted a long-standing recommendation to spray when therewere 15-20 Lygus / 100 sweeps. In 1998 this wasaltered to include mention of nymphs and thepossibility of using a plant-based measurementsystem (25% internally damaged squares). By1999, the nymph fraction was suggested to be 1/3of the total count. After 2000, we have beenteaching the ‘15:4’ threshold and further suggestedthat other factors be considered.

As to termination rules, these have turned out to beless uniform than what would have beenconvenient and more sophisticated modelingand/or statistical / economic approaches will beapplied to these data in the future.



Ellsworth/UA

Many, many people contribute to and support ourresearch and extension efforts. I wish to thankthem all and especially to those that shouldered thebiggest burden of doing the thousands of sweepsnecessary to collect the samples for these studiesand to Virginia Barkley who manages our crews andlaboratory each year.



Ellsworth/UA

Cotton IncorporatedCotton Incorporated

Arizona Cotton Growers AssociationArizona Cotton Growers Association

Arizona Cotton Research & Protection CouncilArizona Cotton Research & Protection Council

Thanks, too, to Cotton Incorporated and especiallyto Dr. Pat O’Leary who supported much of this workas has the two grower organizations in our state,State Support Committee of the Arizona CottonGrowers Association, and the Arizona CottonResearch and Protection Council.

The Arizona Pest Management Center (APMC) aspart of its function maintains a website, the ArizonaCrop Information Site (ACIS), which houses all cropproduction and protection information for our lowdesert crops, including a PDF version of thispresentation for those interested in reviewing itscontent (http://cals.arizona.edu/crops).

Photo credit: J. Silvertooth



Ellsworth/UA

April 15–19, 2007

Asilomar Conference Center

Pacific Grove, CA

http://www.cevs.ucdavis.edu/lygussymposium

2nd InternationalLygus Symposium

One final note of advertisement! For thosescientists, interested growers and consultants, I’dlike to invite you to attend the 2nd InternationalLygus Symposium in beautiful Pacific Grove, CA,hosted at the wonderful Asilomar ConferenceCenter. The location is spectacular on the Montereycoast, and the scientific exchange that takes placewith participants from around the world shouldprove very beneficial to anyone wishing to learnmore about Lygus and their management in anarray of cropping systems. Please join us 15–19April 2007.

Documents

Plant Bug Thresholds in Arizona Cotton Cotton Conferences, Plant Bugs January 10, 2007 Ellsworth, Ly gus thresholds in Arizona 1 JanuaryT 10, 2007 Ellsworth/UA Plant Bug Thresholds