sign in sign up
<<
Home , Centaurea diffusa, Pelochrista medullana, Pterolonche, Pterolonche inspersa

Weed Science Society of America

Distribution, Biology, and Management of Diffuse Knapweed ( diffusa) and Spotted Knapweed (Centaurea maculosa) Author(s): Roger L. Sheley, James S. Jacobs and Michael F. Carpinelli Source: Weed Technology, Vol. 12, No. 2 (Apr. - Jun., 1998), pp. 353-362 Published by: Weed Science Society of America and Allen Press Stable URL: http://www.jstor.org/stable/3988400 . Accessed: 06/08/2014 18:42

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

Weed Science Society of America and Allen Press are collaborating with JSTOR to digitize, preserve and extend access to Weed Technology.

http://www.jstor.org

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions Weed Technology. 1998. Volume 12:353-362

Distribution, Biology, and Management of Diffuse Knapweed () and Spotted Knapweed (Centaurea maculosa)l

ROGER L. SHELEY,JAMES S. JACOBS, and MICHAEL F CARPINELLI2

Abstract: Diffuse knapweed, a biennial or short-livedperennial, and spotted knapweed, a perennial, are taprootedEurasian weeds invading rangelandin the western and .Knapweed (Centaurea spp.) invasion is associated with reductionsin biodiversity,wildlife, and livestock forage, and increased erosion. Spotted knapweed grows to about 1 m and usually has purple , whereas diffuse knapweed is slightly shorter,usually with white flowers. Persistent bracts on diffuse knapweed bear a rigid terminal spine about 8 mm long with four or five pairs of shorter lateral spines. Bracts on spotted knapweed have dark spotted tips. Knapweed managementinvolves a com- bination of containing infestations and control efforts. Hand pulling in areas with small infestations can be effective for controlling spotted and diffuse knapweeds. Picloram applied at 0.28 kg ha-' provides control for about 3 yr. Effective long-term control of knapweeds requires periodic appli- cations of picloram, which are only cost-effective on highly productive range sites with a residual grass understory. About 12 species have been released for knapweed biocontrol. pro- duction has been reduced by 46% by feeding in the flower heads. Although insects have not reduced spotted knapweed densities, they may stress the weed and shift the competitive balance to associated species. Sheep grazing reduces the density of very young seedlings and may limit seedling recruitment into the population. In areas without a residual understory of desired species, revegetation of knapweed-infestedrangeland is required.Components of any integratedweed man- agement program are sustained effort, constant evaluation, and the adoption of improved strategies. Nomenclature: Picloram, 4-amino-3,5,6-trichloro-2-pyridinecarboxylicacid; diffuse knapweed, Cen- taurea diffusa Lam. #3 CENDI; spotted knapweed, Centaurea maculosa Lam. # CENMA. Additional index words: Noxious rangelandweeds, integratedweed management,sustainable weed management,CENDI, CENMA.

INTRODUCTION drought, but can tolerate a wide range of precipitation and temperature (Harris and Cranston 1979). Watson Diffuse and spotted knapweeds are native to grass- and Renny (1974) reported that diffuse knapweed has land steppes of southeastern Europe and Asia Minor. been observed growing at elevations ranging from 150 Diffuse knapweed grows in the eastern Mediterranean to over 900 m. More recently, diffuse knapweed has area, in western Asia, and from the southern part of been found thriving at elevations up to 2,600 m. former U.S.S.R. to western Germany (Rees et al. The native range of spotted knapweed is central Eu- 1996). Diffuse knapweed grows best on fertile, well- rope and east to central , Caucasia, and western watered Cryoborolls, mesic Argiudolls, and mesic Ha- Siberia (Rees et al. 1996). In Europe, spotted knap- pludolls in open and uncultivated sites with a summer weed is most aggressive in the forest steppe on Mol- ' Received for publication June 27, 1997, and in revised form February lisols and mesic Aridisols, but can form dense stands 12, 1998. Published with approvalof the directors,Montana Agricultural Ex- in more moist areas on well-drained including periment Station, as JournalJ-5136. gravel, and in drier sites where summer precipitation 2 Assistant Professor,Post-Doctoral Research Associate, and GraduateRe- search Assistant, respectively, Plant, and EnvironmentalScience, Mon- is supplemented by runoff. Spotted knapweed does not tana State University, Bozeman, MT 59717-3120. compete with vigorously growing grass in moist areas, 1 Letters following this symbol are a WSSA-approvedcomputer code from Composite List of Weeds, Revised 1989. Available from WSSA, 810 East nor with diffuse knapweed in steppic grassland. It oc- 10th Street, Lawrence, KS 66044-8897. curs on the moist end of the range of diffuse knapweed

353

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions SHELEY ET AL.: DIFFUSE AND SPOTTED KNAPWEEDS

w | | ~~~~~~~~~Legend q Diffuse| W Miles knapweedpresent !11 fit,l r Diffusez m knapweed. undetected 0 100 200

Figure 1. The currentdistribution of diffuse knapweed determinedfrom herbariumrecords and interviews with weed authoritiesfrom 15 states in the western United States.

(Harris and Cranston 1979). Lacey et al. (1995) re- 1920. This weed had spreadto 20 counties in the Pacific ported that spotted knapweed has been observed at el- Northwest by 1960 and to 48 counties by 1980. Between evations ranging from 578 to over 3,040 m and in pre- 1980 and the present, the known range of spotted knap- cipitation zones ranging from 20 to 200 cm, annually. weed rapidly increased to include 326 counties in the western United States, including every county in Wash- ington, Idaho, Montana, and Wyoming (Figure 2). DISTRIBUTIONOF DIFFUSEAND SPOTTED KNAPWEED IDENTIFICATION Diffuse and spotted knapweed were introduced to from as contaminantsin Diffuse and spotted knapweeds are deeply taprooted (Medicago sativa L.) (Muller et al. 1988; Roche et al. rosette-forming of the . Basal 1986). Spotted knapweed was also apparentlyintroduced rosette are born on short pedicels and grow up to through discarded soil used as ship ballast (Roche et al. 20 cm long and 5 cm wide. Rosettes are deeply divided 1986). Diffuse knapweed was first recorded in North (once or twice) into lobes on both sides of the center America in in 1907 (Howell 1959), and vein (Figures 3a and 3b). Lobes are oblong with the spotted knapweed was first recorded in Victoria, British broadest part above the middle. Flowering stems are Columbia, in 1893 (Groh 1944). Both knapweeds spread erect 2 to 12 cm tall. Diffuse knapweed has many further in domestic alfalfa and hay before they spreadingbranches, giving it a ball-shaped, were recognized as serious problems (Roche et al. 1986). appearance and mobility, whereas spotted knapweed We have documentedthe geographic spreadof diffuse stems branch on the upper half of the stem and are not and spotted knapweeds in the western United States up ball shaped. Stem leaves for both species are alternate, to 1980 using herbariumrecords compiled by Forcella sessile, have few lobes, or are linear and entire, and are and Harvey (1980), and currentdistribution through in- reduced toward the stem apex. The uppermostleaves are terviews with weed authoritiesin the region (Figure 1). small and simple. Flower heads are ovate to oblong, 6 Diffuse knapweed was first recordedfrom an alfalfa field mm wide and 12 mm long, and are solitary or born in at Bingen, KlickitatCounty, Washington(Roche and Tal- clusters of two or three at the branch ends. Involucre bott 1986). By 1980, it had spread to 26 counties in the bracts are foliaceous, ovate, and yellow-green to brown Pacific Northwest. Between 1980 and the present,diffuse below. Margins have a soft, spinelike fringe with the knapweed had extended its range to include 204 counties center spine longer and spreading on diffuse knapweed in 14 western states. In the United States, spotted knap- (Figure 3c), whereas the center spine is shorterthan the weed was limited to the San Juan Islands, WA, until lateral ones on spotted knapweed. Spotted knapweed

354 Volume 12, Issue 2 (April-June) 1998

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions WEED TECHNOLOGY

Spottedknapweed present Miles K:= ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~ D Spottedknapweed undetected 0 100 200

Figure 2. The currentdistribution of spotted knapweed determinedfrom herbariumrecords and interviews with weed authoritiesfrom 15 states in the western United States. also has a black margin at the bract tips and obvious a loss of 220 elk annuallyin Montanabecause of spotted longitudinal veins (Figure 3d), whereas bract margins of knapweed infestations on winter range. diffuse knapweed are buff or brown. Most flowers of Knapweedinfestation on bunchgrassrangeland is also diffuse knapweed are white, but occasionally, they are detrimental to water and soil resources. Lacey et al. rose to lavender, with about 10 to 15 flowers on each (1989) determinedthat surface water runoff and stream flower head (Stranget al. 1979). Spotted knapweedflow- sediment yield were 56 and 192% higher, respectively, ers are purple to pink, rarely white, with 25 to 35 flow- for spotted knapweed-dominated sites compared to ers/head. Diffuse knapweed flowers from late June to bunchgrass-dominatedsites. Baregroundand water infil- September.Flower heads fall from the plant when seeds trationrates were greateron sites with unclipped bunch- are mature. Seeds are 2 to 3 mm long and dark brown. grass than on those with spotted knapweed (Lacey et al. The pappus is generally absent or slightly less than 1 1989). mm long. Spotted knapweed flowers bloom from June to October.The flower heads remainon the usually plant. Nutritive Content. Diffuse and spotted knapweed ro- Seeds are 3 mm long, brown to with oval, black, pale settes and young flower shoots have some nutritivevalue longitudinallines. Seeds bear a pappus of simple bristles (Fletcher 1961). Nutrient analysis of spotted knapweed that are 1 to 2 mm long and persistent. collected before flowering showed neutraldetergent fiber at 24.2 to 53.0% (dry wt), ether extract 3.1 to 9.0%, BIOLOGYAND ECOLOGY crude protein 6.2 to 18.2%, total nonstructuralcarbo- Impacts. Diffuse and spotted knapweeds reduce live- hydrates 11.0 to 27.5%, ash 4.9 to 9.3%, in vitro dry stock forage. Watson and Renny (1974) found decreased matter digestibility 53.2 to 61.8%, and gross energy bluebunch wheatgrass [ (Parsh) 4,088 to 4,539 callg (Kelsey and Mihalovich 1987). Kel- Love] yield was correlatedwith increased productionof sey and Mihalovich (1987) observed crude protein and spotted knapweed. Bluebunch wheatgrass-roughfescue nonstructuralcarbohydrates were most concentrateddur- (Festuca scabrella Torr.) production was reduced by ing the spring,becoming more fibrouswith lower protein 88% by knapweed invasion (Watson and Renny 1974). and carbohydratelevels as stems matured during the Knapweeds also have a negative impact on wildlife. summer. Diffuse and spotted knapweeds are grazed by Elk (Cervus canadensis) use, as estimated by pellet deer (Odocoileus spp.) and sheep (Ovis aries) (Olson et groups/hectare,was reduced by 98% on spotted knap- al. 1997). These knapweeds also provide nectar and pol- weed-dominated range compared to bunchgrass-domi- len for domestic bees (Apis mellifera) (Watsonand Ren- nated sites (Hakim 1979). Spoon et al. (1983) predicted ny 1974).

Volume 12, Issue 2 (April-June) 1998 355

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions SHELEY ET AL.: DIFFUSE AND SPOTTED KNAPWEEDS

a

6I

//

'ii

b

Figure 3. Diffuse and spotted knapweed rosettes and flower heads. a-diffuse knapweed rosette, b-spotted knapweed rosette, c-diffuse knapweed flower head showing characteristicinvolucre bracts, d-spotted knapweed flower head showing characteristicinvolucre bracts.

356 Volume 12, Issue 2 (April-June) 1998

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions WEED TECHNOLOGY

Phenology. Diffuse knapweed is a semelparous peren- are transportedlong distances in mud, and commonly nial (Nolan and Upadhyaya 1988). It may complete its become attached to or drop into shoes. Seeds of both life cycle in a single year, or live for several years as a diffuse and spotted knapweeds spreadthrough rivers and rosette before flowering. It can also continue to grow along watercourses,and are transportedin crop seed and after producing seeds to flower again the following year hay. (Roche and Roche 1993). Spotted knapweed is a peren- nial that lives up to 9 yr and is capable of producing . Diffuse and spotted knapweed seeds ger- seeds in each year (Boggs and Story 1987). Seed pro- minate over a wide range of environmentalconditions. duction of diffuse knapweed ranges from 11,200 to Watson and Renny (1974) reported germination occurs 48,100 seeds/M2, and for spotted knapweed 5,000 to for both species at temperaturesranging from 7 to 34 C, 40,000 seeds/M2 (Shirman 1981). Site conditions and with optimum germination at about 24 and 19 C for precipitationduring the growing season have the greatest diffuse and spotted knapweed, respectively. Seeds ger- effect on the number of seeds produced per year. More minate in the light and dark; however, maximum ger- seeds are produced during wet years (Shirman 1981). mination occurs in an alternatinglight and darkenviron- Seeds germinate in the fall and early spring when ment (Watson and Renny 1974). They also found that moisture and temperatureare suitable (Watson and Ren- 20 to 40% of diffuse and spotted knapweed seeds, re- ny 1974). Fall and early spring germinatingseedlings are spectively, germinatedwhen removed from flower heads capable of maturing into seed-producing adults in 1 yr at maturity,which improved to 68 and 80% 25 d after (Shirman 1981). Seedlings develop into rosettes, and removal. Optimumemergence occurs at the soil surface maximal root growth occurs at this stage (Watson and with little emergence below 5-cm depth (Watson and Renny 1974). If rosettes do not bolt, they die back to Renny 1974). Seedling mortality averages about 12%, the root crown, which serves as the perennatingpart of but can be as high as 55% (Watson and Renny 1974). the plant over winter. Root crowns form rosettes in the Seeds of diffuse and spotted knapweeds have three early spring and bolt in early May. Diffuse knapweed types of germinationbehavior: nondormant seeds, which produces one stem, rarely two, whereas spotted knap- germinate in the dark; light-sensitive dormant seeds, weed produces one to six stems (Watson and Renny which germinate after exposure to red light; and light- 1974). Flowering buds form in early June, and flowering insensitive dormant seeds, which do not respond to ex- occurs from July through September.Mature seeds form posure to red light (Nolan and Upadhyaya 1988). Indi- by mid-August. Most spotted knapweed seeds are shed vidual plants can produce seeds of all three germination upon maturitywith few overwintering in the seedheads, types. whereas diffuse knapweed delays seed shed. Davis et al. (1993) determined that although spotted knapweed seeds decreased by 95% over a 7-yr period, Spread. Seed (achene) dispersal of diffuse knapweed is approximately40,000 viable seeds/ha remained.Wallan- mainly by wind. Matureplants break off at ground level der et al. (1995) reportedthat up to 22% of spottedknap- and tumble in the wind or become attached to the un- weed seeds can remain viable after passing through the dercarriageof vehicles and equipment(Watson and Ren- digestive tractsof sheep and mule deer (Odocoileus hem- ny 1974). Seeds are released through a small opening at ionus spp. hemionus). In that study, viability of seeds the tip of the flower head when the persistentdry flowers passed throughthe digestive tracts decreased as time af- are dislodged, usually by the action of tumbling (Roche ter ingestion increased. et al. 1986). Seeds are also carried in mud and by ro- Diffuse knapweed does not reproduce vegetatively. dents. In contrast, spotted knapweed plants do not break Spotted knapweed is able to send lateral shoots below off at ground level, and populations are largely extended the soil surface that can form rosettes up to 3 cm from through peripheralenlargement of existing stands (Wat- the parent plant (Watson and Renny 1974). Multiple ro- son and Renny 1974). Bracts of the flower heads open settes on a single spotted knapweedroot crown are com- when dehydrated,2 to 3 wk after maturity,and move- mon. ment of the stem by wind or passing can flick the loosely held achenes up to 1 m from the parentplant . The allelopathic compound, cnicin, has (Watson and Renny 1974). Long-distance transportoc- been isolated from diffuse and spotted knapweed leaves curs when achenes become attachedto passing animals, and shoots (Fletcher and Renny 1963). In vitro experi- or by rodents and birds. Spotted knapweed flower heads ments show this chemical to reduce germinationin crest- also become attached to the undercarriagesof vehicles, ed wheatgrass [Agropyroncristatum (L.) Gaertn.],blue-

Volume 12, Issue 2 (April-June) 1998 357

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions SHELEY ET AL.: DIFFUSE AND SPOTTED KNAPWEEDS bunch wheatgrass,and rough fescue (Kelsey and Locken Inventory and Mapping. The primary objective of 1987). In spring and fall soil samples, Kelsey and Lock- weed inventory and mapping is to accurately identify en (1987) found cnicin in soil samples collected in Au- and delineate lands infested with populations of un- gust only, and in a level not toxic in in vitro experiments. wanted plants (Cooksey and Sheley 1996). The goal is Later, they determinedthat the effects of resource com- to record weed species present, areas infested, weed den- petition were more importantthan allelopathy (Locken sity, rangeland under threat of invasion, soil and range and Kelsey 1987). types, and other site factors pertinentto successful man- agement of knapweed-infested rangeland. Information Disturbance. Diffuse and spotted knapweed densities from an inventory should be incorporatedinto a map that are correlated with the degree of soil disturbance.The shows the location, type, and size of knapweed infesta- greater the disturbance,the higher the density (Watson tions. Accurate mapping is critical in developing an in- and Renny 1974). Once established, these weeds are able tegrated knapweed management program. An example to form monotypic stands because their age class hier- of a weed survey and mapping system is provided by archy allows them to occupy all available niches (Sheley Cooksey and Sheley (1996). and Larson 1996). Disturbance allows for rapid estab- lishment and spread; however, both species are capable Prevention. Preventing diffuse and spotted knapweeds of invading well-managed rangelands(Lacey et al. 1990; from spreadingonto adjacentrangeland is the most cost- Myers and Berube 1983; Tyser and Key 1988). effective managementstrategy. Weed seed dispersal can be limited by not driving vehicles throughweed-infested areas when seeds are present, not grazing livestock in DIFFUSEAND SPOTTEDKNAPWEED weed-infested areas during flowering and seeding or MANAGEMENT holding animals for 7 d before moving to uninfested ar- Management Objectives. Land use objectives must be eas, and using hay that is free of knapweed seeds (Sheley developed before diffuse and spotted knapweedmanage- et al. 1996a). Knapweed spread can also be limited by ment plans can be designed. In many cases, especially detecting and eradicatingweed introductionsearly, min- for large-scale infestations, strictly killing knapweeds is imizing soil disturbance, and containing neighboring an inadequate objective. Sheley et al. (1996b) proposed knapweed infestations (Sheley et al. 1996a). Bounty pro- that a generalized objective could be to develop an eco- grams have been successfully used to encouragethe ear- logically healthy plant community that is weed resistant ly detection of diffuse and spotted knapweed in Mon- and meets other land-use objectives such as livestock tana. forage, wildlife habitat, or recreation. Once the desired Biological Control. Various plant community has been determined, an ecologically naturalenemies have been released based integrated weed management strategy can be de- as biological control agents for diffuse and spot- veloped. ted knapweeds (Table 1) (Rees et al. 1996). Most bio- control techniquesuse insect larvae that damage the host Integrated Weed Management. The magnitudeand root, shoot, , or flower, resulting in reduced seed pro- complexity of the diffuse and spotted knapweedproblem duction. Two seedhead-feeding flies ( indicates that successful managementrequires the adop- and U. quadrifasciata) are well established on diffuse tion of integrated strategies. Integratedknapweed man- and spotted knapweeds. These fly species lay their eggs agement involves the use of several techniquesin a well- inside the knapweed flower buds in June. Their larvae planned, coordinated,and ecologically based strategy to induce galls in the flower heads where they feed on the maintain desired plant communities, or shift plant com- phloem, which can reduce seed productionby as much munities to those that are desired (Sheley et al. 1996b). as 50% (Story et al. 1989). Larvae of the , Metz- Inventory and mapping are the first phases of any inte- neria paucipundctella, feed on the flowers and seeds of grated weed managementprogram. The second phase in- spotted knapweed and may reduce seed production by cludes prioritizing, choosing, and strategically imple- about 20% (Story et al. 1989). menting management techniques for a particularunit. Five root-mininginsect species have been released on The third phase is adopting proper grazing management diffuse and spotted knapweeds. A root moth (Agapeta practices as a portion of the integratedprogram. The in- zoegana) and a root weevil (Cyphocleonusachates) have tegrated weed management program must fit into an been observed to damage roots of spotted knapweed overall range managementplan. plants. The root beetle, , has

358 Volume 12, Issue 2 (April-June) 1998

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions WEED TECHNOLOGY

Table 1. Currentbiological control insects released on diffuse and spotted knapweeds. Agent Common name Weed attacked Type of agent States established Sulphur knapweed moth or yellow- CENMA, CENDIa Root-boringmoth MT, OR, WA winged knapweed root moth Bangansternusfausti Broad-nosed seedhead weevil CENMA, CENDI Seedhead weevil MT, OR, UT Chaetorellia acrolophi Knapweed peacock fly CENMA Seedhead weevil MT, OR Cyphocleonusachates Knapweed root weevil CENMA Root-boring/gallweevil CO, MT, OR, WA Lesser knapweed flower weevil CENMA, CENDI Seedhead weevil MT, OR, WA Larinus obtusus Blunt knapweed flower weevil CENMA Seedhead weevil WA Metzneriapaucipunctella Spotted knapweed seedhead moth CENMA Seedhead moth ID, MT, OR, WA medullana Brown-winged root moth CENMA, CENDI Root-boringmoth Not established in United States inspersa Gray-winged root moth CENDI Root-boring moth MT Sptenopterajugoslavica Bronze knapweed root borer CENDI Root-boring/gallbeetle CA, ID, MT, OR, WA Terellia virens Green clearwing fly CENMA Seedhead fly MT, OR Urophora affinis Banded gall fly CENMA, CENDI Seedhead fly CA, ID, MT, OR, UT, WA UV knapweed seedhead fly CENMA, CENDI Seedhead fly ID, MT, OR, UT, WA

aWSSA-Bayer Code: CENMA-spotted knapweed, CENDI-diffuse knapweed.

been released on diffuse knapweed.Two other root provideda flushof nutrientsavailable to remainingde- , ( and Pterolonche insper- sirablespecies. sa), have been releasedin Montana.Distribution of bio- Cultivation.A single, shallow (7 cm) cultivationwas controlagents is currentlyunderway. ineffectivein controllingdiffuse knapweed, especially if In additionto insect biocontrolagents, there are fun- crownswere incompletelycovered (Atkinson and Brink gal and bacterialpathogens that infect diffuseand spot- 1953; Popova 1960). Cultivationto 18 cm eliminated ted knapweeds. Scierotinia sclerotiorum is a common knapweedwith subsequentvigorous grass growth(Po- soil fungusnative to NorthAmerica that infects diffuse pova 1960).However, Velagala (1996) foundsingle cul- and spottedknapweeds, which can cause wilt anddeath tivationto 20 cm increasedspotted knapweed density undersome conditionsand shift the competitionbalance over the control,but reducedspotted knapweed biomass to associatedgrasses (Ford 1989; Jacobset al. 1996). 1 yr aftertreatment. In that study,cultivation enhanced Spottedknapweed is infectedby the bacteriaPseudo- the establishmentof intermediatewheatgrass [Elytriga monas syringae pv. syringae (Kearing and Nowierski intermedia(Host) Nevski] seedlings. 1997). Pathogensare often associatedwith insectinjury to the plant. Fertilization.Nitrogen fertilizer enhances spotted knap- weed becausespotted knapweed captures newly avail- Burning.A single,low intensityfire does not effectively able resourcesbefore neighboring desirable species (Po- controldiffuse and spottedknapweeds because it is not pova 1960; Story et al. 1989). In a 2-yr study,Sheley hot enoughto preventresprouting from crownsor rees- andJacobs (1997) foundthat picloram plus fertilizerdid tablishmentfrom viable seeds in the soil (Renny and not interactto affect eitherspotted knapweed density or Hughes 1969). A single low intensityfire increasedthe grassyield. However,fertilizer applied at 32 plus 40 kg/ cover anddensity of bothweeds in northernWashington ha of nitrogenand phosphorus,respectively, increased withoutaltering the residual,desirable understory spe- grassyield on those sites with a substantialgrass under- cies (Sheley and Roche 1982). Similarly,spotted knap- story. weed increasedabout sixfold within 2 yr after a con- trolledfire on a forestedsite in Montana.Fires may cre- Grazing. Spotted knapweedcan tolerate defoliation. ate the typeof disturbancethat promotes the colonization However,severe defoliation will reduceroot, crown, and of knapweeds. abovegroundgrowth (Kennet et al. 1992).Low to mod- Herbicideefficacy on diffuse and spottedknapweeds erate levels of grazingof spottedknapweed by cattle may increasewhen appliedafter burning (Lacey et al. (Bos taurus), sheep, and goats (Capra hircus) have been 1995). Picloramapplied at 0.28 kg/ha provided100% observed in Montana.Cattle appearto prefer grasses controlof both weeds 2 yr aftera postburningapplica- over spottedor diffuse knapweed(Lacey et al. 1995). tion (Sheleyand Roche 1982).Interestingly, residual un- Althoughrosettes of the firstyear's growth are nutritious derstorygrass cover and density on burnedplots in- andedible, they are difficult for cattleto eat becausethey creasedover nonburnedplots wherepicloram had been are closely appressedto the ground(Popova 1960). Ma- applied.The authorssuggested that burningmay have tureknapweed plants are fibrous and coarse, which make

Volume 12, Issue 2 (April-June) 1998 359

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions SHELEY ET AL.: DIFFUSE AND SPOTTED KNAPWEEDS them less desirable, and the rough flowering stems along growth stages (Lacey et al. 1995). Percent control de- with spines on the floral bracts can irritategrazing ani- clined to below 83% when this combination mals. was applied at rosette, flowering, and after flowering Controlled grazing duration and repeated grazing by growth stages. sheep of spotted knapweed when associated grasses were Dicamba (1.12 kg/ha) and 2,4-D (2.24 kg/ha) provid- dormant altered the age class distribution of spotted ed inconsistent and short-termcontrol of spotted and dif- knapweed (i.e., fewer but older and larger plants) and fuse knapweeds (Lacey et al. 1995). For long-term con- reduced reproduction (Olson et al. 1997). Because of trol, these must be applied annually until the dormancy, there was minimal impact on associated seedbank is depleted through attrition.Control with di- grasses. Sheep consumed all 1-yr-old and a majority of camba and 2,4-D was similar to that of clopyralid plus 2-yr-old plants, suggesting that long-term grazing may 2,4-D when applied at the bud stage, but lower than that provide control of spotted knapweed by limiting recruit- of picloram 3 yr after application. ment. Angora goats have been successfully used to con- trol spottedknapweed on the Lee Metcalf National Wild- Mowing. Long-term effects of mowing on diffuse and life Refuge in Montana (P. Gonzalez, personal commu- spotted knapweed population densities are unknown. A nication). single mowing at the bud growth stage reducedthe num- ber of stems that producedseeds from about 34 (control) Hand Pulling. Persistent and careful hand pulling can to below 8/M2 on both species (Watsonand Renny 1974). control diffuse and spotted knapweeds (Lacey et al. In that study, mowing at the flower stage reduced the 1995). Since regrowth can occur from both crowns and number of flowering stems to 0.3/m2. Mowing at both viable seeds in the soil, entire plants must be removed the bud and flower stage did not have an additive effect. before they produce seeds each year. Knapweed plants Watson and Renny (1974) also found that mowing at the are best pulled when the soil is wet because the crown flowering stage or both bud and flowering stages reduced is more completely removed. Flowering plants should be the percent germinationof the seeds formed. Secondary transported offsite for disposal in a manner ensuring flowering, with capitula forming, was observed in dif- seeds are not dispersed. fuse knapweed populations.In a greenhouse study, some spotted knapweed plants produced flowers even though Herbicides. Picloram, clopyralid (3,6-dichloro-2-pyri- they were clipped monthly from June through Septem- dinecarboxylic acid), dicamba (3,6-dichloro-2-methoxy- ber, suggesting the response of plants to mowing will benzoic acid), and 2,4-D [(2,4-diclorophenoxy)acetic vary with environmentalconditions (Kennet et al. 1992). acid] effectively control diffuse and spotted knapweeds on rangeland.In an 8-yr study, Davis (1990) found pic- Revegetation. In areas where residual plant species are loram at 0.28 kg/ha provided nearly 100% spotted knap- absent, long-term control of diffuse and spotted knap- weed control for 3 to 5 yr. The period of control tends weeds is unlikely because desirablespecies are not avail- to be shorteron coarse soils or as precipitationincreases able to occupy niches opened by the control procedure (Lacey et al. 1995). Application timing does not affect (James 1992; Sheley et al. 1996b). In these areas, estab- control (C. Duncan and M. Halstvedt, personal com- lishing competitive plants is essential for the successful munication).During the period of control, residualgrass- managementof knapweeds and the restorationof desir- es increased by 200 to 700% (Davis 1990). Sheley and able plant communities. Revegetation with aggressive Jacobs (1997) found that controlling spotted knapweed species has been shown to inhibit the reinvasion of knap- with picloram increased grass yield by an average of weeds (Borman et al. 1991; Hubbard 1975; Larson and 1,500 kg/ha, while providing nearly 100% control for 2 McInnis 1989). However, there is a lack of data on re- yr. Griffith and Lacey (1991) suggested that as site pro- liable and affordablerevegetation methods. ductivity, value of unit months, and rate of spot- Typically, revegetation of diffuse and spotted knap- ted knapweed spread to new areas increased, economic weed-infested rangeland involves late-fall cultivation, returnsincreased relative to picloram treatmentcosts. In followed immediately by a dormant seeding of grass. contrast, picloram treatment became less cost-effective Grass and knapweeds emerge the next spring; however, as knapweed use by cattle increased. knapweeds usually emerge first. At that time, Clopyralid (0.27 kg/ha) and clopyralid (0.21 kg/ha) [N-(phosphonomethyl)glycine]may be applied before plus 2,4-D (1.12 kg/ha) provided spotted knapweed con- grass seedlings emerge to control knapweed and downy trol similar to picloram when applied during bolt or bud brome (Bromus tectorumL.) seedlings. As long as there

360 Volume 12, Issue 2 (April-June) 1998

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions WEED TECHNOLOGY is adequate spring precipitation, both grass and knap- Fletcher,R. A. and A. J. Renny. 1963. A growth inhibitorfound in Centaurea spp. Can. J. Plant Sci. 43:475-481. weed seedlings survive. If grass seedlings survive until Forcella, F and S. J. Harvey. 1980. New and Exotic Weeds of Montana. II: mid-summer,a reduced rate of 2,4-D or mowing can be Migration and Distribution of 100 Alien Weeds in NorthwesternUSA, 1881-1980. Bozeman, MT: Montana State University Herbarium.117 p. applied to weaken knapweed plants. Mowing and/or2,4- Ford, E. J. 1989. Sclerotinia as a mycoherbicide.In P. K. Fay and J. R. Lacey, D applications may need repeating. Grass seedling es- eds. Knapweed Symposium Proceedings. Bozeman, MT: Montana State University. pp. 182-189. tablishmentis enhanced by increasing seeding rate (Ve- Griffith,D. and J. R. Lacey. 1991. Economic evaluation of spotted knapweed lagala et al. 1997). Although the species most effective (Centaurea maculosa) control using picloram. J. Range Manage. 44:42- for revegetation of diffuse and spotted knapweed-infest- 44. Groh, H. 1944. CanadianWeed Survey. 2nd Annual Report. CanadaDepart- ed rangeland depends on site condition, we believe a ment of Agriculture.74 p. healthy, weed-resistantplant community consists of a di- Hakim, S.E.A. 1979. Range Conditionon the ThreemileGame Range in West- ern Montana. M.S. thesis. University of Montana, Missoula, MT. 62 p. verse group of species that occupies most of the niches Harris,P. and R. Cranston. 1979. An economic evaluation of control methods (Sheley et al. 1996b). for diffuse and spotted knapweed in western Canada. Can J. Plant Sci. 59:375-382. Howell, J. T. 1959. Distributionaldata on weedy thistles in western North America. Leafl. West. Bot. 9:17-32. ADOPTINGPROPER GRAZINGMANAGEMENT Hubbard,W. A. 1975. Increased range forage production by reseeding and PRACTICES the chemical control of knapweed. J. Range Manage. 28:406-407. Jacobs, J. S. and R. L. Sheley. 1997. Relationship among Idaho fescue de- foliation, soil water, and spotted knapweed emergence and growth. J. On areas with a competitive grass stand, properman- Range Manage. 50:258-262. agement ensures that grasses remain strong and vigor- Jacobs, J. S., R. L. Sheley, and B. D. Maxwell. 1996. Effect of Sclerotinia sclerotiorum on the interference between bluebunch wheatgrass (Agro- ous, thereby minimizing knapweed encroachment.Prop- pyron spicatum) and spotted knapweed (Centaurea maculosa). Weed er grazing is essential to maintainingcompetitive desired Technol. 10:13-21. James, D. 1992. Some principles and practices of desert revegetationseeding. plants. Sheley et al. (1997) found that a single moderate Arid Lands Newsl. 32:22-27. (< 60%) grass defoliation did not enhance the invasion Kearing, S. A. and R. M. Nowierski. 1997. First reportof stem and bud blight by Pseudomonas syringae pv. Syringae on spotted knapweed (Centaurea of diffuse knapweed into either a pristine bluebunch maculosa Lam.). Plant Dis. 81:113. wheatgrass/needle-and-thread (Stipa comata Trin. & Kelsey, R. G. and L. J. Locken. 1987. Phytotoxic properties of cnicin, a Rupr.)or a crested wheatgrasscommunity. Spotted knap- sequiterpene lactone from Centaurea maculosa (spotted knapweed). J. Chem. Ecol. 13:19-33. weed emergence and growth was enhancedby increasing Kelsey, R. G. and R. D. Mihalovich. 1987. Nutrient composition of spotted Idaho fescue (Festuca idahoensis Elmer) defoliation knapweed (Centaurea maculosa). J. Range Manage. 40:277-281. Kennet, G. A., J. R. Lacey, C. A. Butt, K. M. Olson-Rutz, and M. R. Haf- over 60% (Jacobs and Sheley 1997). Increasing defoli- erkamp. 1992. Effects of defoliation, shading and competitionon spotted ation increased soil moisture availabilityto spottedknap- knapweed and bluebunch wheatgrass.J. Range Manage. 45:363-369. Lacey, C. A., J. R. Lacey, P. K. Fay, J. M. Story, and D. L. Zamora. 1995. weed. To minimize weed invasion, grazing systems ControllingKnapweed in MontanaRangeland. Montana State University should alter the season of use; rotate or combine live- Cooperative Extension Service Circ. 311. 17 p. stock types and pastures, which allows grazed plants to Lacey, J., P. Husby, and G. Handl. 1990. Observationson spotted and diffuse knapweed invasion into ungrazed bunchgrass communities in western recover before being regrazed; and promote litter accu- Montana. Rangelands 12:30-32. mulation (Sheley et al. 1996a, 1997). Lacey, J. R., C. B. Marlow, and J. R. Lane. 1989. Influence of spotted knap- weed (Centaurea maculosa) on surface water runoff and sediment yield. Weed Technol. 3:627-631. Larson, L. L. and M. L. McInnis. 1989. Impact of grass seedlings on estab- LITERATURECITED lishment and density of diffuse knapweed and yellow starthistle.North- west Sci. 63:162-166. Atkinson, T. G. and V. C. Brink. 1953. Progress Report on the Biology and Locken, L. J. and R. G. Kelsey. 1987. Cnicin concentrationsin Centaurea Control of Diffuse Knapweed (Centaurea diffusa Lam.) in British Co- maculosa (spotted knapweed). Biochem. Syst. Ecol. 15:313-320. lumbia. Vancouver,BC: University of British Columbia. Muller, H., D. Schroeder, and A. Gassmann. 1988. Agapeta zoegana (L.) Boggs, K. W. and J. M. Story. 1987. The population age structureof spotted (:Cochylidae), a suitable prospect for biological control of knapweed (Centaurea maculosa) in Montana. Weed Sci. 35:194-198. spotted and diffuse knapweed, Centaurea maculosa Monnet De La Borman, M. M., W. C. Krueger,and D. E. Johnson. 1991. Effects of estab- Marck and Centaurea diffusa Monnet De La Marck (Compositae) in lished perennial grass on yields of associated annual weeds. J. Range North America. Can. Entomol. 120:109-124. Manage. 44:318-326. Myers, J. H. and D. E. Berube. 1983. Diffuse knapweed invasion into range- Cooksey, D. and R. Sheley. 1996. MontanaNoxious Weed Survey and Map- land in the dry interior of British Columbia. Can. J. Plant Sci. 63:981- ping System. Montana State University Cooperative Extension Service 987. MontGuide 9613. 4 p. Nolan, D. G. and M. K. Upadhyaya. 1988. Primaryseed dormancy in diffuse Davis, E. S. 1990. Spotted Knapweed (Centaurea maculosa Lam.) Seed Lon- and spotted knapweed. Can. J. Plant Sci. 68:775-783. gevity, Chemical Control and Seed Morphology. M.S. thesis. Montana Olson, B. E., R. T Wallander,and J. R. Lacey. 1997. Effects of sheep grazing State University, Bozeman, MT 109 p. on a spotted knapweed-infestedIdaho fescue community.J. Range Man- Davis, E. S., P. K. Fay, T. K. Chincoine, and C. A. Lacey. 1993. Persistence age. 50:386-390. of spotted knapweed (Centaurea maculosa) seed in soil. Weed Sci. 41: Popova, A. 1960. Centaurea diffusa Lam., a steppe pasture weed in the Cri- 57-61. mea. Bot. Zh. (Moscow) 45:1207-1213. [English translation] Fletcher,R. A. 1961. A GrowthInhibitor Found in Centaureaspp. M.S. thesis. Rees, N. E., P C. Quimby, Jr., G. L. Piper, E. M. Coombs, C. E. Turner,N. University of British Columbia, Vancouver,BC. R. Spencer,and L. V. Knutson. 1996. Biological Control of Weeds in the

Volume 12, Issue 2 (April-June) 1998 361

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions SHELEY ET AL.: DIFFUSE AND SPOTTED KNAPWEEDS

West. Western Society of Weed Science, USDA-ARS. Bozeman, MT: Sheley, R. L., T. J. Svejcar, and B. D. Maxwell. 1996b. A theoreticalframe- Montana State University. 144 p. work for developing successional weed managementstrategies on range- Renny, A. J. and E. C. Hughes. 1969. Control of knapweed Centaureaspecies land. Weed Technol. 10:712-720. in British Columbia with Tordon herbicides. Down Earth 24:6-8. Shirman, R. 1981. Seed productionand spring seedling establishmentof dif- Roche, B. F, G. L. Piper, and C. J. Talbott. 1986. Knapweeds of Washington. fuse and spotted knapweed. J. Range Manage. 34:45-47. Washington State University Cooperative Extension Service Bull. Spoon, C. W, H. R. Bowles, and A. Kulla. 1983. Noxious Weeds on the Lolo EB1393. 41 p. National Forest. A Situation Analysis Staff Paper. Missoula, MT: U.S. Roch6, B. F, Jr., and C. J. Talbott. 1986. The Collection History of Centaurea Departmentof AgricultureForest Service. 33 p. Found in WashingtonState. AgriculturalResearch Center ResearchBull. Story, J. M., K. W. Boggs, W. R. Good, and R. M. Nowierski. 1989. The XBO978. Pullman, WA: Washington State University Cooperative Ex- seed moth, Metzneriapaucipuntella: its impact on spottedknapweed seed tension Service. 36 p. production and two seedhead slies, Urophora spp. In P. K. Fay and J. Roche, C. T. and B. F Roche. 1993. Identificationof Knapweeds and Star- R. Lacey, eds. Proceedingsof the KnapweedSymposium. Bozeman, MT: thistles in the Pacific Northwest. Pacific Northwest Extension Pub. Montana State University. pp. 172-174. PNW432. 22 p. Strang, R. M., K. M. Lindsay, and R. S. Price. 1979. Knapweeds: British Sheley, R., M. Manoukian, and G. Marks. 1996a. Preventing noxious weed Columbia's undesirablealiens. Rangelands 1:141-143. Tyser, R. W. and C. H. Key. 1988. Spotted knapweed in naturalarea fescue invasion. Rangelands 18:100-101. grasslands:an ecological assessment. Northwest Sci. 62:981-987. Sheley, R. L. and J. S. Jacobs. 1997. Response of spotted knapweed and grass Velagala, R. P. 1996. Using Seed Rate and Plant Densities to Enhance Inter- to picloram and fertilizer combinations. J. Range Manage. 50:263-267. mediate Wheatgrass Establishment in Spotted Knapweed Dominated Sheley, R. L. and L. L. Larson. 1996. Emergence date effects on resource Rangeland.M.S. thesis. Montana State University, Bozeman, MT. 66 p. partitioningbetween diffuse knapweed seedlings. J. Range Manage. 49: Velagala, R. P., R. L. Sheley, and J. S. Jacobs. 1997. Influence of density on 241-244. intermediate wheatgrass and spotted knapweed interference. J. Range Sheley, R. L., B. E. Olson, and L. L. Larson. 1997. Effect of weed seed rate Manage. 50:523-529. and grass defoliation level on diffuse knapweed. J. Range Manage. 50: Wallander,R. T, B. E. Olson, and J. R. Lacey. 1995. Spotted knapweed seed 39-43. viability after passing through sheep and mule deer. J. Range Manage. Sheley, R. L. and B. F Roche, Jr. 1982. Rehabilitationof spotted knapweed 48:145-149. infested rangeland in northeasternWashington. West. Soc. Weed Sci. Watson, A. K. and A. J. Renny. 1974. The biology of Canadian weeds. 6. Abstr. p. 31. Centaurea diffusa and C. Maculosa. Can. J. Plant Sci. 54:687-701.

362 Volume 12, Issue 2 (April-June) 1998

This content downloaded from 158.135.136.72 on Wed, 6 Aug 2014 18:42:18 PM All use subject to JSTOR Terms and Conditions