Yellow starthistle control.
Thomsen, Craig and WA. Williams

Adapted from Range Science Report, No. 30. Department of Agronomy & Range Science, University of California, Davis. 1991

Yellow starthistle, Centaurea solstitialis, is a plant of Old World origin that probably arrived in California in the mid-1800s as a contaminant in alfalfa seed. Since its arrival it has steadily spread and now inhabits about 8 million acres statewide. Yellow starthistle is one of California's worst noxious weeds, infesting rangelands, pastures, hayfields, orchards, vineyards, roadsides, canal banks, and parklands. It has many traits that make it a successful weed including: 1) a large seed output, 2) seeds that germinate over a long period of time, 3) a deep taproot, 4) late spring and summer growth, 5) an ability to quickly regrow after mowing or grazing, 6) and spines that discourage grazing in the reproductive stage.

Life Cycle and Plant Description

Understanding the basics of yellow starthistle biology, and recognizing the various stages of growth are necessary when attempting control. Yellow starthistle has a very long life cycle for an annual plant. Germination is initiated by autumn rains, but plants mature long after most other annuals have completed their life cycle, sometimes not completing their life cycle until late summer or fall. In addition, successive germination occurs long into the growing season.

The seedling stage is the most difficult time to identify the plants. A good way to learn to recognize them is to first locate seedlings under older starthistle skeletons that remain in the field from the previous year's growth. The winter and spring rosettes produce many deeply-lobed leaves. Although there is variability in the size and lobing pattern of the rosette leaves, a good diagnostic character is the large terminal, triangular lobe at the tip of each leaf In general, the rosettes tend to grow close to the ground in open places, but assume an up-right habit when there is an abundance of neighboring vegetation.

During May and June the plant "bolts" and sends up stalks, which give rise to the flower heads. The mature plant reaches a height of one to three feet. The gray-green to bluish-green stems of the mature plant are widely branched, and the entire plant is covered with soft, appressed hairs. Rigid spines project from the bracts that surround the bright yellow flowers.

The seed development stage can be recognized by the absence of the bright yellow pigment that characterize the younger flowers. Two types of seed are produced. The light-colored disk seeds (central portion of the flowerhead) contain short bristly hairs and are dispersed quickly after maturity. The outer seeds lack hairs and persist for months, remaining in the flower heads until harsh weather or some other disturbance breaks them up. The number of seeds produced by individual plants varies widely according to environmental conditions and genetic factors. In studies of starthistle populations from Hopland, Woodland, and Concord, Maddox (1981) reported a range of about 700 to 10,000 seeds per plant. Once seeds are dispersed, many become part of the soil's "seed bank" where they remain until conditions become favorable for germination, or are eliminated through natural means.

Control Considerations

There is no simple recipe for controlling yellow starthistle, and any control program will require a systematic and persistent effort that may take several years or more. The approach that is taken will depend upon such factors as the size of infestation, plant density, location, equipment available, and planned use for the site.

When planning a control program determine whether you are aiming to eradicate, manage, or contain yellow starthistle. Eradicating means to eliminate starthistle from the site, and requires that seed production is halted and the seed reservoir in the soil from previous years is depleted or managed in a way so new seedlings don't survive. It is impossible to completely eradicate large infestations. With persistence and good management, however, dense stands can be reduced to tolerable levels.

Containment is done by delineating boundaries around large infestations and concentrating control efforts on the smaller occurrences that exist outside of the contained areas. By controlling isolated plants or small patches that are the "pioneers," the larger infestation is contained and further weed spread is prevented. As information and experience are gained from controlling small occurrences, better decisions can be made about whether larger areas can also be successfully controlled.

Become acquainted with starthistle's ecology as it applies to your situation. This will enable you to better assess what cultural practices may be encouraging its proliferation. Often, weeds invade and persist in areas because management practices create ecological "voids" that are exploited by invading species. If a species is controlled, but the void remains unfilled, reinvasion will occur by the targeted species or by a different one that may be equally troublesome. In an ecologically-based control program, the replacement of an unwanted "pest" species by desirable species should be a key consideration.

Timing control efforts to various stages of plant growth is absolutely essential. Figure I lists some control measures placed according to stage of starthistle growth. Some of these can be used during several growth stages, but all of them should occur before seed production. Due to weather patterns, site differences and genetic variation, the duration of life cycle stages depicted by the bars will vary considerably. Thus, on-site monitoring is necessary and control activities should be performed accordingly.

Control Methods

There are six categories of control including mechanical (tillage, grubbing, or mowing), fire, chemical (herbicides), biological (insects, plant competition, and live-stock grazing), preventive, and integrated control. The various methods discussed below are based on established principles of weed control, research, and anecdotal information. Research is incomplete, but studies are underway to fill some of the information gaps.

Mechanical. Cultivation with appropriate implements as the seedlings emerge in the fall is an excellent means of removing young plants. On sites where irrigation is available, infested areas can be pre-irrigated prior to autumn rains and then disked to remove germinating seedlings. If this sequence is repeated, much progress can be made in reducing the seed bank. Deeper cultivation in the spring will eliminate well-developed plants for that season. But deep tillage will bring weed seeds to the surface so expect more plants to germinate with the onset of autumn rains.

Hand-weeding or "grubbing" with a hoe or a weed-eater to control yellow starthistle in small areas should not be overlooked as an important part of any integrated weed control program. The easiest time for this is during the seedling or early rosette stages before the taproot has become well-developed. As the plant develops, its potential to regrow from the taproot is increased, and the upper portion of the taproot will need to be dislodged. Periodically monitor the site for more germination or re-growth.

Mowing is a useful method in managing yellow starthistle provided it is well timed and repeated as necessary. Although no replicated studies have been done, preliminary studies indicate that when starthistle is mowed during the early flowering stage regrowth is minimized. Mowing at this stage removes the aerial portion of the plants after much of their root reserves has gone into producing flowers; therefore, less reserve is available for regrowth. Under low soil moisture conditions, this single-event mowing may be sufficient, but plants should be monitored for regrowth and mowed again if significant growth and flowering occurs. When mowed at earlier stages, regrowth should be expected and several additional mowings will be necessary.

All mowing should be done prior to seed production. As mentioned previously, flower pigmentation is a good way to monitor whether seed development has commenced, but if in doubt open a flower and check for mature seeds. If the flowers are bright yellow and have not faded, seed production has not occurred. Mowing after seeds have been produced removes the hedge-like canopies but does not diminish the seed bank and may aid in seed dispersal. In general, mowing will be most effective when soil moisture is low and no irrigation or rain-fall follows the mowing.

Prescribed Burns. In some situations prescribed burns may be an appropriate management tool. The best time to burn is probably the same stage recommended for mowing. Since starthistle is still green during this period, there must be enough dry biomass from other annual plants to carry a fire. Burning permits are available through the California Department of Forestry (CDF).

Chemical. There are many types of herbicides available, most requiring a permit from your County Agricultural Commissioner. If you are unfamiliar with herbicide use, refer to UC Cooperative Extension publication No.1919, Selective chemical Weed Control for specific information. Before using any chemical, carefully read and follow precautions on the label. With yellow starthistle, seedlings are killed by herbicides but new plants will germinate with additional rains. Therefore, post-emergent herbicides are best applied in spring, after the rainy season, when temperatures are warm, soil moisture is high, and plants are actively growing in the late seedling or rosette stage. Non-selective herbicides, like glyphosate, are effective for spot treatments. However, since glyphosate kills nearly all other vegetation, treatment with this chemical is not usually suited for use over large areas. Selective herbicides such as 2,4-D will help control yellow starthistle and leave grasses unaffected. Care should be taken because these materials will also partially control legumes and other broadleaf plants that may be useful to the ecosystem or production goals. In pastures, a 30~ay waiting period is required before animal grazing can be resumed.

Biological Control & Grazing. Insects and a rust pathogen to control yellow starthistle are currently being evaluated to determine their effectiveness in controlling starthistle populations. Field releases have been made and some appear promising, but additional time is needed to assess the long-term effects of the various biocontrol agents.

Livestock routinely graze yellow starthistle before it becomes spiny, and nutritional studies have shown that it is an acceptable component of a ruminant's diet. Well-timed controlled grazing with cattle has been demonstrated to be an effective method for managing large stands of yellow starthistle in annual rangeland (Thomsen et al. 1989, 1990). For three consecutive years, intensive cattle grazing (high stocking densities with short grazing periods facilitated by portable electric fencing) reduced plant densities and seed production in the dense starthistle stands under study. Spiny canopies persisted through summer and fall in the ungrazed paddocks but were greatly diminished in the grazed areas. Grazing was initiated in mid-to late-May after many other resident annual plants had completed their life cycle and produced seed. Since yellow starthistle was green and palatable at this time, animals readily defoliated plants, making use of it as a forage.

The most effective control of yellow starthistle is obtained when grazing occurs during the stem elongation and early floral bud stages, and is repeated several times to remove the regrowth. Most defoliated plants recover quickly, and if rotations are used, animals should be put back about two weeks later to regraze the plants. The actual number of grazings required varies according to soil moisture levels. Thus, the amount and timing of rainfall has a strong influence on starthistle recovery.

Observations from other sites where continuous low-density grazing is practiced indicate that if sheep, cattle, and/or goats are present in sufficient numbers during the late stages of growth and grazing pressure is maintained, animals utilize the plant and also suppress it. On the other hand, if grazing occurs when starthistle is in the rosette stage but is not continued during bolting, yellow starthistle tends to be favored relative to other herbaceous vegetation. Along with yellow starthistle, neighboring plants are also defoliated, and the competition they provide is eliminated. Yellow starthistle's ability to regrow following defoliation exceeds most, if not all other annual plants on California dryland pastures and ranges.

Grazing yellow starthistle is not a viable control option in horse pastures. When eaten by horses in sufficient quantities, it can lead to a potentially fatal disease of the nervous system called "chewing disease." The nature of the disease is such that the plant may be ingested over a period of years before any symptoms are apparent, and at that point it is irreversible.

Preventive. Preventive weed control measures generally refer to doing what is necessary to stop the introduction of new weed species to a specific area. As in containment programs, this includes detection and control of "pioneer" plants before they go to seed along roadsides, fields, pastures, etc. and develop into large infestations. Because spot occurrences and small colonies seem harmless, the tendency is to overlook them; however, this is the way most large infestations begin. Even though yellow starthistle is widespread, there are still many portions of the state where it has yet to invade, but would be well-adapted.

Integrated Control

A combination of control methods will generally be the most effective way of controlling yellow starthistle. A good example is in agricultural areas where starthistle is often absent in cropland, but abundant as a border vegetation between the fence line and roadside. The frequent cultivation and use of herbicides in conventional crop-ping Systems reduces the seed bank and prevents yellow starthistle from reaching maturity.

Combining native perennial grasses and herbicides has potential to suppress starthistle and other weeds that occur in drainage ditches, along roadsides, and borders of agricultural lands. Researchers from UC Davis are currently evaluating different native perennial grass species and herbicide applications to achieve this objective. The aim of this research is to use selective herbicides to aid in grass establishment, and reduce herbicide use to periodic spot treatments once the deep-rooted grasses become established. When herbicide use subsides, broad-leaf plants such as lupines, poppies, and other wildflowers can be incorporated into the system. The investigators view this as a means not only to control weeds, but also to create wildlife habitat for quail and pheasant, improve the aesthetics of the rural landscape, integrate biological diversity onto agricultural lands, and reduce the $40,000 a year ($100.00 per mile) spent on roadside weed control in Yolo County (Bugg, et al., 1991). [See also Components 2(1), 1991.]

According to Bob Roan of the US Soil Conservation Service, lana vetch was used as a competitor against yellow starthistle. In a several acre trial, Roan reported that starthistle was suppressed by the aggressive, sprawling growth of the vetch, and the starthistle that survived was weak. Mixing a robust grass such as Merced rye, red oats, or Briggs barley with the vetch would add an additional measure of competition to further crowd out starthistle. Using a no-till drill allows seeding without turning the soil, and helps keep deeply buried seeds from germinating. A well-timed mowing could be used in combination with the seeding, which would add another stress to starthistle at a critical stage of growth.

These are a few of the many possible combinations that are options for controlling starthistle. Regardless of what methods are used together, effective long-term control requires that 1) seed production is halted, 2) plants emerging from the seed bank are eliminated, and when possible that 3) other, more competitive plants be permanently established that match the ecological niche filled by starthistle to prevent reinvasion.

Conclusion

Although yellow starthistle is a troublesome weed, its useful properties should not be overlooked. Starthistle is a valuable source of summer nectar for bees, and honey produced from it is a premium quality. Yellow starthistle's early and late-season growth, palatability, and resilience make it a useful forage plant to ruminant animals before it becomes spiny. It provides food and cover to wildlife, especially small mammals and birds. Bill and Helga Olkowski of the Bio-Integral Research Farm in Winters have found that the cut stalks and seed provide a supplementary food for their chickens. Also, as an aggressive, colonizing species, starthistle rapidly covers and helps stabilize unprotected soil.

Despite these useful qualities, yellow starthistle remains a significant pest for most farmers and landowners. The stout spines and hedge-like stands of yellow starthistle that remain through summer and fall, and its invasiveness make controlling it a necessary task. It has been estimated that since 1958, yellow starthistle infested land has increased from 1.2 to 7.9 million acres, an increase of 640 percent (Maddox 1985). Yellow starthistle will continue to increase statewide and will be particularly prominent in years with abundant late-season rainfall.

Ongoing research efforts on yellow starthistle control include seed bank studies, mowing, and goat grazing at the UC Davis Agronomy Farm; seed bank studies at UC Sierra Field Station; the use of native grasses in combination with herbicides along roadsides in Yolo County; and biological control at USDA-ARS, Albany, California (contact Dr. Charles Turner).

References

Ashton, F.M. 1987. Selective Chemical Weed Control. University of California Division of Agriculture and Natural Resources, Cooperative Extension Leaflet #1919.

Bryant, D. 1990. Personal communication. Department of Agronomy & Range Science, UC Davis.

Bugg, R.L., J.H. Anderson, J.W. Menke, K. Compton, and W.T. Lanini. 1991. Perennial grasses as roadside cover crops to reduce agricultural weeds in Yolo County. Grasslands 1(1). California Native Grass Association.

California Weed Conference Committee. 1985. Principles of Weed Control in California. Thompson Publications. 474 pp.

Cordy, D.R. 1978. Centaurea species and equine nigropallidal encephalomalacia. In Keeler, R.F., K.R. Van Kampen and L.F. James (eds.) Effects of Poisonous Plants on Livestock. Academic Press, pp. 327-336.

Fowler, M., A.L. Craigmill, B.B. Norman, and P. Michelsen. 1982. Livestock Poisoning Plants of California. University of California Division of Agriculture and Natural Resources, Leaflet #21268.

Maddox, D.M., A. Mayfield, and N.H. Poritz. 1985. Distribution of yellow starthistle (Centaurea solstitialis) and Russian Knapweed (Centaurea repens). Weed Science 33: 315-327.

Maddox, D.M. 1981. Introduction, phenology and density of yellow starthistle in coastal, intercoastal and central valley situations in California USDA-ARS, W-20. 33 pp.

McHenry, W.B., R.B. Bushnell, M.N. Oliver, and R.F. Norris. 1990. Three Poisonous Plants Common in Pastures and Hay: Fiddleneck, Groundsel and Yellow Starthistle. University of California Division of Agriculture and Natural Resources, Pub. #21483.

Roan, B. 1991. Personal communication. U.S. Soil Conservation Service, Auburn, CA.

Thomsen, C.D., W.A. Williams, M.R. George, W.B. McHenry, F.L. Bell, and R.S. Knight. 1989. Managing yellow starthistle on rangeland. Calif. Agric. 43(S):4-6.

Thomsen, C.D., W.A. Williams and M.R. George. 1990. Managing yellow starthistle on annual range with cattle. Knapweed 4(2). Cooperative Extension Newsletter, Washington State University.

Thomsen, C.D., W.A. Williams, and M.R. George. 1991. Preliminary results using sheep grazing to manage yellow starthistle. Knapweed 4(3). Cooperative Extension Newsletter, Washington State University.

For more information write to: Department of Agronomy and Range Science, University of California, Davis, CA 95616. Tel.(916)752-8810.

(DEC.329) Contributed by Craig Thomsen and David Chaney