More info for the terms: allelopathy, autogamy, competition, cover, density, fire management, frequency, interference, mesic, natural, prescribed fire, presence, root crown, xenogamy
Impacts: The control of garlic mustard may be desirable to undisturbed deciduous forests of the eastern and midwestern United States and southern Ontario [3,15,17,49,55]. In forested natural areas, garlic mustard has the potential to dominate the herb layer [41,52,56,91]. Invasion of mature eastern deciduous forests by garlic mustard is notable because these habitats were thought to be relatively resistant to nonindigenous plant invasion, particularly by herbaceous species [43,45,55,56]. From the results of a greenhouse study examining the competitive potential of garlic mustard, Meekins and McCarthy [45] postulated that competition for light within dense garlic mustard stands might inhibit oak regeneration in the understory of eastern deciduous woodlands. However, this same study failed to show greater levels of interspecific competition among garlic mustard, jewelweed, and box elder, 2 potential understory associates.
McCarthy [43] demonstrated removal of garlic mustard from a deciduous forest understory resulted in increased richness and abundance of understory species, especially annuals and woody perennials. Garlic mustard may be particularly detrimental to native spring ephemerals in eastern deciduous forest understories [15]. McCarthy [43] failed to demonstrate a link between the magnitude of garlic mustard infestation and native species diversity. Removal experiments, while providing some insight into possible effects of nonindigenous plant invaders, may be inherently limited in their ability to reflect impacts of invasives on preinvasion communities [84]. Limited and conflicting evidence surrounding the assumption that garlic mustard infestation necessarily results in reduced richness and cover of native herbaceous species points out the critical need for more research in this area.
The allelopathic potential of garlic mustard has received some study, with mixed results. McCarthy and Hanson [44] found little evidence of allelopathic effects of garlic mustard on several plant species studied. They attributed the success of garlic mustard invasiveness strictly to its competitive abilities. Other evidence indicates at least the possibility for allelopathic interference between garlic mustard and neighboring herbaceous plants, as well as the possibility for toxicity against mycorrhizal fungi [35,80]. Roberts and Anderson [64] found a significant negative correlation (r2 = 0.29; P< 0.05) between garlic mustard density in the field and the mycorrhizal inoculum potential of the soil. McCarthy [43] found garlic mustard inhibited establishment of seedlings of other species, yet no quantitative relationship could be discerned between garlic mustard biomass and native species diversity. This finding suggests that the mere presence of garlic mustard depresses native diversity, perhaps an allelopathic effect. Further research is needed to a) determine what mechanisms, if any, are responsible for garlic mustard allelopathy, and b) sort out the relative effects of allelopathy vs. resource competition in interactions between garlic mustard and native plants.
Control: The biology of garlic mustard presents significant challenges to its control because it simultaneously possesses characteristics of native forest herbs such as shade tolerance and relatively large seeds, as well as characteristics often ascribed to weeds such as xenogamy and autogamy, and high seed production and germination under a range of environmental conditions. It is also not impacted by its native herbivores and parasites [3,5,17,44]. While garlic mustard invades relatively undisturbed woodlands, invasion may be expedited by natural and anthropogenic disturbance that removes competing native vegetation. Once garlic mustard becomes established, further dispersal and perpetuation within a particular habitat may require little to no further disturbance [46,55].
Deciduous forest fragments that are isolated in an otherwise predominantly agricultural landscape may be more resistant to garlic mustard invasion, due to limited seed sources and inhibitive dispersal distances [12]. However, in areas with large populations of white-tailed deer, even these insular forest remnants may become colonized by garlic mustard.
As with most invasive plants, deterrence is the most effective strategy against garlic mustard. This includes annual monitoring and removal of all invading plants prior to seed production. Garlic mustard is prolific partly because of its ability to self-pollinate. A single individual can produce large numbers of genetically similar but interfertile progeny, which in turn may colonize even small, local microsite disturbances, leading to a potential garlic mustard outbreak. Allaying invasion may require reducing habitat perturbation in susceptible areas and promoting the health of native plant communities [3].
Garlic mustard population densities may oscillate widely from year to year [56]. Its biennial nature and its seed banking propensity can lead to occasions in which dense stands of garlic mustard appear where none were apparent the year before, and then seemingly disappear the following year only to reappear yet again in subsequent seasons. Further, in years where rosettes are apparently sparse and may evade detection, those monitoring such sites may easily but falsely conclude that garlic mustard is absent. In previously infested areas or areas of suspected susceptibility, careful annual monitoring may be the only way to ensure that garlic mustard is indeed absent from the site.
Once garlic mustard appears within an area, management activities should focus on preventing seed production. While most seeds of garlic mustard tend to germinate during the 1st or 2nd spring following their production, a small number of seeds remain within the seed bank and may germinate over the next several years. Because garlic mustard seed banks may remain viable for up to 6 years, long-term control for a particular stand requires vigilant attention for several consecutive seasons [3,7,14,49]. Even after successful management leads to the apparent absence of garlic mustard, continued periodic monitoring is prudent. A method for destroying seeds of garlic mustard in the soil that would not harm seeds of other species has not been determined [7].
Because of the biennial life-history strategy of garlic mustard, eradication treatments conducted during spring, after seedlings have germinated and before adults can produce viable seed, have the advantage of affecting 2 generations simultaneously [49]. Ideally, this maximizes the kill of new germinants and seedlings, as well as prevents seed production in adults. Since natural mortality is greatest at the seedling stage garlic mustard may be most vulnerable to control efforts during this time [20]. One potential downside to this strategy is that delaying treatment too late into spring risks unwanted effects on native spring emergents.
An alternative approach is to delay management activities until after the 1st growing season to take advantage of significant natural mortality of rosettes. First year garlic mustard mortality at a site in northern Illinois was estimated at greater than 95% between April and November [51]. This strategy may be especially prudent when the control method requires intensive labor, such as cutting or hand-pulling plants, if minimizing quantities of applied chemicals is desired, or simply if costs of more intensive management activities are prohibitive.
Control of garlic mustard has been tested using several different methods. Since a single control method is rarely 100% effective, a combination of more than 1 may often be a useful strategy. Regardless of methodology, treatments for eradication of garlic mustard must be applied over the entire area of infestation to prevent seed production.
Manual or Mechanical Removal: Pulling entire plants may be an effective method for control of garlic mustard. Care should be taken to remove as much of the root system as possible, to reduce resprouting potential. Pulling can cause soil disturbance and redistribute seeds stored within the upper soil horizons. This problem may be mitigated by thoroughly tamping disturbed soil after pulling. Generally speaking, cutting results in fewer disturbances than pulling. However, pulling may be done at any time during the plant lifecycle, while cutting must be performed during the 2nd growing season while the flowering stem is elongating. Due to the labor-intensive nature of cutting and pulling plants, these practices may only be practical in small or lightly infested areas, especially where burning or herbicide application is inadvisable [49,56]. Hand removal may be most useful for preventing establishment of new garlic mustard colonies in previously uninfested areas [43].
Control may be accomplished by cutting flowering stems, i.e. using sickles, clippers, or string trimmers, prior to seed production and dissemination. Cutting as close to ground level as possible appears to be most effective. Nuzzo [49] found that cutting at ground level killed 99% of plants and resulted in virtually no seed production, while cutting at 4 inches (10 cm) resulted in 71% mortality and 98% lower total seed production. Mortality was 6% in control plants during the 3-month study period. Cutting plants prior to full flowering or the onset of seed development may result in production of additional flowering stems from buds located on the root crown [56]. However, waiting until after plants have finished flowering risks dissemination of viable seed. Cut or pulled plant material should consequently be removed from the site and destroyed whenever possible to minimize the risk of inadvertently distributing viable seed [56,70].
Mowing may be similar in effect to cutting, but with more possible negative consequences. Mowing of flowering plants may result in regrowth of new flowering shoots, although this response reportedly diminishes as the season progresses [15]. While mowing may be convenient in large, relatively open areas of infestation such as roadsides, this practice may be more problematic than cutting, as described above. Mowing may promote seed dispersal and is more likely to be indiscriminate regarding which plant species are destroyed. Mowing equipment may also create more disturbed habitat that is likely to be recolonized by garlic mustard [56].
Prescribed Fire: In areas with a fire-tolerant native flora, frequent prescribed burning may deter garlic mustard invasion by both directly killing invading plants, and perhaps in some areas by enhancing growth of native herbaceous competitors and thereby reducing habitat for garlic mustard colonization [49,88]. For more information about using prescribed fire as a management tool to control garlic mustard, see the Fire Management Considerations section of this summary.
Chemical Control: Chemical control of invasive plants such as garlic mustard can be effective, particularly against large areas of infestation or dense monotypic colonies, and especially when considered within the context of an integrated management plan [47,49]. This report briefly examines the effectiveness of selected chemicals for controlling garlic mustard, some issues involved in the timing of application, and potential effects on native plant communities. Use of herbicides in natural areas should be cautiously considered, and appropriate education and training should be sought before proceeding. Particular caution should be exercised with the use of Bentazon or Acifluorfen. Bentazon is very soluble in water and does not bind to soil well, leading to potential groundwater contamination problems. Acifluorfen is toxic to fish, is moderately persistent in soil and kills native grasses and herbs, and can cause serious eye injury [79]. For further information regarding the use of herbicides in natural areas for control of this and other invasive plant species, see the Weed Control Methods Handbook [76].
The effectiveness of 2,4-D against garlic mustard is questionable [56]. Use of 2,4-D in mixtures with other chemicals may improve its effectiveness, but scant evidence is available [15,56].
Application of 1% and 2% glyphosate during the dormant season significantly (p < 0.05) reduced adult garlic mustard cover and density in mesic and dry-mesic upland forest and mesic floodplain forest in northern Illinois, but also damaged other species that were green at the time, especially sedges and white avens [53]. Treatment with foliar-applied glyphosate also significantly (p < 0.05) reduced adult densities of garlic mustard, regardless of spring or fall application, in a northern Illinois oak woodland. Seedling frequency in these same plots was significantly (p < 0.001) reduced by spring application [49].
Dormant-season application of bentazon was less effective at controlling garlic mustard in northern Illinois mesic deciduous forest, but showed none of the nontarget kill associated with glyphosate. At these same sites, application of acifluorfen during dormant season was highly effective againstgarlic mustard, but also killed most native herbaceous vegetation, which was mainly dormant at the time of application[53].
Use of systemic, nonselective herbicides during the growing season may not be practical in some areas due to deleterious effects on native ground-layer competitors. In these cases, dormant season application may be preferable in order to maintain viable populations of native competitors [49]. Nuzzo [49] found no difference in effect between single herbicide application and twice applied treatment to the same generation of plants (spring and fall of the same year, fall and the following spring, or 2 consecutive springs). It was suggested that management efforts focus on single applications to successive generations of plants. Fall herbicide application may be a prudent option when risk of negatively affecting native spring-emergent herbs exists. Higher garlic mustard rosette densities in fall may require higher volumes of applied herbicide to be effective [51].
Mid-summer application of bentazon reduced garlic mustard cover by 94-96% in previously dense stands of garlic mustard rosettes in northern Illinois. Similar applications of acifluoren were less effective, but still significantly reduced garlic mustard cover by 30-46%. Mortality in control plots over the same period was 15%, and not statistically significant. Chemical control activities conducted during the growing season, as above, might be justified when target species densities overwhelm the native flora [52].
Biological Control: Biological control methods for garlic mustard are not yet developed, but investigations are under way. Several insects that are associated with garlic mustard in its native European habitats are being tested to examine their potential effectiveness as control agents [56]. Fungal pathogens may also have some potential use against garlic mustard. For instance, garlic mustard has shown susceptibility to a fusarium root rot (Fusarium solani) [16].