Yellow Sweet Clover (Invasive Exotic Plants of North Carolina)

Yellow Sweet Clover

Melilotus officinalis

Summary ⁷

Melilotus officinalis, known as yellow sweet clover, yellow melilot, ribbed melilot and common melilot, is a species of legume native to Eurasia and introduced in North America, Africa and Australia.

Description ⁸

More info for the terms: dehiscent, perfect, phenology, root crown

This description covers characteristics that may be relevant to fire ecology and is not meant for identification. Keys for identification are available (e.g., [82,98,101,191,235,267,268]).

Aboveground description: Yellow and white sweetclover have very similar growth habits and morphology [11,117]. In their vegetative state, yellow and white sweetclover are difficult to distinguish [98,110]. The most obvious distinction between the two species is flower color, which is yellow for yellow sweetclover and white for white sweetclover [101,107,201,268]. When flowers are dry, however, both may appear cream colored [93].

Sweetclover is a biennial [126,127,158,191], although in rare instances annual and perennial growth forms were observed. At the Matanuska Research Farm in south-central Alaska, sweetclover seed from mid-latitudes and from Alaskan roadsides produced plants that flowered in the first year. Artificial light experiments revealed that annual growth was triggered by long light periods and a lack of darkness [124,125]. In London, Ontario, some sweetclover plants clipped "rigorously" during the growing season flowered in the 3rd growing season. All plants died after flowering (Cavers unpublished data cited in [251]). Studies by Smith [217] showed that annual and biennial white sweetclover growth is controlled by a single gene, which can be altered through a single mutation.

In the 1st year of growth, sweetclover produces a single stem with many branches. Near the end of the 1st growing season, nutrients are allocated below ground to the taproot, and root crown buds are formed. In the 2nd year of growth, sweetclover stem number increases, plants are much larger, and flowers and seeds are produced (review by [220]). More details about phenology and the characteristics of 1st and 2nd year plants are provided below in the Belowground description and Seasonal development sections.

Sweetclover is generally an erect, freely branched plant up to 10 feet (3 m) tall [82,235]. Stems are coarse with alternate, 3-pinnate leaves and axillary flowers [93,158,282]. Leaflets are small, 0.4 to 1 inch (1-2.5 cm), and the pea-like, perfect flowers occur in 30- to 70-flowered racemes that measure 1.5 to 4.7 inches (4-12 cm) long [101,158,191,241]. Legumes are up to 4 mm long, scarcely dehiscent, and typically produce just 1 seed, but may produce 2 [82,98,169,191].

Although yellow and white sweetclover are more alike than different, the following morphological differences are common:

White sweetclover is generally taller, has a more erect form, and produces coarser stems and branches than yellow sweetclover ([10], reviews by [45,220]); white sweetclover may be up to 3 feet (1 m) taller than yellow sweetclover [241].
At peak flowering, white sweetclover racemes are much longer (8-15 times) than those of yellow sweetclover [267].
Leaflets produced by yellow sweetclover are often twice as wide as those produced by white sweetclover [188,267].
Yellow sweetclover legumes are wrinkled, and white sweetclover legumes are veiny [117,188].

Sweetclover growth form and morphology are variable, not only because many different cultivars, forms, and ecotypes were introduced in North America but also because growth characteristics can be influenced by environmental conditions. Plant height increases with increasing day length, and low temperatures can limit flower production (review by [251]). For descriptions of some sweetclover cultivars, see the following references: [219,273,285].

Belowground description: Sweetclover produces a taproot with secondary fibrous roots and bacterial nodules [11,101,268]. Taproots are semiwoody, and lateral roots can be extensive (reviews by [45,254]). Lateral roots may extend 6 to 8 inches (15-20 cm) from the taproot (review by [251]). In experimental fields in Columbus, Ohio, 1st-year sweetclover taproots penetrated over 4 feet (1.2 m) deep. Second-year plants had roots up to 5.5 feet (1.7 m) deep in July. Bacterial nodules occurred on roots as deep as 4 feet (1.2 m). Increased root length between these 1st- and 2nd-year plants may not reflect plant age but site differences. Most sweetclover plants attained their maximum root length in the 1st year. Root systems were shallower in dry than moist soils [289].

Bare [10] reports that yellow sweetclover roots are generally shorter but spread farther than white sweetclover roots. First-year white sweetclover dug from a field in Columbus, Ohio, in early May produced taproots that averaged 9.3 inches (23.6 cm) long and 0.3 inches (0.8 cm) in diameter [289]. On a sandy site near Central City, Nebraska, researchers described the root system of a 1st-year white sweetclover plant that was 3 feet (1 m) tall with 15 stems. Near the root crown, the taproot measured 1.5 inches (4 cm) in diameter. Diameter tapered to about 1 cm by about 1.5 feet (0.5 m) deep and remained that size to the maximum taproot depth of 5 feet (1.5 m). Most lateral roots also reached 5 feet (1.5 m) deep, but rarely did they extend more than 2 feet (0.6 m) from the taproot. Sublateral roots were abundant, and generally the soil up to 5 feet (1.5 m) deep beneath white sweetclover was "well filled with roots" [275].

Impacts and control ⁹

More info for the terms: competition, cover, density, fire management, forb, forbs, invasive species, natural, nonnative species, prescribed fire, presence, restoration, root crown, succession

Generally sweetclover is most invasive in grasslands and riparian areas. Sweetclover is commonly invasive in the upper Midwest and Great Plains regions [45,115]. In many prairies, sweetclover is associated with displacement of native species by limiting sunlight and moisture and changing nutrient availability (reviews by [45,222]). Some have referred to white sweetclover as the prairie "restorationists nightmare" [203]. When Wisconsin's plant and natural area experts were surveyed, white sweetclover ranked 20th and yellow sweetclover ranked 24th out of 66 invasive plants having negative ecological impacts in native communities [196].

Although invasive sweetclover populations are common in the Midwest and Great Plains, they are not restricted to these areas. In montane grasslands in Rocky Mountain National Park, Colorado, sweetclover has spread rapidly. Between May and the end of the growing season in 1998, sweetclover populations expanded 3 feet (0.8 m). In 1999, population boundaries extended 6 feet (1.8 m) beyond those mapped in May 1998 [296]. In a New Jersey state-agency publication, sweetclover is reported in globally rare plant communities along the Delaware River. Agency officials cautioned that sweetclover has the potential to form dense stands, prevent native plant establishment, alter community structure, and disrupt succession [221]. As of 2005 in Alaska, white sweetclover was ranked among the states top 10 invasive plants based on potential distribution, ecological impacts, dispersal ability, and feasibility of control (review by [225]). Dense patches of white sweetclover also occur where narrow endemics, Williams' milkvetch (Astragalus williamsii) and Setchell's willow (Salix setchelliana), grow in Alaska [37].

While sweetclover can be invasive, in many areas it is not a priority for management and is restricted to disturbed sites. The eastern region of the Forest Service considers sweetclover "moderately invasive" and indicates that native species displacement is only occurring on a local scale [253]. Sweetclover is only "occasionally invasive" according to the Virginia Department of Conservation and Recreation [266]. In Canada, white sweetclover was number 55 in a prioritized list of 81 species thought to seriously reduce biodiversity in natural areas [27]. Based on a review of available literature and solicited opinions of Canadian botanists, yellow and white sweetclover were described as "primarily species of disturbed sites" with "limited" impacts but abilities to "compete well" in natural areas [286]. A model that incorportated climatic ranges, biological traits, and habitat preferences predicted that sweetclover posed a moderate risk for establishment and proliferation in Riding Mountain National Park, Manitoba, but in their review, researchers noted that sweetclover is already invasive in some Canadian natural areas and is an "ephemeral" threat in prairies, open woodlands, and along stream banks [180].

Site conditions including potential natural vegetation type, elevation, soils, and climate are most likely to affect invasibility by sweetclover. Sweetclover is generally more restricted to disturbed sites in forested and alpine ecosystems than in grassland habitats. In mixed-grass prairies of North Dakota's Theodore Roosevelt National Park, yellow sweetclover was related more to native vegetation type than disturbance [139]. In mixed woods in northern Saskatchewan's southern boreal forest region, sweetclover occurred only along roads and not on sites logged or burned in the last 15 years. Sweetclover was classified as a "lesser threat" and a low priority for management or control [242]. Yellow sweetclover occurred only in disturbed areas above and below timberline in subalpine fir/grouse whortleberry and/or Idaho fescue-slender wheatgrass (Festuca idahoensis-Elymus trachycaulus) habitats in the Northern Rocky Mountains [276]. In mature pinyon-juniper stands in Jarvies Canyon in Daggett County, Utah, yellow sweetclover is not considered invasive and is restricted to roads and other frequently disturbed sites [77].

Impacts on vegetation:
Native grasses and forbs: The majority, but not all, greenhouse and field studies show that sweetclover negatively impacts native grass and forb recruitment and growth. In greenhouse experiments, C3 and C4 prairie grasses were seeded with and without yellow sweetclover in soil collected from the field in Fort Riley, Kansas. Grass abundance was lower with than without yellow sweetclover. Species richness and diversity increased over time without yellow sweetclover but decreased with it. Although most grasses germinated in the presence of yellow sweetclover, grass seedling survival was low. When seeded into established cover of western wheatgrass (Pascopyrum smithii) or smooth brome (Bromus inermis), yellow sweetclover germinated and grew well. In the greenhouse, yellow sweetclover grew rapidly, quickly shading grass seedlings. Several months into the experiments, soil moisture was lower in containers with than without yellow sweetclover, and in the early developmental stages, yellow sweetclover took up large amounts of nitrogen [47].

In montane grasslands in Rocky Mountain National Park, Colorado, abundance and cover of nonnative species were significantly greater in sweetclover-invaded patches, and abundance and cover of native species were significantly greater in uninvaded patches (P<0.001). Amounts of bare ground were always greater in invaded than uninvaded sites, and sweetclover patches occurred in "seemingly undisturbed meadows well beyond disturbance edges". Invaded sites supported the most grasses and uninvaded sites supported the most forbs, suggesting that community structure was altered by the presence of sweetclover [296].

Several native prairie forbs (pride of Ohio (Dodecatheon meadia), Canadian lousewort (Pedicularis canadensis), tall cinquefoil (Potentilla arguta), and Virginia mountainmint (Pycnanthemum virginianum)) were negatively associated with white sweetclover (P<0.01) in southeastern Wisconsin's tallgrass Chiwaukee Prairie. There were 859 white sweetclover plants in 40 disturbed (old roadbed) quadrats and none in 60 undisturbed quadrats. White sweetclover was positively correlated with soil potassium, phosphorus, pH, and moisture (P<0.01). Researchers suggested that native species may be unsuccessful in competition for light with white sweetclover, which often reaches over 5 feet (1.5 m) tall in the study area. The shallow, gravelly, highly mineral soils on old roadbeds could also inhibit native species establishment and growth [183].

Along the Healy and Nenana Rivers in interior Alaska, native seedling recruitment was significantly reduced in the presence of white sweetclover (P<0.05). In plots with white sweetclover, recruitment of natives was about half that in plots without white sweetclover. Native seedling mortality was 50% more likely in plots with than plots without white sweetclover. Shading by white sweetclover was considered the likely reason for reduced native species recruitment, but in greenhouse experiments, shading did not increase growing-season mortality of the native species evaluated, although winter mortality was greater for some shade-grown than sun-grown seedlings. In another experiment conducted outdoors, 2 native legumes, field locoweed (Oxytropis campestris) and alpine sweetvetch (Hedysarum alpinum), grew and survived well in sparse or dense 1st-year white sweetclover. These experiments indicate that the mechanism(s) by which white sweetclover limits native species recruitment is not obvious [225].

In Badlands National Park in southwestern South Dakota, yellow sweetclover had a "consistent" and "strong" positive relationship with native and nonnative plant species cover in a sparse vegetation type. Vegetation and plant interactions were monitored in the sparse vegetation type and mixed-grass prairie type. There were no consistent relationships between native or nonnative vegetation and yellow sweetclover in the mixed-grass prairie. In the sparse vegetation type, where the number of hospitable establishment locations is limited, yellow sweetclover acted as a nurse plant. Invasions by sweetclover could result in the loss of the unique sparse vegetation type through conversion to prairie [262].

Woody vegetation: Sweetclover may restrict woody plant establishment and growth. In at least one location, this restriction may serve to control a more aggressive nonnative species. Along the Hassayampa River in central Arizona, sweetclover seedlings were abundant in the dry surface zone following winter flooding. High cover of sweetclover seedlings "preempted" establishment of saltcedar; thus native community dominance and structure were sustained [237]. On a burned site in northern Arizona, growth of ponderosa pine seedlings was lower on sites seeded with yellow sweetclover than on unseeded sites. Two-year-old ponderosa pine seedlings were planted in the burned area in April of the 1st postfire growing season. Yellow sweetclover was seeded on one of the plantation sites 3 months after ponderosa pine seedlings were planted. After 2 years, ponderosa pine seedlings on the unseeded site averaged 12 inches (31 cm) tall and 1.4 cm in diameter and on the seeded site averaged 9.5 inches (24.4 cm) tall and 1.2 cm in diameter. Ponderosa pine seedling survival was not different between seeded and unseeded sites [57].

Impacts on soil nutrients: Because sweetclover is a nitrogen fixer, soil nitrogen levels may be greater on sites invaded by sweetclover. Increased nitrogen levels on sites invaded by sweetclover could alter species compositions (review by [277]), especially in nitrogen-limited ecosystems. Within 4 years of seeding yellow sweetclover on a rangeland impacted by drought and heavy grazing in western South Dakota, total soil nitrogen was significantly (P<0.01) greater on plots with than without yellow sweetclover [172]. During field studies in northern Utah, researchers found that up to 29% of nitrogen fixed by white sweetclover was transferred to crested wheatgrass (Agropyron cristatum) and was transferred distances of at least 10 inches (25 cm) [68]. In a review, researchers estimated that sweetclover could release inorganic N to mineral soil at a rate of about 12 kg/ha/year, which is 2 to 3 times higher than nitrogen fixation in native rangelands without sweetclover. Researchers suspected that increased nitrogen from sweetclover would favor invasive species over stress-tolerant, long-lived perennials adapted to low nitrogen accumulation rates. The researchers concluded their review with a plea to discontinue use of sweetclover in roadside revegetation and rangeland improvement until more is known about sweetclover's effects on ecosystems and where these effects may be most or least detrimental [141].

In Badlands National park in western South Dakota, yellow sweetclover's effects on nitrogen mineralization and nitrification rates were different in sparse vegetation and mixed-grass prairie. In the prairie, nitrogen mineralization and nitrification rates were not different on sites with high and low yellow sweetclover cover. In sparse vegetation, however, nitrogen mineralization and nitrification rates were higher on sites with high versus low sweetclover cover. Sparsely vegetated sites occur on highly erodible Aridisols and Entisols, and total vegetation cover is 5% to 10%. Prairie vegetation occurs on deep, well drained, calcareous Mollisols, and total vegetation cover is 45% to 100% [263].

In montane grasslands in Rocky Mountain National Park, Colorado, available ammonium and nitrate were significantly less (P<0.002 and P<0.09, respectively) on sites invaded by sweetclover than on uninvaded sites. Available ammonium and nitrate were measured for 2 growing seasons. Invaded and uninvaded sites were essentially equal in past disturbance patterns and in exposure, slope, and aspect [294]. On the Nenana River in Alaska, the soil in areas with white sweetclover did not differ in ammonium, nitrate, or total nitrogen content from that of soils taken from uninvaded areas [37].

Impacts on ecosystem processes: In restored tallgrass prairie at Fermilab's National Environmental Research Park in Batavia, Illinois, cumulative daily exchange of carbon was much lower when sweetclover dominated than when it did not dominate aboveground biomass. In 2005 when white sweetclover occurred as a rosette and did not dominate aboveground biomass, the net ecosystem exchange was 437.7 g carbon/m². In 2006 when white sweetclover bolted, flowered, and dominated the aboveground biomass, and net ecosystem exchange was 239.8 g carbon/m². Because white sweetclover senesced by late July or early August in 2006, the photosynthetically positive active days were reduced by 42%, which may partly explain the reduced carbon exchange. While a dramatic decrease in carbon exchange occurred when white sweetclover dominated, in both study years, net ecosystem exchange was high compared to other published values for North American grasslands [81].

Impacts on wildlife habitat: When native and nonnative grasslands in eastern South Dakota and western Minnesota were compared, grassland bird species richness was lower in nonnative than native grasslands. Nonnative grasslands were dominated by smooth brome (Bromus inermis), yellow sweetclover, and intermediate wheatgrass (Thinopyrum intermedium) and supported 40% to 60% fewer bird species than native prairie [9].

Impacts on insects: A diversity and abundance of insects utilize sweetclover (see Insects) ([195], Stevens 1976 cited in [227]). In Indiana and Wisconsin, sweetclover is an important nectar source for the federally endangered Karner blue butterfly ([83], review by [85]). In Capitol Reef National Park, Utah, researchers think that sweetclover may be increasing the bee carrying capacity in riparian areas. Bees were collected from 11 to 31 July from 10 blooming plant taxa. From white sweetclover, 126 bees representing 19 taxa were collected. From yellow sweetclover, 50 bees representing 14 taxa were collected. Pollinator competition among native and nonnative plants was likely low during this sampling period, when there were few blooming species and many bees. Because sweetclover provided nectar and pollen when floral diversity was low and flowers were few at Capitol Reef, bees may have been supported longer with sweetclover than without [244].

Impacts on agriculture: Sweetclover may negatively impact agriculture. White sweetclover is associated with 28 viral plant diseases including beet curly tip, cucumber mosaic, and tobacco streak (review by [204]). If sweetclover infects wheat crops and is still green at harvest, wheat can take on a sweetclover odor; this is referred to as "sweetclover taint" (review by [251]).

Control: Timing is important to successful control of sweetclover. Controlling sweetclover before plants flower is important for seed bank depletion, and damaging plants when carbohydrate stores are lowest reduces the chances of recovery. In fields in Columbus, Ohio, sweetclover was "nearly impossible" to kill after large sweetclover crown buds were produced between August and November [289]. At the Agronomy Research Center in West Lafayette, Indiana, yellow sweetclover total nonstructural root carbohydrates were highest from November to December and were lowest in May. Defoliation in June caused a rapid decline in root carbohydrates, which lasted about 2 weeks [142].

When sweetclover is targeted for control, no matter what method is employed, the potential for other invasive species to fill the void must be considered [19]. Similarly, sweetclover may invade after other weedy species have been removed. Near Little Manatee River State Park in Hillsborough County, Florida, white sweetclover dominates the weedy flora in areas around a riverfront development project that were chemically and mechanically treated to eradicate cogon grass (Imperata cylindrica) [170]. Control of biotic invasions is most effective when it employs a long-term, ecosystem-wide strategy rather than a tactical approach focused on battling individual invaders [148].

Fire: For information on the use of prescribed fire to control sweetclover, see Fire Management Considerations.

Prevention: It is commonly argued that the most cost-efficient and effective method of managing invasive species is to prevent their establishment and spread by maintaining "healthy" natural communities 148,213 and by monitoring several times each year [114]. Maintaining the integrity of the native plant community and mitigating the factors that enhance ecosystem invasibility are likely to be more effective than managing solely to control the invader [102].

Weed prevention and control can be incorporated into many types of management plans, including those for logging and site preparation, grazing allotments, recreation management, research projects, road building and maintenance, and fire management [255]. See the Guide to noxious weed prevention practices [255] for specific guidelines in preventing the spread of weed seeds and propagules under different management conditions.

Cultural control: Sweetclover abundance is often reduced when perennial vegetation cover is high. A review suggests that white sweetclover can be eliminated within about 2 years of establishing perennial species cover [49]. In a smooth brome-dominated old field in Regina, Saskatchewan, field experiments suggested that yellow sweetclover growth may be limited more by belowground than aboveground competition. Yellow sweetclover seedlings were smallest when grown with neighboring smooth brome plants intact. Yellow sweetclover seedlings were slightly larger in plots where the aboveground biomass of smooth brome was removed. Yellow sweetclover seedlings were significantly larger (P<0.05) than those previously described in plots where only the belowground smooth brome biomass was removed and in plots where all biomass was removed [70].

Physical or mechanical control: Hand-pulling, cutting, and mowing can be useful for controlling sweetclover. Hand-pulling has been successful for controlling small sweetclover populations on Nature Conservancy preserves across the country [250]. Hand-pulling is most effective when the ground is moist (early spring or late fall) and complete root removal is most likely. At these times, stress on associated vegetation should be low (reviews by [35,277]). Another review recommends that 1st-year sweetclover be pulled after the root crown has developed [45]. However, failure to remove the entire root could mean plant survival, since sweetclover is "nearly impossible" to kill after large crown buds are produced from August to November [289].

Results from cutting and mowing treatments to control sweetclover are mixed. Cutting may be most effective if done before large amounts of carbohydrates are stored (usually late summer), and cutting may be more effective on 2nd-year than 1st-year sweetclover plants. In fields in Wisconsin, growth of 1st-year sweetclover was monitored with and without cutting. Cutting occurred on 16 August, 2 September, 18 September, and 18 October. The largest increases in root size and carbohydrate storage occurred between the 18 September and 18 October sampling dates. Cutting on 18 September produced the greatest reductions in root size, weight, and available carbohydrates and nitrogen [216]. Along the Nenana River in Alaska, cutting 1st-year white sweetclover plants to a height of 1 inch (2.5 cm) did not decrease plant density but did decrease seed production. Cutting 2nd-year plants reduced both plant density and seed production. Cut plants produced 25 seeds/ft², and uncut plants produced 3,293 seeds/ft². However, only selective cutting would be appropriate management in Alaska, where mowing would likely damage native vegetation [38]. On a loess prairie in northwestern Missouri, about 30% of 2nd-year sweetclover sprouted after being cut as close to the ground as possible (Ladd 1987 personal communication cited in [55]).

Although a review suggests that sweetclover rarely sprouts if cut close to the ground before seeds are formed [35], mowing along roads seems to favor sweetclover populations (review by [251]); and on annually summer mowed and hayed plots in the tallgrass Konza Prairie in Kansas, yellow sweetclover persisted and was dominant in some years [105]. Stems should be removed from the treatment area if cutting occurs during or after flowering, to minimize the chance of adding viable seed to the seed bank (review by [45]).

Frequent and/or summer mowing may be most effective for sweetclover control. In a hay meadow in Lake County, Illinois, frequent (7-8 times/year) mowing reduced the abundance of white sweetclover [203]. In a Wisconsin tallgrass prairie, mowing in July substantially reduced white sweetclover [129]. However, late-June mowing in a restored prairie in northern Illinois did not reduce white sweetclover. Productivity of white sweetclover increased on the mowed area [193]. White sweetclover root growth was rapid in September near Ames, Iowa. Mowing just before this time may reduce the volume of overwintering roots and increase overwintering mortality. The volume of white sweetclover roots for plants mowed in late August averaged 51% of that of unmowed plants. Overwinter mortality for plants mowed in late August averaged 57.5%, for plants mowed in early September averaged 40%, for plants mowed in early October averaged 4.5%, and for unmowed plants was 1.5%. During the study period in this area, the winter was severe with several hard freezes in the absence of snow cover [157].

Biological control: Although no biocontrols have been released to control nonnative sweetclover, native sweetclover weevils may provide some control where they occur in large numbers [286]. Biological control of invasive species has a long history that indicates many factors must be considered before using biological controls. Refer to these sources: [261,290] and the Weed control methods handbook [250] for background information and important considerations for developing and implementing biological control programs.

Chemical control: Herbicides provide control of sweetclover. Herbicides may be most effective on 1st-year sweetclover (review by [251]) and when used with other control methods. In early May following a fall fire in little bluestem prairie in Mason County, Illinois, the burned site was covered with a "carpet" of white sweetclover seedlings. Seedlings were successfully controlled with herbicide [210]. In Alaska, several herbicide treatments were tested on white sweetclover populations. Most herbicide types and rates decreased white sweetclover biomass and seed production. Only 1 herbicide treatment eliminated seed production [38].

Herbicides are effective in gaining initial control of a new invasion or a severe infestation, but they are rarely a complete or long-term solution to weed management [21]. See the Weed control methods handbook [250] for considerations on the use of herbicides in natural areas and detailed information on specific chemicals.

Integrated management: Sweetclover abundance can be reduced through integrated management designed to encourage native vegetation and limit sweetclover growth and reproduction. In restored prairies in southern Manitoba's Beaudry Provincial Park, white sweetclover cover decreased over time as plots were seeded, burned, and selectively mowed. White sweetclover cover averaged 10.6% on sites seeded 4 to 8 years earlier, 3.4% on sites seeded 9 to 12 years earlier, and 0.4% on sites seeded 13 to 16 years earlier. In the restoration area, early spring prescribed fires occurred every 5 years, and white sweetclover patches were mowed repeatedly [160].