When a weed species is not seen on a site, one typically assumes it is not present. This assumption can be incorrect. If a weed has gone to seed once in recent years, the weed probably is present in the form of a live seed, despite the mature plant being absent. The seed from noxious weeds can remain viable (alive) for at least two years, and for some species 20 years or longer.
Live seeds that germinate and grow into a mature plant originate from a pool of seeds called the seedbank. There are two general types of seedbanks: transient and persistent. A transient seedbank is shortlived with all seeds either germinating or dying within one germination (growing) season following the seed’s maturation (ripening). A persistent seedbank lasts for two or more growing seasons.
To develop and implement successful weed control and management programs, landowners and managers must understand the seedbank, including how long it may persist without additional inputs. The seedbank, including its probable longevity, reflects past and future weed problems, even if no weeds germinated and grew during the current year.
Figures 1a and 1b show how seed can persist for one year or longer under adverse conditions and establish a significant weed problem upon germination. Drought in the early 2000s resulted in Chimney Dam Reservoir being very low and perennial pepperweed (Lepidium latifolium) establishing on the exposed lakebed. By 2007, the lake had been full for at least one year, killing the mature plants and preventing germination during that period (Figure 1a).
Figure 1a. Area of Chimney Dam Reservoir inundated for 12 to 18 months or longer. April 9, 2007.
Figure 1b. Taller bright green plants are seedlings of perennial pepperweed. June 12, 2007.
When the water level receded in 2007, tens of thousands of seeds of perennial pepperweed germinated and seedlings emerged (Figure 1b). In this case, the only source of seed was the lakebed’s seedbank (see Schultz 2011). For this site, the death of a mature population of weeds did not preclude its rapid return.
Understanding the Seedbank
Seedbanks are not restricted to the soil. They may occur above ground with seed stored in the plant litter (dead material on the soil surface) or attached to the plant. For example, Medusahead (Taeniatherum caput-medusae) seed germinates best in the dense litter that persists from one year to another. After germination, the root grows toward the soil. The leaves and stems develop only after the root reaches the soil. Soil-based seedbanks have two spatial locations: the soil surface and deep burial. Seeds on the soil surface typically have a much shorter lifespan than deeply buried seed, because they are exposed to many organisms and processes that cause their death. These include rodents, insects, soil-borne pathogens, UV radiation and other mechanisms.
Viable seeds in the seedbank are either active or dormant. Active seeds are available for immediate germination if proper germination conditions exist. Readily germinable seeds, however, may be a small part of the total seedbank. Seed dormancy may be either innate or induced. Innate dormancy is due to a biological, physical or chemical property that prevents germination. Induced dormancy results from an environmental change after seed dispersal that induces dormancy. The dormant state can persist even after the inducing condition is removed.
The seedbank is like a bank account. There is a balance (live seeds) with annual deposits and withdrawals (Figure 2). The largest input to the account is the current year’s seed production (called seed rain). Additional seed may be dispersed to the site by wind, overland water flow, animals, vehicles or other mechanisms. Losses (withdrawals) include germination (with or without eventual establishment) and mortality from many sources, including physical damage by implements, pathogens or fungi, predation by rodents or insects, or an unfavorable environment for growth. When inputs exceed losses, the seedbank becomes larger and the potential for a large weed population increases. Successful weed management programs focus on reducing the seedbank by reducing inputs and/or increasing losses so they exceed inputs.
Once a seedbank develops, at least some of the seed can be redistributed, both horizontally and vertically (Figures 3 and 4). Horizontal movement can be classified into three general types: within a field, between fields and between regions. Regional movement is often related to large-scale human activities. The key point is there are many mechanisms that can move viable seed across a landscape. Before implementing management actions one should consider their potential effect on the seedbank and its movement. Seedbanks typically develop a vertical distribution based upon how farmers work the soil (Figure 4). No till agriculture results in most of the seed residing at or near the soil surface: the physical location of a seed where it is most susceptible to being lost from the seedbank.
The “safe site” is an important concept for seed biology. All sites do not equally promote seed germination and seedling survival. A seed’s mere presence does not guarantee it will germinate and survive. For a seed to germinate it must have optimal contact with the surrounding soil particles. Good soil-seed contact maximizes the transfer of water from the soil to the seed, which improves both germination and seedling survival. For example, the round seed of Indian ricegrass (Achnatherum hymenoides) does not germinate well in silty soil. The soil particles are flat, which leads to little contact with the seed. Many round grains of sand, however, can provide multiple contact points with the soil for the round ricegrass seed.
Management may be able to manipulate the presence/absence of safe sites to promote or retard germination long enough to reduce the seedbank. Figures 1a and 1b provide two extreme examples of manipulating safe sites for seeds. If the reservoir had stayed full for three years most of the perennial pepperweed seed probably would have died because seed from this species is believed to be shortlived. The recently exposed lakebed, however, created an optimal site for germination – moist, warm and shallow burial – and subsequent herbicide treatment as there was no canopy to intercept the chemical. Almost no perennial pepperweed plants were present in the treated area from 2008 through 2010.
Seed Production and Longevity of Noxious Weeds in Nevada
Long-term management and control of noxious weeds in Nevada (see NAC 555 for current list) requires understanding and manipulating their seedbanks. Two important factors land managers must understand are 1) the potential input of seed each year and 2) how long that seed may survive in the seedbank. This interaction determines the size of the seedbank. Few studies have documented seedbank density for noxious weeds found in Nevada, but when conducted they found tens to hundreds of thousands of seeds per square yard (Table 1). Most noxious weed species can produce thousands of seeds per plant, and some over a million seeds per plant (Table 1). Weed infestations typically have population sizes between tens to thousands of individuals per acre; therefore, seed production can range from millions to billions of seeds per acre.
All of the species in Table 1 (and probably all other noxious weeds) produce some seed that carries over for several years. Salt cedar has the lowest longevity, with well under 10 percent of the seed alive one year after dispersal. One mature tree, however, can produce 500,000 seeds. If only 10 percent survive, that results in about 50,000 live seeds the following year. Most likely, some of this seed will move from its dispersal site and spread the infestation to new areas.
Once a noxious weed produces seed, a long-term management and control problem exists. This is particularly true for those species whose seed can survive for five years or longer. For example, Mayweed chamomile, a widespread annual species in Paradise Valley, can easily produce 5,000 seeds per plant, with 6 percent still viable after 11 years in the soil. A small percentage of those will remain alive after 25 years. This means that after 11 years, 300 seeds may remain capable of producing new plants. If only 1 percent of those 300 seeds survive the next 14 years, three seeds will be available to replace the single mother plant that produced the seeds 25 years earlier. For most weed species, complete weed control for several or more years does not eliminate the species. Weed control must continue for at least as long as the seed’s longevity.
Managing the Seedbank
To successfully manage the seedbank, one must prevent seed set whenever possible. If inputs to the seedbank are not eliminated, or at least dramatically reduced, a large seedbank will persist and facilitate reinfestation of treated sites and spread to nearby uninfested areas. Efforts to control the seedbank must be sustained for years to be successful. Research with lambsquarters (Chenopodium album) in Colorado found that a six-year effort to control the weeds reduced the seedbank 94 to 99 percent. After one year without control, the seedbank increased to 90 percent of its pre-control size.
The best approach to reducing the seedbank is to prevent weeds from setting seed. This may occur through mowing, grazing, herbicide treatment or other actions that prevent the seed from developing and ripening. Avoid management actions that have a high probability of moving viable seed from known infestations to uninfested or minimally infested areas. Composting manure typically decreases seed viability. Animals that are fed forages that contain weed seed should be quarantined for three to five days and fed weed-free hay to ensure their feces is free of viable seed. After equipment is used in infested fields, it should be cleaned before use in uninfested areas. Mud that contains weed seed can adhere to vehicles and be moved long distances. Reducing tillage leaves more seed at the soil’s surface, which increases the seed’s risk of mortality.
Increased seed mortality accompanied by reduced seed input is needed to reduce the seedbank. Planting and maintaining a high density of robust desired plants can reduce both the number of individual weed plants and their size. This usually reduces seed production and inputs to the seedbank. When conditions permit, it may be possible to promote widespread seed germination (Figure 1b) and subsequently control the seedlings – the most vulnerable growth stage. Finally, weed managers must remember that successful control of mature weeds does not eliminate the problem. Once a weed has set seed one time, or seed is transported into the area, it is only a matter of time before the weed reappears. Out of sight should not be interpreted to mean the weed is no longer present.
Baskin, C.C. and J.M. Baskin. 1998. Seeds. Ecology, Biogeography, and Evolution of Dormancy and Germination. Academic Press. San Diego, CA. 666 p.
Clements, D.R., D.L. Benoit, and C.J. Swanton. 1996. Tillage effects on weed seed return and seedbank composition. Weed Science 44:314-322.
Cline, D.C. Juricek, R.G. Lym, and D.R. Kirby. 2008. Leafy spurge (Euphorbia esula) control with Aphthona spp. affects seedbank composition and native grass reestablishment. Invasive Plant Science and Management 1:120-132.
Davis, E.S., P.K. Fay, T.K. Chicoine, and C.E. Lacey. 1993. Persistence of spotted knapweed (Centaurea maculosa) seed in the soil. Weed Science 41:57-61.
Di Tomaso, J.M. 1998. Mpact, Biology and Ecology of Saltcedar (Tamarix spp) in the Southwestern United States. Weed Technology 12:326-336.
DiTomaso, J.M and E.A. Healy. 2007. Weeds of California and other Wester States. Volumes 1 and 2. University of California, Division of Agriculture and Natural Resources. Publication 3488.
Duncan, C.L. and J.K. Clark. 2005. Invasive Plants of Range and Wildlands and their Environmental, Economic and Societal Impacts. Weed Science Society of America. Lawrence, KS. 222 p.
Harper, J.L., J.T. Williams and G.R. Sager. 1965. The behavior of seed in soil. I. The heterogeneity of the soil surface and its role in determining the establishment of plants from seed. Journal of Ecology 53:273-286.
Kay, Q.O.N. 1971. Biological flora of the British Isles. Anthemis Cotula L. Journal of Ecology. 59:623-636.
Schultz, B. 2011. Differential herbicide effectiveness on adjacent populations of young (seedling) and mature perennial pepperweed (Lepidium latifolium). Journal of the NACAA 4:2. Available at: NACAA. Accessed 10 December.
Sheley, R.L. and J.K. Petroff. 1999. Biology and Management of Noxious Rangeland Weeds. Oregon State University Press. Corvallis, OR. 438 p.
Young, J.A., D. E. Palmquist, and R. Blank. 1998. The ecology and control of perennial pepperweed (Lepidium latifolium L.). Weed Technology 12:402-405.