Weed Management in Wildflower Plantings

Leon S. Warren Jr., Research Associate

Wildflowers are planted and maintained on approximately 3500 roadside acres in North Carolina. Under the direction and leadership of William (Bill) Johnson, the North Carolina Department of Transportation (NCDOT) Roadside Environmental Unit has developed one of the most successful, well known wildflower management programs in the United States. From its inception in the middle 1980's to the fall plantings of 2001, the many colors that blanket the state each year have provided millions of travelers with beautiful scenery and welcome breaks from the hustle and bustle of today's society. People of all ages enjoy the wildflower settings; sometimes so strongly that they relay their feelings to the governor of our state.

Most of the wildflower species selected throughout the years have been nonnative plants that display outstanding colors. In general, these plants are not very competitive with the aggressive native plants that inhabit the state. Winter annual weeds such as vetch species, shepherd's-purse, Carolina geranium and Virginia pepperweed can quickly invade fall plantings. Summer annual weeds such as common lambsquarters and various species of crabgrass, pigweed and morningglory help keep pressure on newly germinated seedlings. Mature wildflower beds that are several years old must combat perennial weeds such as dogfennel, horseweed, white clover and curly dock. Weed management is a never-ending battle for NCDOT personnel.

The most effective weed management tool for establishing wildflower plantings has been Methyl Bromide fumigation. Though not all weeds are controlled (legumes such as vetch and clover species for example), Methyl Bromide has been an excellent choice for bed establishment. However, the Environmental Protection Agency (EPA) has determined that Methyl Bromide is an ozone depleting substance and must be banned in the near future. Table 1 shows the production schedule.

Other soil fumigants on the market today do not offer the weed control consistency that is characteristic of Methyl Bromide. In research conducted by J. Bryan Unruh at the University of Florida, Telone II provided excellent nematode and insect control but poor weed control. To increase weed efficacy, Telone II is often used with Chloropicrin or Vapam. Chloropicrin alone also provided poor general weed control and has been used with Methyl Bromide or Telone II. Vapam or Metam, and Basamid are similar products that require favorable environmental factors such as soil moisture and temperature to chemically convert to biocidal active ingredients. Iodomethane (Methyl Iodide) is not currently registered but potentially can match or surpass Methyl Bromide for control of weeds, nematodes and fungi. Methyl Iodide is not considered hazardous to the ozone layer, because it degrades very quickly in the presence of light.

As effective as Methyl Bromide is, weeds quickly invade fumigated areas with the help of wind, water, birds and mammals. Surveys conducted in the summer of 2001 provided insight as to the weed problems faced by NCDOT division personnel each year (Tables 2 and 3). Many weeds are adapted throughout the state from the coastal plains to the mountains as evidenced by the surveys. The prevalence of these species makes wildflower establishment an extremely challenging task.

The NCDOT does not rely entirely on Methyl Bromide for weed control and has supplemented this treatment with various preemergence (PRE) and postemergence (POST) herbicides. Roundup remains a viable option for control of existing weeds before wildflower germination. However, Roundup offers no residual control, and weed species continually germinate with adequate soil moisture. Pennant Liquid + Treflan is a preplant incorporated (PPI) tank mix with tolerance to several wildflower species. This treatment controls annual grasses and some small seeded broadleaf weeds. Pennant Liquid + Treflan does not adequately control many of the troublesome weeds listed in tables 2 and 3.

To expand wildflower weed control options, Dr. Harold Coble of North Carolina State University (NCSU) conducted field research trials in 1996 with the objective of finding tolerant PPI, PRE and POST herbicides for 11 wildflower species. Twenty-one different herbicides were evaluated. In 1999 and continuing today, Dr. Coble, along with Dr. Fred Yelverton, began a comprehensive set of greenhouse and field experiments studying 29 wildflower species and their effects to 31 PPI, PRE and / or POST herbicides. In addition, weed control data have been recorded for many of the troublesome weeds. The ultimate objective of this research is to develop a comprehensive wildflower weed management program in the absence of Methyl Bromide and present the information in a practical, user-friendly format that is of value to NCDOT field personnel (Table 4 ).

In the Table 4 example, fall seeded blackeyed-susan displayed acceptable tolerance to the four listed herbicides at the given rates. Tolerance ratings were based on a 1 to 9 scale with 1 being complete kill, 6 representing maximum acceptable injury and 9 being no visible injury symptoms. Blackeyed-susan tolerance to the herbicides was 6 or higher in greenhouse trials and /or statistically similar to nontreated field plots. The weed species listed under each herbicide were taken from the NCDOT problem weed surveys. These weeds are either stated on the herbicide labels or were evaluated in NCSU trials and given control ratings of 90% or higher. Good to excellent weed control should occur under optimum conditions and proper timings. In the above example, Eptam 7EC provides morningglory control, but would not be an option when applied in fall seeded plantings. NCDOT field personnel must be proficient in identifying seedling grass and broadleaf weeds for proper herbicide selection and must also determine for themselves the amount of acceptable wildflower injury that can be tolerated because of these applications.

In general, research has shown that there are more PPI and PRE herbicide choices per wildflower species than POST options for weed control (Tables 5 and 6). Of the 14 NCDOT divisions throughout the state, vetch species, curly dock and Carolina geranium were listed as troublesome weeds in wildflower beds in 12, 6 and 8 divisions, respectively. These commonly occurring weeds germinate in the fall and hopefully can be managed as a group. By controlling these weeds PRE in the fall, a greater selection of wildflowers is available for planting as compared to the limited species available if targeting POST control.

In conclusion, Methyl Bromide will not be replaced with a single herbicide but by a comprehensive weed management program. Soil fumigants may still be the first line of defense, but as with Methyl Bromide, herbicides will be needed to handle any weed problems as they develop. From a weed competition and aesthetic standpoint, PPI and PRE control timings are the best choices as compared to POST control. Weeds are not only unsightly; they usually outcompete the wildflowers for nutrients, water and sunlight. NCDOT field personnel must identify the problem weeds in each wildflower bed in order to establish an effective weed management program in wildflower plantings.

Leon S. Warren Jr. is a research associate under Dr. Fred Yelverton in the Crop Science Department at NCSU. He may be reached at 919-515-5651, fax 919-515-5315, e-mail leon_warren@ncsu.edu