Invasive Plant Report

Control of Invasive Grasses in the Myakka River Watershed Results Report: 9/30/04

Inventory: 45 of the original 46 cogongrass sites were checked and measured in September 2004. 128 of the 132 discovered after the grant proposal was submitted were also measured (see cogongrass spreadsheet).

The original goal was to eliminate 90% of the cogongrass sites. We have since learned that elimination of sites using current available treatment methods in native habitat may not be possible. This month we discovered sites cogongrass growth in sites that had been clear for up to three years. We must now reassess our definition of “eliminated.”

In preparing a presentation for the upcoming Florida dry prairie conference I interviewed staff involved in cogongrass control in public lands. Every resource manager I spoke with told me they could prevent their cogongrass sites from spreading but that they had very little success with eliminating sites. We have tried several different treatment strategies, all with increasing success, but have not yet found the magic potion. We tried 2% glyphosate, then 1% glyphosate, then 3% glyphosate. We read that Arsenal was a promising candidate for control, and tried the combination of 2% glyphosate and 1% Arsenal. We continued to get regrowth though not as much as with glyphosate alone. Upon recommendation from a reliable contractor, we added Quest, a water conditioner that is supposed to increase absorption by plant roots, and decreased the Arsenal to 1/2%. We are still getting re-sprouts. All treatments have been documented at each site. I believe we get better results from a higher application rate so the plan for the immediate future is to use 2% glyphosate, 1% Arsenal, 1/2 % Quest, and 1/2% surfactant.

Additionally we have used mowing, burning, and digging. Mowing is used when the palmetto and shrubs are too thick and too tall to spray the grass. Sometimes shrub height may exceed five feet. Cogongrass grows back faster than native vegetation and can be treated when it is about 18 inches tall. We also mow sites when there are layers of dead cogongrass that prevent treatment of green growth underneath.

Burning was advantageous in palmetto/shrub growth to thick or too high to treat with herbicide effectively. Of course it was important to treat the area with herbicide as soon as the grass was knee-high. We found that if too small of sprouts were treated, the leaves would not absorb enough herbicide to kill the underground growth and the grass came back quickly with little impact on reducing density. We also discovered if we burned grass in bloom or seed, the seed was consumed by the flames. This could be a very effective method of preventing spread of seed, as long as there are no unburned patches of seeding grasses. For burned areas make receptive seedbeds for seeds from adjoining properties. We also observed that burn zones traditionally burned in January and February had higher incidence of cogongrass invasion than those usually burned in late spring and summer. We found that grass treated in December was effective but there was little effect on grass treated in late January and February. We have more volunteers January through March than any other months of the year, so we continued to treat cogongrass by digging up the sprouts during those months. We also discovered that in the summer when sites were inundated, the young cogongrass sprouts could be pulled up in the mud with rhizomes intact.

The number of newly discovered cogongrass sites was daunting. I could never have imagined how widespread the grass was in remote areas of the park. Had it not been for the atvs and the number of volunteers recruited for the project we would never have discovered most of them.

Due to the number of hours necessary to monitor and treat so many sites scattered over such a large area I had to prioritize the goals of this project. I decided to get cogongrass under control before putting intensive efforts into managing Hymenachne. We have continued to treat and monitor West Indian marshgrass when it was in the area of cogongrass destinations. We also removed new infestations manually when encountered. But I soon came to realize there was not enough manpower to target both grasses intensively. It took the entire month to assess the cogongrass sites and so I had to forego visiting and documenting Hymenachne sites. The hurricanes and extreme flooding were also a hindrance to completing assessment of both grasses.

William Overholt, Assistant Professor of Entomology at the Indian River Research and Education Center in Ft. Pierce, Florida, recruited a graduate student that has been studying Hymenachne and the affect of the seedbug Ichnodermus varigatus on the grass. Park staff will continue to assist the research center with the project to determine the potential of the Myakka bug as a biological control for this invasive grass.

Recruitment and training of volunteers was successful, especially in the winter and spring months. It was difficult to find volunteers interested in work the hot and wet summer months or early fall. Forty-six volunteers participated in this project.

We have been working with staff from Sarasota County Resource Management on strategies for controlling invasive grasses on our adjoining properties. Their contractor treats cogongrass on park property in near the park border; park staff treats cogongrass on county property when encountered near our border. We are also developing a no- vehicles strategy during the time of year Hymenachne is seeding. We are closing roads this fall to all vehicles until danger of spreading the seed has passed. Though we have have asked our staff and contractors not to travel roads south of State Road 72 while the grass is in seed, these measures would be ineffective if Sarasota County did not cooperate by restricting their employees and contractors from traveling between the two properties.

The invasive plant monitoring system has been very successful. We could not have done it without it. There are more sophisticated systems in place elsewhere but conversations with natural systems managers reveal that most are faulty in design as they rely on one person to enter data into a database and make it available to staff involved in exotics control. My presentation at the upcoming Florida Dry Prairie Conference in Sebring will share the system with other managers of natural areas that contend with cogongrass. The system will also be posted in the resource management section of Myakka Friends website at www.MyakkaRiver.org.

Seed from cogongrass along road shoulders and adjoining private lands also threatens park lands. We have found that road departments in both Sarasota and Manatee couties are willing to treat cogongrass growing along road shoulders. We have also begun contacting park neighbors to make them aware of the problem and and two cases to treat their patches of cogongrass.

A manual was developed for park staff illustrating category 1 and 2 exotics that have been found in the park and describing specific treatments for each plant. The manual was converted to html and posted in the resource management section of the Friends website. The pages include information on this project including a copy of this report. A source page lists references on the web and in literature to find additional information on cogongrass. We have also developed a message board so that people involved in treating cogongrass can post their observations.

Examples of Treatment histories of cogongrass sites:

Site #1

  • 10/5/98 5’x10’ Treated w/ 1% Roundup
  • 2/1/99 no cogongrass found
  • 6/10/99 small patch N of original site Treated 1% Roundup
  • 9/3/99 small patch, hand pulled and sprayed (herbicide not specified)
  • 10/26/99 Treated 100’x100’ w/ Roundup (ranger probably misidentified other grasses as cogongrass)
  • 6/26/00 15 plants treated w/ Roundup
  • 6/21/01 10-20 plants Treated w/ 1/5% Roundup
  • 8/15/01 10 x 10 sparsely scattered 3% Roundup
  • 4/18/02 10 x 25 3% Roundup
  • 7/31/02 sparce regrowth 2% Roundup
  • 11/15/02 1 x 2 & scattered sprigs 2% Rodeo & 1% Arsenal
  • 3/27/03 Area burned; no cogongrass present
  • 8/7/03 10 x 25 2% glyphosate, 1% Arsenal
  • 3/23/04 no cogongrass present
  • 6/7/04 1 plant Treated w/ 2% glyphosate, .5% Quest, .5% Arsenal, .3% surfactant
  • 9/16/04 no cogongrass present but has come up 150’ west of site & named 1B

Site #3

  • 10/5/98 10 x 20 Treated w/ 1% Roundup
  • 3/29/99 10 x 20 Treated w/ Roundup (% not specified)
  • 4/16/99 pulled 42 plants
  • 8/28/99 10x20 Treated w/ 2% Rodeo
  • 10/8/99 5x5 Treated w/ 2% Roundup
  • 6/26/00 no cogongrass
  • 9/12/00 10x15 Treated w/ 2% Rodeo
  • 10/31/00 5x10 Treated w/ 2% Roundup
  • 8/31/01 2x2 Treated w 15% Roundup
  • 4/10/02 no cogongrass
  • 7/25/02 no cogongrass
  • 3/11/03 no cogongrass The site was checked about once a month but not recorded as we thought this site was eliminated.
  • 9/16/04 5x6 New growth aprx 200 feet east of original site.
  • 9/20/04 5x6 Treated w/ .5% Arsenal, 2% glyphosate, .5% Quest, .5% Induce

Cogongrass Observations

Cogongrass site #3 had no growth for three years, then popped up about 100 feet from original site. Cogongrass site #1 had no growth for 18 months then popped up about 75 feet from original site.

Seeds are consumed by fire when burned. So an alternative to picking and bagging seeds could be to burn patch of grass. However, if there is more grass in seed in the vicinity, that will leave open ground for germination of wind dispersed seeds. Areas usually burned in January and February seem to have more infestation of cogongrass than those traditionally burned with growing season fires.

The majority of sites are found along edges of communities where palmetto and shrubs are higher than other parts of the zone rather than in clearings (fire lines, marsh edges, etc). Very few occur on road shoulders, clearings, or in improved pasture. Cogongrass growing in the shade looks very different that what grows in sunny areas. The blades are wider and deeper green in color. Cogongrass growing in flooded or muddy areas pulls up with rhizomes intact, though the one long, feeder root that goes deeper into the ground may break off.

Small seedlings sprayed repeated come back in full force immediately. We need to wait until the grass is at least knee-high to spray with herbicide so the leaves can absorb enough herbicide to affect the rhizomes.

The best time to detect new sites is in late March and early April when the bright-white flower plumes and seedheads are easy to spot from a distance. We have discovered many sites 500 feet or more off roads and trails from the back of a pickup truck or in a tractor in spring. A tractor is good for monitoring cogongrass in prairie. You can detect it quite a distance from the height of the cab, and can carry herbicide and tools for marking and monitoring. We have found that manual systems of recording and monitoring sites work better than sophisticated computerized systems. People are too busy to enter the data and therefore it is not available to all field staff at all times. Schedule 80, white pvc pipe, 5-foot high, with top 4” spray-painted bright red over a wooden stake has been most effective for marking sites. The pvc can be easily removed from the stake when the zone is burned, and the wooden stakes do not puncture tractor tires like rebar.

One way to identify cogongrass is that it usually grows just a couple of inches taller than the palmetto, even in palmetto 5-foot high. When cogongrass is mowed in the spring, a blooming response is triggered. The bloom stalks will come from ground level even if when the leaves are less than 3” tall.