Weed treaters

Birds, hikers and tire tread. Those are just some of the vehicles that non-native plant species use to travel to new environments where they choke out wildlife habitat, and unbalance and destroy thousands of acres every day. Each year, according to the Federal Highway Administration (FHWA), the unwelcome guests cost $23 billion in damage to the environment, crops and people.

Because invasive species can spread undetected quickly, local, state and federal agencies are banning together to share resources and develop new technologies to stop invasive weeds from growing uncontrollably. The technologies address a range of weed management issues from soil sampling and native seed collection to satellite and hyperspectral sensor imagery for mapping.

Hyperspectral sensors

Jeffrey Pettingill, county weed superintendent for Bonneville County, Idaho, and president of the Meridian-based Idaho Weed Control Association, says invasive weeds cost his state $500 million every year. In a state where 70 percent of the land is owned by the federal government, invasive weeds can go undetected until they become a financial and technical nightmare to eradicate. Uprooting weeds at an early stage costs approximately $20 per acre. If the infestation is obvious to the naked eye simply by flying over an area, the cost can rise to $100 per acre and take 10 years to eliminate, Pettingill says.

In 2001, Bonneville received a three-year, $800,000 grant from the National Aeronautics Space Administration (NASA) to determine the most cost-effective technologies to detect invasive weeds. Bonneville tried using satellite imagery to map weeds, but the smallest image, or pixel, NASA satellites captured was 30 meters. Because the county was trying to detect weeds when the plant patches were still dinner table size, satellite imaging was unfeasible.

Aerial imagery taken with hyperspectral sensors, or digital cameras with the ability to detect more colors than the human eye, has shown promise. The sensors, which are roughly the size of a washing machine cylinder, can be mounted to a plane and collect images 11,000 feet in the air. In partnership with Idaho State University, Boise State University and the University of Idaho, Bonneville County began testing the technology to detect leafy spurge in 2001. The Eurasian native was brought to the United States in 1897 and thrives in many soil and environmental conditions where it chokes out trees and grasses. It also can cause blindness in humans and is poisonous to cattle.

“We found leafy spurge at this one particular point [on the ground], then went to the imagery and found that point and told the computer, ‘This is what a spectral profile of leafy spurge looks like, go find it everywhere else,’” says Dr. Nancy Glenn of Idaho State University in Pocatello. The coordinates locate the leafy spurge and then are sent back to the county.

The method is so accurate for detecting leafy spurge that if a 3.5-square-meter area has 40 percent weed coverage, the researchers will find it 90 percent of the time. The sensors achieve varying levels of success with different species, however, higher accuracies can be obtained with more data manipulation, Glenn says.

At present, the cost of imaging a 40-square-kilometer flight line is $6,000 to $8,000, but Glenn believes eventually the sensors will be able to be mounted on a satellite rather than an airplane, which will cut the technology’s cost. “We also can keep costs down by partnerships between universities, states, counties and federal folks,” she says, noting that with the $800,000 Bonneville County grant, with the exception of $7,500 for 15 computers and the flight costs, most of the funds were devoted to graduate students conducting the analysis. “Our next step is to convince people that this works so that land managers can do this on a more regular basis.”

The final leg of Bonneville County’s research is under way. County employees are compiling an integrated plant management plan that weed control employees can present to property owners with a hyperspectral map that delineates botanic invaders on their property. The plan also includes strategies to bring invasive species under control, such as removing plants with a shovel, using pesticides or introducing a new species to control the non-native species.

Weed management software

In February 2005, Arlington, Va.-based The Nature Conservancy (TNC) released software for invasive species management called the Weed Information Management System (WIMS). The program is a Microsoft Access relational database that anyone can download from the Web. WIMS can be loaded onto a hand-held computer, such as a Pocket PC with attached global positioning system (GPS) unit, to collect data in the field.

“The thing that makes it attractive for the field is to have the handheld with background image loaded and with the GPS, you can pinpoint yourself on the map and easily do weed mapping and capture associated data,” says Mandy Tu, TNC’s invasive species ecologist. The WIMS allows users to track three types of data records to include weed occurrences, assessments and treatments. It generates reports that can be exchanged between multiple users to North American Weed Management Association standards and can be written to shape files for geographic information systems (GIS) mapping.

The conservation group, which owns and manages approximately 15 million acres in the United States, conducted a weed survey a few years ago to discern the extent of invasive weeds on TNC lands. Of the field staff that responded, only 5 percent indicated that weeds were not a problem. “We were evaluating what our goals were,” Tu says, “and it came up that weeds were one of our top three threats.” In addition, TNC staff also concluded that they had no efficient method for keeping track of the number, location or treatment of the invasive species. Field personnel gathered weed information on paper and kept track of it in GIS, a spreadsheet or in a notebook.

In 2002, TNC staff decided they needed a specialized database that could help site managers track weed-related data. Idaho’s Boise Valley District provided a database, and TNC hired a programmer to modify it. In the end, the application cost about $200,000 in money and time to develop.

WIMS was tested over two years on TNC preserves. The largest trial took place through a U.S. Fish and Wildlife Service partnership in which more than 5,000 acres of Ohio’s Ottawa National Wildlife Refuge were mapped. The test showed that centralizing the information was beneficial to weed managers who could better prioritize their actions and share them with others, Tu says.

Coming soon

Other technologies that could help weed managers currently are being developed with the help of grants from the federal government, says Bonnie Harper-Lore, restoration ecologist for the FHWA. They range from soil sampling technology to harvesting equipment that can be used to gather native plant seeds. Once the technology is created, it will become available for public use.

Moving invasive weeds into previously weed-free areas through construction top soil impacts all levels of government, Harper-Lore says. Brigham Young University in Utah is working on soil sampling technology that will allow top soils to be tested for weed seeds before the soil is placed on construction sites. The technology, which will produce sample results within two weeks of testing, is expected to be available this summer.

Another problem being addressed through research is the shortage of native seed stock, which is important in replanting cleared areas with native species before the area is taken over by noxious weeds. To address the issue, FHWA has awarded a research grant to an ecological consulting group in Texas to develop equipment that can be used to harvest small remnant areas of native seed found in places such as road sides, state and county parks, and wildlife refuges. The machine specifications include developing a harvester that will not harm surrounding wildlife and operate on smooth or rough terrain. A cleaning feature also was included in the grant specifications so that seed is not inadvertently transferred where it is not wanted. A report from the research is expected in two years.

Early detection and rapid response system

Gina Ramos, senior weeds specialist for the U.S. Bureau of Land Management, has been developing a National Early Detection and Rapid Response System for Invasive Plants for the Federal Interagency Committee for the Management of Noxious and Exotic Weeds, a partnership between 16 federal agencies with direct invasive plant management and regulatory responsibilities in the United States and territories. The rapid response system model has five main elements, including detection and reporting, identification, assessment, planning, and response.

The driving idea behind the system is prevention, Ramos says. “We’re pushing the idea of a national database, to actually monitor where outbreaks are, as in a fire, so that we can channel resources.” The early detection and rapid response system, which could be a reality in two years, also will serve as a national clearing-house of information for practitioners and the public and will include a toll-free number for reporting invasive plant sightings.

“We’re definitely at a period of time where we are now willing to cross the political boundaries with equipment and knowledge to help each other out,” Harper-Lore says. While technology and information-sharing infrastructure are being developed, weed managers like Pettingill advocate strong weed control programs that include all levels of government and the public. “Weeds have no respect for fences or property rights,” Pettingill says.