A status report on "non-chemical control of knapweed", as of Jan., 2000.

Tim Seastedt, Department of EPO Biology and Institute of Arctic and Alpine Research, University of Colorado, Boulder, CO 80309

"...in the long run, the only real hope against invasion is a public that values the creatures that belong where they are." (Bright 1998. pg 226)

A Candid Look at the Current Status of Research

The following paragraph was included in an email response I received from a former student.

"... it looks like ecologists are all over the board on invasive species research. Tackle the weed or repair the ecosystem? Better to attack the main population of weeds or better to attack isolated outcroppings of the invasive weed? There is way more (often differing) advice than most land managers have time to read and sort through, and only a few ecologists offer sound management plans based on their research (usually not in the top journals), and even fewer folks publish whether those sound management plans worked (and this usually occurs in the really obscure journals, which aren't listed in mainstream citation indices). Some of the conflicting advice is pretty funny - one paper talks about how it is going to look at "environmental weeds" - i.e. weeds that can invade undisturbed natural areas, and then four paragraphs later says that disturbance is always needed before invasion happens ... Other authors accept generalities which are just plain wrong...

At least I am figuring out where I stand on the issues, but who else would want to sort through this mess?"

These words have been reiterated by a much larger group of scientists. A report by the Office of Technology Assessment of Congress stated, "Although much information on non-indigenous species exists, overall it is widely scattered, sometimes obscure, and highly variable on quality and scientific rigor" (OTA 1993). That quote was repeated by Hobbs and Humphries (1995) for an article in Conservation Biology, who went on to make the following points:

* An integrated program of prevention, detection, early control and ecosystem management...is required. The current emphasis on control measures implemented once a species has become a major problem, and these approaches are inadequate to tackle the problem. (Emphasis mine)

* Much of the plant invasion problem is caused by socioeconomic rather than ecological factors. Attempts to treat the problem will fail unless the underlying causes are identified. (Emphasis mine). Some weed problems may be untreatable under current land uses. In these cases, successful treatment requires radical changes in land use.

* A framework for establishing management goals and priorities needs to be established based upon the relative ecological or economic values of different areas. Weed problems are too ubiquitous to allow for ad hoc management decisions.

Some of us have become fascinated by invasive species because we see them as an intellectually challenging area of ecological research. Others have become involved because these species appear to be a threat to the survivorship and abundance of preferred, often native species, and, finally, some concerned individuals have become involved because they see some of the treatments applied to these invasive weeds - be these treatments pesticides or introduction of nonnative biocontrols - as potentially larger threats to the environment than the weeds, themselves. Regretfully, we need to add a fourth class of individuals, the profiteers. Weed control is a billion-dollar business. Some individuals in the chemical industry are refreshingly honest about this: "...we think that if we can raise the awareness of this issue, we'll sell more herbicides" (statement of Dow Chemical employee to Robert Devine (1998, pg 252).

"...the dominant perspective of identifying a problem, then seeking to eradicate that problem in one fell swoop is unrealistic...Until we get past the instant gratification mentality of wanting that quick fix, we will always be at odds with the Earth." (Jablonski 1998)

This report presents the case for non-chemical management of diffuse knapweed found on public lands within a broader ecosystem management framework. Ecological, human health, and economic information support the concept of minimal use of pesticides on public land. Concurrently, increased proactive management activities are clearly needed in most cases to constrain the negative impacts of invasive species. Weed management programs must always contain two components: 1) mechanisms for reduction of unwanted species and 2) mechanisms for the persistence or enhancement of numbers and composition of desired species. Weed control in public lands often means removing species from a matrix of native, desirable plant species. I argue that traditional weed management, which evolved in the context of agronomic systems, has focused much too heavily on the first component of management. (This is because the second component, "the crop" was always implicit to the management activity. On public lands, the remaining suite of plant species represents "the crop".) Ecosystem management provides the scientific basis and context for achieving the ultimate goals that are seldom achieved on public lands by traditional integrated pest management (IPM) techniques. Classical IPM has failed because it has yet to incorporate effects of human impacts into its management approach.

Simple solutions to invasive species problems rarely exist. Nonchemical controls do include environmental risks to natural ecosystems, and I do not minimize or understate those risks here. However, what is painfully clear to us as scientists is that the solutions - realistically, the compromise solutions - require the application of principles of ecology and ecosystem science that are unappreciated and certainly underutilized by the public and public agencies. To paraphrase George Wooten, a botanist in Oregon, the control of invading species requires comprehensive activities beyond the scope of mere symptomatic treatments that have characterized previous government weed management projects. At a minimum, these incomplete activities doom most current IPM efforts to failure. At worst, both human and ecosystem health are damaged by this management.

Land managers do what they do to be successful. This document has been written to not only emphasize that "dead weeds" are not sufficient criteria for success, but to argue that sustainability, maintenance and enhancement of biological diversity and human health, and enhancement of ecological services obtained from our public lands should be the criteria with which we judge management activities. The solutions include the education of managers in underutilized techniques, a recognition of the critical need for monitoring in all management programs, and the need for a collaboration between managers and scientists to exploit the potential offered by adaptive ecosystem management.

If I were to ask one of my colleagues who has conducted research on vegetation composition of the prairies, "How can you convert a grassland to a area dominated by herbaceous broadleaf (forb) plants?" I'm almost certain the answer would be, "Dump nitrogen on it." Those of us who have conducted fertilizer addition experiments in grasslands have almost always observed a change from grasses to forbs when nitrogen is applied in large amounts.

We've inadvertently dumped nitrogen on the Front Range. Increased nitrogen availability in soils can result from many sources, such as autos, industry, feedlots, and even fire suppression, and our activities in the Front Range of Colorado have contributed to the potential for substantial nitrogen inputs to natural areas. In the spring of 1999, we added additional nitrogen to knapweed-dominated plots east of Boulder. To our surprise (and perhaps we shouldn't have been), we found no response of the plants to the nitrogen fertilizer. The only conclusion possible is that sufficient nitrogen was available (given the availability of water and other resources), so that additional nitrogen was in excess of plant needs. This experiment strongly argues that we have via our direct and indirect land management activities "dumped nitrogen on it". We therefore should expect to see forbs as the dominant plants instead of grasses. However, why should that group of forbs be dominated by the noxious plant, diffuse knapweed?

Diffuse knapweed is just one of many hundreds of non-native plant species that found its way to the Front Range, but unlike most of those other species, it has flourished and moved into areas that don't appear to be heavily disturbed by human activities. Small-scale disturbances provided by gophers, prairie dogs, or other animal burrowing activities are sufficient as sites for knapweed. Diffuse knapweed possesses a number of growth characteristics that give it a competitive advantage given that the once dominant grasses have been reduced and/or are no longer strong competitors for light, water, or other soil resources. In fact, the plant may exhibit characteristics that "keep the soil the way it likes it". This means that once the plant gets established, it is going to remain in the site until we intervene. As mentioned in the introduction, just killing the weed isn't likely going to be sufficient to solve the problem. However, somehow, somewhere, we must find a mechanism to reduce the existing high densities of the plant if we're going to ultimately restore the ecosystem.

There are lots of ways to kill knapweed. One very cost-effective way is to let other organisms do the job for us. Biocontrols are agents that have negative effects on target organisms like weeds. The trick in using biocontrols is to find something that harms the target plant but doesn't harm the other plants (or, at least the desirable plants). This isn't always an easy task. However, with knapweed, scientists have spent over two decades identifying, studying and releasing insects that harm knapweed but don't appear to feed in any significant way on native plant species.

On the Front Range of Colorado, the potential for reducing the vigor and abundance of knapweed using biocontrols appears very strong. We've got several species of insects capable of reducing the size of the knapweed plants as well as the number of seeds produced per plant.

Right now, the biocontrols consist of two picture-winged flies that develop within the flowers of the weed, a weevil that devours the seed heads of knapweed, and a root-feeding beetle that specializes on feeding on the juvenile (rosette) stage of the plant. In the presence of other vegetation, these biocontrols may be sufficient to reduce knapweed densities to levels where it's no longer a threat to any native species and no longer a nuisance to land managers (i.e., doesn't clog up fences with tumbleweeds). We're going to need one or two more years to be sure of this fact, but the data look very, very positive. One very good sign is that one of the flies, which was so rare it took us until 1999 to positively identify it on the knapweed, now appears to be abundant, and total fly densities on knapweed have doubled since we began monitoring in 1997. We've seen the root-feeding beetle do substantial damage to knapweed elsewhere, and the seed head feeding weevil has reduced seed production of knapweed to almost zero at a site where it's abundant. The increases we've seen in these insects, as well as the literature on the effects of these insects, suggests that control is possible.

So, if we just were just dealing with knapweed within pastures of the Front Range, our work may almost be done. It sounds too easy, and it is.

First, we haven't solved the problem with the conditions that allowed knapweed to become common in the first place. Should our biocontrols diminish in number along with knapweed densities, a large seed input of knapweed (from tumbleweeds originating from elsewhere) could result in another outbreak of the weed. Also, even if we're successful at controlling diffuse knapweed, another weed will likely take its place (This is true of any management practice that focuses only on killing the weed). Further, along trails, roadways, at sites that have been bulldozed, etc., knapweed can grow at a pace that exceeds the ability of biocontrols to control plant densities. These sites will produce tumbleweeds that will re-infect grassland areas as well. More drastic, proactive methods of control are needed in these areas. Often, expensive treatments like mowing or herbicides appear to be the only feasible way to control knapweed in these sites. What's needed is a treatment that (like the biocontrols), when applied once, it has the ability to exclude subsequent return of the weed.

Recall that we found our knapweed sites to no longer be sensitive to additional nitrogen. So, what element, if any, is now limiting the growth of this vegetation? Our data indicate that knapweed requires high amounts of soil phosphorus to grow to large size. Young soils in general, and disturbed sites in particular, may provide the phosphorus knapweed needs to grow into the tumbleweeds that make this species capable of invading new sites. Therefore, if we can add something to the soil that binds up plant-available phosphorus, we may find the "Achilles heel" of this weed. Also, there may be native plants that are very good at competing for phosphorus. If we could identify these and plant them along disturbed areas, the knapweed would be deprived of its limiting nutrient.

We know that we can bind up phosphorus; we're not sure that we can do it under field conditions or by using an inexpensive salt that does not harm other living things. This is a current area of research in our laboratory.

Summary. Biocontrol insects appear capable of reducing knapweed densities to low levels on the prairies of the Front Range. Whether this will be shown to be a reality should be known in one or two more years of monitoring and study of these insect effects.

The only way a forb like diffuse knapweed should be able to dominate in what once was a strongly grass-dominated ecosystem is because soil chemistry has been altered. Finding the mechanisms to return soil chemical characteristics back to those favoring grasses is believed critical to a long-term sustainable solution for reducing diffuse knapweed (and perhaps many other undesirable forbs) on these areas. Current research focuses on reducing plant-available phosphorus in the soil.

Literature Cited

Bright, C. 1998. Life Out of Bounds: Bioinvasion in a Borderless World. W.W. Norton & Company, New York.

Devine, R. 1998. Alien Invasion: America's Battle with Non- Native Animals and Plants. National Geographic Press, Washington, D.C.

Hobbs, R.J. and S. E. Humphries. 1995. An integrated approach to the ecology and management of plant invasions. Conservation Biology 9:761-770.

Jablonski, L.C. 1998. Get into step with earth. Fort Collins Coloradoan, June 24, 1998, p. A6.