On Saturday I hiked up Spruce Gulch, off Left Hand Canyon Road in Boulder, to see the weed research and eradication projects being conducted there. Tim Seastedt, a professor of ecology and evolutionary biology at the University of Colorado, says the landowners asked him and his colleagues to help control weeds without using herbicides. The owners didn’t want chemicals in the water supply.

We parked along a meadow where a student has installed rain-out devices to determine how cheatgrass (a winter annual) responds to different levels of precipitation. Seastedt said these devices keep 50 percent of precipitation off the grasses growing below them. Cheatgrass is classified as a noxious weed in Colorado, but it is on List C, which means the state will help cities and counties more effectively manage cheatgrass, if they wish to do so. In another experiment, the meadow was mowed regularly to 2 inches, which caused cheatgrass to yield to storksbill, another winter annual.

Up around the bend, we stopped to look at this patch of western wheatgrass. Plant ecologist David Buckner pointed out that there was little to no cheatgrass here because invasive species like cheatgrass cannot compete with native perennials. This conclusion has been borne out at other study sites in this area by planting spotted knapweed seedlings along with perennials. Only a few of the seedlings survived.

The chimney and foundation below are all that’s left of a homestead and blacksmith’s forge, destroyed in the Overland fire. Seastedt said his research team was planning on using the homestead as a staging site for their experiments.

Farther up the road, we passed the access for running water at the site, which Seastedt said is most helpful for conducting this kind of research. It is a siphon system, and he described how they had to pour many, many gallon jugs of water into a small hose in order to restart the siphon one year.

Despite the weeds invading this land, many native species still survive, such as these creeping mahonia, pussytoes (with white flowers), and horsetail. Ancient forests of horsetail were one of the sources of coal.

We walked through a fence that marked the boundary of the area where the landowner hand-pulled spotted knapweed. Seastedt and his colleagues also do hand-pulling of spotted knapweed, but they have introduced five different species of insects (biological controls) to make the job easier. The insects do not eradicate the plant—that would eradicate their food supply—but they do discourage it to the point that native plants can compete with spotted knapweed in disturbed areas.

Insects have also proven effective in reducing the spread of dalmation toadflax in Spruce Gulch. I asked Seastedt if the team studied the insects to ensure they didn’t spread to other plants. He said they had tried to force the weevils to eat penstemon, a close relative of toadflax, and the weevils didn’t like it.

As we hiked up to the ridge, we could see how the 1988, 2003, and 2011 fires had affected the area. Buckner said that studies of burned areas that have been “restored” show less diversity of plants than areas left to recover on their own.

Restoration is an ever-evolving science, as that conversation showed, and I for one hope that future “management” of ecosystems will use the lightest touch possible. We can’t avoid management in the future because we have altered ecosystems around the world so drastically, but we can point those ecosystems in a direction that will allow them to recover on their own.