Here is an interesting article from the University of Florida Extension that makes the point that herbicides are of relatively low toxicity compared to many chemical we come in contact with every day. Take a minute to give it a read. Done? OK, well they do make some good points, but bear in mind that this article only discusses the acute toxicity (short term effects) of these chemicals to mammals (here rats serve as stand-ins for humans). So in most cases you're more likely to live if you drink herbicide than if you drink bleach or down a bunch of Tylenol. Even table salt will kill you faster than a few of these herbicides.
While this article does briefly address the long-term impacts of herbicide, such as the Agent Orange disaster in Vietnam, it largely ignores the often unknown long-term impacts these chemicals might have on human health. For example, a child probably isn't going to be killed by eating lead tainted paint chips, but it is well documented that their long-term mental development will suffer. This important impact is simply not well addressed by the way most herbicides are tested. This article also fails to mention that 2,4-D, the component of Agent Orange inevitably contains some carcinogenic dioxins as a result of the manufacturing process. Sure, there is a lot less dioxin than in Agent Orange proper, but considering that (according to Wikipedia) 2,4-D is the most commonly herbicide used in the world, perhaps we should be a little concerned.
So this is good information if you're just passing out these chemicals at your next cocktail party, but what about their effects on the rest of the environment? Sure, its relatively safe to drink glyphosate, but we wouldn't go spraying Mountain Dew or bleach all over a pasture or crop field. Who knows what the impacts would be to insects, birds, soil organisms and the fish downstream from this field? Would our groundwater become contaminated with caffeine (with the accompanying day-glow yellow color)? Even good old table salt can make soil unable to support (most) plant life.
We know herbicides are dangerous, and there are likely many negative impacts that we are just not aware of. Yet we need them to get our job done; they are valuable and effective tools. For example, the above mentioned client has a lot of buckthorn, honeysuckle, autumn olive and some of the more aggressive native shrubs such as sumac and poison ivy.
These plants are having several direct negative environmental impacts. First, they are reducing the habitat quality for other plants and animals. The effects range from suppressing herbaceous plants that provide food for insects and birds to blocking out the sunlight to such an extent that oak seedlings cannot germinate. If things keep going as they are, 50 years from now this historic oak-hickory woodland won't have any oaks in it.
Another environmental impact is the erosion resulting from the dense shade under these non-native plants. The steep hillside this woodland is on has a LOT of bare soil, and is clearly eroding away. The silt from this hillside then moves downstream and silts in our wetlands and streams, killing native wetland plants and animals and encouraging algae blooms and the nasty invasive reed canary grass.
This sounds pretty bad, and clearly something must be done to rectify this situation! (Sadly this site is typical of most woodlands in the Midwest.) The solution is to kill off the invading brush (and herbaceous plants like garlic mustard and Japanese hedge parsley) and allow native vegetation to recover (or push it along with some interseeding).
Shrubs, by definition, resprout readily when they are cut to the ground, so in order to kill this invasive brush we have to herbicide the stumps or repeatedly cut the resprouts down every few weeks for a couple of years until the energy stores in the roots are exhausted. The latter is feasible if you have just a few invasive shrubs, as is probably the case in most residential yards in the midwest, but in a 1.5 acre woodland, cutting those hundreds, maybe thousands of stems again until those shrubs all die would triple or quadruple the project costs. Meanwhile, in the short term at least, we would be exacerbating the erosion problem and impact to native plants as we repeatedly visit the site and scour the entire area finding every re-sprouting bush to cut down again.
While we'd be happy to have the work, I think the increase in the cost of the project would put it out of the means of most landowners. If we do our brush clearing work in the winter when the ground is either frozen or covered in snow, we can nearly eliminate the erosion impacts caused by our work. Furthermore, if we're careful with the use of herbicide, we could have very little "collateral damage" to surrounding plants. We use glyphosate whenever it is feasible because this chemical is easily diluted by water (such as snow) and breaks down in the soil fairly quickly, so it's not around to impact plants, insects or soil organisms, or even wash away and accumulate in our streams and wetlands.
Even in cases when we are spraying broadcast herbicide to kill off weeds on a site before we seed in a prairie, its easy to see that the short term impacts of herbicides is a lesser environmental evil than the alternatives. Consider the regular, long-term application of herbicide to crop fields or well (overly) manicured lawns. Or alternately, the incredible benefit to wildlife that a prairie planting has compared to a rough lawn, pasture or weedy pasture.
Ecological Restoration work is like a doctor working on a critically ill patient. We know that the radioactive chemicals used in chemotherapy are dangerous, but if they can kill the cancer and allow the patient to recover and return to health then they are a most necessary and critical tool needed to save a life. Our natural areas are in critical condition too. We need to carefully and consciously apply herbicides as part restoration work so that we can restore health to our natural areas for the wildflower, wildlife and innumerable ecosystem services they provide. Herbicides are toxic, but maybe aren't as bad as we sometimes think they are, and certainly better than the alternative.
