Non-target effects of broadleaf herbicide on a native perennial forb: a demographic framework for assessing and minimizing impacts.
Invasive species are one of the leading threats to biodiversity worldwide. Therefore, chemical herbicides are increasingly used to control invasive plants in natural and semi-natural areas. Little is known about the non-target impacts of these chemicals on native species. We conducted an experiment to test the demographic effects of the herbicide picloram on a native dominant forb, arrowleaf balsamroot Balsamorhiza sagittata. As reported in earlier studies, picloram did not change leaf area of balsamroot in the short term (5 years). However, a single application of picloram dramatically reduced flowering and seed set, and these effects have persisted for at least 4 years after spraying. Matrix population models based on these data suggest that arrowleaf balsamroot is slowly declining in the presence of weeds, rapidly declining in the presence of herbicide, but might experience a release period after the herbicide decays and before weeds reinvade. Therefore, from the perspective of persistence of this native wildflower, herbicide use can be beneficial or detrimental, depending on how often a site is sprayed and how long the herbicide persists in the soil. In this system, spraying intervals of more than 10 years may be beneficial, but intervals of less than 5 years will be detrimental. In general, our results emphasize the importance of finding ways to combine herbicide use with other weed control techniques to maximize return intervals and minimize non-target impacts. Synthesis and applications. Herbicide use in natural areas differs fundamentally from herbicide use in agricultural areas, because the non-target species are not directly under management control. Herbicides inevitably will impact non-target species due to limitations in selectivity. Herbicide persistence and reapplication intervals are key factors determining the demographic impacts of herbicides on native plants. Understanding how persistence and reapplication timing interact with native plant demography allows us to develop management prescriptions that minimize non-target effects in natural areas. To date, this interface of ecology and management has received little attention from practitioners or researchers.