Establishment and seed production of native forbs used in restoration

Abstract

The importance of incorporating native wildflowers into seed mixtures for disturbed land revegetation projects is widely known and accepted. However, further assessment of weed management approaches is a necessary step to successfully establish and produce native wildflower seed. We examined the impact of pre and post-emergence herbicides alone and in combination with hand weeding on 5 wildflower species [slender white prairie clover (Dalea candida(Michx). ex Willd), blanketflower (Gaillardia aristata Pursh), fuzzy tongue penstemon (Penstemon eriantherus Pursh var. eriantherus), silverleaf phacelia (Phacelia hastata Douglas ex Lehm.), and prairie coneflower (Ratibida columnifera (Nutt.) Woot. & Standl)] under greenhouse and field conditions. Herbicides evaluated included Treflan (trifluralin) 189 l/ha, Lorox (linuron) 1.121 kg/ha., Permit (halsulfuron) 91 g/ha., Plateau (imazapic) 560 g/ha, and Prowl (pendimethalin) 4.2 l/ha. The objectives of this study were to 1) determine wildflower seedling tolerance to post-emergence herbicides, 2) evaluate the effect of pre-and post-emergence herbicides on native wildflower seedling establishment, weed control, and wildflower seed production. For objective 1 a randomized block design was used and repeated twice. A Monte Carlo resampling assessed herbicide damage and a randomized block design analysis of variance (ANOVA) assessed herbicide impact on fresh and dry biomass. Results indicated that the D. candida and R. columnifera were minimally affected by herbicide treatments, while G. aristata and P. hastata were strongly affected, the first by linuron and halosulfuron and the last by halosulfuron and imazapic. Objective 2 assessed hand weeding and pre and postemergence herbicide effects on native wildflowers. A randomized block design was used to assess wildflower establishment, percentage cover, yield, and seed germination and viability, along with weed community composition and cover as a function of weed management approach. Data were analyzed with a randomized block design analysis of variance (ANOVA) to test for differences in wildflowers seedling emergence, percent cover of wildflowers, and seed yield. Wildflower species responded uniquely to weed management, indicating caution should be used when applying herbicides to the tested species. Specifically, emergence of P. eriantherus, D. candida and P. hastata were negatively affected by trifluralin, indicating this herbicide may not be suited for the tested wildflowers.