The effects of herbicides and tillage on green ash Fraxinus pennsylvanica growth and herbaceous vegetation composition in a reforested field along the Saline River, Illinois, USA

Published source details

Baer S.G. & Groninger J.W. (2004) Herbicide and tillage effects on volunteer vegetation composition and diversity during reforestation. Restoration Ecology, 12, 258-267

Background

Green ash Fraxinus pennsylvanica is commonly used in reforestation of agricultural lowlands in the midwestern and eastern USA. The effects of two herbicides (glyphosate and sulfometuron methyl) and tillage on green ash establishment and growth, and herbaceous vegetation composition, was assessed on a reforested former cultivated field in Illinois, eastern USA.

Action

Study area: The study was conducted in a poorly drained former agricultural field on the Saline River flood plain in southern Illinois, USA. The site was cleared of forest in the 1960s and cultivated, but left fallow in 1998 and then enrolled in the U.S. Department of Agriculture's Wetlands Reserve Program (WRP) that provides landowners financial incentive to remove wetlands from agricultural production and restore them for the purposes of improving water quality and wildlife habitat.

In 1999, during the first year of study, vegetation composition in untreated control plots (i.e. no tillage and no herbicide application) was dominated by three species accounting for more than 50% of plant cover: a perennial non-native grass (Johnsongrass Sorghum halepense, 18% cover), a non-native annual grass (crabgrass Digitaria sanguinalis, 18% cover), and a native perennial forb (horseweed/Canadian fleabane Conyza canadensis, 14% cover). Twenty-six species of other grasses and forbs were also present, each 1–3% cover, or less.

Experimental design: Tillage (no-till and tillage) and herbicide (untreated control, pre-emergence herbicide, post-emergence herbicide) treatments were assigned to strips within four 91.5 × 18.2 m blocks. Strips were assigned to one of the tillage treatments and one of the herbicide treatments (six treatment combinations replicated four times).

Tillage treatment strips were disked to a depth of about 15 cm on 4 May 1999. Twenty, 1-year-old green ash seedlings (approximately 30 cm tall) were planted (1.5 m apart) in a row down the centre of each strip. Herbicide treatments were randomly assigned to three 18.2 × 30.5 m vertical strips. Before budbreak (28 May 1999), the pre-emergence herbicide, sulfometuron methyl (Oust), was applied using a vehicle-mounted sprayer at a rate of 14 mg/m² solution in a water carrier. On 8 July 1999, glyphosate (1.5% RoundupPro) solution in a water carrier was applied to 1.37 m diameter circles around each ash seedling, the seedlings were shielded from the herbicide with a 10.2 cm diameter stovepipe.

Vegetation responses: Three years later, tree growth and herbaceous composition around each tree were recorded. Ash stem height and basal diameter were measured in the winter of 2001. The identity and percent cover of herbaceous species within a 0.5 m² area around each ash seedling were estimated in August of 2001.

Consequences

Green ash growth and survival: After two growing seasons, ash growth was enhanced by herbicides application relative to the untreated control, but there was no difference in performance between glyphosate and sulfometuron methyl. Tillage had no effect on ash growth.

Herbaceous composition: In year 3, 42 plant species were recorded, mostly widespread agricultural weeds and common early successional natives found throughout the region. Dominant species included a native perennial grass, broomsedge Andropogon virginicus, and a non-native annual grass, D.sanguinalis. Forb composition was typical of abandoned agricultural land, with Canadian fleabane Conyza canadensis, Canadian goldenrod Solidago canadensis and musk thistle Carduus nutans, all common.

Native species cover was greater in no-till treatments with herbicide, whilst non-native cover was reduced. Native and non-native cover was unaffected by tillage in the absence of herbicide, there were no differences among the herbicide treatments with tillage.

Perennial cover was higher in the no-till (59%) than tillage treatment (38%). Annual/ biennial cover had higher cover in tilled soil (63%) compared with untilled soil (41%).

Species responses: Cover of species that were present in more than 20% of the experimental units were analyzed for treatment responses. Of the native species (Bidens cernua, C.canadensis, ragweed Ambrosia sp., Solidago sp., Asclepias syriaca, Verbena urticifolia, Panicum dichotomiflorum, A.virginicus, Campsis radicans and juniper Juniperus virginiana) only two exhibited strong responses. Ragweed cover was significantly lower in the pre-emergence herbicide (0.4%) treatment than the control (9%) or post-emergence herbicide (15%) application regardless of tillage. Broomsedge had significantly higher cover in the pre-emergence herbicide strips (65%) than the post-emergence herbicide (39.5%) strips or untreated control (30%). Over all herbicide treatments, broomsedge cover was significantly higher in the absence of tillage (57%) compared with tilled strips (33%).

Of the non-native species examined (Commelina communis, C. nutans, Vicia sp., Allium vineale, Echinochloa crusgalli, D.sanguinalis, Sorghum halepense, Setaria glauca, Cyperus esculentus and rose Rosa multiflora), only D.sanguinalis was strongly influenced by treatment, with lowest cover in untilled soil treated with pre-emergence herbicide.

Diversity and richness: Higher diversity occurred in the tillage compared to no-till strips but was lower in the sulfometuron methyl herbicide strips relative to the control and glyphosate treatments. Native species diversity did not differ among treatments, whereas non-native species, in the presence of herbicides, diversity was higher in the tilled strips.

Total (native and non-native) richness were higher in the control and glyphosate treatments than the sulfometuron methyl treatment. There was lower diversity in the sulfometuron methyl treatment due to greater broomsedge cover.

Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: http://www.blackwell-synergy.com/journal.asp?ref=1061-2971