Effects of plant diversity on invasion of weed species in experimental pasture communities

Studies have shown that weed invasion into grasslands may be suppressed if the resident plant community is sufficiently diverse. The objective of this study was to determine whether increased forage plant diversity in grazed pasture communities might be associated with reduced weed abundance both in the aboveground vegetation and soil seed bank. Data were collected from a pasture experiment established in 1994 in Missouri, USA. The experiment consisted of 15 m x 15 m plots sown with Festuca arundinacea Schreb. or Bromus inermis Leysser as a base species in mixtures of 1, 2, 3, 6, or 8 forage species. The plots were grazed by cattle during each growing season from 1998 to 2002. Aboveground plant species composition in each plot was measured using a point step method. Soil cores were collected in 1999 and 2002, and the species composition of germinable weed seeds in plots were evaluated by identifying seedlings as they germinated over an 8-week period. Species diversity was measured using several indices: species richness (S), Shannon-Wiener diversity index (H'), and forage species evenness (J). Aboveground weed abundance in plots was unrelated to forage species richness (S), but weed abundance declined as the evenness (J) of resident forage species increased in mixtures. The species composition of mixtures may have affected weed abundance. Weeds both in the soil seed bank and aboveground vegetation were less abundant in mixtures that contained F. arundinacea compared with mixtures that contained B. inermis. Although variables like forage plant productivity may also suppress weed abundance in pastures, our resuits suggest that maintaining an evenly distributed mixture of forage species may help suppress weeds as well. (C) 2004 Elsevier GmbH. All rights reserved.