Tillage system effects on weed communities in a 4-year crop rotation under Mediterranean dryland conditions

Effective weed management has been identified as a limiting factor in the adoption of conservation tillage systems. in view of this fact, weed communities were characterized in order to evaluate similarities and differences under three tillage systems: chisel (minimum tillage), moldboard plow (conventional tillage), and no-tillage. The study was carried out near Barcelona throughout a 4-year (1997-1998 to 2000-2001) rotation of winter crops (pea: Pisum sativum L., wheat: Triticum aestivum L.. wheat-barley: Hordeum vulgare L.) on a deep silt loam Calcixerollic Xerochrept soil. Analyses of variance based on a randomized complete block design were performed on total aboveground weed biomass (g m(-2)), monocotyledon and dicotyledon biomass, and the biomass of each species. Three main effects were considered: tillage, block and year (crop). Data were combined in order to compute one Shannon-Wiener diversity index for each year and tillage system. The year (crop) effect was the most important factor, followed by the year (crop) x tillage interaction; winter pea had the lowest weed biomass. Tillage system had no effect on total weed biomass. The mean biomass of Avena sterilis L. was greatest under minimum tillage. while those of Diplotaxis erucoides (L.) DC and Sonchus tenerrimus L. were greatest under no-tillage. The highest Shannon-Wiener diversity index was obtained for the no-tilled weed community, and was related to the highest richness and highest evenness. As tillage system had no effect on weed biomass and although the no-tillage system had the richest and most diverse weed community, the weed flora would not constitute an obstacle to an increasing use of no-tillage system for cereal production in the study area. (C) 2003 Elsevier Science B.V. All rights reserved.