Ecological interpretation of weed flora dynamics under different tillage systems

In northern Italy, on soil managed with three different tillage systems (conventional tillage, ridge tillage, and no-tillage) and submitted to standard cultural practices (crop rotation, and chemical weed control), the weed vegetation was assessed at the beginning of the trial (1987) and after six, and eight years. The aims were to evaluate (1) the effect of tillage systems on the weeds; and (2) the possibility of linking the floristic changes under reduced disturbance to the theory of ecological succession. The weeds were categorised according to life-forms (biological groups), periodicity types (ecophysiological groups), dispersal types and seed longevity. Data were analysed using Sorenson's Indices of Similarity, the Independence test, and Principal Components Analysis. The tillage systems profoundly altered the weed community: in undisturbed soils the importance of the geophyte and hemicryptophyte species, and among the annuals, Digitaria sanguinalis, Conyza canadensis and Kickxia elatine increased, as well as that of the wind-dispersed weeds. The species linked to disturbance were annuals and in particular Amaranthus spp., Chenopodium album and Echinochloa crus-galli. After eight years the floristic evolution in the reduced tillage system can be interpreted on the basis of ecological succession. The community that has formed assumes, from the quantitative point of view, characteristics of a pioneer community of secondary succession with a predominance of annual species and a large number of wind-dispersed plants. Qualitatively there is a movement towards a more mature community that could become similar to that of the woodland edge, with more perennial species, shrubs, and bird-dispersed plants. The implications of these conclusions are discussed in terms of weed management. (C) 1997 Elsevier Science B.V.