Natural Mechanisms of Weed Suppression of cv. ‘Elstar’ Apple Trees

Abstract

The objective of this field trial was to investigate causes of the reduced weed population/vegetation in the herbicide strip of fully-grown cv. ‘Elstar’ apple trees. The vegetation in the herbicide strip of cv. ‘Elstar’ with 150 assessments was compared with that under cv. ‘Golden Delicious’, ‘Gala’ and ‘Braeburn’ apple trees on the same rootstock (M 9) at the same location at Klein-Altendorf, University of Bonn, Germany (51º N). To determine the amount of light available to the vegetation in the herbicide strip, light interception of the tree canopy was measured. The 300 ceptometer and the 100 sunfleck measurements with the ‘point quadrat method’ (chessboard) in September after herbicide application in May showed a significant 10% (ceptometer) and 35% (sunfleck) light reduction in the herbicide strip under ‘Elstar’ apple trees relative to ‘Golden Delicious’ in the adjacent tree row. The tree strip under fully-grown cv. ‘Gala’ and ‘Braeburn’ apple trees was statistical significantly more densely covered with weeds rated as 3.8–4.3 in comparison with that of cv. ‘Elstar’ rated as 1.7–1.9 on a scale of 1 (0–20%) to 5 (80–100% surface covered); this finding applied to both chemical (IP) and mechanical (organic) weeding. Common groundsel (Senecio spp.) and daisy (Bellis perennis) were less abundant in the tree strips under fully-grown ‘Elstar’ apple trees than under fully-grown ‘Golden Delicious’ apple trees of the adjacent row. The indirect weed species selection as a result of longtime glyphosate application as described in the literature was confirmed; this herbicide application allowed the subsequent germination of seeds of common chickweed (Stellaria media) and common groundsel (Seneccio spp.). In the present experiment, cuckooflower (Cardamine hirsuta) and meadow grasses (Poa spp.) dominated. Seeds in trays placed in the strip under the more shading cv. ‘Elstar’ and the less shading cv. ‘Golden Delicious’ tree canopy showed no difference in germination rate. Similarly, seeds of garden cress (Lepidium sativum) showed no difference in germination rate in trays with substrate mixed with cv. ‘Elstar’ or cv. ‘Golden Delicious’ leaves (1/10 v/v) thereby discounting an allelopathic cause. The present results of weed suppression by the cv. ‘Elstar’ trees after herbicide application are in line with those reported of vigorous apple cv. ‘Ingrid Marie’ on M2 rootstock in the literature in the same region. This weed suppression by fully-grown cv. ‘Elstar, Michielsen’ appears not only cultivar dependent, but may be related to shading by the canopy of vigorous fruit trees, irrespective of cultivation (IFP with herbicide or organic production with mechanical weeding) and possibly influenced by further factors.