Kaya Altop, Emine/0000-0002-0987-9352 WOS: 000367410500002 Weeds limit hazelnut productivity through competition and interference. Field experiments were conducted from spring 2010 through 2013 to evaluate string trimming alone or combined with herbicides in hazelnut orchards to control mugwort (Artemisia vulgaris L.), purple nutsedge (Cyperus rotundus L), and burning nettle (Urtica urens L.). Treatments were mowing with a string trimmer (ST) alone, or ST followed by glyphosate, glyphosate + diflufenican, and glyphosate + carfentrazone-ethyl. These combinations also were applied with pendimethalin and oxyfluorfen as pre-emergence applications. The experiments were conducted in Fatsa-Ordu, Turkey, on a sandy clay soil with multi-stemmed trees (ocak). The combination treatments improved weed control compared to ST alone. Glyphosate plus carfentrazone-ethyl application 15 days after ST was the most effective treatment for control of these weeds. This combination was more effective in reducing seed bank reserves of A. vulgaris, C rotundus and U. urens than herbicides applied alone or in mixtures throughout all seasons. Hazelnut yield was not significantly affected by herbicide treatments in 2010 or 2011. Hazelnut yield increased during the latter half of the experimental period, in 2012 and 2013. In a mixed population of A. vulgaris, C. rotundus and U. urens, 89 plants per m(2) caused 12.1% yield loss. When the density increased to 256 plants per m2, yield losses increased to 29.7%. Hazelnut treated with ST followed by glyphosate plus carfentrazone-ethyl had the highest financial return. Post-emergence weed management systems in hazelnut should include ST followed by glyphosate plus carfentrazone-ethyl applications for maximum returns. (C) 2015 Elsevier Ltd. All rights reserved.