Epsilon Open Archive

Abstract

The aim of this work was to study the influence of biological and technical factors on the effect of flame weeding, as a basis for reducing the energy consumption and increasing the effective ground speed required for good weed control Responses of weeds and test plants were evaluated in the field. Effects of propane dose and ground speed were described by logistic models. The sigmoidal dose-response and speed-response curves imply that propane dose and the ground speed can be adjusted to the required control effect, the weed flora and the developmental stage of the plants, A 95% reduction in susceptible annual weed species, with 0-4 true leaves, was achieved at propane doses of 10-20 kg ha^-1, and the weeds were completely killed at 20-50 kg ha^-1 (900-2300 MJ ha^-1). Considerably higher doses were needed at later stages and for more tolerant species. The most tolerant species could not be controlled with one treatment, regardless of dose. Flamers with different characteristics were studied Flamers with covered burners were generally more effective than an open flamer, especially on larger plants and tolerant species. The effective ground speed was generally higher for flamers with a higher fuel input, but only up to a certain level, A flamer with a relatively high propane consumption of 34 kg h^-1 per metre working width (440 kW m^-1) allowed an effective ground speed of 8 km h^-1 when smaller plants were treated, whereas the effective ground speed was 2,6 km h^-1 for a flamer with the more usual burner power of 12 kg h^-1 m^-1 (150 kW m^-1). Temperatures from the flamers were measured above the ground under weed-free conditions in the laboratory. There was generally a high correlation between different thermal parameters, e. g. the maximum temperature and the temperature sum, measured from the flamers in the laboratory, and the weed reduction in the field, although discrepancies were found, Flame weeding is useful although the method could be further improved.