Comparison of flaming and infrared radiation techniques for thermal weed control

The effectiveness of two types of thermal weeder was evaluated in field experiments on Sinapis alba L. as the test species. Both thermal weeders used propane combustion to provide heat from either a covered flamer or an infrared radiator. The flamer showed better performance than the infrared radiator on plants at the four-leaf stage, but the opposite was true on plants at the cotyledon stage. Both thermal weeders required an effective dose of propane of about 60 kg ha -1 to obtain 95% reduction of plants at the zero- to two-leaf stage. The effective speed was slightly higher for the flame weeder than for the infrared radiator, mainly because the propane consumption per unit working width was higher for the flamer. Temperatures under the thermal weeders were measured by thermocouples in the laboratory. At equivalent propane doses, the flamer gave higher temperatures than the infrared radiator at I cm above ground, but temperatures were similar at 3.5 cm height. There was a highly positive correlation between the temperature sum and control of S. alba by flaming and infrared radiation based on temperatures measured at the upper position. It is concluded that evaluating thermal weeders based solely on indiscriminate temperature measurements is unreliable : proper comparison has to take into account the weed species, plant size and the propane consumption per unit working width.