Cover Crop Termination Timing on Rice Crop Production in a No-Till System

Measuring shikimic acid accumulation in response to glyphosate applications can be a rapid and accurate way to quantify and predict glyphosate-induced damage to sensitive plants. The objective of this paper was to evaluate the effect of cover crop termination timing by glyphosate application on rice (Oryza sativa L.) yield in a no-till system. A factorial experiment, arranged in a split-plot design, was conducted for 2 yr. Treatments consisted of cover crops (main plots) and timed herbicide applications (subplots) to these cover crops (30, 20, 10, and 0 d before rice planting). There was a decrease in rice yield from 2866 kg ha(-1) to 2322 kg ha(-1) when the herbicide was applied closer to the rice planting day. Glyphosate application on cover crops increased shikimate concentrations in rice seedlings cultivated under palisade grass (Brachiaria brizantha), signal grass (B. ruziziensis), guinea grass (Panicum maximum), and weedy fallow (spontaneous vegetation) but not under millet (Pennisetum glaucum), which behaved similarly to the control (clean fallow, no glyphosate application). Glyphosate applications in the timing intervals used were associated with stress in the rice plants, and this association increased if cover crops took longer to completely dry and if higher amounts of biomass were produced. Millet, as a cover crop, allowed the highest seedling dry matter for upland rice and the highest rice yield. Our results suggest that using millet as a cover crop, with glyphosate application far from upland rice planting day (10 d or more), was the best option for upland rice under a no-tillage system.