Comportamento de herbicidas pré-emergentes sob semeadura de arroz irrigado em palha de azevém

Abstract

Establishment of no-tillage system in lowland rice has been developed using the ryegrass (Lolium multiflorum) as the main cover crop for this purpose. Cover crops can affect the herbicide dynamics, either by herbicide straw retention or by modification in soil conditions such as soil moisture and temperature. Also, cover crops may affect weed emergency, such as barnyadgrass (Echinochloa sp.), one of the most important weed in lowland rice areas. To date, research in this subject was not performed considering specificity of rice in lowland areas. Therefore, this research was conducted to study: (i) the herbicide dynamics on no-tillage system through a literature review (Chapter 1); (ii) Barnyardgrass control and density, at field scale, by the herbicides clomazone, imazapyr+imazapic, oxyfluorfen, pendimethalin and quinclorac applied on different ryegrass straw quantities (0, 1.5, 3 and 4.5t ha-1); herbicide interception by straw; and rice grain yield (Chapter 2); (iii) the barnyadgrass response, in greenhouse conditions (GH), using factors explored in chapter 2 plus different periods of simulated rain after application (2, 7 and 14 days after application) (Study 1), and herbicide washoff in ryegrass straw (Study 2) (Chapter 3). Field experiments were performed in two locations and two years, and in GH replicated two times. In field, the experiment was conducted in randomized complete block design with four replications in factorial arrangement 6 × 4, with first factor constituted by herbicides and second factor by ryegrass straw quantities. In GH, the research was performed under completely randomized design with three factors, using the two factors of Chapter 2 plus the simulated rain periods (Study I). The washoff experiment was conducted using the maximum level of ryegrass straw. In field, barnyadgrass control was superior in the straw treatments than in bare soil for herbicides oxyfluorfen, pendimethalin and quinclorac in the first year of study, despite the fact that the spray volume interception by straw was 26, 41 and 67% (average among years and locations) when straw levels were 1.5, 3 and 4.5 t ha-1, respectively; control with clomazone and imazapyr+imazapic was not influenced by the ryegrass straw. There was no significant herbicide by straw level interaction for barnyadgrass density, observing that in herbicide absence, the straw resulted in greater barnyardgrass density than in plots of bare soil. Rice grain yield decreased as more straw was present on the soil surface in two locations and two years of study (Chapter 2). In GH, barnyadgrass control was greater than 90%, independent of treatment (Study 1). The herbicide washoff in straw was in follow order: pendimethalin = clomazone < imazapyr = imazapic < quinclorac. Therefore, two first herbicides presented superior environmental losses and/or straw sorption, and the lasts one were more recalcitrant and were more easily washable from ryegrass straw (Chapter 3). Based on these results, oxyfluorfen, pendimethalin and quinclorac efficacy was positively affected by ryegrass straw on soil surface in years that occur a rain delay after application (first year of study); in years that rain occur soon after application, no influence of straw was reported (second year of study). Furthermore, the herbicides that potentially will potentially reduce efficacy, in case of rainfall delays, are clomazone and pendimethalin, although good efficacy have been seen in greenhouse conditions.