Effects of High-risk Herbicides on Bacterial Community Structure in Tobacco Field Soils with Different Textures

A field experiment was conducted on two types of tobacco planting soils (sandy loam and clay loam) in Bijie, Guizhou, aiming to explore the effects of three high-risk herbicides in preceding crops (atrazine, quinclorac and nicosulfuron) on the structure of soil bacterial microbial communities, thereby providing scientific basis for herbicides application and soil microenvironment conservation in tobacco-grain rotation fields. The results showed that after the application of three herbicides at conventional doses, the bacterial counts in both types of soils was inhibited, and the bacterial communities in soils with different textures responded differently to herbicide residues. Among them, the most evident inhibition occurred in the atrazine treatment of sandy loam, and the nicosulfuron treatment of clay loam, respectively. In both types of soil textures, Actinobacteria was the dominant phylum, but different herbicides had varying effects on the relative abundance of bacteria at the phylum level. At the genus level, quinclorac and atrazine treatments had the greatest effects on the relative abundance of each genus in sandy and clay soils, respectively. Meanwhile, herbicide treatment, and its interaction with soil texture significantly influenced the soil bacterial community diversity, but soil texture has a significant influence only on the Chao 1 index. Among them, in sandy loam the nicosulfuron treatment exhibited the highest Chao1 index, while the quinclorac treatment exhibited the lowest. In clay loam, only the nicosulfuron treatment resulted in a significantly lower Chao1 index compared to other treatments. In clay loam, the nicosulfuron treatment was lower than that of the other treatments. For the suitable selection and application of herbicides in tobacco-grain integrated areas, it is advisable to take into consideration the differences of soil texture and their effects on soil micro-ecological responses.