The effects of weed residue management on pests, pest damage, predators and crop yield in upland rice in Cote d'Ivoire

The effects of weed residues on the abundance and activity of generalist arthropod predators and insect pests and on rice yields in upland rice were studied. Weed residues were placed in the experimental plots in three ways: (1)residues left in situ where weeds are uprooted; or (2) placed in small piles throughout the field; or (3) removed from the field but heaped at the edges to be burnt when dry. The study investigated the effects of residue disposal in upland rice under research station conditions designed to simulate conditions in farmers' fields. Spider activity was significantly greater in plots with scattered residues and strips of residue and least in plots from which residues had been removed. Staphylinid beetle activity was significantly greater in plots with a supplementary weed mulch than in other treatments. By contrast, spider density was greater when residue was piled or used as mulch rather than when scattered but the smallest populations occurred in the no-residue plots. Carabid beetles were significantly more abundant in plots with strips or piles of weeds and staphylinid beetles were most abundant in the mulched plots. Ants were not affected by weed residue placement. Among the ten pest groups studied, the abundance of only Chaetocnema sp., was significantly greater in plots with strips and piles than in other treatments, populations of other pests being unaffected by weed residue placement. In 1994, mean percentage of deadhearts (stem-borer damage) was significantly greater in the residue-free plots than in other treatments but in 1995 the differences between treatments were not significant. In both years, grain yield was not affected significantly by weed residue placement but was lowest in the residue-free plots.