Improving estimation of weed suppressive ability of upland rice varieties using substitute weeds

Ideal weed-competitive rice (Oryza sp.) varieties are high yielding under both weed-free and weedy conditions, and have strong weed-suppressive ability (WSA). WSA is the ability to suppress weed growth and reduce weed seed production, benefiting weed management in both the current and the subsequent growing seasons. WSA is determined by assessing weed biomass under weedy conditions. However, natural weed growth in field conditions is heterogeneous, resulting in high experimental error in phenotyping studies. Using crops as substitute weeds is one approach for improving the accuracy of estimation of WSA. Four upland-rice experiments were conducted using 14 diverse varieties in Benin between 2009 and 2011 to examine if the use of substitute weeds can improve estimation of WSA, and help identify varieties with strong WSA. Two rice varieties (Aus 257, which was also included in 14 varieties tested, and IR 1552, a purple-leaved rice) and cowpea (cv. KVx396-18) were tested as substitute weeds, and compared with natural weed infestation. At 50-62 days after rice sowing, weed biomass (weeds, rice or cowpea) were measured as an indicator of WSA. While WSA was more heritable in plots with substitute weeds than with natural weed infestation, there was no large difference in heritability among substitute weeds. In only one of the four experiments was the effect of variety on WSA not consistent between substitute weeds and natural weeds. But in this experiment, Aus 257 consistently showed strong WSA in both plots with substitute weeds and plots with natural weed infestation. Superior WSA of Aus 257 was also observed in the other experiments, and associated with higher biomass accumulation. These results indicate that the use of substitute weeds can offer an efficient selection approach for improving weed-suppressive ability of upland rice. Aus 257 can be used as donor for developing varieties with strong weed-suppressive ability. (C) 2014 Elsevier B.V. All rights reserved.