Abstract:
Most of the presently grown cultivars of rice (Oryza sativa L.) have poor
competitive ability against weeds if compared with traditional cultivars. Therefore,
effective weed control is an important but often costly and difficult management
practice in rice crop production. If weed competitive ability of rice cultivars could be
improved by changing the plant morphology and some agronomic practices, crop
losses and the cost of weed control in rice production would be reduced. Three field
experiments were conducted to identify weed competitive morphological traits of rice
and to evaluate the effect of delaying starter fertilizer application on weed competition
in rice.
The nature of rice-barnyardgrass competition varies greatly with rice cultivar
morphology. In an addition series experiment, the semi-dwarf, erect leaved, high
yielding cultivar BG 350, which possesses much of the plant characteristics required
for a high yield was affected much more than the cultivar BG 94-2, which has a tall
stature, high leaf area index and good seedling vigor.
In the varietal testing experiment, as predicted by the sensitivity analysis,
cultivars with rapid height growth, high leaf area growth and plant dry weight
increase during seedling stage, height and leaf area at maturity were associated with
high competitive ability. The weed competitive ability of tested cultivars varied from
relatively high in cultivars PPL and BG 94-2, moderate in BW 267-3 to poor in BG
1611 and BG 350.
The delay of starter fertilizer application by 14 days in cultivar BG 350
improved its competitive ability against barnyardgrass (BYG). Under BYG competition,
a 14-day delay of starter fertilizer application reduced the percent crop loss by
weed competition 6.64 and 22.99% under the normal planting density in the Maha
and Yala cultivation seasons, respectively.
The crop-weed competition model INTERCOM simulated yield losses for the
tested cultivars followed the trends observed in field experiments. However the
simulated values were always higher than the observed values.
