Genotypic variation in grain yield and flowering pattern in terminal and intermittent drought screening methods in rainfed lowland rice

• Published in: Field crops research 2015-04, Vol.175, p.26-36
• Main Authors: Monkham, T., Jongdee, B., Pantuwan, G., Sanitchon, J., Mitchell, J.H., Fukai, S.
• Format: Article
• Language: English
• Subjects: Aerobic rice; Delay in flowering; Drought resistance; Intermittent stress; Rice maturity group; Water deficit
• ISSN: 0378-4290; 1872-6852
• DOI: 10.1016/j.fcr.2015.02.003

•Grain yield reduction in intermittent drought screening was greater (23–33%) in earlier genotypes while it was 10–19% for the whole population.•Intermittent drought screening method can be effective in discriminating among genotypes when duration of each drought period is such that drought level is appropriate.•Under intermittent drought condition genotypes that exhibited shorter delay in flowering had smaller grain yield reduction percentage and delay in flowering was greater in earlier flowering genotypes than in later flowering genotypes.•There was no association between delay in flowering and flowering duration in terminal drought. Rainfed lowland rice in the Mekong region is mostly grown in drought-prone areas, and drought resistant varieties could be selected using either terminal drought (severe water stress after flowering) or intermittent drought (frequent mild stress) screening. Delay in flowering is considered as a putative trait for selection of drought resistant genotypes, and its effectiveness as an indirect selection trait and its association with genotype's flowering duration, i.e. the time taken from the first to last flower in a plot, was evaluated under the two drought screening methods. These screening methods together with flood control were used at two locations, Chum Phae and Ubon in north-east Thailand using a total of 70 genotypes with different growing duration. Delay in flowering was estimated for the two drought screening methods and flowering of individual panicles within a plot was recorded to determine flowering duration of each genotype in the flood and terminal drought conditions. Mean reduction in grain yield was 52–55% under the terminal drought screening while it was 10–19% for the whole population and 23–33% for early flowering genotypes in the intermittent drought. Delay in flowering under intermittent drought condition at both locations was associated with yield reduction; genotypes that exhibited shorter delay in flowering had smaller yield reduction percentage and larger number of panicles. At Chum Phae, where grain yield reduction was associated with flowering time, genotypes with longer flowering duration under terminal drought had smaller yield reduction, and this was found even among genotypes that flowered about the same time under flood condition. However, there was no association between delay in flowering and flowering duration. It is concluded that delay in flowering appears promising as a selection criter