Genotype x Environment Interaction and Yield-Stability Analyses of Rice Grown in Tropical Inland Swamp

Adesola L. NASSIR, Omolayo J. ARIYO
2011 Notulae Botanicae Horti Agrobotanici Cluj-Napoca  
Twelve rice varieties were cultivated in inland hydromorphic lowland over a four year-season period in tropical rainforest ecology to study the genotype x environment (GxE) interaction and yield stability and to determine the agronomic and environmental factors responsible for the interaction. Data on yield and agronomic characters and environmental variables were analyzed using the Additive Main Effect and Multiplicative Interaction (AMMI), Genotype and Genotype x Environment Interaction, GGE
more » ... t Interaction, GGE and the yield stability using the modified rank-sum statistic (YSi). AMMI analysis revealed environmental differences as accounting for 47.6% of the total variation. The genotype and GxE interaction accounted for 28.5% and 24% respectively. The first and second interaction axes captured 57% and 30% of the total variation due to GXE interaction. The analysis identified 'TOX 3107' as having a combination of stable and average yield. The GGE captured 85.8%of the total GxE. 'TOX 3226-53-2-2-2' and 'ITA 230' were high yielding but adjudged unstable by AMMI. These two varieties along with 'WITA 1' and 'TOX 3180-32-2-1-3-5' were identified with good inland swamp environment, which is essentially moisture based. The two varieties ('TOX 3226-53-2-2-2' and 'ITA 230'), which were equally considered unstable in yield by the stability variance, ?2i, were selected by YSi in addition to 'TOX 3107', 'WITA 1', 'IR 8' and 'M 55'. The statistic may positively complement AMMI and GGE in selecting varieties suited to specific locations with peculiar fluctuations in environmental indices. Correlation of PC scores with environmental and agronomic variables identified total rainfall up to the reproductive stage, variation in tillering ability and plant height as the most important factors underlying the GxE interaction. Additional information from the models can be positively utilized in varietal development for different ecologies.
doi:10.15835/nbha3915591 fatcat:wbf5w2at55hbbp3cmdm3pk6cay