Importance of mega-environments in evaluation and identification of climate resilient maize hybrids (Zea mays L.)

Multi-location experiments on maize were conducted from 2016 to 2019 at ten locations distributed across two agro-climatic zones (ACZ) i.e., ACZ-3 and ACZ-8 of Karnataka, India. Individual analysis of variance for each location-year combination showed significant differences among the hybrids; similarly, combined analysis showed a higher proportion of GE interaction variance than due to genotype. Mega-environments were identified using biplot approaches such as AMMI, GGE, and WAASB methodologies for the years 2016 to 2019. The BLUP method revealed a high correlation between grain yield and stability indices ranging from 0.67 to 1.0. Considering all three methods together, the three location pairs Arabhavi-Belavatagi, Bailhongal-Belavatagi, and Hagari-Sirguppa had three occurrences in the same mega-environment with a value of 0.67, and these location combinations consistently produced winning genotypes. Among the common winning genotypes identified, it was G7 during 2016 and 2017 and G10 during 2018 and 2019, based on WAASBY. The likelihood of Arabhavi-Nippani, Hagari-Mudhol, and Dharwad-Hagari occurring in the same mega-environment is minimal because they did not share the same winning genotype, with the exception of a small number of events. Despite being in the same agro-climatic zone, Arabhavi, Hagari, and Mudhol rarely had a winning genotype in common. An agro-climatic zone is grouped based on climatic and soil conditions which doesn’t consider GE interaction of cultivars thus, releasing the cultivars for commercial cultivation considering mega environments pattern would enhance the yield for the given target region.

2016至2019年，在印度卡纳塔克邦的两个农业气候区（agro-climatic zones, ACZ）即ACZ-3与ACZ-8内的10个试点地点，开展了玉米多地点试验。针对各地点-年份组合单独开展方差分析，结果显示各玉米杂交种间存在显著差异；联合方差分析结果表明，基因型×环境（Genotype × Environment, GE）互作方差的占比高于基因型本身的方差占比。本研究采用AMMI、GGE、WAASB等双标图分析法，对2016至2019年的数据开展了超级环境（mega-environments）的识别工作。最佳线性无偏预测（Best Linear Unbiased Prediction, BLUP）法分析结果显示，籽粒产量与稳定性指数间存在0.67至1.0的高度相关性。综合三种分析方法的结果，Arabhavi-Belavatagi、Bailhongal-Belavatagi与Hagari-Sirguppa这三组地点对，在同一超级环境中重合的次数达3次，对应相关系数为0.67，且此类地点组合始终可筛选出优胜基因型（winning genotypes）。基于WAASBY指标，本次研究筛选出的共通优胜基因型为：2016年与2017年为G7，2018年与2019年为G10。Arabhavi-Nippani、Hagari-Mudhol与Dharwad-Hagari这三组地点对，处于同一超级环境的概率极低，因为除极少数案例外，它们并未共享共同的优胜基因型。尽管同属一个农业气候区，但Arabhavi、Hagari与Mudhol三地极少拥有共同的优胜基因型。农业气候区的划分仅依据气候与土壤条件，未考虑品种的基因型×环境互作效应。因此，结合超级环境分布模式开展品种商业化推广与种植，可有效提升目标区域的作物产量。