甘肃省马铃薯区试品种产量性状和环境鉴别力评价分析

为筛选甘肃省2018—2019年马铃薯(Solanum tuberosum L.)品种区域试验中丰产、稳产和适应性强的品系,同时评价试点的代表性和鉴别力,本研究收集整理2018—2019年在甘肃省安定区、秦州区、渭源县、永昌县、临夏县、庄浪县6个试点种植的各育种单位育成的15个品系的产量数据,用主效可加互作可乘模型(AMMI)和基因型与环境互作分析模型(GGE)联合对产量进行基因型与环境互作分析,评价品系的丰产性、稳定性、适应性和环境的鉴别力。方差分析结果表明,马铃薯品系的产量在基因型效应(G)、环境效应(E)和互作效应(G×E)中均呈现极显著差异(P<0.01),环境效应平方和占总方差平方和最大,其次是基因型效应,互作效应占比最小。AMMI分析结果显示,丰产稳产的品系是0730-185;品系凯薯2号、0730-156和0730-185在永昌县的适应性较好,庄薯8号在安定区的适应性较好,GN-99、W1在渭源县的适应性较好,GN-99、GN-122和W1在临夏县的适应性较好;AMMI环境的鉴别力由强到弱依次为2018年的永昌县、2018年的安定区、2019年的临夏县等。GGE分析结果显示,丰产稳产的品系是GN-99;品系凯薯2号在永昌县的适应性最好,品系GN-122在渭源、安定区、秦州区、庄浪县、临夏县适应性最好;GGE环境的鉴别力由强到弱依次为2018年的永昌县、2019年的临夏县、2019年的安定区等。利用AMMI模型和GGE模型共同对甘肃省马铃薯区域试验的产量数据联合分析可以更好地评价并筛选优良品种。两种模型对基因型丰产性分析的准确性取决于其解释变异的大小,AMMI对基因型稳定性和环境鉴别力的分析优于GGE,GGE对基因型适应性分析优于AMMI。两种模型鉴定出高产稳产品种(系)有GN-99和0730-185;高适应性的品系试点组合有GN-122-临夏县、GN-122-庄浪县、GN-122-秦州区、GN-122-安定区、GN-122-渭源县、凯薯2号-永昌县;环境的鉴别力最好的两个试点为永昌县、安定区,其他试点均属于以上两试点的品种生态亚区。本研究结果为马铃薯新品种推广和后续的区域试验提供了参考。

Evaluation and Analysis of Yield Traits and Environmental Discriminability of Potato Varieties in Regional Trial of Gansu Province

Data analysis was conducted on the regional testing of potato (Solanum tuberosum L.) varieties in Gansu Province from 2018 to 2019. The aim of present research was screening high-yield, stable and highly adaptable lines as well as representativeness test sites for future study. In this study, 15 potato lines were bred by breeding institutions in Gansu Province, planted in Anding, Qinzhou, Weiyuan, Yongchang, Linxia and Zhuanglang of Gansu Province in 2018 and 2019. The interactions between genotype and environment were analyzed by AMMI (Additive main multiplicative interaction) and GGE (genotype + genotypes and environment interactions) to evaluate the productivity, stability and adaptability of the lines and the discriminability of the environment. The results of variance analysis showed that yield of potato lines were significantly different on genotype effect, environmental effect and interaction effect between genotype and environment (P<0.01). The square sum of environmental effect on the plot yield accounts for the largest square sum of total variance, followed by genotype effect, and interaction effect accounts for the least. AMMI analysis showed that the high-yielding and stable lines were 0730-185; Highly adaptable lines were KaiShuNo.2, 0730-156 and 0730-185 in Yongchang, were ZhuangShuNo.8 in Anding, were GN-99 and W1 in Weiyuan, were GN-99, GN122 and W1 in Linxia; The orders of discriminability of test-environment from strong to weak were 2018-Yongchang, 2018-Anding, 2019-Linxia, etc. GGE analysis showed that the high-yielding and stable lines was G5; most adaptable lines were KaishuNo.2 in Yongchang, and GN-122 in Weiyuan, Anding, Qinzhou, Zhuanglang and Linxia. The orders of discriminability of test-envirnment from strong to weak are 2018-Yongchang, 2019-Linxia, 2019-Anding, etc. AMMI model and GGE model were used to jointly analyze the yield data of potato regional testing in Gansu Province, which could better evaluate and select the superior varieties. The accuracy of the two models in the analysis of genotype productivity depends on the size of their explained variation. AMMI is superior to GGE in the analysis of genotype stability and discriminability of environment, and GGE is superior to AMMI in the analysis of genotype adaptability. The high-yield and stable lines distinguished by the two models were GN-99 and 0730-185. Combination of highly adaptable lines and test-environment were GN-122-Linxia, GN-122-Zhuanglang, GN-122-Qinzhou, GN-122-Anding, GN-122-Weiyuan, and KaishuNo.2-Yongchang. The two test-environment with the highest discriminability were Yongchang and Anding, and all the other test-environment belong to the mega-environment of the above two test-environment. The results provided the reference data for extension of new potato varieties and subsequent regional testing.