GGE biplot analysis of multi-environment trials of NS maize hybrids

Abstract

GGE biplot analysis is a method used to identify mega-environments (i.e grouping, division of locations, which are examined in a joint mega-environment), evaluate the genotypes within these mega-environments and select the most discriminating and the most representative locations within a single mega-environment. This paper analyses twenty hybrids at five locations in 2006 and 2007. For the first year, mega-environment analysis showed a greater similarity between locations Rimski Šančevi, Pančevo and Srbobran classifying them into one group (the potential mega-environment), whereas Sombor and Sremska Mitrovica were classified into another group. In 2007 grouping was different from the previous in that Rimski Šančevi and Sombor comprised one mega-environment, whereas Sremska Mitrovica and Pančevo comprised the other, while Srbobran was separated from both groups. Discrepancies in formation of the mega-environments can be explained by different climatic conditions in two research years. This was confirmed by analyzing the ideal environment where the discrepancies between those two years are also reported. The results in this study suggest that more reliable data on the mega-environments formation should be sought in multi-year studies of the certain genotypes in a number of locations.

Multilokacijski ogledi za prinos zrna izvode se svake godine kako bi se ocenila reakcija hibrida kukuruza na različite agroekološke uslove. Osnovni problem ovog načina ispitivanja hibrida jeste prisustvo interakcije genotip-spoljna sredina. Uzimajući u obzir interakciju, GGE biplot analiza pruža vizuelni prikaz ponašanja hibrida u različitim sredinama. Dvadeset hibrida kukuruza je ispitivano u toku 2006. i 2007. na pet lokaliteta po godini. Rezultati su predstavljeni preko GGE biplotova u cilju grupisanja sredina u tzv. megasredine i izdvajanja idealne sredine. Značaj izdvajanja megasredine ogleda se u odsustvu unakrsne interakcije koja remeti njihov rang, pa se hibridi mogu odabrati na osnovu prosečnih vrednosti određene osobine. Određivanje tzv. idealne sredine može biti od velikog značaja za izbor najreprezentativnijeg lokaliteta za testiranje, koji takođe ima i najveću moć diskriminacije hibrida.