Genotype by environment (G × E) interaction analysis for seed yield and other contributing traits in linseed (Linum usitatissimum L.) across conventional and zero budget natural farming production systems in north-western Himalayas

Development of high-yielding and stable linseed (Linum usitatissimum L.) genotypes in Himachal Pradesh under zero budget natural farming system (ZBNF) is an absolute need. A study was conducted to evaluate 30 elite genotypes of linseed for yield and stability across sixteen different environments over two years using Additive main effects and a multiplicative interaction model. For seed yield the environment, genotype and genotype x environment interaction (GEI) effects were highly significant (p <0.001), with contributions to total observed variation of 89.74, 0.81 and 8.08%, respectively. Out of the two production systems, ZBNF was observed to show consistently poor mean yields as compared to the conventional system across all locations. However, locations Kangra and Dhaulakaun performed better under the ZBNF system than Palampur and Bajaura. ZBNF was also identified as less discriminating with weak interactive forces. As per Eberhart and Russell model, the most stable and high-yielding genotype was Surbhi whereas, as per AMMI model, the ASV and GSI values indicated Giza-7 (G18), KL-285 (G16), KL-311 (G1) and Surbhi (G28) as stable and high in grain yield. For yield attributing traits such as primary branches per plant genotypes Janki (G27) and KL-314 (G4), for secondary branches per plant genotypes KL-263 (G10) and Janki (G27), for seeds per capsule genotypes KL-279 (G13), Binwa (G30) and Him Alsi-2 (G19), for trait biological yield per plant Jeewan (G22), for harvest index KL-236 (G6) and Him Alsi-1 (G26) and for 1000 seed weight genotypes Him Alsi-1 (G26) and KL-285 (G16) showed high stability along with high mean performance and therefore could be selected. Genotype KL-284 (G15) showed specific adaptation under ZBNF system of Bajaura and is therefore recommended for production under respective environments after further evaluation.