Wheat Yield and Important Traits Influenced by Interaction of Potassium and Irrigation Levels Evaluated at Number of Locations in the Country by AMMI Analysis and Non-Parametric Measures

AMMI analysis of treatments consisted of levels of potassium with irrigations observed highly significant effects of locations, treatments, and T×L interactions for wheat yield. About 53.4% of the total variations in yield values was due to locations followed by 26.3% and 10.8% by treatments and interactions effects. Further analysis found 59.7% contributed by AMMI1 while 17.2% and 9.4% by AMMI2 and AMMI3 components for thousands grain weight as total of first two components cumulative to 76.9% of the total variation. The sums of squares for G×E signal and noise were 56.7% and 43.3% of interaction effects for grains per spike as the sum of squares of T×L signal was 2.58 times of treatments effects and IPC1 alone was 3.54 times the treatments effects. Measures ASV and ASV1 recommended T6, T5, T4 for wheat yield while measures utilized 81.6% of interaction sum of squares whereas MASV and MASV1 measures based on 98.4% identified T3, T5, T8, T4 treatments. Maximum average for thousands grains weight; GAI selected T8, T9, T6 whereas as per HM values treatments T5, T2, T8 would be more desirable. Grains per spike found the measures RPGV and HMRPGV settled for T8, T9, T5 treatments. Non parametric measures for yield observed Si1 selected T3, T2, T5 as opposed to T6, T4, T1 by Si2 values. T6, T4, T1 genotypes considered by Si3 Si4 measure considered T6, T4, T1 next Si5 for T6, T3, T4 and Si6 pointed towards T6, T4, T8 genotypes while Si7 favored T6, T1, T4 genotypes. Composite measures for thousands grains weight found NPi(1) for T3,T4,T7 while as per NPi(2) for T4,T3,T7, NPi(3) T4,T3,T2, NPi (4) found T4, T5, T7 as suitable treatment combinations. Multivariate hierarchical clustering as per Ward’s method for wheat yield observed first irrigation level with three potassium levels formed a cluster and other irrigation levels with potassium application remained in other one. At the first node of demarcation for thousands grains weight IPC5 exhibited MASV with MASV1, ASV1, IPC4, ASV, Si1 Si2 Si3 Si4 Si5 Si6 Si7 NPi(1), CV in one side and mean, GAI, PRVG, IPC1, HM, IPC1, NPi(2) NPi(3) NPi (4) on other side. The performance of treatments based on AMMI and non-parametric measures would be more meaningful for identification of suitable irrigation and potassium levels for wheat sustainable production.