Understanding crop yield requires navigating the intricate interaction between genotype and environments (GxE), which presents challenges for accurate predictions due to the plant's plastic responses. To study the environmental impacts on grain yield, we addressed two questions: Q1) Do environmental factors affect the consistency (R2trait) of trait performance? Q2) Can physiological relationships between traits across field environments be represented by simulation (APSIM-wheat)? To explore the GxE effect on trait performance of yield (R2yield), we utilise a multi-environment trials (MET) dataset featuring 220 genotypes and 45 environments over three years, five locations and three managements. The management includes combinations of two treatments: total nitrogen fertiliser application (high-HN: 220 kg N ha-1, low-LN: 110 kg N ha-1) and fungicide application (with-WF, without-NF). Resulting three managements are HN_WF, HN_NF and LN_NF. R2trait is calculated using standardised major axis (SMA) regression between two environments grouped by three environmental factors: management, location and year. Trait-trait Pearson correlation (r) was calculated between nine traits in both field and simulation dataset. Our results reveal: 1) significance differences in R2yield between all environmental factors, notably with HN_WF reducing R2yield comparing to HN_NF, indicating varying genotypic responses to fungicide. 2) simulation results confirm a trade-off between grain number and grain protein, but highlight discrepancies in dry matter allocation to grain yield and straw dry matter at maturity between fields and simulation. These insights underscore the indispensable role of understanding environmental influences on trait performance for precise phenotype prediction and crop modelling.