Grain yield and agronomic traits, key breeding goals for most crops, are under complex genetic control and subject to environmental interactions. Plant breeders require that the genetic architecture of agronomic traits be identified in locally relevant germplasm. Our goal was to conduct a genome‐wide association study to identify quantitative trait loci for eight agronomic traits in a diversity panel containing 402 Pacific Northwest winter wheat (Triticum aestivum L.) lines. Phenotypic evaluations of yield, yield components, phenology, and plant height were conducted under drought, rainfed, and irrigated conditions in 2012, 2013, and 2014. Linkage disequilibrium analysis, using 6492 single nucleotide polymorphism (SNP) markers from the 9K SNP array, determined that 70% of markers were in significant LD with other markers and formed 539 LD blocks with an average distance of 3.5 cM. Population structure analysis was conducted using 719 SNP tags that represent marker distribution. Three genetic subgroups were identified in the population, explaining 3 to 38% of the total phenotypic variation. A mixed linear model with 3653 SNP markers of known genetic position, the first two membership coefficients (Q = 2) of population structure and a compressed kinship (Kc = 192) was used in the association analysis. In this study, a total of 94 marker‐trait associations that were significant (false discovery rate < 0.05) were identified. Multiple trait associations for yield and component traits were identified on chromosomes 2B, 4B, 5A, 5B, 6A, and 7B. These findings highlight the possibility of using historical recombination in northern latitude winter wheat to identify genomic regions associated with desired agronomic traits.