Genotype by environment interactions (GxE) complicate selection in common bean (Phaseolus vulgaris L.). Crop models can play a valuable role by helping plant breeding programs to better understand GxE. The objectives of this study were to evaluate agronomic, morphological, and phenotypic traits of a recombinant inbred lines population derived from the inter-gene pool cross [Jamapa (Mesoamerican) x Calima (Andean); RIJC] across five environments and generate data to validate a gene based eco-physiology model using an independent population (RISR) from the cross of Stampede x Redhawk. Field trials were conducted across North Dakota, Florida, Puerto Rico, Colombia (Popayan and Palmira), and Nebraska from 2011 to 2013. Resolvable row-column designs and RCBD with three replications and two-row plots were used to evaluate the populations. Analysis of variance was performed using the PROC MIXED procedure of SAS. Genotype main effect and GxE interaction (GGE) biplots were assessed for seed yield components and RISR were compared to the RIJC population. The results suggested different mega-environments depending on the trait of interest. Locations relatively more homogenous can be clustered and North Dakota usually stands alone. The biplots allowed detecting stable genotypes or subsets which were best adapted to mega-environments. Moderate to high narrow-sense heritability estimates (0.55 to 0.87, 0.25 to 0.76 and 0.56 to 0.69 for phenological traits, seed yield components and other agronomic traits, respectively), were observed suggesting various traits such as flowering time, physiological maturity, seeds per pod, plant height, among others, may be used as selection criteria to improve common bean. The populations responded relatively more similar for most of the traits assessed in North Dakota. However means across locations for RIJC differ significantly from RISR grown alone in North Dakota. Seed yield losses for RISR population in drought conditions were 54.3% and 59.0% in 2012 and 2013, respectively. This study will help developing the next generation gene-based crop model along with a high-resolution linkage map and identification of potential candidate genes controlling various traits. Ideal genotypes suited for specific mega-environments can be designed. These new techniques should shorten the cycle needed to develop superior varieties by implementing efficient early generation selection.