Multivariate analyses reveal key traits in the identification of nitrogen‐use‐efficient cultivars of common beans

Abstract

Common bean (Phaseolus vulgaris L.) cultivars with higher nitrogen-use efficiency (NUsE) are considered to be the main strategy to reduce the dependence on nitrogen (N) fertilizers. In this sense, the objective of this study was to identify key traits and common bean cultivars with higher NUsE using multivariate statistical approaches.Forty carioca common bean cultivars were analyzed in four environments under absence-(0 kg N ha −1 ) and presence N (40 kg N ha −1 ) applied as top-dressing. A total of 15 NUsE-related traits were evaluated. Multivariate analyses of variance identified the presence of a three-way interaction of genotype, N level, and environment, showing that the differential behavior of cultivars depended on the N level and environment evaluated. Strong multicollinearity was observed among the total set of traits studied; however, after removal criteria used, a collection of eight traits remained for subsequent analyses. The path analysis indicated that the dry matter biomass could be used for the indirect determination of NUsE in contrasting N conditions. The cultivars IAC-Alvorada, IPR Sabiá, and IPR Saracura performed the best from those selected in both N conditions by factor analysis. The cultivars that stood out positively in this study can be used as parents in breeding programs aimed at developing cultivar with higher NUsE and less dependence on N fertilizers.