Phenotypic characterization of native quinoa accessions using vegetation indices

Loayza-Loza H.; Álvarez-Cáceres A.; Céspedes-Florez E.; Arrosquipa-Quispe J.-C.; Mayta-Anco M.; Gutiérrez-Rosales R.-O. · Idesia (2026) · DOI: 10.4067/0718-3429-idesia-2026-44-e06

Resumen

Accurate phenotyping of quinoa (Chenopodium quinoa Willd.) is essential for identifying agronomic traits of interest and strengthening conservation and breeding programs. In this study, 25 native accessions from the CICA-FCA-UNSAAC Germplasm Bank were evaluated using multispectral images acquired with a multispectral camera mounted on an unmanned aerial vehicle (UAV). The objective was to establish associations between vegetation indices and agronomic variables, particularly grain yield. Twenty-four vegetation indices were calculated, and analysis of variance (ANOVA) along with Tukey’s tests were performed, identifying ten indices with significant differences among accessions and up to seven statistically distinct groups. Furthermore, it was observed that the date corresponding to 86 days after sowing showed the highest correlations between vegetation indices and grain yield, highlighting nine indices with coefficients of determination (R2) greater than 0.65. Individually, seven accessions showed particularly high linear correlations (R2 up to 0.90). These results demonstrate that spectral indices can accurately reflect phenotypic variability and predict crop yield, validating the use of multispectral sensors mounted on UAVs as an efficient, objective, and low-cost tool for quinoa phenotyping in the Andes.

Instituciones: Universidad Nacional Agraria La Molina; Universidad Nacional de San Antonio Abad del Cusco; Universidad Nacional de San Agustín de Arequipa; Universidad de Huánuco