Application of NDVI-Based Crop Sensor in Alfalfa Selection for Improving Breeding Process

Abstract

Alfalfa (Medicago sativa) is a globally important forage crop; however, improvements in its biomass yield have stagnated due to its complex genetic architecture and the costly, labor-intensive phenotyping. This study evaluated the potential of the normalized difference vegetation index (NDVI) to predict biomass yield and enhance selection efficiency in alfalfa breeding programs. Specifically, nineteen alfalfa experimental populations (AEXP 1–19) and one control cultivar (OS 66) were evaluated over two growing seasons in Croatia. NDVI was measured at four development stages using a GreenSeeker sensor and compared with forage yield, dry matter yield, and plant height. NDVI values varied significantly among genotypes, years, and growth stages, ranging from 0.23 to 0.87, and increased consistently from early to late vegetative phases. Strong positive correlations were observed between NDVI and forage yield (r = 0.543–0.843) and plant height (r = 0.537–0.738) at early vegetative, late vegetative, and early bud stages. Conversely, NDVI at the mid-vegetative stage correlated negatively with yield and height (r = –0.622 to –0.794). High-performing populations (AEXP 2, AEXP 15, AEXP 18) also exhibited the highest NDVI values. NDVI is a reliable, non-destructive indicator for early selection of high-yielding alfalfa genotypes, although multi-location validation is advised to confirm its broader applicability.

Plant phenotyping relevance

NDVIセンサーによる非破壊的な生育・収量形質推定を育種選抜へ適用し、収量や草丈との相関で妥当性を評価しており、フェノタイピング手法が中心である。

Code and data availability

The paper reports NDVI (GreenSeeker), yield, and plant-height measurements on 19 alfalfa experimental populations plus a control, analyzed with STAR v2.0.1. No public phenotype dataset, images, code, models, or supplement with such assets is described. The Data Availability Statement states the original contributions (

No evidence-backed public reproduction asset is currently recorded.