Reliable assessment of vegetation indices under field conditions is essential for crop monitoring and precision agriculture. This study quantified the environmental sensitivity of the normalized difference vegetation index (NDVI) in five triticale varieties using standardized field acquisition protocols in temperate conditions. Multispectral imagery was collected across all developmental stages with a Mapir Survey3W sensor under controlled midday conditions to minimize shadow effects, generating NDVI, SAVI, and EVI2 datasets. A total of 15 acquisition sessions were conducted between 9 April and 9 July 2025, with concurrent measurements of soil and atmospheric parameters, including soil temperature, moisture, pH, nitrogen, phosphorus, potassium, illumination, air temperature, and relative humidity. Multivariate linear regression assessed the contributions of individual factors, reduced models (up to three predictors), and the full model to NDVI variability. Individual factors explained 2–50% of the variance, while reduced models accounted for up to 70%, with relative humidity identified as the most influential variable. The full model, including all predictors, explained a substantially higher proportion of variance; however, this result may reflect overfitting due to the limited sample size and interdependent predictors. These findings indicate that NDVI is strongly influenced by environmental conditions, and multivariate modeling enhances interpretation and predictive capability. The framework may improve reproducibility and comparability of multispectral field measurements, although further validation across multiple sites and growing seasons is recommended.
Atanasov et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: