Abstract
<title>Abstract</title> <p>Purpose Frost represents a recurrent climatic hazard in Andean agroecosystems, exerting severe physiological stress on crops through intracellular ice formation and membrane rupture. In the Andean region of Ecuador, nocturnal frosts occur frequently and severely damage crops (e.g., maize) cultivated by smallholder farmers. Aim This study assessed the spectral response of Zea mays var. Zhima (maize) under frost conditions at the Nero Experimental Farm (3100 m a.s.l.), employing unmanned aerial vehicle (UAV) imagery to monitor canopy vigor across five phenological stages. Methods The Normalized Difference Vegetation Index (NDVI) was calculated from aerial surveys. To understand the relationship between this vegetation index, climatic conditions, and management treatments (plant density and nitrogen fertilization), linear mixed-effects models were fitted. Aerial images were processed using Agisoft Metashape, and statistical analyses were performed in R. Results The most parsimonious model revealed sharp declines in NDVI that coincided with minimum temperatures and high relative humidity. Statistical analysis confirmed that climatic factors exerted stronger effects on NDVI dynamics than fertilization or planting density. Minimum temperature (χ² = 1204.3, P < 0.001) and maximum relative humidity (χ² = 657.3, P < 0.001) emerged as the dominant predictors. These findings indicate the limited resilience of Zhima maize under extreme thermal stress conditions, emphasizing the need for adaptive management strategies and breeding programs to enhance frost tolerance in Andean production systems. Conclusion This research underscores the utility of UAV-based NDVI monitoring as a diagnostic tool for quantifying frost impacts and guiding climate adaptation strategies in vulnerable highland agriculture.</p>