Evaluating Maize Genotype Performance under Low Nitrogen Conditions Using RGB UAV Phenotyping Techniques

Area of Science:

• Agricultural Science
• Plant Breeding
• Remote Sensing Technology

Background:

• Maize yield in Africa is often limited by low soil nitrogen availability.
• Traditional breeding for high-yield maize under field conditions is slow and expensive.
• Remote sensing and high-throughput plant phenotyping (HTPP) offer efficient alternatives.

Purpose of the Study:

• To assess red-green-blue (RGB) and multispectral indices for maize phenotyping under low nitrogen.
• To evaluate ground-based and unmanned aerial vehicle (UAV) HTPP measurements for yield prediction.
• To compare HTPP indices with agronomic data for improved yield assessment models.

Main Methods:

• Utilized RGB and multispectral indices (NDVI, SPAD) for HTPP from ground and UAV platforms.
• Collected data under managed low-nitrogen conditions for 64 maize genotypes.
• Developed multivariate regression models combining HTPP indices and agronomic traits (ASI, AD, PH).

Main Results:

• Ground-based RGB indices (hue, GGA, NDLab) and UAV-based indices (GGA, CSI) showed strong yield correlations.
• SPAD values, especially in the vegetative stage, correlated closely with grain yield.
• Multivariate models incorporating RGB indices achieved higher R² (>0.60) than those using only agronomic or sensor data (R² >0.50).

Conclusions:

• RGB-based HTPP shows significant potential for enhancing maize yield assessments, particularly under nitrogen-limiting conditions.
• Genotypic variation in low-nitrogen performance and grain yield loss index (GYLI) was substantial.
• High yield under optimal conditions does not guarantee performance under low nitrogen, highlighting the need for specific breeding strategies.