Developing early-estimating normalized difference vegetation index calibrations for grain yield and technological quality of bread wheat in semi-arid rainfed conditions

Abstract

In this study, the usability of in-season estimated yield (INSEY) and optical-read sensor Normalized Difference Vegetation Index (NDVI) at the Zadoks30 stage (Z 30) in rainfed conditions for predicting bread wheat grain yield and technological quality was evaluated. Data from nitrogen fertilizer application field trials conducted in eight consecutive years in 14 environments were used to develop regression equations to predict yield and some quality attributes in rainfed conditions. The trials were divided into two groups, low NDVI (LNE) and high NDVI (HNE), according to the magnitude of NDVI. Technological bread quality parameters and yield were higher in the HNE. The increase in grain protein content (GPC) and macro SDS (MSDS) sedimentation against nitrogen rates became significant beyond 60 kg N ha⁻¹ in the LNE and 30 kg N ha⁻¹ in the HNE. Linear relationships occurred between NDVI and observed values of grain yield (R² = 0.743, p<0.001) and GPC (R² = 0.963, p<0.001). The use of NDVI and INSEY values at Z 30 stage facilitated the development of equations capable of predicting grain yield, GPC, and gluten quality. The developed equations can be used in the nitrogen fertilization strategy during the tillering stage to achieve specific yield and protein.