Winter Wheat Yield Prediction Using Normalized Difference Vegetative Index and Agro‐Climatic Parameters in Oklahoma

• DOI: 10.2134/agronj2017.03.0133
• Publication Date: 2017-10-12T21:58:28Z
• Authors: Ning Zhang; Chen Zhao; Steven M. Quiring; Jinlin Li
• Keywords: Anthesis; Winter wheat; Growing degree-day; Stepwise regression

Core Ideas Normalized difference vegetative index has a stronger correlation with yield than moisture and temperature indices. Optimal winter wheat model includes normalized index, variables. All three variables have different periods during which they are important. Model can accurately predict one month before harvest. Gridded data outperform station‐based for county‐level prediction. This article develops predicting ( Triticum aestivum L.) variations in Oklahoma, based on vegetation, moisture, conditions. A common structure is identified using stepwise regression vegetation indicator (normalized NDVI) jointing anthesis stages (March April), at emergence period (October November), (temperature TI) emergence, (October, March, April). The final accounts ∼70% of the variation be used to forecast yields 1 mo Spatially, it performs best northern central portions Oklahoma belt. performance similar regardless used. correctly predicted least 9 14 counties every year, case all counties. Our results also demonstrate that gridded meteorological generally outperforms prediction county level. methods this study applied identify most significant growth other regions.