Relationships between Blue‐ and Red‐based Vegetation Indices and Leaf Area and Yield of Alfalfa

Abstract

The need for site‐specific yield assessments of alfalfa (Medicago sativa L.) has spurred interest in developing methods to remotely sense biomass at harvest. Relationships between reflectance‐based vegetation indices (VIs) and yield and yield‐components of alfalfa have not been fully characterized. The objectives of this study were to evaluate the relationships between blue‐ and red‐reflectance based VIs and canopy variables such as leaf area index (LAI), mass shoot−1, shoot length, and alfalfa yield. Canopy reflectance was obtained with two reflectance spectrometers 1 d before each of five harvests in 2005 within rainfed and subsurface drip‐irrigated alfalfa. Blue‐ and red‐based normalized difference vegetation indices (NDVIs) and wide dynamic range vegetation indices (WDRVIs) at three levels of a near‐infrared (NIR) reflectance‐scalar (‘α’ = 0.1, 0.05, and 0.01) were calculated and regressed on alfalfa canopy variables. A quadratic‐plateau model was used to determine when VIs no longer detected yield increments. Both blue‐ and red‐based NDVIs and WDRVIs exhibited significant (P < 0.0001) saturative responses to LAI, yield components, and dry matter (DM) yield. Decreasing α widened the estimable yield range (0–1.82 vs. 0–2.76 Mg ha−1 and 0–2.60 vs. 0–3.74 Mg ha−1, respectively) of both blue‐ and red‐based WDRVIs. Significant (P < 0.0001) yield regression models within the effective range of the VIs (<Yieldmax) were found within two harvests in 2005 and when data were pooled across all harvests. These results indicate that the use of a NIR reflectance‐scalar can extend the range of herbage biomass (to 3.74 and 2.76 Mg ha−1, respectively) within which blue‐ and red‐based indices may be used to estimate alfalfa yield.