Estimating Absorbed Photosynthetic Radiation and Leaf Area Index from Spectral Reflectance in Wheat¹

TLDR

Plant growth models require estimates of leaf area and absorbed photosynthetically active radiation, and previous work has shown that spectral reflectance, PAR absorption, and leaf area index are interrelated. This study aims to develop a procedure that uses spectral reflectance to simultaneously estimate absorbed PAR and leaf area index in wheat. The authors derive absorbed PAR from the normalized difference between near‑infrared (800–1100 nm) and red (600–700 nm) canopy reflectances, then use this ND along with four leaf‑angle classes to estimate LAI across three geographic sites. The ND correlates with PAR absorption, with bare‑soil reflectance and NIR scattering as key influences, and the method predicts LAI with a mean standard deviation of 0.5–0.9 (0–6 LAI) but becomes less sensitive above LAI 6 due to diminishing ND response.

Abstract

Abstract Some plant growth models require estimates of leaf area and absorbed radiation for simulating evapotranspiration and photosynthesis. Previous studies indicated that spectral reflectance, absorption of photosynthetically active radiation (PAR), and leaf area index (LAI) are interrelated. The objective of this study was to establish a procedure by which spectral reflectance can be used to simultaneously estimate PAR absorption and LAI. A method is presented for estimating the quantity of absorbed PAR by wheat ( Triticum aestivum L.) plants and their LAI based on the normalized difference (ND), transformation of the near infrared ( ρ n = 800 to 1100 nm) and red ( ρ r = 600 to 700 nm) canopy reflectances. The results, from a theoretical analysis and field measurements, indicated that ND correlates with the fraction of PAR absorbed by wheat canopies. Bare soil reflectance and scattering of near infrared radiation by foliage elements were the major factors that influenced the relation between ND and PAR absorption. The estimated PAR absorption values, based on the ND, and four classes of assumed leaf angles (45°, 60°, 75°, and spherical), were used to indirectly evaluate LAI of wheat for three different geographical locations. The standard deviation on mean predicted to measured LAI's for the three locations varied from 0.5 to 0.9 for a range of 0 to 6 LAI. The method is considerably less sensitive in predicting LAI above 6.0 since the sensitivity of ND to changes in LAI becomes small (<0.01), due to small changes in canopy reflectance.