Estimation of leaf area in pecan cultivars (Carya illinoinensis)

Abstract

Nondestructive and mathematical approaches of modeling can be very convenient and useful for plant growth estimation. The objective of this research was to develop a simple, accurate and nondestructive predictive model for leaf area (LA) estimation in different pecan cultivars (Carya illinoinensis) commonly found in Argentina. Linear and exponential regression equations were fitted and evaluated for pecan cultivars 'Desirable', 'Harry Super', 'Kernodle', 'Mahan', 'Mahan-Stuart', 'Shoshoni', 'Stuart' and 'Success' (grafted on seedling pecan rootstocks, 24 years old, planted at 15 × 15 m) using length (L) and width (W) alternatively or both L and W in a stepwise analysis as nonforced independent variables. Stepwise regression analysis using L and W variables for each cultivar fitted the data better than L or W alone. A general equation for all cultivars showed high accuracy (R2=0.93, p<0.0001). However, for certain cultivars, the general equation provided lower precision than the model for each cultivar. Therefore, we conclude that LA can be estimated through the general or individual model for each cultivar.