Estimation of leaf area in sweet cherry using a non-destructive method

Abstract

Leaf area measurement can be a time consuming process and requires sophisticated electronic instruments. The objective of this research was to develop a simple, accurate, non-destructive and time saving predictive model for leaf area (LA) estimation in sweet cherry trees. Linear regression equations were fitted and evaluated for three cultivars and two training systems using alternatively the length (L), the width (W) and their product (L*W) as independent variables. Regression using L*W variable fitted the data better (R2 = 0.994) than L or W (R2 = 0.863 and 0.787, respectively). The slopes using L*W as the explanatory variable were between 0.6776 and 0.6442 for different combinations of cultivar and training system. Combinations of cultivar – training system showed different slopes (P<0.05), except for «tatura-Bing» and «vase-Lapins». A general equation had a slope of 0.6612 with R2 = 0.993 (slightly lower than considering all combinations of cultivar and training system). Validation of the general equation using extra data from a «Lapins/ Mahaleb» orchard showed high accuracy (R2 = 0.9826), but underestimated LA. However, the general equation can be used for predicting LA for practical purposes, such as estimating Leaf Area Index of commercial orchards.

Keywords

• leaf area
• prunus avium
• non
• destructive estimation
• leaf length
• leaf width