Non-destructive method for estimating Annona muricata leaf area using linear leaf dimensions

Resumo

Estimating leaf area (LA) in fruit plants using non-destructive methods is a valuable parameter, since leaves function as the primary interface between plants and the environment and directly contribute to photosynthetic capacity. Therefore, the objective of this study was to determine allometric equations to estimate the leaf area of the soursop using linear dimensions (length and width) of the leaves. Eight hundred and twenty-five soursop leaves of the ‘Morada’ cultivar were randomly collected from healthy seedlings. Measurements of length (L), width (W), and leaf area (LA) were carried out by means of digitized images, followed by the calculation of the product between length and width (LW), length squared, and width squared. Linear, linear non-intercept, power, and exponential regression models were applied, based on the L and W dimensions of the leaves, to estimate the LA of the species. The best models and equations were chosen based on Pearson’s highest coefficients of determination and correlation, Willmott’s agreement index, Akaike’s lowest information criterion, root mean squared error, and mean absolute percentage error. LA can be estimated from the LW of leaves using the allometric equations ŷ = 0.72×LW (linear model without intercept) and ŷ = 0.69×LW^1.00 (power model). Allometric equations are accurate and safe for estimating LA in soursop seedlings, allowing repeated measurements throughout development and contributing to agronomic studies.