Mathematical models for leaf area estimation of brazilian cultivars of goiabeira-serrana (Acca sellowiana)

Abstract

Feijoa [Acca sellowiana (O. Berg.) Burret] is a fruit-bearing species native to southern Brazil and northeastern Uruguay. The knowledge of total leaf area is important for evaluating plant management, studying plant physiology and estimating phytosanitary damage. The mathematical models used to estimate feijoa leaf area did not consider differences in leaf area between cultivars. The objective of this work was to evaluate, adjust, and validate mathematical models for estimating leaf area in four feijoa Brazilian cultivars. Data was collected over two harvest periods; the first to establish the mathematical models and the second to validate them. The leaves of the cultivars SCS411 – ‘Alcântara’, SCS412 – ‘Helena’, SCS414 – ‘Mattos,’ and SCS415 – ‘Nonante’ were collected from Active Germplasm Bank of Epagri, in São Joaquim, SC, Brazil. Leaf area (AF), length (C), width (L), and shape index were recorded, and the data analyzed using linear and quadratic regression models. The estimation of leaf area from the mathematical model presents high precision and simplicity of execution. The equation y=0.7345x+0.0298, where x is the product length x width of the leaf (L*C), presented the best results.