Models for estimating leaf area in the ‘Palmer’ mango

Techniques for measuring leaf area are basic for evaluating plant growth in the mango. As such, the aim of this study was to determine the leaf area of the ‘Palmer’ mango using mathematical models proposed by the present study, and compare the results of the proposed models with models available in...

Full description

Saved in:
Bibliographic Details
Published in:Agro@mbiente on-line Vol. 13; p. 279
Main Authors: Santos da Silva Junior, José Luiz, Rodrigues, Marcos Sales, Braga, Gabriella Amaral, Regis, Ester Silva
Format: Journal Article
Language:English
Published: Universidade Federal de Roraima 06-12-2019
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Techniques for measuring leaf area are basic for evaluating plant growth in the mango. As such, the aim of this study was to determine the leaf area of the ‘Palmer’ mango using mathematical models proposed by the present study, and compare the results of the proposed models with models available in the literature for other mango cultivars. The mango leaf was simulated as a function of leaf length (L) and width (W) using two distinct geometric models: an ellipse and a rosacea petal. Models found in the literature and determined for other cultivars, were also tested. The values for leaf area were obtained using the ImageJ software and taken at their actual value; these were later compared with the values achieved by the geometric models. The models were tested for quality of prediction through cross-validation. The models proposed in the present study were not superior to the best models found in the literature. The model LA = 3.80 + 0.67 (LW) achieved the best performance, with a mean absolute percentage error (MAPE) of 3.78%. Using only length, the best model was LA = 0.0142C2 + 6.1902C - 49.444, with a MAPE of 4.07%. The use of mathematical models proved to be a suitable option for estimating leaf area in the ‘Palmer’ mango. Moreover, the use of R2 as the only form of model quality assessment can lead to errors in choosing the best model.
ISSN:1982-8470
1982-8470
DOI:10.18227/1982-8470ragro.v13i0.5642