Models for prediction of individual leaf area of forage legumes

Abstract

Leaf area is an essential variable for the quantification of other important leaf characteristics in physiological studies of plants, such as normalized photosynthetic rate and normalized phosphorus content. That is one of the reasons for the need of fast and accurate methods to estimate leaf area. The objective of this work was to fit linear or non-linear regression models to predict the individual leaf area of six species of forage legumes, based on digital images analyzed with the package LeafArea, R software. In a field experiment, 100 leaves were randomly collected from the following species: Crotalaria juncea (L.), Canavalia ensiformis (L.), Cajanus cajan (L.), Dolichos lablab (L.), Mucuna cinereum (L.), and Mucuna aterrima (Piper & Tracy) Merr., in which the central leaflet length and width were measured. Afterwards, digital images of each leaf were processed in R software for leaf area estimation. These estimates were used to fit leaf area prediction models; in fact, seventy leaves were used to fit the models; the rest of them were used for model validation. For the six species, the complete second-degree polynomial model, or derivative submodels, can be used to predict leaf area as a function of length and width of the central leaflet, presenting R2 above 0.98 and percentage absolute mean error below 9%. In these models, the effect of leaf width is generally greater than the leaf length. The R package LeafArea showed to be a very efficient tool for the estimation of leaf area through the execution of the software ImageJ, with high precision and easy calibration.