Journal of Taiwan Agricultural Research

Author:Lit-Fu Chan, Hsiu-Ying Lu, Meng-Li Wei, and Chun-Tang Lu

Abstract:

Plant leaf area is a useful observation in physiological and agronomic research. The development of a model for leaf area estimation requires a non-destructive method which is reliable and rapid. The objective of this study was to test twenty-four models for leaf area prediction in rice. Relationships between leaf area and a leaf area coefficient are used, but the coefficients K do not have the same dimension in all models. Some models require the measurement of both the length and the maximum width, others only use the length. Some models require the measurement of all plant leaves, others only measure a few number of leaves. Four rice cultivars, Taikeng 8, Taikeng 9, Taikeng 10 and Tainung 67 were used for model development. Experimental results indicated that different coefficients need not to be calculated depending on leaf position and tiller position. This model predicts the leaf area of an individual plant most accurately, but requires extra time and effort. It seemed that the leaf morphology differ from flag and lower leaves in rice plants, but it is not necessary to use the different K values to predict the total leaf area of an individual plant. The choice of a model depends mainly on the objectives of the investigation. If a large number of genotypes had to be evaluated (e.g. in a breeding programme), an overall average K value of 0.71 from the model with length and width measurements can be used for total leaf area prediction irrespective of the genotype considered. If a limited number of genotypes are studied but a large number of plants are sampled (e.g. to evaluate the influence of environmental factors upon growth), the models consuming less time and effort can be used. These models use the length and width of a few number of leaves (e.g. all the lower leaves, the 2^nd~4^th leaves or the 2^nd~3^rd leaves) only, or even require the measurement of length only for all the leaves, all the lower leaves or the first three lower leaves. The more rapid methods provided low accurate estimate of leaf area. The measurement of length is easier under field conditions than width and can be easily extended to the area estimation for non-expanded leaves.

Key words:Rice, Leaf area, Non-destructive measurement, Accuracy of Prediction, Model

Download: PDF Links