Perbandingan Model Regresi untuk Pengukuran Luas Daun Gandum di Daerah Tropis

Abstract

The growth of wheat plants can be determined by measuring leaf area index and net assimilation rate. Both measurements require leaf area data. Measurement of leaf area of wheat in Indonesia requires a method that is not only accurate, but also easier and cheaper. One of them is the regression method. The purpose of this study was to determine an accurate regression equation model in predicting wheat leaf area. This research was conducted from March to June 2021 at the Experimental Station and Plant Production Technology Laboratory, Faculty of Agriculture, UNPAD, Jatinangor, Sumedang. The materials used in this study were various leaf area printing papers from wheat plants aged 14 days after planting (DAP), 28 DAP, and 42 DAP. The regression equation was assembled from the relationship between leaf area with leaf width and length, then compared with the actual leaf area that measured by scanning. The results showed that the linear, quadratic, cubic, and logarithmic regression equations had a coefficient of determination of more than 90% to predict leaf area, at the age of 14, 28, and 42 DAP, as well as all plant ages. Quadratic regression had a limit of data that can be entered, so it needed circumspection in using the formula. Cubic regression tended to have better accuracy in predicting leaf area at 14, 28, and 42 DAP, but the accuracy was the same as other regression equations at all plant ages.