The present work establishes regression models to estimate leaf area of Curcuma alismatifolia and Curcuma zedoaria. To use these of species as ornamental plants is necessary to establish adequate cultivation techniques. Thus, the determination of leaf area is very important, once it is used to evaluate plant response to environmental factors and crop techniques. The use of prediction models to estimate leaf area is a simple, accurate and nondestructive method. At the stage of flowering, a hundred leaves of C.alismatifolia ('Pink' and 'White') and C.zedoaria were collected for each species and cultivar. Maximum length (L), maximum width (W) and real leaf area (RLA) were measured with a leaf area meter (LI-3100). The relation between RLA and the L, W and the product of length by width (LW), was studied through linear regression models. The models RLA = 0.59048 LW (C.alismatifolia 'Pink'), RLA = 6.08410 + 0.52161 LW (C.alismatifolia 'White') and RLA = 0.70233 LW (C.zedoaria) are adequate to estimate leaf area.
thai tulip; zedoary; floriculture; linear regression