Calibration and validation of regression model for non-destructive leaf area estimation of saffron (Crocus sativus L.)

Leaf area measurements are required in several agronomical and physiological studies. Usually, there is an interest for measurement methods that are simple, quick and that will not destroy the leaf. The objectives of this study were to establish equations to estimate leaf area (LA) by using length (L), width (W) and dry weight (DW) of saffron leaves. Leaves of different sizes were collected from the experimental area at different time intervals. Leaf area was measured with an automatic measuring device and leaf dimensions were determined with a ruler. An equation for estimating the leaf area from L and W was developed and validated with the area of leaves collected during different periods. Regression analyses of LA versus L, W, LW and DW led several models that could be used for estimating the area of individual saffron leaves. A exponential model having L as the independent variables [LA = 191.33e(L)0.0037] provided the most accurate estimate (R2 = 0.9373, RMSE = 27.68) of saffron leaf area. Validation of the regression model showed that the correlation between measured and simulated values by the use of this equation was very high.