Prediction of saffron crocus corm mass by geometrical attributes

Saffron (Crocus sativus L.) is the most precious spice plant in the world. The only method for the reproduction of saffron is Cultivation of corm. It is necessary to know relationships between mass and size of saffron corms in order to design and develop of planting, harvesting, grading and handling equipments. The objective of this research was to determine the most accurate models to predict saffron corm mass based on geometrical attributes. Models for predicting the mass of corms were classified as follows: (1) single or multiple variable regressions of saffron corm mass and dimensional characteristics, (2) single or multiple variable regressions of saffron corin mass and projected areas and (3) single regression of saffron corm mass based on its actual Volume and calculated volume assuming the saffron corm shapes of sphere, oblate spheroid and ellipsoid. Three sets of corms were considered from Torbat-e-Heydareyeh, Kashmar, and Gonabad fields in Khorasan province of Iran. Some physical characteristics that were needed for the saffron corm mass modelling were determined for the three regions corms. The results showed that there were not significant differences between mean values of mass, height, major diameter, geometric mean diameter and projected areas of saffron corms for the three regions. However, there were significant differences between saffron corm minor diameter, volumes, corm density, bulk density and porosity for the three regions. Therefore, the mass of the corms was modelled based on the height, major diameter and projected areas irrespective to the regions, whereas in the third classification models, the mass of the corms was modelled for each region. The statistically significant fitted models were selected based on the higher coefficient of determination (r(2)) and lower regression standard error (RSE). The results of mass modelling showed that the prediction of saffron corm mass based on the major diameter (M=0.0017a(2.5168), r(2) = 0.94, p < 0.05) and the first projected area (M = 0.0019 (PA(1))(1.2916), r(2) = 0.96, p < 0.05) were the most appropriate models. The model that predicted the mass of saffron corm based on the estimated Volume of an ellipsoid shape was found to be the most appropriate. The developed models and data of physical characteristics of saffron corms could be used for designing and developing the saffron corm grader and other cultivation equipments. (C) 2009 Elsevier B.V. All rights reserved.