English

Abstract

This study was conducted in the Iraqi Kurdistan region in January 2014 to determine the individual leaf area of oleander (Nerium oleander L.) by easy, accurate, inexpensive, and nondestructive method. Simple, multiple and exponential regression analyses were used by length (L) and width (W) and their combinations as independent variables and with leaf area as dependent variable to determine more accurate models (high coefficient of determination and less MSE). The results showed that the best fitting models that show more accurate estimation of oleander leaf area, compared to other models, were the simple linear regression that depends on length multiple width for Koya and Erbil cities and the total leaves of the two cities plants. On the other hand, the best fitting multiple linear equations were those which depend on square length and square width for Koya city and the total leaves of the two cities plants, whereas for Erbil city the best model was that depends on leaves with square length and width. Multiple linear regressions were the more accurate among the models, followed by simple linear regression, whereas the exponential model had the lowest accuracy. All coefficients of regressions values were found to be significant at the P < 0.0001 level.