The most appropriate statistical distribution for studding effect of land use changes on distribution of Oak trees (Quercus brantii) canopy in Zagros forests of Ilam Province

Regarding to extensive forest land use change in Iran and especially central Zagros region, this study paid to the effects of land use changes on the statical distribution of tree’s canopy in this region. For this purpose three 10 ha mass located in the city of Ilam was considered that there are including contains mass of the northern edge of the Goljar village with a background of long-term agriculture, the southern edge of the Goljar village with a background of destruction (extracting of stones, sand and Agriculture), and the area is located about a mile from the east Goljar village as witness stand. For this purpose statistical functions beta, gamma, Johnson, Weibull, normal and lognormal, were used which they are frequently used in biological studies. Then ranking functions and significant tests were performed with Anderson-Darling goodness of fit test. The results showed that in the Long-term land use change to agriculture, various long-term damage and witness area are the most appropriate function to model the distribution of trees were including lognormal, weibull and beta respectively. The results of this study indicate that in effect of land uses based on statistical distributions (Beta, Gamma, Johnson, Weibull, Normal and lognormal) in the different canopy classes has been created dispute, and these changes demand a variety of management for Protection and development of these masses.