Modeling loss of water resources after cutting out trees in the Hyrcanian forest of Iran

The crisis of water shortage and the impact of El Nino in the past few years, are the most important issues that should be considered. The aim of this study was to introduce optimal models for monitoring water reserves in the trees with respect to water cycle in Glandrood mixed-beech forests, Noor. Therefore, after sampling of disks for 2- 5 meters of trunk length of 174 individuals, wood samples with fixed dimensions were oven-dried. After weighing each tree trunk in the field and difference between the wet mass and dry biomass considering the ratio of drought index, the whole water reserves were measured. To develop the models for predicting response variable, simple, multiple linear and nonlinear regression analysis were used. The results showed that simple and linear models had no considerable accuracy to predict the response variable. In contrast, power function, exponential and curve-models predicted the response variable with the high accuracy. Power function was the best model and that was developed to predict the water reserve value with the highest accuracy. Using this equation provides possibility for predicting the loss of water resource related to water cycle after cutting out the tress from the studied forests.