Mapping Brant's oak (Quercus brantii Lindl.) mortality using geostatistical methods in Dasht-e Barm, Fars province

This research was conducted to study the spatial distribution of Brant's oak (Quercus brantii Lindl.) mortality using geostatistical prediction and mapping approaches in Dashte-e Barm, Fars province. Field sampling was performed based on a 500m×500m systematic random grid and 1200 m2 rectangular forest plots. Different geostatistical methods were used for plotting anisotropic empirical semivariogram and surface creation. Results of cross validation showed that ordinary kriging with spherical model achieved superior results. The models were used for wall-to-wall prediction maps with four classes, including 60% mortality. The 25-60%, mortality class occupied the largest area (3827 ha), whereas the 60% tree mortality across the entire study site was shown to be less than 25%. This research concluded that Geostatistical kriging methods could be applied to predict and map missing tree mortality values in forest stands. Our analysis suggests that these methods can be used to generate prediction and probability maps in zagros oak stands for overarching goals such as forest mortality, pest and disease managements.