On the potential of kriging for estimation and mapping of forest plantation stock (Case study: Beneshki plantation)

This research was conducted to investigate on spatial structure and estimation of tree attributes in a forest plantation in the Caspian region of Iran. Field sampling was performed based on a 50m×125m systematic grid in a maple stand (Acer velutinum Boiss.) at age of 18 years using circular samples of 200m2 area. Overall, 96 sample plots were measured in 63 hectare and 14.25 hectare was inventoried as full census area, as well. Experimental variograms for forest stem basal area, stem density and tree height attributes were calculated and plotted using the geo-referenced inventory plots. The calculated variograms of basal area and height showed a high spatial autocorrelation, which fitted by spherical models. However, stem density showed a large amount of nugget effect. Based on variography results, optimal sampling distance for stem basal area and density obtained 165m and 350m, respectively. Estimations for basal area and height were made by ordinary block (15m×15m) kriging and cross-validation results showed that all the estimations are accurate. Furthermore, the estimated kriged mean of basal area showed no significant difference to the real mean in the full census area, while kriging in term of statistical precision was two times better than sampling. Therefore, geostatistics is able to capture and describe the spatial variability as well as estimates tree attributes (not stem density) in this kind of forest plantation, accurately.