The prediction and distribution of general yield classes of Sitka spruce in Scotland by empirical analysis of site factors using a geographic information system

Abstract

A framework has been developed to predict growth of Sitka spruce (Piceasitchensis (Bong.) Carr.) at a regional scale throughout Scotland based on an analysis of yield, site, and climatic factors. Using the general yield class (GYC) system for predicting forest production (m3•ha−1•year−1), the statistical model was integrated with a geographic information system to predict tree growth at a resolution of 1 km2. Site factor data from 487 sample sites were analysed along with the associated climate data derived from monthly 30-year means recorded at meteorological stations throughout the country. Multiple regression was used to develop and validate the model, which accounted for 59% of the variation in observed GYC. Standard errors of prediction were determined by analysis of variance to provide confidence limits for the 1-km2 GYC predictions. The distribution of these estimates of production was plotted in the form of a digital map for Scotland, housed within the geographic information system and providing a facility to readily analyse and summarise regional information.