Top height has been an essential metric in forest practice and forest science over the last century. Despite the many issues that have been risen in regard to this metric, top height is still popular and useful these days. Crucially, top height is used as the engine of many growth and yield models worldwide, which are the base of national or regional forest planning. Unfortunately, predictions from most of these top height models do not come with an uncertainty assessment, which may have a non-trivial impact on decision making and therefore on forest policy. Our contribution seeks to identify the historical and technical reasons for this systematic omission and to explore the consequences of ignoring uncertainty in this context. To do this, the most popular modelling approaches are described along with the technical issues related to uncertainty. Then an alternative method to easily derive the uncertainty of top height predictions through a modification of the Generalized Algebraic Difference Approach is introduced. Finally, yield predictions are computed with M1, a top height-dependent yield model used to forecast timber production at the national level in Great Britain, both considering uncertainty in top height estimates and ignoring it.
No datasets are available for this submission.
No license information is available for this submission.