Abstract
Knowledge about the growing stock and the cutting potentials of stands, as well as predictions of growth and yield, are essential aspects of forest management planning. Growth predictions are obtained using complex simulation systems, whose accuracy and precision are difficult to predict. The uncertainty of growth and yield predictions, as well as the uncertainty of the stand-level inventory data behind the predictions, are not usually taken into account sufficiently in the planning process. Furthermore, the uncertainty resulting from the increasing use of updated inventory data as planning data should be studied in connection with current forest planning practices. However, the lack of suitable comprehensive re-measured study data sets with true planning data must also be noted. This dissertation provides new knowledge on the uncertainty related to forest management data. It also addresses the possibilities to use updated forest inventory data as forest management planning data and evaluates assessment methods for predicting the uncertainty of forest management data. Furthermore, four alternative simulation methods are evaluated as regards their ability to generate assessment errors in forest management planning data for further research. The usability of updated forest management planning data is evaluated also by looking at the suitability of the proposals for forest management operations as derived from the updated data. The accuracy and precision of stand-level inventory were found to be moderate, although the costs and time spent in field work are considered to be fairly high. The variation between measurers was substantial. This variation in stand-level inventory data should be noted in forest management planning. The assessment of uncertainty of updated forest management data was approached by means of two different methods, i.e. by modelling observed (past) errors and by applying the k-nearest neighbor method with multiobjective optimization. The uncertainty assessments of growth and yield predictions using these methods were found to be feasible with large stand data. The main advantage of the studied methods is in that both bias and accuracy can be assessed. However, the methods require independent contemporary data, which is their main drawback. Modelling observed (past) errors and k-nearest neighbor are quite easy to apply in forest simulation systems if only contemporary models and distance functions are estimated. The utilization of both methods does not considerably add the calculation time even when dealing with growth predictions for large areas. Stand-specific predictions of uncertainty were also found to be satisfactory. According to the results of this study, updated stand inventory data can be used as a forest management planning data with respect to the accuracy of the updated stand characteristics. Updated stand inventory data were also found feasible with respect to treatment proposals when the mean stand characteristics and regulations and recommendations of current forest management practices were considered. However, treespecific data are considered to be a slightly more suitable in this context.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.