Doctoral Degrees (Forest and Wood Science)

Permanent URI for this collection

https://scholar.sun.ac.za/handle/10019.1/545

Browse

Browsing Doctoral Degrees (Forest and Wood Science) by browse.metadata.advisor "Astrup, Rasmus"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Industrial forestry compartment characteristics: the effect on end of rotation processes and the forestry supply chain
    (Stellenbosch : Stellenbosch University, 2022-03) Ackerman, Simon Alexander; Talbot, Bruce; Astrup, Rasmus; Stellenbosch University. Faculty of AgriSciences. Dept. of Forest and Wood Science.
    ENGLISH ABSTRACT: Industrial plantation management aims to produce timber at the end of its rotation of a desired size to maximise the volume on the site, through the application of intensive silviculture. These silvicultural operations maximise this volume primarily through maintaining an acceptable level of tree size and tree size distribution uniformity. With tree size strongly related to timber harvesting productivity, uniformity of the trees on the stand should also benefit these end of rotation processes. For this reason, a series of experiments were undertaken to test the effect of tree size uniformity of mechanised cut to length (CTL) harvesting. These were firstly, to understand the fine scale effect of tree-by-tree mechanised CTL harvesting productivity and overall costs and cost sensitivity for two different sized harvesting machines (medium sized Ponsse Beaver and larger Ponsse Bear). Secondly, to explore the consequences of varying tree distributions on the productivity of mechanised CTL harvesters by comparing the marginal and weighted mean productivity at clear-felling age. Lastly, to investigate the systems effect of mechanised CTL harvesting of marginal first thinning trees by a combination harvester-forwarder machine in a fully selective thinning operation. In the first investigation, a long-term data set of harvesting machine data were collected and analysed using a non-linear mixed effects modelling approach to model the productivity for each of the different sized machines applied to the machine-based tree data. Results showed clear differences between the productivity of these two machine sizes and a clear decline of productivity when these machines reached their technical tree size limit. In terms of machine costs, the higher capital cost machine, the Bear, was only slightly greater to that of the Beaver. This was due to the high productivity of the Bear in the tree sizes harvested by this machine. Cost sensitivity further reinforced this; with the Bear’s capital cost being offset by the machine’s productivity, even at high interest rates. This investigation showed that both machines could effectively operate in any tree size, although one could infer that the smaller machine is not technically suited to large trees, and these could damage the machine in the long term. The second investigation used the models and data from the first investigation and compared these marginal productivity results to a weighted mean productivity when using the tree distribution rather than assuming a uniform distribution of trees in a stand. This investigation showed the disparity between using a mean tree size to predict harvesting productivity vs the tree size distribution. This disparity equates up to 8 m3 PMH-1 for the smaller Ponsse Beaver. This component highlights where remote sensing technology can be used to measure stand tree distributions more accurately than current enumeration techniques. The third part of the study investigated the application of a small-scale combination harvester forwarder in small and poor form first thinning trees. The investigation highlighted the challenges faced with these small and poorly formed trees that often the standing tree volume is not effectively translated to timber extracted to roadside. This necessitated a quantification of this timber ‘waste’ and an adjustment of the harvesting component’s productivity to take this into account. Further to this, since the machine forms a harvesting and forwarding system (the harvesting and forwarding functions cannot operate at the same time), the systems productivity was calculated, with the overall productivity limited by the least productive function. This final experiment aligned the previous two, where heterogeneous tree size effect of CTL harvesting is quantified. The overall objectives of the study were met, and it was found that tree size uniformity does affect CTL harvesting productivity. To this end, this needs to be matched against investments in more intensive silviculture. The study also highlighted the connection with the need for accurate stand characterisation to efficiently predict tree metrics for efficient timber harvesting.