FORESTSCAN

 

Terrestrial laser scanning technology for multi-resource forest inventories

DOWNLOAD 2009 REPORT AS PDF (pdf 327Kb) 
DOWNLOAD 2009 REPORT IN IRISH (pdf 326Kb)  

PROJECT TEAM
Prof. Maarten Nieuwenhuis, University College Dublin*
Charles Harper, University College Dublin
Martin van Leeuwen, University College Dublin
Mark Tarleton, PTR Ltd
Enda Keane, TreeMetrics Ltd
Garret Mullooly, TreeMetrics Ltd
Taye Mengesha, University College Dublin

* Email:_maarten.nieuwenhuis@ucd.ie

COMPLETION DATE: April 2012

BACKGROUND
Ireland is a high cost economy, and this is reflected in the delivered-in price of wood. Research and development aimed at reducing costs, and thereby increasing competitiveness in roundwood production, is the key driver behind this project. Effective planning and decision-making in modern multi-purpose and sustainable forest management requires up to date and accurate data in order to evaluate the range of potential options for the future management and utilisation of a forest.

OBJECTIVES
The three main objectives of the study are:

  •     An investigation of the basic principles of terrestrial laser scanning technology and its applicability to (multi-resource) forest inventories.
  •     An evaluation of existing data analysis software for forestry applications. Development of new software, for a range of applications, will be investigated.
  •     A cost-benefit analysis for the introduction and use of this technology over a range of potential inventory applications
     
    PROGRESS
    Research continued with two stands of Sitka spruce (Picea sitchensis (Bong.) Carr.) of different ages at Clonmel and Kinnity scanned early in the year. Measuring different aged stands addresses the effect of size and density of the stand on estimating diameter and height using the laser scanner. Trees in the stands within a 15 m radius of the laser scanner were scanned before and after removing the lower branches (up to a height of 3 m). Sample trees of different diameters were felled and prescribed measurements taken from the felled trees. The resulting point cloud data were extracted using FARO Scene software, and were processed with AutoStem software to derive dbh, tree height and number of stems for estimating basal area, volume and stand density. The results were compared with the manually measured parameters (i.e. obtained after felling the trees).
    Generally, tree diameters in the scanned forest are estimated from the point cloud data with acceptable accuracy, except in the upper stem due to the occlusion by branches, and in the lower stem (i.e. the first 1.5 m) due to interference from ground unevenness and the irregular shape of the lower stem. The number of trees per plot and the diameters of a number of trees were under-estimated in some cases due to occlusion of the stem by other trees in the stand. Pruning did not significantly improve diameter estimation. The position of trees within the plot (i.e. distance from the scanner) did not show a significant effect on the accuracy of diameter estimation.
    In the autumn, PTR and TreeMetrics initiated a second phase of measurement and scanning operations in the Clonmel plots. This will provide a second set of sample plot data representing two full growing seasons between Phase 1 and Phase 2 measurement and scanning operations. These data will be used to study the use of terrestrial laser scanning technology in forest crop growth/yield monitoring.
    Martin van Leeuwen completed his Masters thesis entitled: On the use of laser range-finding techniques for forest inventory studies. It includes interesting work on the concepts of branch recognition algorithms and the complexities involved in attempting to train software to recognise real branches on the tree and exclude artefacts.

    ACTIVITIES PLANNED
  •     Other plots in different terrain, broadleaved stands and stands which are ready for clear felling will be identified to be measured during the second phase of scanning.
  •     Scanning will be done in the previous plots to monitor annual growth by measuring and scanning plots before and after growing seasons.
  •     Scanned data will be processed and analysed.
  •     Paper writing and presentations at workshops and international conferences (e.g. Silvilaser 2010 in Germany).
  •     Identify stands which were previously scanned with aerial LiDAR and ready to be harvested will be scanned by terrestrial LiDAR for comparison.

    OUTPUTS
    Mengesha, T. and Nieuwenhuis, M. 2009. Retrieving forest parameters using laser scanning technology [Poster Presentation]. The 9th international conference on LiDAR applications for assessing forest ecosystems. Texas A&M University, College Station, Texas USA
    Van Leeuwen, M. 2009. On the use of laser rangefinding techniques for forest inventory studies. Unpublished MSc thesis, University College Dublin, Ireland.
    Nieuwenhuis, M. 2009. COFORD's PLANSFM Research Programme: Planning and Implementation of Sustainable Forest Management. Presentation to the COFORD Council, AFBI Field station, Hillsborough, Co Down, 24 July 2009.
    Harper, C. and Nieuwenhuis, M. 2009. PLANSFM - Planning and implementation of sustainable forest management [Poster presentation.] UCD School of Agriculture, Food Science and Veterinary Medicine Research Day, 8 December 2009.

     


    2008 report

    download 2008 report as pdf (pdf 314Kb)  

    PROJECT TEAM
    Prof. Maarten Nieuwenhuis, University College Dublin*
    Martin van Leeuwen, University College Dublin
    Mark Tarleton, PTR Ltd.
    Enda Keane, TreeMetrics Ltd.
    Garret Mullooly, TreeMetrics Ltd.
    Taye Mengesha, University College Dublin

    * Email: maarten.nieuwenhuis@ucd.ie

    COMPLETION DATE: December 2011

    BACKGROUND
    Ireland is a high cost economy, which is reflected in the delivered-in price of wood. Research and development aimed at reducing costs and thereby increasing competitiveness in roundwood production is the key driver behind this project. Effective planning and decision-making in modern multi-purpose and sustainable forest management requires up to date and accurate data in order to evaluate the range of potential options for the future management and utilisation of a forest.

    OBJECTIVES
    The three main objectives of the study are:

  •     An investigation of the basic principles of terrestrial laser scanning technology and its applicability to (multi-resource) forest inventories;
  •     An evaluation of existing data analysis software and the development of new software, for a range of forestry applications;
  •     A cost-benefit analysis for the introduction and use of this technology over a range of potential inventory applications.

    PROGRESS
    An investigation of the state of the art of laser scanning has been conducted and the report is currently being written. The investigation includes terrestrial, aerial and space-borne LIDAR techniques, and their relative advantages and potential to support each other. It is understood that the combined application of these three techniques could enhance conducting forest inventories both on a local and global scale.
    A survey has been completed and highlighted the data analysis software for terrestrial laser scanning data from forests. Other work includes the development of branch detection algorithms. This will be included in written work resulting from the development of a new software algorithm that is designed to detect and map a tree's branching structure.
    Following the measurement of the first 19 plots, PTR and TreeMetrics have submitted their independently compiled data sets to University College Dublin for processing and comparative analysis.

    ACTIVITIES PLANNED
  •     The data collected from the 19 plots will be analysed.
  •     Based on the results of the analysed data and manually measured parameters in the 19 plots, other sites will be selected by including various species of forests at different ages.
  •     The terrestrial scanning and measurement of sample trees by callipers will be carried out together with TreeMetrics and PTR.
  •     Work will commence on improving the reported weakness of the Autostem software together with the software developers from Dresden University and TreeMetrics.
  •     A research paper will be prepared on the results of the first data analysis.
  •     A second phase of field work will commence in spring 2009.

    OUTPUTS
    Nieuwenhuis, M. 2008. FORESTSCAN - Terrestrial laser scanning technology for multi-resource inventories. Irish Timber and Forestry 17(3): p 32 - 35.