Please use this identifier to cite or link to this item:
Type of publication: Straipsnis konferencijos medžiagoje kitose duomenų bazėse / Article in conference proceedings in other databases (P1c)
Field of Science: Miškotyra / Forestry (A004)
Author(s): Mozgeris, Gintautas;Bikuvienė, Ina;Jonikavičius, Donatas
Title: The opportunities and challenges of using airborne laser scanning for forest inventories in Lithuania
Is part of: Rural Development 2017 [elektroninis išteklius]: Bioeconomy Challenges : Proceedings of the 8th International Scientific Conference, 23-24th November, 2017, Aleksandras Stulginskis University. Akademija : Aleksandras Stulginskis University
Extent: p. 698-702
Date: 2017
Note: eISSN 2345-0916; eISBN 9786094491283
Keywords: forest inventory;laser scanning;compartment;stand volume
Abstract: The aim of this study was to test the usability of airborne laser scanning (ALS) data for stand-wise forest inventories in Lithuania based on operational approaches from Nordic countries, taking into account Lithuanian forest conditions and requirements for stand-wise inventories, such as more complex forests, unified requirements for inventory of all forests, i.e. no matter the ownership, availability of supporting material from previous inventories and high accuracy requirements for total volume estimation. Test area in central part of Lithuania (area 2674 ha) was scanned using target point density 1 m-2 followed by measurements of 440 circular field plots (area 100–500 m2). Detailed information on 22 final felling areas with all trees callipered (total area 42.7 ha) was made available to represent forest at mature age. Updated information from conventional stand-wise inventory was made available for the whole study area, too. A two phase sampling with nonparametric Most Similar Neighbor estimator was used to predict point-wise forest characteristics. Total volume of the stand per 1 ha was predicted with an root mean square error of 18.6 %, basal area – 17.7 %, mean diameter – 13.6 %, mean height – 7.9 % and number of tree – 42.8 % at plot-level with practically no significant bias. However, the relative root mean square errors increased 2–4 times when trying to predict forest characteristics by three major groups of tree species – pine, spruce and all deciduous trees taken together. Main conclusion of the study was that accuracy of predicting volume using ALS data decreased notably when targeting forest characteristics by three major groups of tree species
Affiliation(s): Kauno kolegija
Vytauto Didžiojo universitetas
Žemės ūkio akademija
Appears in Collections:Universiteto mokslo publikacijos / University Research Publications

Files in This Item:
marc.xml7.25 kBXMLView/Open

MARC21 XML metadata

Show full item record

Page view(s)

checked on Jan 6, 2020


checked on Jan 6, 2020

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.