Date of Award

Spring 3-24-2020

Semester of Degree


Document Type

Open Access Thesis

Degree Name

M.S. in Forest and Natural Resources Management


Forest and Natural Resources Management

Major Professor

Eddie Bevilacqua

Steering Committee Member

Diane Kiernan

Steering Committee Member

Jane M Read


Accurate and reliable methods of assessing forest regeneration are necessary to improve forest inventories and assist management decisions. This research evaluates the effectiveness of high spatial resolution imagery from unmanned aerial systems (UAS) to assess abundance and structure of forest regeneration. Data were collected for 696 young Norway spruce (Picea abies) trees to establish field-based census. UAS digital stereo imagery was collected at three altitudes, two flight speeds and four flight azimuths, for a total of 24 separate missions. Using two orthomosaic programs, orthoimages and Digital Elevation Models (DEM) were created. Number, location and size distribution of Norway spruce trees were derived from UAS products through manual and automated processes and compared to field measurements. Manual tree detection and position estimates produced best results with 93% accuracy, while automated tree detection was only 63% accurate. Significantly strong correlations (R2 > 55%) between UAS crown estimates and field measurements were obtained.