Analysis of secondary forest succession using LIDAR analysis in the southern Appalachians Brian D. Kloeppel1, Robbie G. Kreza1, Marcus C. Mentzer1, Tucker J. Souther1, and Ryan E. Emanuel2 1Department of Geosciences and Natural Resources, Western

1
Analysis of secondary forest succession using
LIDAR analysis in the southern AppalachiansBrian
D. Kloeppel1, Robbie G. Kreza1, Marcus C.
Mentzer1, Tucker J. Souther1, and Ryan E.
Emanuel21Department of Geosciences and Natural
Resources, Western Carolina University,
Cullowhee, NC 287232Department of Forestry and
Environmental Resources, North Carolina State
University, Raleigh, NC 27695
Introduction
30 Year
50 Year
Secondary forest succession is a natural process
as forested ecosystems develop after disturbance.
The rate and extent of succession is difficult
and expensive to quantify since southern
Appalachian forested areas are large and
challenging to assess due to complex topography
and a varied land use history. Airplane-based
LIDAR, LIght Detection And Ranging, image data
became available when North Carolina was the
first state in the USA to have complete state
LIDAR coverage after the devastating floods
caused by Hurrican Floyd in 1999. Please see
Figure 1. We wish to use both indirect and direct
forest structure measurements to determine tree
height and forest structure during forest
successional changes.
Number of Returns
90 Year
70 Year
Figure 3. Each plot in the 30, 50, 70, or 90
year-old stands was arranged similarly with
marked and repeatable locations for the
measurement of each LIDAR pixel (entire plot),
forest density and basal area (entire plot), tree
height (of three tallest plot trees), leaf area
index (on each of 9 subplots), and GPS readings
(on each plot corner and plot center).
Figure 6. Stand age versus LIDAR predicted tree
height and field measured (clinometer) tree
height in four montane oak-hickory stands from 30
to 90 years old.
Tree Height (feet)
Figure 1. Following Hurricane Floyd's flooding
damage in 1999, North Carolina embarked on a
joint effort with FEMA to re-map the state's
flood zones. A by-product of this work is
detailed elevation data collected by airborne
LIDAR sensors. As a result, North Carolina became
the first state in the country with complete
LIDAR coverage.
Within the Balsam Mountain Preserve study site,
four montane oak-hickory forest stands were
identified across a chronosequence representing
differing successional stages (30, 50, 70 and 90
years since disturbance). Ground-based data
collection for each plot included tree species
and diameter (measured with a DBH tape). In
addition, in each plot the three tallest trees
were identified and height was measured using a
Vertex Laser Hypsometer and a clinometer. Leaf
area index was measured using the LAI 2000 leaf
area index meter. The remotely sensed tree height
data predictions were developed using ArcGIS.
Figure 5. Histograms showing frequency of returns
per LIDAR tree height value in each of the four
study stands.
Objectives
Results

• The objectives of this research were to
• Determine if the LIDAR imagery available in
  Jackson County, North Carolina can accurately
  predict tree height as measured in the field
• Determine what impact stand age (30, 50, 70, and
  90 year-old stands) and the concomitant change in
  forest structure has on our ability to accurately
  predict tree height using LIDAR imagery
• Predict the leaf area index (LAI) of these stands
  from field-based measurements using the LAI 2000
  instrument

Figure 7. Stand age versus leaf area index as
measured with a LAI 2000 in October 2010 in four
montane oak-hickory stands ranging from 30 to 90
years old.
Figure 8. Stand age versus tree stem density and
basal area in four montane oak-hickory stands
from 30 to 90 years old. Tree density, stand
basal area, and leaf area index impact LIDAR
returns by reflecting light pulses.
Methods
Conclusions
This research was conducted at Balsam Mountain
Preserve (BMP) in Jackson County in western North
Carolina. BMP is an 1800 hectare mountain site
(4400 acres) that was originally owned by
Champion International Paper Company which was
then purchased in the late 1990s for an upscale
housing development containing 354 home lots. At
that time, half of the remaining land that was
not being developed was placed into a
conservation easement which formed the non-profit
BMP trust. This research was conducted on land in
the BMP trust. The past land use history was
critical to provide us with a range of stand ages
in the mountain oak forest type with dominant
species including chestnut oak (Quercus montana),
northern red oak (Quercus rubra), hickory (Carya
spp.), sourwood (Oxydendrum arboreum), and red
maple (Acer rubrum). Please see Figure 2 below.

• Based on our analyses to date
• Our age chronosequence has resulted in stands of
  varying density, basal area, leaf area index, and
  tree height.
• Preliminary LIDAR estimates show that tree height
  is underestimated by 7 to 25 depending upon
  stand age and tree height.
• Sympatric analyses of LIDAR tree height and stand
  structure data at four other locations in North
  Carolina at Appalachian State University,
  UNC-Penmbroke, Johnson C. Smith University, and
  Livingston College will be analyzed at a workshop
  at North Carolina State University in May 2011.

Acknowledgements
We thank the National Science Foundation for
financial support (Award DEB-1110742). We thank
Balsam Mountain Preserve for permission to sample
on their property, especially Michael Skinner,
Ron Lance, and Blair Ogburn. We thank Cody
Amakali at Appalachian State University for LIDAR
data processing.
Figure 4. The vegetation layer was calculated by
subtracting the difference in height of the bare
earth data from the first return data layer.
Plot locations within each stand are approximate
and were chosen at random.
funding provided by
Figure 2. Forest cutting history at Balsam
Mountain Preserve by the former land owner,
Champion International. Years indicate the most
recent cut for each stand.
Figure 9. Co-authors TJ Souther, Robbie Kreza,
and Marcus Mentzer at Balsam Mountain Preserve.