overrepresentation of nearly 8 percent. The wide variation
in height error can be explained by the difficulties in using a
hand-held instrument in such extreme terrain conditions as
well as the extreme slope at the sites. Also, given the great
heights of the objects in the field and the proximity of other
trees as obstacles, it was often difficult to get an adequate
distance from the measured trees to obtain the measurements.
Although it is difficult to assign significant statistical impor-
tance to the errors between the field data and the lidar data
due to the intentionally small sample size, it is worth noting
that five of the seven measurements suggest that the lidar data
underestimated the actual height which is consistent with
findings of missed tree crowns described by Zimble
et al
.
(2003). It is worth noting that given the difficulty in obtaining
valid field measurements, it is entirely feasible that the lidar
data should become the reference data for broad forest canopy
height measurement in the future.
The trees that could be accessed did not have large trunks
proportional to their height, leading to speculation that these
trees are not old growth, but likely successional trees that
have been growing since logging practices ended in the late
1920s and early 1930s. The favorable environmental condi-
tions in the Southern Appalachians and geological conditions
that created fertile soils in the region have allowed these trees
to grow unthreatened by humans for nearly a century.
The topographical conditions of each site are also of
interest. All but one site are below or very near the threshold
for “low elevation” conditions described by Madden
et al
.
(2002) as below 2,500 ft. (762 m) with the average elevation
at the ten sites as 554.1 m. Slopes at the ten sites were varied
from relatively flat conditions (10.3°) to severe (80.5°), with
the average slope for the ten sites measuring 37.8°. Eight of
the ten sites had a generally Northwest, North, or Northeast
orientation, which is expected because these aspects pres-
ent the moistest conditions for tree growth. Interestingly,
the tallest potential tree site was found to have a Southwest
facing aspect, usually associated with hot and dry conditions,
i.e., a less favorable environment for tree growth. Based on
the
GRSM
Vegetation Community Database created by Mad-
den
et al
. (2004) as part of the
USGS NPS
National Vegetation
Inventory program, the overstory communities at the ten sites
were identified and split into three main categories: Pines
(PIs = Eastern White pine successional, PI = Southern Yellow
pine, PIs-T = Eastern White pine with Hemlock), Southern
Appalachian Cove Hardwoods (CHx = S. Appalachian Cove
Hardwoods, CHxA-T = S.A. Cove Hardwoods/Acid type with
Hemlock), and Oaks (OmH/T = Submesic to mesic oak/Hard-
woods with Hemlock). Five of the seven accessible sites were
visually inspected and found to be hardwoods, four tulip
poplar, and one white oak. The remaining two trees were
identified as pines. Figure 6 shows point cloud representa-
tions of each tree site as visualized by the
USFS
FUSION/LDV
Lidar utility tool.
Conclusions and Recommendations
This study demonstrates the use of lidar data as an effective
tool for detecting individual trees of extreme height within a
complex forest community in an area of rugged terrain. Using
GIS
and other data processing workflows, a methodology was
developed to parse and query large volumes of data to obtain
verifiable results. In all, ten tree sites were detected using this
unique methodology with heights ranging from 55 m to 59 m.
Sites 1 and 2 would be the tallest trees ever measured in the
eastern US, if they prove to be as tall as the lidar data indicate
(both 59 m) and all ten trees are taller than the tallest mea-
sured tree in the Tennessee portion of the
GRSM
(53.8 m). These
ten tree sites varied in their tree species, general overstory
community type, and terrain slope, but had similarities in
regards to elevation and terrain aspect at the tree sites. Field
measurements of these individual trees was made difficult
given the rugged terrain and density of vegetation at each
site. Regardless, the findings of this research are sufficient to
require further investigation of these tree sites, especially the
three sites that could not be accessed in the field at the time of
this work.
This tree site information can be passed on to arborists and
park managers at the
GRSM
to better inform them regarding the
locations of potentially historical, record-setting trees. This
information can be used to promote park visitation, as well as
provide specific areas for further ecological niche research. It
is recommended that these ten areas be revisited by arborists
skilled in accurate tree measurement techniques to verify the
lidar-derived data and further research be conducted at these
sites to gather more environmental and ecological informa-
tion, including soil types, soil moisture, precipitation, tem-
perature, and understory vegetation.
This work has resulted in a listing of the tallest trees in the
Tennessee portion of the Great Smoky Mountains National
Park. These tree heights are among the tallest ever recorded in
the Eastern portion of the United States. We have shown how
remote sensing technologies, specifically lidar, can be utilized
to create projects of public interest and to develop a dialogue
between the general public and the scientists engaged in
remote sensing research.
Acknowledgments
This work was based on data collected as part of the
USGS
Geospatial Program in support of the National Map. The
authors wish to express their gratitude to the National Park
Service for their support for this work as well as to the anony-
mous reviewers for their insightful comments.
References
American Forests. 2013. Tree Facts, URL:
org/discover-forests/tree-facts
(last date accessed: 16 March 2015).
Andersen, H.E., S.E. Reutebuch, and R.J. McGaughey, 2006. A
rigorous assessment of tree height measurements obtained using
airborne lidar and conventional field methods,
Canadian Journal
of Remote Sensing
, 32(5):355–366.
Bruckheimer, I., 2011. Great Smoky Mountains National Park:
Environmental suitability for trees >180 ft. in height, URL:
8b8630a83a09ef250b8a57d7095&mode=view
(last date accessed
16 March 2015).
CRMS-UGA, 2011. Acquisition of lidar for the Tennessee portion
of Great Smoky Mountains National Park and the Foothills
Parkway ARRA LIDAR Task Order-GSM Project Metadata; USGS
Contract: G10AC0015; Proposed Number: NM-ARRA-0073.
Husch, B., C. Miller, and T. Beers, 1972.
Forest Mensuration
, Second
edition, New York: Ronald Press Company.
Jensen, J.R. 2007.
Remote Sensing of the Environment, An Earth
Resource Perspective
, Second edition, Upper Saddle River, New
Jersey: Pearson-Prentice-Hall.
Lawson, A. 2010. New heights reached with technology: World’s
tallest Eucalyptus treefound with lidar and GIS, URL:
http://
, Esri,
Redlands, California (last date accessed: 16 March 2015).
Li, W., Q. Guo, M.K. Jakubowski, and M. Kelly, 2012. A new method
for segmenting individualtrees from the lidar point cloud,
Photogrammetric Engineering & Remote Sensing
, 78(1):75–84.
Madden, M., R. Welch, T. Jordan, P. Jackson, R. Seavey, and J. Seavey,
2004. Digital Vegetation Maps for the Great Smoky Mountains
National Park, Center for Remote Sensing and Mapping Science.
412
May 2015
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
