ArticlePDF Available

Terrestrial Laser Scanning for Vegetation Sampling

Authors:
  • Sterling College Vermont

Abstract

We developed new vegetation indices utilizing terrestrial laser scanning (TLS) to quantify the three-dimensional spatial configuration of plant communities. These indices leverage the novelty of TLS data and rely on the spatially biased arrangement of a TLS point cloud. We calculated these indices from TLS data acquired within an existing long term manipulation of forest structure in Central Oregon, USA, and used these data to test for differences in vegetation structure. Results provided quantitative evidence of a significant difference in vegetation density due to thinning and burning, and a marginally significant difference in vegetation patchiness due to grazing. A comparison to traditional field sampling highlighted the novelty of the TLS based method. By creating a linkage between traditional field sampling and landscape ecology, these indices enable field investigations of fine-scale spatial patterns. Applications include experimental assessment, long-term monitoring, and habitat characterization.
Sensors 2014, 14, 20304-20319; doi:10.3390/s141120304
sensors
ISSN 1424-8220
www.mdpi.com/journal/sensors
Article
Terrestrial Laser Scanning for Vegetation Sampling
Jeffrey J Richardson *, L. Monika Moskal and Jonathan D. Bakker
School of Environmental and Forest Sciences, University of Washington, Box 352100,
Seattle, WA 98195-2100, USA; E-Mails: lmmoskal@uw.edu (L.M.M.); jbakker@uw.edu (J.D.B.)
* Author to whom correspondence should be addressed; E-Mail: jeffjr@uw.edu;
Tel.: +1-206-221-6391; Fax: +1-206-685-3091.
External Editor: Assefa Melesse
Received: 8 August 2014; in revised form: 9 October 2014 / Accepted: 17 October 2014 /
Published: 28 October 2014
Abstract: We developed new vegetation indices utilizing terrestrial laser scanning (TLS)
to quantify the three-dimensional spatial configuration of plant communities. These indices
leverage the novelty of TLS data and rely on the spatially biased arrangement of a TLS
point cloud. We calculated these indices from TLS data acquired within an existing long
term manipulation of forest structure in Central Oregon, USA, and used these data to
test for differences in vegetation structure. Results provided quantitative evidence of a
significant difference in vegetation density due to thinning and burning, and a marginally
significant difference in vegetation patchiness due to grazing. A comparison to traditional
field sampling highlighted the novelty of the TLS based method. By creating a linkage
between traditional field sampling and landscape ecology, these indices enable field
investigations of fine-scale spatial patterns. Applications include experimental assessment,
long-term monitoring, and habitat characterization.
Keywords: density; patchiness; forest; remote sensing; grazing; fire
1. Introduction
Vegetation field sampling in community ecology has changed little in more than a century; the core
metrics of species abundance and richness still provide the foundation of most studies. This is of
no surprise, as these measures of community structure are easy to measure by a human observer,
OPEN ACCESS
Sensors 2014, 14 20305
requiring only a few simple tools and knowledge of the local flora [1]. Although highly informative,
these measures do not assess spatial configuration. The complexity and difficulty of assessing
three-dimensional spatial configuration in the field has slowed the integration of landscape ecology,
and its emphasis on linking spatial patterns with ecological processes [2], with traditional community
ecology. Most methods of assessing spatial configuration are limited to coarse scale information from
categorical maps or discrete points [3] that do not align with the scale of ecological field sampling [4].
Furthermore, while field methods such as stem mapping can provide spatial information, they are time
consuming, impractical for small plants, and provide only two-dimensional information [5]. Enhanced
methods of measuring spatial configuration in the field would thus be a welcome supplement to
measurements of vegetation richness and abundance, and could improve our understanding of
ecological processes.
Terrestrial Laser Scanning (TLS, also known as Terrestrial LiDAR) can measure the
three-dimensional spatial configuration of vegetation at extremely fine scales. The TLS instrument is
a portable scanning laser rangefinder that can record millions of discrete three-dimensional points
corresponding to the locations of solid objects around the scanner. TLS differs from Airborne Laser
Scanning (ALS, also known as Aerial LiDAR) in that it collects a much higher density of points in a
much smaller area. The majority of the energy of ALS is directed at the top of the canopy, while the
energy of energy from TLS is directed at the bottom or side of the canopy, depending on canopy height
and configuration. To date, most TLS studies have sought to use these point data to recreate
two-dimensional stem maps [6,7], estimate leaf area index (LAI) and other conventional metrics [8,9],
or perform forest inventory [10,11]. Studies are increasingly focused on producing highly detailed
three-dimensional models of individual trees [12,13]. These studies focus on using TLS as a tool to
arrive at traditional measures of vegetation structure, usually attempting to increase the efficiency
and/or precision of these traditional measures.
The novelty of a TLS dataset over any remotely sensed dataset previously available allows for the
development of novel ways of quantifying vegetation spatial configuration [14], but little research has
demonstrated truly new ways of doing so. A few recent studies have begun to explore novel measures
of vegetation density with TLS by examining whether three dimensional sections of space contain
vegetation, either through voxelization of TLS point clouds [15] or by calculating a measure of
vegetation density in 6.25 m3 sections of space [16]. These studies develop methods to correct for the
spatial biases inherent to a single TLS scan: occlusion by objects near to the scanner, which block the
line of sight to objects further away, and beam divergence, which reduces the effective point density
the farther one moves from the scanner. If one is attempting to accurately map vegetation density
across a defined landscape, spatial biases present a serious problem, and either correction measures or
the addition of additional stitched point clouds from additional nearby scan locations are required [16].
In this study, we present a different approach to quantifying vegetation spatial configuration by
accepting the inherent spatial bias of TLS data. Instead of attempting to describe spatial configuration
across a landscape, we develop two novel indices of vegetation density and patchiness taken from the
perspective of a single point in space, the scan location. These indices leverage the novelty of TLS to
create a novel way of quantifying vegetation spatial configuration, but due to their novelty, there is no
Sensors 2014, 14 20306
direct comparisons to metrics obtained using existing field methods. Therefore, rather than performing
a traditional validation, we demonstrate the applicability of these indices by comparing them to
two traditional measures of vegetation structure within an existing long-term experiment. Finally,
we suggest some potential applications of these TLS based indices. These indices give ecologists new
and powerful tools to quantify and analyze vegetation spatial configuration at fine scales in the field
while allowing spatial linkages to traditional methods of vegetation sampling.
2. Methods
2.1. Derivation of Indices
In their raw form, data collected by a TLS instrument consist of millions of discrete points in
Cartesian coordinates (X, Y, Z), putting the scanner at the origin (0, 0, 0). Conversion of the X, Y, Z,
points to a spherical coordinate system (easily done using a script in R software, for example) allows
easier manipulation of the data by describing each point relative to a fixed location. Spherical
coordinates describe each point in three-dimensional space by its distance (r) from the origin, the
zenith angle (θ) measured from directly overhead and relative to a vertical line through the origin, and
the azimuthal angle (ϕ) orthogonal to θ. Figure 1 provides sample visualizations of these data in
representative open and closed canopy forests. The top row shows a photo-mosaic view of the two
forests from the perspective of the scanner. The spherical coordinates r and θ are plotted in the bottom
graphs. The plots suggest that more points are clumped at small r in the closed forest compared to the
open forest, and suggest that the distribution of r is highly related to vegetation density. However,
these plots are not quantitative measures of forest spatial configuration. We suggest that vegetation
density be quantified via the Three-dimensional Vegetation Density Index (3DI):
3DI =
,
where  is the median distance from the scanner for all points with zenith angle between θ and θ.
Within the specific range of θ, 3DI assesses a three-dimensional space visualized as a cone rotated
around its apex. This index can be easily compared among scans.
Variation in vegetation density around the scanner provides an assessment of vegetation patchiness,
and thus we also propose the three-dimensional Patchiness Index 3PI:
3PI = ( σ
μ
),
where σ and µ are the standard deviation and mean, respectively, of the set: , for ϕ [0, 0 + a, …,
360 a] where , is the median of all r with azimuthal angle between ϕ and ϕ+a and zenith angle
between θ and θ, and a is the azimuthal angle increment determining the number of bins to use in
the index. Note that σ/µ is synonymous with the coefficient of variation.
Sensors 2014, 14 20307
Figure 1. Visualizations of two representative plots, an open forest plot (left), and a closed
forest plot (right). The top rows show a photo-mosaic, captured by the TLS scanner,
at each plot location. The bottom row shows all points collected by the TLS instrument
plotted with their distance from the scanner, r, on the horizontal axis and zenith, θ, on the
vertical axis. Note that the blue lines correspond to 90° of zenith angle (parallel to ground).
2.2. Application at Starkey Experimental Forest and Range
TLS data were collected at Starkey Experimental Forest and Range (SEFR), in the Blue Mountains
of north-central Oregon in August of 2009 (Figure 2). The TLS scan locations were located within
forested areas of a long-term experiment begun in 2000. Three blocks were thinned and burned in
2000, while three others were untreated. In each block, seven experimental treatments were installed:
six grazing treatments within a large fenced area, and one unfenced adjacent treatment.
We chose two treatments within each of the six blocks that we believed would show the most
variability in vegetation structure: the treatment that had excluded grazing and browsing since 2001,
hereafter referred to as ungrazed, and the nearby unfenced treatment that had experienced continuous
grazing and browsing by cattle, deer, and elk for the duration of the experiment, hereafter referred to as
grazed. In each roughly 0.8 ha treatment, two TLS scan locations were randomly selected under the
stipulation that they be >50 m from each other and >20 m from the treatment boundary (Figure 2).
Sensors 2014, 14 20308
Figure 2. The study area where TLS scans were obtained. (A) shows the state of Oregon
with the extent of (B) enclosed by the white box. (B) shows the Starkey Experimental
Forest and Range. Red polygons are thinned and burned blocks, and green polygons are
untreated blocks. The black rectangle is the extent of (C), which shows the experimental
design for an untreated block. Scans were collected at locations represented by yellow dots.
The dots within one of the six contiguous rectangles are in the ungrazed plot, while the
grazed plot is the unattached rectangle.
In total, 24 scans were obtained (four scans in each block, two in each grazed and two in each
ungrazed treatment). TLS scans were obtained using a Leica Scan Station 2 with the following
specifications: a complete scan in 360° of azimuth (ϕ) and 135° of zenith (θ; the 45° of zenith
corresponding to the area below the scanner is blocked from the scanner’s view) at a point spacing of
2.5 cm × 2.5 cm at 30 m from the scanner. The scanner was controlled using Cyclone software (Leica
Geosystems, Heerbrugg, Switzerland) which also allowed the raw data to be exported into a text file
for further processing.
We also obtained field data collected by the U.S. Forest Service at the same plots that describe the
structural components of the shrub understory. Shrubs were defined by a list of 28 species. These data
were collected at multiple subsample locations within each treatment (number of subsamples ranged
from 11 to 20). At each subsample location, shrub stem density was counted within a 4 m by 4 m
square. Shrub cover was counted along a 10 m line, where a minimum of 2.5 cm of shrub canopy
intercepted the transect. Mean values of shrub density and shrub cover for each of the 12 treatments
were calculated from the subsamples within that treatment. All statistical analyses were performed in R
version 2.11.1 using the “aov” function.
Sensors 2014, 14 20309
3. Results
To apply the two indices described above, we first needed to select an appropriate range of θ for
which to summarize r. While any number of increments could have been used, we found that dividing
the 135° of zenith into 15 bins provided a good overview of the variability; smaller increments masked
the overall trends in spatial pattern (results not shown).
For all angle bins, vegetation density as measured by 3DI was greater in thinned and burned than
untreated plots (Figure 3). At the smallest zenith angle (closest to ground) the difference is small
because of the relatively short distance the laser travels before it reflects off the ground. We
concentrated on 3DI within a θ angle range between 81° and 90° as this range allowed a large swath
of vegetation to be sampled while including a minimum of ground points, as the perspective is close
to horizontal in all directions. A two-factor split-plot ANOVA (Table 1; Figure 4) indicated that
3DI differed significantly between forest treatments (p < 0.01) but not between grazing treatments
(p = 0.54). The forest treatment × grazing interaction was also not significant (p = 0.64).
Figure 3. The Three-dimensional Vegetation Density Index (3DI) computed using
15 zenith angle bins for representative plots that were untreated and plots that were thinned
and burned. Standard error bars are shown.
Sensors 2014, 14 20310
Table 1. Split-plot analysis of variance for the Three-dimensional Vegetation Density
Index (3DI). Block Treatment is Thinning and Burning/untreated, Plot Treatment is
Grazed/ungrazed. Note that the mean of two 3PI values were used in the ANOVA,
computed using two separate scans within each replicate.
Source df Sum of Squares Mean Square F-value p
Whole Plot Factor
Block Treatment 1 128.69 128.69 44.55 <0.00
Error 4 11.55 2.89
Subplot Factors
Plot Treatment 1 7.39 7.39 0.44 0.54
Block X Plot 1 4.28 4.28 0.25 0.64
Error 4 67.38 16.84
Figure 4. Mean values of the Three-dimensional Vegetation Density Index (3DI) for zenith
angles between 81° and 90° for grazed and ungrazed plots that were either thinned and
burned or untreated. Standard error bars are shown.
We used the same θ increment to assess vegetation patchiness around the scanner. Figure 5 displays
a visualization of patchiness for representative grazed and ungrazed plots that were thinned and
burned. Median values of r are plotted for each of the 15 bins of ϕ, and 3PI values are given.
A two-factor split-plot ANOVA (Table 2; Figure 6) indicated that 3PI did not differ between forest
treatments (p = 0.86) but tended to differ (p = 0.07) between grazed and ungrazed treatments. The
forest treatment × grazing interaction was not significant (p = 0.94).
Sensors 2014, 14 20311
Figure 5. Visual representations of TLS data in a thinned and burned and grazed plot
(above) and a thinned and burned and ungrazed plot (below). On the left are overhead
views of raw TLS points rendered in Cartesian coordinates. The scan location is at the
center, points greater than scanner height (2 m) are shown in green, and points less than
scanner height are shown in beige, black areas had no points collected. The radius of each
extent is 80 m. On the right are polar plots showing the median r for 18 different azimuthal
angles ranges. The three-dimensional Vegetation Patchiness Index (3PI) is calculated for
each plot using zenith angles from 81° to 90°. Mean and standard deviation used to
calculate 3PI are: grazed (4.1, 1.7), ungrazed (23.4, 19.9).
Table 2. Split-plot analysis of variance for the Three-dimensional Vegetation Patchiness
Index (3PI). Block Treatment is Thinning and Burning/untreated, Plot Treatment is
Grazed/ungrazed. Note that the mean of two 3PI values were used in the ANOVA,
computed using two separate scans within each replicate.
Source df Sum of Squares Mean Square F-value p
Whole Plot Factor
Block Treatment 1 0.0011 0.0011 0.35 0.86
Error 4 0.12 0.031
Subplot Factors
Plot Treatment 1 0.25 0.25 6.25 0.07
Block X Plot 1 0.00023 0.0023 0.0056 0.94
Error 4 0.16 0.040
Sensors 2014, 14 20312
Figure 6. Mean values of the Three-dimensional Vegetation Patchiness Index (3PI) for
zenith angles between 81° and 90° for grazed and ungrazed plots that were either
thinned and burned or untreated. Standard error bars are shown.
Figure 7 shows mean values of shrub stem density calculated from the U.S. Forest Service field
collected data. Two-factor split-plot ANOVAs (Tables not shown) indicated that stem density did not
differ between forest treatments (p = 0.78), grazed and ungrazed treatments (p = 0.17), or the forest
treatment × grazing interaction (p = 0.71). Figure 8 shows mean values of shrub cover calculated
from the U.S. Forest Service field collected data. Shrub cover differed significantly between forest
treatments (p = 0.01) but not between grazing treatments (p = 0.68) or the forest treatment × grazing
interaction (p = 0.83).
Figure 7. Mean values of field measured shrub stem density in grazed and ungrazed plots
that were either thinned and burned or untreated. Standard error bars are shown.
Grazed Ungrazed
Stem Density
(
m
2)
0246810
Thinned and Burned
Untreated
Sensors 2014, 14 20313
Figure 8. Mean values of field measured shrub cover in grazed and ungrazed plots that
were either thinned and burned or untreated. Standard error bars are shown.
4. Discussion
The ability to quantify fine scale patterns using a simple, repeatable methodology is of practical
use in the myriad experimental designs that investigate vegetation structure. The application of these
vegetation density and patchiness indices at SEFR shows how these techniques can be used to
investigate quantifiable differences in spatial configuration. Although we did not test pre-defined
hypotheses about spatial configuration at SEFR, our results indicate that thinning and burning affected
vegetation density but not patchiness, whereas grazing affected vegetation patchiness but not density.
However, more detailed understanding of the development of these spatial patterns would require
additional study.
The observations based on the field collected measures of vegetation structure (Figures 7 and 8)
help to support the validity of the indices developed in this study, while at the same time highlighting
the limitations of traditional measures. The measure of shrub cover is most similar to 3DI, as both are
responsive to the amount of vegetation present. In both cases, significant differences were observed
between treatments that were thinned and burned compared to those that were not. There is an
important difference between the two measures: namely that 3DI samples all vegetation types within
a given angle bin, while measures of shrub cover don’t account for cover of trees or herbaceous
vegetation that may also be within a height stratum of interest. The lack of a comparable field based
measure to 3PI further highlights the novelty of this new tool for investigating vegetation structure.
The intent of these indices is not to replicate these traditional measurements, but rather to summarize
TLS data in a new way with the potential to provide new insights relative to those available from
traditional measurements.
The unique perspective of these two indices may require a shift in the way remote sensing scientists
typically interpret remotely sensed data products. It is common to process remotely sensed data to
describe variables in contiguous raster cells across a landscape, such as in LAI derived from ALS [17]
or the normalized difference vegetation index (NDVI) derived from Landsat [18]. These products have
Sensors 2014, 14 20314
little to no spatial bias, as each grid cell is exposed to the same energy, either from a laser or the sun.
The TLS indices derived here are more akin to field based remote sensing techniques such as
hemispherical photographs, which use information from a single point location to estimate variables
such as LAI and solar radiation for a single point in space [19], and are therefore spatially biased to
that single point in space. Just as in a hemispherical photograph, vegetation that is near to the scanner
will strongly affect these indices. Thus, it is key that these indices are treated as a sample of nearby
vegetation rather than reflecting a census of the vegetation surrounding the scan location.
5. Applications
5.1. Long Term Monitoring
The long term nature of the SEFR experiment highlights the power of TLS methods and these
indices for long term monitoring. For example, a logical extension would be to obtain repeated scans
in the same locations and quantify how density and patchiness change over time. As a benefit, TLS
scanning does not modify the environment, as the instrument can be set up outside the area of interest.
A TLS based monitoring effort might be able to supplant destructive sampling altogether, as TLS has
been used to estimate biomass [20,21] and chlorophyll content [22]. Using a single instrument to
quantify spatial configuration and biophysical attributes of vegetation without modifying the community
would be a powerful new tool, especially considering the cost and time required for long-term monitoring
efforts like LTER and NEON [23,24].
5.2. Integration with Traditional Field Sampling
An investigator seeking to integrate TLS-based methods with traditional sampling might use transects
or Daubenmire style quadrats to assess species richness and abundance at each TLS scan location. This
is especially efficient because there is a period of time after the scan commences during which data
could be collected in areas already scanned. An index such as Shannon’s Diversity would provide a
single metric which could then be easily compared or integrated with the TLS derived indices.
The area surrounding the scanner could also be separated into quadrants in order to link even finer
scale patterns in richness and abundance to vegetation density and patchiness. An example sampling
strategy is shown in Figure 9. 3DI and 3PI can be easily modified to assess density and patchiness
within a quadrant by calculating each index using the totality of the data within the quadrant, and in the
case of 3PI, dividing each quadrant into a discrete number of ϕ bins. Investigators should be wary of
pseudoreplication when dividing a plot into multiple sub-samples for experimental assessment [25],
but the technique would be robust for studies where examining fine-scale spatial variability of plant
community structure is the principal focus.
It is also possible using TLS to collect point clouds corresponding to the exact spatial location
of quadrats or belt transects rather than the neighborhood represented by a quadrant. The indices
developed in this study should not be used to assess these point clouds, as the indices are built on the
unique geometry of point clouds collected using the full field of view of the TLS scanner. The main
assumption of these indices is that each increment of ϕ contains an equal number of laser pulses with
equivalent spacing and geometry. Laser pulse density declines with distance from the laser; to be
Sensors 2014, 14 20315
comparable, a field collected belt transect would need to increase in width as distance from the
scanner increased. Other methodologies, such as biomass estimation based on estimated canopy
volume [26], may be more appropriate for direct comparison of TLS point clouds clipped to the extent
of field measurements.
Figure 9. Suggested sampling strategy for linking field measurements of species abundance
and richness with vegetation spatial configuration indices (3DI and 3PI). The image is an
overhead view of a TLS point cloud, rendered by height with the tallest points white and
the lowest points black. Superimposed on the image are the TLS instrument location (red
dot), quadrants of the scanning area (delineated by solid lines), vegetation transects (dotted
lines), and Daubenmire style vegetation quadrats (green squares).
5.3. Applications for Habitat Characterization
Another clear application of these methods is habitat characterization, one of the most common
ecological applications of structural measurements of vegetation. Unlike airborne LiDAR, TLS has
yet to be applied to this purpose [14]. The fixed location of the TLS produces a three-dimensional
representation of the surrounding space, similar to how an organism may perceive the space. For
example, the experiment at SEFR sought to measure the influence of ungulate grazing on forest
structure. The indices provide a quantification of the patterns an ungulate may experience in those
spaces, which may then inform the resulting ecological processes that affect the observed results.
If the significantly greater patchiness observed in the grazed treatment is attributed to modification
by ungulates, the perspective of the TLS, 2 m above the ground and at ungulate eye level, may aid in
understanding how ungulates affect vegetation structure when the complete long-term results of this
experiment are later analyzed.
Sensors 2014, 14 20316
By adjusting the ϕ and θ bins, investigators can tailor the quantification of vegetation structure to
an appropriate scale for their organism(s) of interest. While ungulate grazers may interact with their
environment over relatively coarse scales that leverage the entire field of view of the TLS, smaller
organisms such as insects may modify their environments at very fine scales, thus requiring many
small ϕ bins, while canopy dwelling organisms may necessitate that only θ bins above the horizontal
plane are included in analysis. Furthermore, if a scan with a high point density using the full field of
view is performed, much of these decisions and analyses can be performed after the field campaign by
selecting subsets of the full TLS dataset.
5.4. Applications to Non-Forested Systems
While we used a forest ecosystem to demonstrate the vegetation density and patchiness indices, they
would be robust in many different ecosystems. Grasslands, for example, are difficult to assess
structurally because the plants are so small and numerous. Small scale measurements of density are
often restricted to visual estimates of percent cover [27], while patchiness estimates require remotely
sensed data that may be too coarse in scale [28]. TLS can provide measurements of density and
patchiness because its laser beams are focused enough to sample fine structures such as individual
blades of grass. Practitioners may want to modify the increments and ranges of θ and ϕ as well as
constraining the range of r to better suit their system and accurately sample variation at the scale of
interest. Grasslands, for example, may more efficiently be sampled at locations very close to the
scanner but with a larger number of ϕ increments because vegetation varies at a finer scale.
5.5. Avenues for Future Research
One significant research need is how these indices, and TLS based methods in general, can be
applied in areas with highly sloped terrain. On a slope, the TLS instrument cannot sample equally in all
directions of ϕ. On the uphill side, more laser beams will reflect off the ground, while the opposite is
true on the downhill side. Without correction, this can affect both indices, as point density will be
higher on the uphill side regardless of the vegetation present. In this study, the plot locations were all
located on relatively flat topography, but the effects of slope can still be seen. The open forest in
Figure 1 provides an example of non-flat terrain; the photomosaic shows ground sloping slightly uphill
to the left and downhill to the right. The pattern is also reflected in the plots of r and θ, where distinct
“ground lines” can be seen. Slopes should be avoided while using these methods, unless modifications
and corrections are developed. For example, it is possible to angle the scanner to match the slope as is
sometimes performed with hemispherical photography [29] or to construct a digital elevation model
from the raw points and adjust the indices in post-processing. Constraining the area sampled to only
short distances away from the scanner may limit the effect of slope.
We encourage other researchers to creatively develop TLS-based methodologies. The vegetation
density and patchiness indices developed in this study are a first step toward realizing the power of the
instrument as a tool for radical innovation in ecological data collection.
Sensors 2014, 14 20317
6. Conclusions
In conclusion, we have shown how an emerging technology, TLS, can be used to produce precise
measurements of vegetation spatial configuration through two new indices of vegetation density and
patchiness. The fine scale at which the TLS operates allows a linkage between traditional vegetation
sampling and landscape ecology, which has been mostly constrained to coarse scale examination of
raster and point data. TLS methodologies of vegetation sampling are also well suited for experimental
assessment, long-term monitoring, and habitat characterization, and integrate well with traditional
methods of sampling vegetation communities. These indices can be applicable in many different
ecosystems and represent one of the first practical methods of assessing spatial configuration in
three-dimensions at fine scales in the field.
Acknowledgments
This work benefited from contributions by Michael Hannam, Megan Davis, Brendan Boyer,
Tom Hinckley, Greg Ettl, and Daniel Schwartz. Data, housing, and field access was provided by the
USDA Forest Service PNW Research Station’s Le Grande Forestry and Range Sciences Laboratory.
This work was supported by NSF IGERT grant #654252, NSF University of Washington I/UCRC
Center for Advanced Forest Systems grant #0855690, the Precision Forestry Cooperative, the Remote
Sensing and Geospatial Analysis Laboratory, and UW Student Tech Fee equipment grant.
Author Contributions
Jeffrey J Richardson organized the project, conducted the field work, performed the analysis, and
was the principal writer. L. Monika Moskal provided guidance on TLS application and Jonathan D. Bakker
provided guidance on vegetation sampling applications and statistics.
Conflicts of Interest
The authors declare no conflict of interest.
References
1. Daubenmire, R. A canopy-coverage method of vegetational analysis. Northwest Sci. 1959, 33,
43–64.
2. Turner, M.G. Landspace ecology—The effect of pattern on process. Annu. Rev. Ecol. Syst. 1989,
20, 171–197.
3. Gustafson, E.J. Quantifying landscape spatial pattern: What is the state of the art? Ecosystems
1998, 1, 143–156.
4. Kotliar, N.B.; Wiens, J.A. Multiple scales of patchiness and patch structure—Hierarchical
framework for the study of heterogeneity. Oikos 1990, 59, 253–260.
5. Shaw, D.C.; Franklin, J.F.; Bible, K.; Klopatek, J.; Freeman, E.; Greene, S.; Parker, G.G.
Ecological setting of the wind river old-growth forest. Ecosystems 2004, 7, 427–439.
Sensors 2014, 14 20318
6. Tansey, K.; Selmes, N.; Anstee, A.; Tate, N.J.; Denniss, A. Estimating tree and stand variables
in a corsican pine woodland from terrestrial laser scanner data. Int. J. Remote Sens. 2009, 30,
5195–5209.
7. Liang, X.L.; Litkey, P.; Hyyppa, J.; Kaartinen, H.; Vastaranta, M.; Holopainen, M. Automatic
stem mapping using single-scan terrestrial laser scanning. IEEE Trans. Geosci. Remote Sens.
2012, 50, 661–670.
8. Moorthy, I.; Miller, J.R.; Hu, B.X.; Chen, J.; Li, Q.M. Retrieving crown leaf area index from an
individual tree using ground-based lidar data. Can. J. Remote Sens. 2008, 34, 320–332.
9. Moskal, L.M.; Zheng, G. Retrieving forest inventory variables with terrestrial laser scanning (tls)
in urban heterogeneous forest. Remote Sens. 2011, 4, 1–20.
10. Maas, H.G.; Bienert, A.; Scheller, S.; Keane, E. Automatic forest inventory parameter determination
from terrestrial laser scanner data. Int. J. Remote Sens. 2008, 29, 1579–1593.
11. Ducey, M.J.; Astrup, R. Adjusting for nondetection in forest inventories derived from terrestrial
laser scanning. Can. J. Remote Sens. 2013, 39, 410–425.
12. Raumonen, P.; Kaasalainen, M.; Akerblom, M.; Kaasalainen, S.; Kaartinen, H.; Vastaranta, M.;
Holopainen, M.; Disney, M.; Lewis, P. Fast automatic precision tree models from terrestrial laser
scanner data. Remote Sens. 2013, 5, 491–520.
13. Delagrange, S.; Jauvin, C.; Rochon, P. Pypetree: A tool for reconstructing tree perennial tissues
from point clouds. Sensors 2014, 14, 4271–4289.
14. Vierling, K.T.; Vierling, L.A.; Gould, W.A.; Martinuzzi, S.; Clawges, R.M. Lidar: Shedding new
light on habitat characterization and modeling. Front. Ecol. Environ. 2008, 6, 90–98.
15. Harman, C.J.; Lohse, K.A.; Troch, P.A.; Sivapalan, M. Spatial patterns of vegetation, soils, and
microtopography from terrestrial laser scanning on two semiarid hillslopes of contrasting lithology.
J. Geophys. Res. Biogeosci. 2014, 119, 163–180.
16. Ashcroft, M.B.; Gollan, J.R.; Ramp, D. Creating vegetation density profiles for a diverse range of
ecological habitats using terrestrial laser scanning. Methods Ecol. Evol. 2014, 5, 263–272.
17. Richardson, J.J.; Moskal, L.M.; Kim, S.-H. Modeling approaches to estimate effective leaf area
index from aerial discrete-return lidar. Agric. For. Meteorol. 2009, 149, 1152–1160.
18. Anderson, G.L.; Hanson, J.D.; Haas, R.H. Evaluating landsat thematic mapper derived vegetation
indexes for estimating aboveground biomass on semiarid rangelands. Remote Sens. Environ. 1993,
45, 165–175.
19. Parker, G.G.; Lefsky, M.A.; Harding, D.J. Light transmittance in forest canopies determined using
airborne laser altimetry and in-canopy quantum measurements. Remote Sens. Environ. 2001, 76,
298–309.
20. Hilker, T.; van Leeuwen, M.; Coops, N.C.; Wulder, M.A.; Newnham, G.J.; Jupp, D.L.B.;
Culvenor, D.S. Comparing canopy metrics derived from terrestrial and airborne laser scanning in
a Douglas-fir dominated forest stand. Trees-Struct. Funct. 2010, 24, 819–832.
21. Lefsky, M.; McHale, M. Volume estimates of trees with complex architecture from terrestrial
laser scanning. J. Appl. Remote Sens. 2008, 2, 1–19.
22. Eitel, J.U.H.; Vierling, L.A.; Long, D.S. Simultaneous measurements of plant structure and
chlorophyll content in broadleaf saplings with a terrestrial laser scanner. Remote Sens. Environ.
2010, 114, 2229–2237.
Sensors 2014, 14 20319
23. Long Term Ecological Research Network. Available online: http://www.lternet.edu/ (accessed on
2 July 2014).
24. National Ecological Observatory Network. Available online: http://www.neoninc.org/ (accessed
on 2 July 2014).
25. Hurlbert, S.H. Pseudoreplication and the design of ecological field experiments. Ecol. Monogr.
1984, 54, 187–211.
26. Loudermilk, E.L.; Hiers, J.K.; O’Brien, J.J.; Mitchell, R.J.; Singhania, A.; Fernandez, J.C.;
Cropper, W.P.; Slatton, K.C. Ground-based lidar: A novel approach to quantify fine-scale fuelbed
characteristics. Int. J. Wildland Fire 2009, 18, 676–685.
27. Spehn, E.M.; Joshi, J.; Schmid, B.; Diemer, M.; Korner, C. Above-ground resource use increases
with plant species richness in experimental grassland ecosystems. Funct. Ecol. 2000, 14, 326–337.
28. Baldi, G.; Paruelo, J.M. Land-use and land cover dynamics in South American temperate
grasslands. Ecol. Soc. 2008, 13, 6.
29. Gonsamo, A.; Pellikka, P. Methodology comparison for slope correction in canopy leaf area index
estimation using hemispherical photography. For. Ecol. Manag. 2008, 256, 749–759.
© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/4.0/).
... TLS is well-suited to determine understory vegetation characteristics such as the density of foliage and amount of open area [42][43][44]. The depth of view and openness of a location can be quantified by looking at how far each pulse travels, and if a pulse is not returned at all [45,46]. TLS can replicate the use of a cover board at fixed locations in a non-subjective manner, as well as return a robust estimate of the total depth and openness of a plot. ...
... The number of studies using TLS for habitat characterization is increasing, as well as the sophistication of the approaches. For example, a form of the DCB metric has been previously used to quantify forest structure [45,46], but this study is the first we are aware of that directly applies this approach to habitat quantification. There has been previous work using TLS for understory habitat classification [58,59], as well as excellent R packages that have been developed to quantify the viewshed specifically for habitat characterization [60]. ...
Article
Full-text available
The Canada lynx is listed as a threatened species, and as such, the identification and conservation of lynx habitats is of significant concern. Lynxes require areas with high amounts of horizontal cover made up of ground vegetation. Lidar offers a robust method of quantifying vegetation structure, and airborne lidar has been acquired across large areas of potential lynx habitat. Unfortunately, airborne lidar is often not able to directly measure understory horizontal cover due to occlusion from the upper branches. Terrestrial lidar does directly measure understory horizontal cover and can be used as training data for larger area models using airborne lidar. In this study, we acquired 168 individual terrestrial lidar scans (TLS) across 42 sites in north-central Washington state. We generated metrics from the single-scan TLS plots using depth maps, a digital cover board, and voxels. Using our TLS metrics as the training data for the airborne lidar acquired for the entire Loomis State Forest, we were able to produce a model using xgboost with 85% accuracy. We believe our study shows that single-scan TLS plots can be used effectively to quantify fine-scale forest structure elements relevant to species habitat, to then inform larger area models using airborne lidar.
... For instance, ref. [3] explores the measurement of vegetation canopy structures using laser scanning, showcasing the method's capability to capture detailed canopy data in situ without direct contact. The authors of [4] introduce innovative vegetation indices derived from terrestrial laser scanning, which effectively quantify the 3D spatial organization of plant communities. Another example is [5], which presents an analysis of a 3D laser tracking scanner system, focusing on its immunity to the effects of ambient sunlight and its geometric resolution. ...
Article
Full-text available
This paper proposes an advancement in the application of a Technical Vision System (TVS), which integrates a laser scanning mechanism with a single light sensor to measure 3D spatial coordinates. In this application, the system is used to scan and digitalize objects using a rotating table to explore the potential of the system for 3D scanning at reduced resolutions. The experiments undertaken searched for optimal scanning windows and used statistical data filtering techniques and regression models to find a method to generate a 3D scan that was still recognizable with the least amount of 3D points, balancing the number of points scanned and time, while at the same time reducing effects caused by the particularities of the TVS, such as noise and entropy in the form of natural distortion in the resulting scans. The evaluation of the experimentation results uses 3D point registration methods, joining multiple faces from the original volume scanned by the TVS and aligning it to the ground truth model point clouds, which are based on a commercial 3D camera to verify that the reconstructed 3D model retains substantial detail from the original object. This research finds it is possible to reconstruct sufficiently detailed 3D models obtained from the TVS, which contain coarsely scanned data or scans that initially lack high definition or are too noisy.
... What is required is a reasonable sample of nutrient concentrations of consumed tissue types and a three-dimensional assessment of the frequency distribution of those types in the plant canopy. Recent developments with terrestrial laser scanning (Li et al., 2014;Muller et al., 2010;Richardson et al., 2014) could vastly accelerate this process over traditional methods. The total biomass of consumed plant species is needed if the conversion of the proportion of biomass consumed to population density is desired, but total biomass is far easier to estimate than the biomass of dozens of tissue types. ...
Article
Full-text available
The discipline of ecology seeks to understand how ecosystems, communities, and populations are regulated. A ubiquitous mechanism of population regulation of consumers is that capturing energy and nutrients in sufficient quantities for survival and reproduction becomes more difficult as population density increases. Extensive evidence has revealed that populations of large herbivores are often regulated by density dependence, defined as the reduction in the per‐capita population growth rate that occurs as populations grow large. Diminished body mass of individuals has been repeatedly observed in high‐density populations, implicating compromised nutrition as the primary cause of density dependence. However, there is no general explanation for why these nutritional deficiencies occur. Recent work demonstrated that reduced food intake rates resulting from the functional response of herbivores to depleted plant biomass does not provide a sensible explanation for density dependence because rates of food intake of herbivores are often insensitive to changes in plant biomass. A new model of feedbacks from plant biomass to herbivores shows how reduced nutrition of herbivores can result from increased dilution of nutrients in the plant tissue they consume as populations grow, even when their rate of consumption of plants remains constant. The model contains parameters that can be scaled to body mass, allowing unusually general predictions. The model shows that convex, concave, and linear relationships between the per‐capita growth rate and population density can arise from the effects of depletion of plant biomass by herbivore foraging. The model is the first to explicitly include spatial variance in the nutritional quality of plants as a general driver of herbivore population dynamics. I show how regulation of herbivore abundance by plant nutrients can occur, even when a large fraction of the consumable plant biomass remains uneaten, providing a simple, mechanistic explanation for bottom‐up control of population dynamics of primary consumers in a “green world.”
... O TLS pode ser utilizado para medições de inventário florestal (altura das árvores, diâmetro a altura do peito, densidade, área basal estimativas de volume de madeira) e caracterização do dossel (projeções, número de clareiras, distribuição das árvores). Além disso, o sensor terrestre também auxilia na obtenção de dados ecológicos (propriedades físicas das folhas, processos de transpiração e diversidade de micro-habitat) (DASSOT et al., 2011).O TLS é um scanner empregado como forma de avaliação de parâmetros florestais de modo não destrutivo, permitindo que objetos sejam digitalizados de forma rápida e precisa, gerando nuvem de pontos tridimensionais (3D)(RICHARDSON et al., 2014). A tecnologia TLS, apresenta alto potencial para pesquisas ecológicas e de dinâmica florestal(TROCHTA et al., 2017).Os equipamentos laser scanner terrestre geralmente são montados em um tripé ou plataforma estável. ...
Article
Full-text available
As técnicas de sensoriamento remoto permitem que os manejadores e gestores florestais coletem de forma eficiente, rápida e precisa variáveis difíceis de obter em larga escala. O LiDAR terrestre é uma tecnologia com alto potencial para coleta de dados florestais, como a medição da biomassa, estrutura e diversidade florestal. Diante disso, o objetivo do trabalho foi apresentar uma revisão bibliográfica sobre o potencial da tecnologia LiDAR terrestre como uma ferramenta alternativa de sensoriamento remoto na obtenção de dados e informações florestais. Os scanners terrestres (TLS) permitem que sejam realizadas representações 3D da distribuição espacial dos elementos do dossel florestal, além de fornecer métricas sem amostragem destrutiva das árvores, especialmente útil para estudos de levantamento de biomassa. Entretanto, a sua utilização tem sido limitada devido a falta de recursos financeiros para investir em tecnologias de ponta, falta de capacitação e treinamento para operar e analisar os dados coletados, infraestrutura insuficiente para armazenamento e processamento de grandes volumes de dados e custo elevado. Para superar essas limitações, é importante que haja maior investimento em pesquisa e desenvolvimento. Com o aumento da conscientização sobre a importância do monitoramento florestal, espera-se que essas limitações possam ser superadas no futuro.
... Examples of a standalone ground-based remote sensing survey include utilization of terrestrial laser scanning (TLS) to generate microtopography (Harman et al., 2014;Nouwakpo et al., 2016;Stovall, Diamond, et al., 2019), vegetation structure (Ashcroft et al., 2014;Grau et al., 2017;Kong et al., 2016;Richardson et al., 2014), and soil erosion and transport (Ballesteros-Canovas et al., 2015;Eitel et al., 2011;Longoni et al., 2016) models at very high resolution (i.e., millimeters to centimeters). Due to deployment height and viewing angle, TLS is particularly specialized in monitoring understory vegetation structure, including tree stems (Heinzel & Huber, 2017) and biomass , evaluating post-disturbance understory changes , and refining other laser scanning platforms (Crespo-Peremarch et al., 2018, 2020Luscombe et al., 2015). ...
Article
Full-text available
Understanding the severity and extent of near surface critical zone (CZ) disturbances and their ecosystem response is a pressing concern in the face of increasing human and natural disturbances. Predicting disturbance severity and recovery in a changing climate requires comprehensive understanding of ecosystem feedbacks among vegetation and the surrounding environment, including climate, hydrology, geomorphology, and biogeochemistry. Field surveys and satellite remote sensing have limited ability to effectively capture the spatial and temporal variability of disturbance and CZ properties. Technological advances in remote sensing using new sensors and new platforms have improved observations of changes in vegetation canopy structure and productivity; however, integrating measures of forest disturbance from various sensing platforms is complex. By connecting the potential for remote sensing technologies to observe different CZ disturbance vectors, we show that lower severity disturbance and slower vegetation recovery are more difficult to quantify. Case studies in montane forests from the western United States highlight new opportunities, including evaluating post‐disturbance forest recovery at multiple scales, shedding light on understory vegetation regrowth, detecting specific physiological responses, and refining ecohydrological modeling. Learning from regional CZ disturbance case studies, we propose future directions to synthesize fragmented findings with (a) new data analysis using new or existing sensors, (b) data fusion across multiple sensors and platforms, (c) increasing the value of ground‐based observations, (d) disturbance modeling, and (e) synthesis to improve understanding of disturbance.
... The key to using single point TLS scans for forest structure characterization is likely to focus on deriving different types of metrics that still relate to forest structure beyond trying to replicate conventional sampling techniques (e.g., basal area and mean diameter). Richardson et al. [51] proposed the Three-dimensional Vegetation Density Index (3DI), which is a metric based on median distance traveled of each pulse between defined scan angles from the scanner. This method can quantify differences between stands with markedly different forest structure such as an area of dense understory vegetation compared to a stand where heavy cattle grazing removed much of the understory vegetation. ...
Article
Full-text available
We developed new measures of structural complexity using single point terrestrial laser scanning (TLS) point clouds. These metrics are depth, openness, and isovist. Depth is a three-dimensional, radial measure of the visible distance in all directions from plot center. Openness is the percent of scan pulses in the near-omnidirectional view without a return. Isovists are a measurement of the area visible from the scan location, a quantified measurement of the viewshed within the forest canopy. 243 scans were acquired in 27 forested stands in the Pacific Northwest region of the United States, in different ecoregions representing a broad gradient in structural complexity. All stands were designated natural areas with little to no human perturbations. We created “structural signatures” from depth and openness metrics that can be used to qualitatively visualize differences in forest structures and quantitively distinguish the structural composition of a forest at differing height strata. In most cases, the structural signatures of stands were effective at providing statistically significant metrics differentiating forests from various ecoregions and growth patterns. Isovists were less effective at differentiating between forested stands across multiple ecoregions, but they still quantify the ecological important metric of occlusion. These new metrics appear to capture the structural complexity of forests with a high level of precision and low observer bias and have great potential for quantifying structural change to forest ecosystems, quantifying effects of forest management activities, and describing habitat for organisms. Our measures of structure can be used to ground truth data obtained from aerial lidar to develop models estimating forest structure.
... Terrestrial Laser Scanning (TLS) is an active remote sensing technique, which can be used to measure structural parameters such as height, diameter, above-ground biomass and Leaf Area Index (LAI) from LiDAR-derived 3D point clouds (Atkins et al., 2018;Béland et al., 2014;Calders et al., 2015;Strahler et al., 2008). It has advanced swiftly in the last decade Dassot et al., 2011;Owen et al., 2021;Richardson et al., 2014). Although the use of this technology to study lianas has lagged behind that of trees (Krishna , TLS now enables distinguishing liana stems from trees in point cloud data with great precision and accuracy (Bao et al., 2018;Krishna Moorthy et al., 2019). ...
Article
1. Lianas (or vines) are woody climbers that root in the ground and utilise the structure of neighbouring trees to ascend into the forest canopy. These plant forms are found in forests globally, although they are particularly prevalent seasonal and lowland forests of the neotropics where they make up a significant proportion of woody species and stems. Lianas infest host trees and can dominate forest canopies, often described as structural parasites, lianas compete intensely with trees for sunlight and belowground resources without investing in their own supporting structure. 2. Research interest in lianas has increased substantially in recent decades, though they remain understudied compared to other plant forms. Chapter 1 reviews the increasingly comprehensive and geographically broad evidence base documenting the competitive effect of lianas upon host trees. Liana infestation has a significant and detrimental impact on tree growth, mortality, reproduction, and regeneration. Ultimately this reduces the capacity of trees to sequester and store carbon, a globally important ecosystem service provided by tropical forests. Furthermore, some evidence suggests that lianas are disproportionately affecting shade-tolerant carbon dense tree genera, contributing to a decline in these species. Reports of increasing liana abundance in the neotropics adds to the urgency for further research into the extent, magnitude, and mechanisms by which lianas effect the carbon balance in tropical forests. 3. Studying tropical forest canopies is challenging, as such relatively little is known about the distribution of lianas in forest canopies. Remote sensing is an emerging research method that has overcome some of the shortfalls associated with standard field surveys and can provide new and critical insights into liana ecology. Chapter 2 provides the first systematic assessment of the progress in the use of remote sensing to further understand the (i) spatial and temporal distributions, (ii) structure and biomass, (iii) responses to environmental conditions, and (iv) diversity, of lianas. The possibilities offered by new and future advances in remote sensing technology to study lianas, and the further data requirements needed, are then considered. For unanswered research questions to be resolved, liana ecology needs remote sensing. 4. Chapter 3 is the first study to report findings from a large-scale liana removal experiment for which there is more than 3-4 years of data. I consider eight census years of data from an ongoing liana-removal experiment in Gigante, Panama to address the paucity of research into how lianas impact carbon accumulation in trees with differing life histories and functional traits. Tree biomass growth was 49.21% lower in liana-infested plots over the eight-year period. This is attributed to a growth release in low wood density pioneer species in the four years after removal, which shifts towards high wood density shade tolerant species dominating growth contributions to overall forest carbon in the latter four census years. My findings also support the notion that severity of crown infestation dictates the magnitude of liana effect on tree biomass growth. This study highlights the need to quantify the contribution of species functional groups to forest carbon balance in order to better understand potential future liana effects. With reports of increasing liana abundance, my findings present a worrying picture of the resilience of tropical forests to persist as a functioning global carbon sink. 5. Synthesis. Our understanding of liana ecology and the relationship between these plant forms and carbon balance in tropical forests has increased rapidly. Observational studies, removal experiments, repeated plot censuses and the rapidly evolving applications of remote sensing have provided new insights into the nature of lianas and their interactions with the tropical forests they infest. This study adds to these foundations for future liana research by presenting a thorough review of existing literature, a critical analysis of the integration of remote sensing and an assessment of disproportionate lianas effects on tree functional groups. With the current state of tropical forest decline, urgent knowledge gaps must be addressed now more than ever.
... Terrestrial Laser Scanning (TLS) can be used to measure structural parameters such as height, diameter, aboveground wood volume and Leaf Area Index (LAI) from LiDAR-derived 3D point clouds (Atkins et al., 2018;Béland et al., 2014;Calders et al., 2015;Strahler et al., 2008). This remote sensing technique has advanced swiftly in the last decade Dassot et al., 2011;Owen et al., 2021;Richardson et al., 2014). Although the use of this technology to study lianas has lagged behind that for trees ( . ...
Article
Full-text available
Lianas (woody vines) are abundant and diverse, particularly in tropical ecosystems. Lianas use trees for structural support to reach the forest canopy, often putting leaves above their host tree. Thus they are major parts of many forest canopies. Yet, relatively little is known about distributions of lianas in tropical forest canopies, because studying those canopies is challenging. This knowledge gap is urgent to address because lianas compete strongly with trees, reduce forest carbon uptake and are thought to be increasing, at least in the Neotropics. Lianas can be difficult to study using traditional field methods. Their pliable stems often twist and loop through the understorey, making it difficult to assess their structure and biomass, and the sizes and locations of their crowns. Furthermore, liana stems are commonly omitted from standard field surveys. Remote sensing of lianas can help overcome some of these obstacles and can provide critical insights into liana ecology, but to date there has been no systematic assessment of that contribution. We review progress in studying liana ecology using ground‐based, airborne and space‐borne remote sensing in four key areas: (i) spatial and temporal distributions, (ii) structure and biomass, (iii) responses to environmental conditions and (iv) diversity. This demonstrates the great potential of remote sensing for rapid advances in our knowledge and understanding of liana ecology. We then look ahead, to the possibilities offered by new and future advances. We specifically consider the data requirements, the role of technological advances and the types of methods and experimental designs that should be prioritised. Synthesis. The particular characteristics of the liana growth form make lianas difficult to study by ground‐based field methods. However, remote sensing is well suited to collecting data on lianas. Our review shows that remote sensing is an emerging tool for the study of lianas, and will continue to improve with recent developments in sensor and platform technology. It is surprising, therefore, how little liana ecology research has utilised remote sensing to date—this should rapidly change if urgent knowledge gaps are to be addressed. In short, liana ecology needs remote sensing.
... One of the research on vegetation mapping using terrestrial method is Terrestrial Laser Scanning for Vegetation Sampling (Richardson et al. 2014). This research using Terrestrial Laser Scanner (TLS) at Starkey Experimental Forest and Range and obtain the point cloud. ...
Article
Full-text available
The 3D city/landscape model is digital representation of real environment that can be used for planner or landscape architecture in urban development planning. That model’s more focused on building, whereas vegetation model’s also needed for urban planning simulation. This research purposes are to map existing vegetation and to generate individual trees model in some level of details. The research area is campus of ITB Jatinangor and the used data are orthophoto and DSM from UAV-Photogrammetry technology. Manual segmentation, classification, and NDSM generation process can provide tree information (position, crown diameter, species, height)-as 3D vegetation modeling input. It’s also necessary to provide classification, information, detail level, and visualization of vegetation model according to landscape architecture analysis needs. This research results are 3D vegetation models in LoD 1-3 with differents information based on appearance, geometry, semantic, and topology aspects of CityGML. Models then tested qualitatively based on visualization and sun shadow analysis. For visualization, the used data only able to generate LoD 1 and 2 vegetation model and the minimum LoD required for sun shadow analysis is LoD 2. Terrestrial data, which provide the real form and size of each tree part, is needed to generate LoD 3 vegetation model.
Chapter
The geographical dimensions of landscapes require the collection and the processes of the geometric attributes of patches like perimeters, shape, size, distance between other elements, fractal dimension, etc.
Article
Full-text available
The reconstruction of trees from point clouds that were acquired with terrestrial LiDAR scanning (TLS) may become a significant breakthrough in the study and modelling of tree development. Here, we develop an efficient method and a tool based on extensive modifications to the skeletal extraction method that was first introduced by Verroust and Lazarus in 2000. PypeTree, a user-friendly and open-source visual modelling environment, incorporates a number of improvements into the original skeletal extraction technique, making it better adapted to tackle the challenge of tree perennial tissue reconstruction. Within PypeTree, we also introduce the idea of using semi-supervised adjustment tools to address methodological challenges that are associated with imperfect point cloud datasets and which further improve reconstruction accuracy. The performance of these automatic and semi-supervised approaches was tested with the help of synthetic models and subsequently validated on real trees. Accuracy of automatic reconstruction greatly varied in terms of axis detection because small (length < 3.5 cm) branches were difficult to detect. However, as small branches account for little in terms of total skeleton length, mean reconstruction error for cumulated skeleton length only reached 5.1% and 1.8% with automatic or semi-supervised reconstruction, respectively. In some cases, using the supervised tools, a perfect reconstruction of the perennial tissue could be achieved.
Article
Full-text available
We present the point cloud slicing (PCS) algorithm, to post process point cloud data (PCD) from terrestrial laser scanning (TLS). We then test this tool for forest inventory application in urban heterogeneous forests. The methodology was based on a voxel data structure derived from TLS PCD. We retrieved biophysical tree parameters including diameter at breast height (DBH), tree height, basal area, and volume. Our results showed that TLS-based metrics explained 91.17% (RMSE = 9.1739 cm, p < 0.001) of the variation in DBH at individual tree level. Though the scanner generated a high-density PCD, only 57.27% (RMSE = 0.7543 m, p < 0.001) accuracy was achieved for predicting tree heights in these very heterogeneous stands. Furthermore, we developed a voxel-based TLS volume estimation method. Our results showed that PCD generated from TLS single location scans only captures 18% of the total tree volume due to an occlusion effect; yet there are significant relationships between the TLS data and field measured parameters for DBH and height, giving promise to the utility of a side scanning approach. Using our method, a terrestrial LiDAR-based inventory, also applicable to mobile- or vehicle-based laser scanning (MLS or VLS), was produced for future calibration of Aerial Laser Scanning (ALS) data and urban forest canopy assessments.
Article
Full-text available
Light detection and ranging (lidar) sensors, both at the terrestrial and airborne levels, have recently emerged as useful tools for three-dimensional (3D) reconstruction of vegetated environments. One such terrestrial laser scanner (TLS) is the Intelligent Laser Ranging and Imaging System (ILRIS-3D). The objective of this research was to develop approaches to use ILRIS-3D data to retrieve structural information of an artificial tree in a controlled laboratory experiment. The key crown-level structural parameters investigated in this study were gap fraction, leaf area index (LAI), and clumping index. Measured XYZ point cloud data from a systematically pruned tree were sliced to retrieve laser pulse return density profiles, which subsequently were used to estimate gap fraction. LAI, and clumping index. Gap fraction estimates were cross-validated with traditional methods of histogram thresholding of digital photographs (r2 = 0.95). LAI estimates from lidar data were corrected for the confounding effects of woody material and nonrandom foliage distribution and then compared with direct LAI measurements (r2 = 0.98. RMSE = 0.26). The methods developed in this research provide valuable lessons for application to field-level TLS data for structural parameter retrievals. Successful demonstration of analysis protocols to extract crown-level structural parameters like gap fraction. LAI, and clumping index from TLS data will be important for detailed assessments of 3D canopy radiative transfer modeling and likely will lead to more robust inversion algorithms.
Article
Full-text available
Ecologists need data on animal-habitat associations in terrestrial and aquatic environments to design and implement effective conservation strategies. Habitat characteristics used in models typically incorporate (1) field data of limited spatial extent and/or (2) remote sensing data that do not characterize the vertical habitat structure. Remote sensing tools that directly characterize three-dimensional (3-D) habitat structure and that provide data relevant to organism-habitat interactions across a hierarchy of scales promise to improve our understanding of animal-habitat relationships. Laser altimetry, commonly called light detection and ranging (lidar), is a source of geospatial data that can provide fine-grained information about the 3-D structure of ecosystems across broad spatial extents. In this review, we present a brief overview of lidar technology, discuss recent applications of lidar data in investigations of animal-habitat relationships, and propose future applications of this technology to issues of broad species-management and conservation interest.
Article
Full-text available
In the Río de la Plata grasslands (RPG) biogeographical region of South America, agricultural activities have undergone important changes during the last 15-18 years because of technological improvements and new national and international market conditions. We characterized changes in the landscape structure between 1985-1989 and 2002-2004 for eight pilot areas distributed across the main regional environmental gradients. These areas incorporated approximately 35% of the 7.5 × 10 5 km² of the system. Our approach involved the generation of land-use and land cover maps, the analysis of landscape metrics, and the computation of annual transition probabilities between land cover types. All of the information was summarized in 3383 cells of 8 × 8 km. The area covered by grassland decreased from 67.4 to 61.4% between the study periods. This decrease was associated with an increase in the area of annual crops, mainly soybean, sunflower, wheat, and maize. In some subunits of the RPG, i.e., Flat Inland Pampa, the grassland-to-cropland transition probability was high ( pG→C = 3.7 × 10-2), whereas in others, i.e., Flooding Pampa, this transition probability was low ( pG→C = 6.7 × 10-3). Our description of the magnitude, direction, and spatial distribution of land-use and land cover changes provides a basis from which to develop spatially explicit scenarios of land cover change.
Article
We develop a hierarchical model of heterogeneity that provides a framework for classifying patch structure across a range of scales. Patches at lower levels in the hierarchy are more simplistic and correspond to the traditional view of patches. At levels approaching the upper bounds of the hierarchy the internal structure becomes more heterogeneous and boundaries more ambiguous. At each level in the hierarchy, patch structure will be influenced by both contrast among patches as well as the degree of aggregation of patches at lower levels in the hierarchy. We apply this model to foraging theory, but it has wider applications as in the study of habitat selection, population dynamics, and habitat fragmentation. It may also be useful in expanding the realm of landscape ecology beyond the current focus on anthropocentric scales.
Article
Nondetection of trees is a serious problem for the use of terrestrial laser scanning (TLS) in forest inventory applications. The use of multiple coregistered scans can reduce nondetection but may not eliminate it, and it carries substantial field and post-processing costs. We examined and extended previously developed theoretical approaches to modeling nondetection. The results suggested that tree size as well as multiple stand structural characteristics may be factors, but the theoretical models do not lend themselves to empirical estimation. We then used distance sampling techniques to identify detection probabilities and develop adjusted estimates for trees per hectare and basal area in nine forest stands in southern Norway. The results compared favorably with field estimates based on fixed-area plots. The estimated detection probabilities indicate that correction for nondetection is needed unless the search for trees is limited to very small distances from the scanner. Distance sampling appears promising when TLS is used in the context of temporary-plot forest inventories.
Article
Shrublands in semiarid regions are heterogeneous landscapes consisting of infertile bare areas separated by nutrient rich vegetated areas known as resource islands. Spatial patterns in these landscapes are structured by feedbacks driven by the transport of water and nutrient resources from the intershrub space to areas below shrubs, and the retention of these resources to locally drive productivity and tight biogeochemical cycles. Most understanding of plant-soil feedbacks is based predominantly on studies of low topographic gradient landscapes, and it is unclear whether the patterns of association between soils and vegetation, and the autogenic processes that create them, also occur on more steeply sloping terrain. Here we analyze the spatial patterns of soils, vegetation, and microtopography on hillslopes of contrasting lithology (one granite at 16°, one schist at 27°) in the Sonoran desert foothills of the Catalina Mountains. We also describe a method of extracting vegetation density from terrestrial laser scanning point cloud data at 5 cm × 5 cm scales and find that it correlates well with soil organic carbon measurements. Vegetation was associated with microtopographic mounds (relative to the mean slope) extending 0.3 m downslope and 1.8 m (schist) and 0.9 m (granite) upslope on the study hillslopes. Soils below the shrub canopies exhibited 2-3 times more soil organic matter and 2-4 times higher hydraulic conductivity than the interspaces. Soils enriched with organic matter were found to extend at least two canopy radii downslope of woody shrubs, but not upslope. These plumes were clearest in the lower gradient granite site where vegetation mounds created distinct patterns of microtopographic convergence and divergence. At the steeper schist site, microtopography appeared to have a weaker control on topographic flow accumulation. Collectively, our findings suggest that the spatial structure of association between soils and microtopography and vegetation on these slopes exhibit many of the features observed in lower gradient areas. However, microtopography and soils are more asymmetric along the downslope axis of the hillslopes than lower gradient areas and vary with lithology. Alluvial and colluvial processes are likely more important in shaping vegetation and soil dynamics on hillslopes, and these factors need further consideration in scaling results to the landscape level.
Article
1. Vegetation structure is an important determinant of species habitats and diversity. It is often represented by simple metrics, such as canopy cover, height and leaf area index, which do not fully capture three-dimensional variations in density. Terrestrial Laser Scanning (TLS) is a technology that can better capture vegetation structure, but methods developed to process scans have been biased towards forestry applications. The aim of this study was develop a methodology for processing TLS data to produce vegetation density profiles across a broader range of habitats. 2. We performed low-resolution and medium-resolution TLS scans using a Leica C5 Scanstation at four locations within eight sites near Wollongong, NSW, Australia (34.38–34.41oS, 150.84–150.91oE). The raw point clouds were converted to density profiles using a method that corrected for uneven ground surfaces, varying point density due to beam divergence and occlusion, the non-vertical nature of most beams, and for beams that passed through gaps in the vegetation without generating a point. Density profiles were evaluated against visual estimates from three independent observers using coarse height classes (e.g. 5–10m). 3. TLS produced density profiles that captured the three-dimensional vegetation structure. Although sites were selected to differ in structure, each was relatively homogeneous, and yet we still found a high spatial variation in density profiles. There was also large variation between observers, with the RMS error of the three observers relative to the TLS varying from 16.2% to 32.1%. Part of this error appeared to be due to misjudging the height of vegetation, which caused an overestimation in one height class and an underestimation in another. 4. Our method for generating density profiles using TLS can capture three-dimensional vegetation structure in a manner that is more detailed and less subjective than traditional methods. The method can be applied to a broad range of habitats—not just forests with open understoreys. However, it cannot accurately estimate near-surface vegetation density when there are uneven surfaces or dense vegetation prevents sufficient ground returns. Nonetheless, TLS density profiles will be an important input for research on species habitats, microclimates and nutrient cycles.
Article
Plant structure and chlorophyll content strongly affect rates of photosynthesis. Rapid, objective, and repeatable methods are needed to measure these vegetative parameters to advance our understanding and modeling of plant ecophysiological processes. Terrestrial laser scanners (TLS) can be used to measure structural and potentially chemical properties of objects by quantifying the x,y,z coordinates and intensity of laser light, respectively, returned from an object's surface. The objective of this study was to determine the potential usefulness of TLS with a green (532nm) laser to simultaneously measure the spatial distribution of chlorophyll a and b content (Chlab), leaf area (LA), and leaf angle (LAN). The TLS measurements were obtained from saplings of two tree species (Quercus macrocarpa and Acer saccharum) and from an angle-adjustable cardboard surface. The green laser return intensity value was strongly correlated with wet-chemically determined Chlab (r2=0.77). Strong agreement was shown between measured and TLS-derived LA (r2=0.95, intercept=−1.43, slope=0.97). The TLS derived LANs of both species followed a plagiophile LAN distribution, and the measured angles of the cardboard surface allowed us to quantify that these LAN values were strongly correlated with TLS derived angles (r2=1.0, intercept and slope=0.98). Our results show that terrestrial laser scanners are feasible for simultaneous measurement of LA, LAN, and Chlab in simple canopies of small broadleaved plants. Further research is needed in more complex and larger canopies.