remote sensing

Article
Aboveground Carbon Stock Estimation Based on Backpack
LiDAR and UAV Multispectral Imagery at the Forest Sample
Plot Scale
Rina Su 1,2 , Wala Du 1,3, *, Yu Shan 4 , Hong Ying 4 , Wu Rihan 4 and Rong Li 4

1 The Institute of Grassland Research of CAAS, Hohhot 010010, China; 20172104185@mails.imnu.edu.cn

2 College of Resource and Environmental Sciences, Inner Mongolia Agricultural University,
Hohhot 010018, China
3 Forest and Grassland Disaster Prevention and Iitigation Field Scientific Observation and Research Station of
Inner Mongolia Autonomous Region, Arshan 137400, China
4 The College of Geographic Science, Inner Mongolia Normal University, Hohhot 010022, China;
yushan@imnu.edu.cn (Y.S.); hongy864@nenu.edu.cn (H.Y.); wurh651@imnu.edu.cn (W.R.);
20224016019@mails.imnu.edu.cn (R.L.)
* Correspondence: duwala@caas.cn; Tel.: +86-186-8602-5858

Abstract: Aboveground carbon stocks (AGCs) in forests play an important role in understanding
carbon cycle processes. The global forestry sector has been working to find fast and accurate methods
to estimate forest AGCs and implement dynamic monitoring. The aim of this study was to explore
the effects of backpack LiDAR and UAV multispectral imagery on AGC estimation for two tree
species (Larix gmelinii and Betula platyphylla) and to emphasize the accuracy of the models used.
We estimated the AGC of Larix gmelinii and B. platyphylla forests using multivariate stepwise linear
regression and random forest regression models using backpack LiDAR data and multi-source remote
sensing data, respectively, and compared them with measured data. This study revealed that (1) the
diameter at breast height (DBH) extracted from backpack LiDAR and vegetation indices (RVI and
GNDVI) extracted from UAV multispectral imagery proved to be extremely effective in modeling for
estimating AGCs, significantly improving the accuracy of the model. (2) Random forest regression
Citation: Su, R.; Du, W.; Shan, Y.; models estimated AGCs with higher precision (Xing’an larch R2 = 0.95, RMSE = 3.99; white birch
Ying, H.; Rihan, W.; Li, R.
R2 = 0.96, RMSE = 3.45) than multiple linear regression models (Xing’an larch R2 = 0.92, RMSE = 6.15;
Aboveground Carbon Stock
white birch R2 = 0.96, RMSE = 3.57). (3) After combining backpack LiDAR and UAV multispectral
Estimation Based on Backpack LiDAR
data, the estimation accuracy of AGCs for both tree species (Xing’an larch R2 = 0.95, white birch
and UAV Multispectral Imagery at the
R2 = 0.96) improved by 2% compared to using backpack LiDAR alone (Xing’an larch R2 = 0.93, white
Forest Sample Plot Scale. Remote Sens.
2024, 16, 3927. https://doi.org/
birch R2 = 0.94).
10.3390/rs16213927
Keywords: backpack LiDAR; UAV multispectral imagery; aboveground carbon stock (AGC); multiple
Academic Editors: Qinghua Guo and
stepwise linear regression (MSLR); random forest regression (RF)
Brenden E. McNeil

Received: 4 September 2024

Revised: 28 September 2024
Accepted: 18 October 2024 1. Introduction
Published: 22 October 2024
Forest ecosystems play a crucial role in terrestrial ecosystems, acting as massive
global carbon reservoirs that contain 80% of the terrestrial carbon stored in aboveground
biomass [1]. The aboveground carbon stock (AGC) of forests is a key parameter for assessing
Copyright: © 2024 by the authors.
the carbon sequestration capacity and carbon balance above the forest soil layer, making
Licensee MDPI, Basel, Switzerland. it vital to understand the role of forests in carbon cycling and climate change [2]. As the
This article is an open access article issue of climate change becomes increasingly severe, accurately estimating and monitoring
distributed under the terms and changes in forest AGCs are essential for developing sustainable ecosystem management
conditions of the Creative Commons and policies to mitigate climate change [3].
Attribution (CC BY) license (https:// However, accurately estimating forest AGCs is a challenging task. Although traditional
creativecommons.org/licenses/by/ field survey methods are precise and reliable, they are time-consuming and labor-intensive,
4.0/). making it difficult to apply them over large areas or to fully cover diverse ecosystem

Remote Sens. 2024, 16, 3927. https://doi.org/10.3390/rs16213927 https://www.mdpi.com/journal/remotesensing

Remote Sens. 2024, 16, 3927 2 of 23

types. As a result, remote sensing-assisted datasets have gradually become an important

supplementary approach, providing spatially and temporally continuous information over
extensive regions [4]. Remote sensing data sources such as LiDAR and UAV multispectral
imagery have high-resolution and wide-area surface information, offering the possibility of
estimating carbon stocks over large areas [5–8]. However, despite numerous studies show-
ing that LiDAR or optical data alone can predict forest AGCs, the accuracy obtained from
these methods without support from field measurements remains a challenge. Therefore,
calibration and validation of remote sensing-assisted prediction techniques still require
field measurement data.
Light detection and ranging (LiDAR) data, as a remote sensing data source, can be
used to rapidly and accurately obtain elevation information and vertical structure data for
vegetation cover [9] and are among the latest remote sensing technologies for forest carbon
accounting [10]. Backpack LiDAR, such as LiBackpack, a new type of portable LiDAR, offers
high capacity, accessibility, and flexibility in route selection and can obtain high-quality 3D
dense point clouds in forests with different vegetation structures [11]. However, despite
the accuracy of backpack LiDAR for measuring tree diameter [12,13], identifying the best
variables suitable for different tree species remains challenging, especially in the absence
of canopy spectral information, where LiDAR data classification accuracy for tree species
under complex vegetation conditions is limited [14]. On the other hand, UAV multispectral
imagery includes multispectral information, aiding in studying the spectral characteristics
of different tree species. Early research revealed that the high-resolution image texture
features of optical remote sensing data are strongly correlated with forest biomass [15,16],
and the vegetation indices obtained from optical data typically reach saturation at relatively
low biomass values [17,18]. UAV multispectral imagery has better spatial, spectral, and
temporal resolution compared to other optical datasets for similar data volumes and data
collection costs [19], and their time-series data provide high-quality information on seasonal
changes in forests. Additionally, UAV multispectral data are effectively used for forest
resource monitoring and dynamic management [20]. Recent studies suggest that combining
LiDAR and optical sensors is a feasible approach for estimating biomass and carbon storage
in both plantations and natural forests [21]. Brown et al. [22] showed that modeling with a
large amount of field measurement data added to data fusion can improve the estimation
of forest AGBs and AGCs. Kim et al. [23] stated that combining spectral information with
attributes derived from LiDAR data is more suitable for assessing the AGB and AGC than
using optical images or LiDAR data alone. However, finding a method for accurately
estimating forest AGCs with fewer field measurements to establish regression models is
currently challenging.
Extracting vegetation information from remote sensing imagery and integrating it
with ground-measured data for modeling has become an effective and popular method
for obtaining regional forest AGCs. The study mentioned the use of parametric and non-
parametric models. Multiple stepwise linear regression (MSLR) represents the traditional
parametric model, assuming a linear relationship between predictive variables and the
variable being predicted. However, this assumption limits the inherent nonlinearity of the
relationship between them and requires a large sample size [24]. On the other hand, random
forest regression (RF) is a nonparametric model that does not assume a specific distribution
for the samples. It can handle complex nonlinear relationships and high-dimensional prob-
lems and has been proven effective for estimating forest AGCs [25]. Additionally, machine
learning techniques aid in combining data from different sources to improve outcomes [26].
In this study, L. gmelinii and B. platyphylla, two typical tree species, were chosen as
the study species due to their ecological significance and notable differences in carbon
storage [27]. Liu et al. explored the carbon storage capacity of Xing’an larch and birch
forests, suggesting that they have an important role in boreal forests [28]. Larix gmelinii,
an important economic tree species, grows in cold and dry conditions and has a high
carbon stock. In contrast, compared with L. gmelinii, B. platyphylla plays a different role in
ecosystems and has distinct growth environments and characteristics, leading to potentially
Remote Sens. 2024, 16, 3927 3 of 23

significant differences in carbon storage. By studying these two typical species, we aim
to deepen the understanding of the variability in forest carbon storage among different
tree species, thereby providing more accurate estimation models to support forest resource
management and conservation.
Despite these developments, comprehensive validation of the accuracy of forest AGC
estimation is lacking. This study is dedicated to addressing this challenge. We hypothesize
that the combination of forest vertical variables with horizontal variables from optical
images through allometric relationships can be used to accurately estimate forest AGCs
at the plot scale with minimal field measurement data. Therefore, the objectives of this
study are: (1) to assess the effectiveness of using backpack LiDAR and UAV multispectral
imagery in estimating aboveground carbon stocks (AGC) for L. gmelinii and B. platyphylla;
(2) to compare the predictive accuracy of different models; and (3) to validate whether
integrating multi-source data can enhance the accuracy of AGC estimation.

2. Materials and Methods

2.1. Study Area
The research area is located in the Dural National Forest Farm, Arxan city, Hing-
gan League, Inner Mongolia Autonomous Region of China (119◦ 28′ –120◦ 01′ E, 47◦ 15′ –
47◦ 35′ N), which is a comprehensive forest farm integrating both natural and artificial forests
(Figure 1—Some of the sampling sites are located outside of the yellow box labeled as
the UAV-LS working area. However, in the actual study, the sample plots used in the
estimation of AGCs were those within the yellow boxes). This region, situated in the
Hinggan League of the Inner Mongolia Autonomous Region of China, is an important
natural ecological conservation area. The total area of the Dural Forest Farm is 49,812
hectares, 33,466 hectares of which are designated for forestry, accounting for 67% of the
total area. The forest coverage rate is as high as 48.3%, with an altitude ranging from 792 to
1495 metres. The region experiences a cold temperate continental monsoon climate, with an
average annual temperature of 1.48 ◦ C, an average temperature of −25.6 ◦ C in the coldest
month, and an average temperature of 16.6 ◦ C in the hottest month. The average annual
precipitation is 437 mm. The main tree species of the forest are B. platyphylla and L. gmelinii,
along with a minority of Populus davidiana and Pinus sylvestris. Most of the area has good
site conditions, with similar forest types distributed in concentrated and contiguous areas,
which is beneficial for forest management and operation [29].

2.2. Inventory Data

Within the study area covered by airborne LiDAR data, we conducted two field sur-
veys for tree information collection at the plot scale in July 2021 and July 2022 (Figure 2). In
this study, we used field-measured data from 435 rectangular plots, each measuring 0.04 ha
(230 plots from larch forests and 205 plots from birch forests). Additionally, considering the
distributions of L. gmelinii and B. platyphylla, these plots were established to provide a good
statistical representation of these two main tree species. Given the relative rareness of tree
species in the study area, our data collection specifically targeted these two tree species.
Within each plot, the tree heights were measured using a laser rangefinder. Trees with a
diameter at breast height (DBH) greater than 5 cm were marked at a height of 1.2 m using a
DBH measuring tape. Simultaneously, the coordinates of the plot’s center point and its four
corners were obtained using a GNSS receiver with differential satellite station technology,
which was employed for accurate positioning.
Currently, most forest biomass estimation studies are based on existing allometric
growth equations selected according to the study area or equations fitted from the analysis
of felled trees [30]. Allometric growth equations are fundamental not only for calculating
the biomass of various tree organs and vegetation carbon storage but also for estimating
tree carbon sequestration rates and potentials [31]. The diameter at breast height (DBH)
and tree height (H) have been used as predictive variables for estimating aboveground
biomass (AGB) [32,33]. Therefore, this study estimated the AGB of individual L. gmelinii
Remote Sens. 2024, 16, 3927 4 of 23

and B. platyphylla trees in the Dural Forest Farm based on DBH (cm) and H using allometric
growth equations for the dominant tree species of the Greater Khingan Range in Inner
Mongolia [34,35] (Table 1). The total AGB of the forests in the study area was calculated as
shown in Equation (1):
Remote Sens. 2024, 16, x FOR PEER REVIEW 4 of 25
AGBTotal = WStem + WBranch + WLea f + WBark (1)

Figure1.1. Location
Figure Location of
of the
the study
study area.
area. The
The study
study area
area is
is located
located entirely
entirely within
within Chinese
Chinese territory,
territory,with
with
research activities strictly confined to ecological and environmental aspects.
research activities strictly confined to ecological and environmental aspects.

2.2. Inventory
Table Datagrowth equations for Inner Mongolia L. gmelinii and B. platyphylla.
1. Allometric
Within the study area covered by airborne LiDAR data, we conducted two field sur-
Allometric Growth Equation of Larix Allometric Growth Equation of Betula
Organ
veys for tree information collection at the plot scale platyphylla
in July 2021 and July 2022 (Figure 2).
gmelinii
In this study, we used field-measured data from 435 rectangular plots, each measuring
Stem WStem = 0.01258(D2 H)0.99331 WStem = 0.02853(D2 H)0.89278
0.04 ha (230 plots from larch forests and 205 plots from birch forests). Additionally, con-
Branch WBranch = 0.00136(D2 H)1.02797 WBranch = 0.00278(D2 H)1.02568
sidering
Leaf the distributions of L.2gmelinii
WLeaf = 0.01009(D H)0.64543 and B. platyphylla,
WLeaf =these plots2 H)
0.01545(D were established to
0.61265
provide
Bark a goodWstatistical
Bark = representation
0.002307(D 2 H)0.70655 of these twoW main
Bark =tree species.
0.02392(D 2 H)Given
0.71131 the relative

Drareness
representsofthetree species
diameter in the
at breast study
height (DBH)area, our
of the tree;data collection
H represents specifically
the tree targeted
height; WStem these
represents the
two tree
biomass species.
of the Within
tree trunk; each
WBranch plot, the
represents tree heights
the biomass were measured
of the branches; usingthea biomass
WLeaf represents laser range-
of the
leaves; and WBark represents the biomass of the bark.
finder. Trees with a diameter at breast height (DBH) greater than 5 cm were marked at a
height of 1.2 m using a DBH measuring tape. Simultaneously, the coordinates of the plot’s
center the L.
Forpoint andgmelinii
its four and B. platyphylla
corners forests using
were obtained in theaDural
GNSSForest
receiverFarm,
withwe referred
differential
to the standard conversion factors for L. gmelinii and B.
satellite station technology, which was employed for accurate positioning. platyphylla forests in the Greater
Khingan Range, which are 0.4948 and 0.5018, respectively [36], to convert AGB to AGC
Remote Sens. 2024, 16, 3927 5 of 23

(aboveground carbon) [37], as shown in Equations (2) and (3). Table 2 provides a summary
of the forest parameters obtained from field measurements.

AGCLarix = 0.4948 × AGBTotal (2)

Remote Sens. 2024, 16, x FOR PEER REVIEW 5 of 25
AGCBetula = 0.5018 × AGBTotal (3)

(a) (b)

(c) (d)
Figure 2.
Figure 2. Parameter
Parametermeasurements
measurements inin
thethe
study area:
study (a) Setup
area: of base
(a) Setup stations.
of base (b) Tree
stations. (b)height meas-
Tree height
urement. (c) Obtaining the coordinates. (d) DBH measurement and recording.
measurement. (c) Obtaining the coordinates. (d) DBH measurement and recording.

Currently,
In mostAGB
the formula, forest biomass estimation studies are based on existing allometric
Total represents the total aboveground biomass, AGCLarix is the
aboveground carbon stock of L. gmelinii,the
growth equations selected according to andstudy
AGCarea or equations fitted from the analysis
Betula is the aboveground carbon stock of
of felled trees [30]. Allometric
B. platyphylla forests. growth equations are fundamental not only for calculating
the biomass of various tree organs and vegetation carbon storage but also for estimating
tree carbon sequestration rates and potentials [31]. The diameter at breast height (DBH)
and tree height (H) have been used as predictive variables for estimating aboveground
biomass (AGB) [32,33]. Therefore, this study estimated the AGB of individual L. gmelinii
and B. platyphylla trees in the Dural Forest Farm based on DBH (cm) and H using allome-
tric growth equations for the dominant tree species of the Greater Khingan Range in Inner
Mongolia [34,35] (Table 1). The total AGB of the forests in the study area was calculated
Remote Sens. 2024, 16, 3927 6 of 23

After simplifying Equations (1)–(3), the final equations for calculating the measured
aboveground carbon (AGC) for L. gmelinii and B. platyphylla forests are obtained, as shown
in Equations (4) and (5):
0.8955
AGCLarix_Measured = 0.0248 × ( D2 H ) (4)
0.8622
AGCBetula_Measured = 0.034 × ( D2 H ) (5)
In the formulas, AGCLarix measured represents the measured aboveground carbon
stock of L. gmelinii; AGCBetula measured represents the measured aboveground carbon
stock of B. platyphylla forests; D is the measured diameter at breast height (DBH); and H is
the measured tree height.

Table 2. Summary statistics of the field-measured forest parameters.

Larix gmelinii Betula platyphylla

Variable
Min Max Mean Std Min Max Mean Std
H (m) 4 16 11.89 2 6 16.9 13.43 2.25
DBH (cm) 5.5 35 15 4.28 5.3 29.8 12.58 4.37
AGB (Kg) 3950 320,540 64,490 40,480 6590 231,500 56,580 39,570
AGC (Kg) 1960 158,610 31,910 20,030 3310 116,170 28,390 19,850

2.3. Remote Sensing Data

2.3.1. Backpack LiDAR Data
The Backpack Laser Scanning (BLS) system is a backpack-mounted device that inte-
grates LiDAR with other sensors. The simultaneous localization and mapping (SLAM)
algorithm allows for the rapid and continuous acquisition of LiDAR point cloud data, even
as the robot or device moves through an unknown environment. The SLAM algorithm
enables a robot or device to autonomously create a map and accurately locate itself without
knowing its initial position, facilitating navigation in unknown environments.
In this study, we employed the LiBackpack DGC 50 backpack laser scanner developed
by Beijing Greenvalley Technology Co., Ltd. (Beijing, China) (Figure 3). This device
efficiently captures tree point cloud data within plots using a designed “S”-shaped hiking
route for data collection [38]. The choice of the LiBackpack DGC 50 was based on its effective
data collection capability and reliable data quality assurance. During data collection, the
operator connects to the backpack laser scanner via a smartphone to monitor the number
of satellite signals and real-time point cloud scanning status. To ensure the quality of the
point cloud data, steady movement is required during collection, especially at turns, to
ensure data accuracy and completeness.
To acquire point cloud data with geographic coordinates, a GNSS receiver (CHC
iRTK 5) is used in an open area outside the plot with stable Global Navigation Satellite
System signals. This receiver acquires the absolute geographic coordinates of a point using
satellite differential techniques, and a base station is set up at this point to gather static data.
Finally, the raw point cloud data, trajectory files, and GNSS static data are imported into
GreenValley International Li-Fuser BP software (Digital Green Earth; Beijing, China) for
processing to obtain point cloud data with absolute geographic location information.
The LiBackpack DGC 50 laser scanner and Li-Fuser BP software were selected for their
superior capabilities in rapid data acquisition and processing. The backpack LiDAR data
were collected by an operator who walked along an ‘S’-shaped route, ensuring compre-
hensive coverage of each plot. During data collection, the operator monitored the data
quality in real-time via a mobile connection, ensuring the accuracy and completeness of the
point cloud data. The scanning path was meticulously designed to cover the entirety of
each designated plot, ensuring that the collected data was precisely aligned with the AGC
measurement points.
 data were collected by an operator who walked along an ‘S’-shaped route, ensuring com-
prehensive coverage of each plot. During data collection, the operator monitored the data
quality in real-time via a mobile connection, ensuring the accuracy and completeness of
the point cloud data. The scanning path was meticulously designed to cover the entirety
Remote Sens. 2024, 16, 3927 7 ofthe
23
of each designated plot, ensuring that the collected data was precisely aligned with
AGC measurement points.

Figure3.3.LiBackpack
Figure LiBackpackDGC50
DGC50backpack
backpackLiDAR
LiDARscanning
scanningsystem.
system.

Remote Sens. 2024, 16, x FOR PEER REVIEW 8 of 25

Whenprocessing
When processingthe thebackpack
backpackLiDAR LiDARdata,data,the
theinitial
initialsteps
stepsinvolve
involve copying
copyingthe the raw
raw
fielddata
field dataand
andexporting
exportingthe theimage
imagedata.
data. Next,
Next, Hi-Target
Hi-Target Business
BusinessCenter
Centersoftware
software(Beijing
(Beijing
Huatai Kejie Information
Huatai Information Technology
TechnologyCo.; Co.;Beijing,
Beijing, China)
China) is used
is usedto convert
to convertthe static data
the static
data
into into
GNSS GNSS
files files in RINEX
in RINEX format.format. The Insta360
The Insta360 studiostudio
softwaresoftware (Shenzhen
(Shenzhen Qianhai Qianhai
Shad-
Shadowstone
owstone InnovativeInnovative Technology
Technology Co.; Shenzhen,
Co.; Shenzhen, China)China)
is thenisutilized
then utilized
to convertto convert
the re-
the required
quired videovideo
format format
fromfrom the backpack
the backpack LiDAR LiDAR
into into
.MP4. .MP4. Afterward,
Afterward, Li-Fuser
Li-Fuser BP
BP soft-
software is used
ware is used to process
to process thethedatadata coordinates,
coordinates, calibrating
calibrating thethe relative
relative coordinates
coordinates from from
the
the mobile trajectory to the geodetic coordinates,
mobile trajectory to the geodetic coordinates, with all coordinates with all coordinates resolved using the
using the
CGCS2000
CGCS20003-degree
3-degreeGK GK CM CM 120E
120E system
system (Environment
(Environment SystemSystem Research
Research Institute,
Institute, ESRI).
Each
Each plot
plot covers
covers an an area
area of of 0.04
0.04hectares,
hectares, and and the
theLiDAR
LiDARscan scanensures
ensurescomprehensive
comprehensive
coverage
coverageof ofall
alltrees,
trees,providing
providingcomplete
completedata dataonontree
treeheight,
height,DBH,DBH,and andcanopy
canopystructure.
structure.
The
The processed
processed Backpack
Backpack data data were
were preprocessed
preprocessed using using LiDAR360
LiDAR360 V5.3 V5.3 software
software
(Digital Green Earth Ltd.; Beijing, China). The point cloud data were
(Digital Green Earth Ltd.; Beijing, China). The point cloud data were clipped according to clipped according
to
thethe sample
sample range,
range, andand after
after removing
removing redundancy
redundancy andandnoise,noise,
the the
datadata
werewere filtered.
filtered. The
The ground points were classified, and a digital elevation
ground points were classified, and a digital elevation model (DEM) was generated model (DEM) was generated
through
throughirregular
irregulartriangular
triangularmesh meshinterpolation.
interpolation. Based
Based onon thetheDEM,
DEM,the thepoint
pointcloud
cloudwas was
normalized
normalized to obtain seed points, and individual trees were segmented using theseseed
to obtain seed points, and individual trees were segmented using these seed
points
pointstotoacquire
acquireground-based
ground-basedpoint pointcloud
clouddata.
data.Figures
Figures44and and55display
displaythetheprofile
profilepoint
point
cloud data of the collected L. gmelinii and B. platyphylla plots, visualized
cloud data of the collected L. gmelinii and B. platyphylla plots, visualized by elevation. by elevation.

Figure4.4.Point
Figure Pointcloud
clouddata
dataof
ofL.
L.gmelinii
gmeliniisample
sampleplot
plotprofiles
profilescollected
collectedvia
viabackpack
backpackLiDAR.
LiDAR.
Remote Sens. 2024, 16, 3927 8 of 23

Figure 4. Point cloud data of L. gmelinii sample plot profiles collected via backpack LiDAR.

Figure5.5.Point
Figure Pointcloud
clouddata
dataof
ofB.
B.platyphylla
platyphyllasample
sampleplot
plotprofiles
profilescollected
collectedby
bybackpack
backpackLiDAR.
LiDAR.

2.3.2.
2.3.2.UAV
UAVMultispectral
MultispectralData Data
On
On11 11July
July2021,
2021,multispectral
multispectralimagery
imagerywas wascollected
collectedacross
across sixsix
transect
transect areas
areas using a
using
Feima
a Feima D200
D200 multirotor
multirotor UAVUAV (Pegasus
(Pegasus Robotics
RoboticsTechnology
Technology Co.;
Co.;Shenzhen,
Shenzhen, China)
China)and a
and
V300
a V300fixed-wing
fixed-wing UAV UAVequipped
equippedwithwitha D-CAM2000
a D-CAM2000 multispectral sensor
multispectral (Pegasus
sensor Robotics
(Pegasus Ro-
Technology Co.; Shenzhen, China) (shown in Figure 6). The
botics Technology Co.; Shenzhen, China) (shown in Figure 6). The sensor captured sensor captured six spectral
six
bands,
spectralnamely
bands, blue,
namely green,
blue,red,
green,red-edge, near-infrared,
red, red-edge, and panchromatic
near-infrared, and panchromatic bands.bands.
The
flight operation
The flight was designed
operation was designedat an ataltitude of 383ofm383
an altitude with an 80%
m with an flight path path
80% flight overlap and
overlap
60% side overlap. The UAVs were equipped with an inertial navigation
and 60% side overlap. The UAVs were equipped with an inertial navigation system (IMU),
Remote Sens. 2024, 16, x FOR PEER REVIEW system
9 of 25 (IMU),
providing
providingaaspatial
spatialresolution
resolutionof of0.02
0.02m. m. The
The acquired
acquired sensor
sensor images
images were were loaded
loadedintointo
pix4dmapper
pix4dmapper 4.5.6 official version software (Pix4D Company, Switzerland) toallow
4.5.6 official version software (Pix4D Company, Switzerland) to allowthethe
feature
feature point matchingalgorithm
point matching algorithmtoto match
match thethe different
different images,
images, andand the aerial
the aerial triangulation
triangula-
method and and
tion method beam method
beam methodleveling algorithm
leveling algorithm were
wereusedusedtotoobtain
obtainthe
the multispectral
multispectral images
of the UAV-piloted flight area. During field data collection, we used
images of the UAV-piloted flight area. During field data collection, we used RTK-GNSS RTK-GNSS equipment
to precisely measure the corner and center points of each plot, ensuring
equipment to precisely measure the corner and center points of each plot, ensuring centi- centimeter-level
meter-level
accuracy in accuracy in the coordinates.
the coordinates. These coordinates
These coordinates were thenwere then converted
converted to the same to thecoordinate
same coordinate system as the UAV imagery (CGCS2000) to maintain
system as the UAV imagery (CGCS2000) to maintain spatial consistency across different spatial consistency
acrosssources.
data different Indata sources.10.7,
ArcMap In ArcMap 10.7, the UAV-acquired
the UAV-acquired multispectralmultispectral
images images
were cropped
according to the boundaries of each plot, ensuring that the resulting imagesimages
were cropped according to the boundaries of each plot, ensuring that the resulting only included
only included pixels within the plot area. Additionally, feature point matching algorithms
pixels within the plot area. Additionally, feature point matching algorithms were applied
were applied to spatially align the UAV imagery with ground-based data and further op-
to spatially align the UAV imagery with ground-based data and further optimized using
timized using bundle adjustment techniques to ensure precise spatial correspondence be-
bundle adjustment techniques to ensure precise spatial correspondence between datasets.
tween datasets.

(a) (b)
Figure 6.
Figure 6. (a)
(a)Pegasus
PegasusV300
V300drone andand
drone (b) (b)
multispectral D-CAM2000
multispectral sensor.sensor.
D-CAM2000

The six
The six spectral
spectralbands
bands of of
UAVUAVmultispectral imagery
multispectral contain
imagery rich vegetation
contain infor- infor-
rich vegetation
mation, which
mation, whicharearecrucial
crucialfactors in estimating
factors forest
in estimating carbon
forest storage.
carbon This study
storage. Thisselected
study selected
five bands closely related to vegetation (blue, green, red, and near-infrared) and used
five bands closely related to vegetation (blue, green, red, and near-infrared) and used
ArcGIS 10.7 to extract band information for the main tree species plots in the study area.
ArcGIS 10.7 to extract band information for the main tree species plots in the study area.
The spectral information resulting from the reflection, absorption, and scattering of solar
radiation by the forest canopy, along with chlorophyll content, is an important variable
for AGB/AGC modeling [39–41]. The seven vegetation indices derived from band combi-
nations, including NDVI, RVI, DVI, EVI, GNDVI, NDRE, and SAVI, effectively reflect veg-
etation growth and health and are closely related to AGB/AGC [42–48]. In this study, these
vegetation indices were extracted and coupled with LiDAR structural variables to estab-
Remote Sens. 2024, 16, 3927 9 of 23

The spectral information resulting from the reflection, absorption, and scattering of solar
radiation by the forest canopy, along with chlorophyll content, is an important variable
for AGB/AGC modeling [39–41]. The seven vegetation indices derived from band com-
binations, including NDVI, RVI, DVI, EVI, GNDVI, NDRE, and SAVI, effectively reflect
vegetation growth and health and are closely related to AGB/AGC [42–48]. In this study,
these vegetation indices were extracted and coupled with LiDAR structural variables to
establish a model for quantitatively estimating AGC density (see Table 3). ENVI 5.3 soft-
ware was used to perform calculations on the aforementioned bands of UAV multispectral
imagery, computing various types of vegetation indices for modeling and analysis when
estimating forest vegetation carbon storage.

Table 3. Vegetation Index Formulas.

Vegetation Index Descriptions Equations References

ρ N IR −ρ RED
NDVI Normalized Vegetation Index NDV I = ρ N IR +ρ RED
[42]
ρ N IR
RVI Ratio Vegetation Index RV I = ρRED [43]
DVI Difference Vegetation Index DV I = ρ N IR − ρ RED [44]
EVI Enhanced Vegetation Index 2.5×(ρ −ρ RED ) [45]
EV I = (ρ +6×ρ N IR −7.5×ρ
N IR RED BLUE +1)
ρ N IR −ρGREEN
GNDVI Green Normalized Vegetation Index GNDV I = ρ N IR +ρGREEN
[46]
ρ −ρ N IR
NDRE Normalized Redside Vegetation Index NDRE = ρRedEdge
RedEdge + ρ N IR
[47]
SAVI Soil-Adjusted Vegetation Index (ρ −ρ ) [48]
SAV I = 1.5 × (ρ N+IRρ RED
N IR RED +0.5)

2.4. Predictive Model

Forest AGC was estimated using combinations of three types of data variables: LiDAR
variables (LVs), optical variables (OVs), and a combination of LiDAR and optical variables
(LVs + OVs). Predictive variables from LiDAR data (2 variables) and vegetation index
data (7 variables) were used along with corresponding AGC field data to estimate the
forest AGC of L. gmelinii and B. platyphylla. The models were run with the measured forest
AGC and remotely sensed data-derived indices as dependent and independent variables,
respectively. The impact of different modeling methods on result quality varies [49]. In this
study, two types of multiple stepwise linear regression (MSLR) models and random forest
(RF) regression models were used to predict AGCs.

2.4.1. Multiple Stepwise Linear Regression Model

A multiple linear regression model using a stepwise selection of predictive variables
was employed to predict the relationship between AGCs obtained from remote sensing
datasets and variables. Stepwise regression is a parametric algorithm that screens variables
and establishes an optimal regression equation. In the stepwise regression modeling pro-
cess, predictive variables are input into the regression equation one by one based on given
statistical standards. In each step of the analysis, the variable with the highest correlation
with the dependent variable is first entered into the regression equation, followed by the
introduction of other variables one by one to establish the model. The MSLR model has
been widely applied in the estimation of forest AGB and AGCs [50–53].
However, these techniques are not conducive to establishing complex relationships
between biophysical parameters and remote sensing matrices. Machine learning technology
is a powerful nonlinear regression method that can serve as an alternative to traditional
methods for dealing with complex and nonlinear problems. Machine learning algorithms
can integrate data from different sources [26].

2.4.2. Random Forest Regression Model

The random forest (RF) method is a classification and regression method that uses
regression trees to predict the relationships between variables and is widely used in biomass
carbon storage prediction [54,55]. It is often effective in predictive models [56]. In this
Remote Sens. 2024, 16, 3927 10 of 23

algorithm, a subset of training data is randomly selected to construct a decision tree. The
remaining part of the training data is used to estimate the error of each tree. At each node of
the tree, a set of predictive variables is randomly selected to determine the split. Hundreds of
trees are constructed in a similar manner, and the final prediction is formed by aggregating the
predicted values of all the trees [19]. RF has been proven to reduce overfitting and systematic
errors [57], rank important variables, and generate independent measurements of prediction
error [58]. It has been shown to be more accurate than linear regression models [59].
In this study, the Random Forest (RF) model utilized data from 210 Xing’an larch
and 180 white birch trees, selecting one-third of the total number of variables (70 for
larch and 60 for birch) as the basis for splitting at each node. The input data consisted
of structural attributes derived from LiDAR, such as diameter at breast height (DBH)
and tree height (H), along with vegetation indices extracted from UAV multispectral
imagery, including the Ratio Vegetation Index (RVI) and Green Normalized Difference
Vegetation Index (GNDVI). The model was trained and validated using a 70/30 split
ratio, with cross-validation employed to ensure the robustness and stability of the model’s
performance. Model accuracy was evaluated through multiple metrics, including the
root mean square error (RMSE), coefficient of determination (R2 ), and mean absolute
error (MAE). Furthermore, the model assessed the relative importance of each variable by
measuring the increase in prediction error when a particular variable was omitted, which
facilitated model optimization and highlighted the most influential predictors.

2.5. Accuracy Evaluation

To validate the effectiveness of the models, in this study, we randomly selected
160 plots from 230 field-measured L. gmelinii carbon storage plots and 144 plots from
205 field-measured B. platyphylla carbon storage plots to construct regression models. The
remaining 70 and 61 plots were used to evaluate the predictive accuracy of our established
forest AGC estimation models. Widely used statistical indicators were employed to assess
the accuracy of the forest AGC estimation models. For each model, we calculated the
coefficient of determination (R2 , Formula (6)), root mean square error (RMSE, Formula (7)),
relative root mean square error (RRMSE, Formula (8)), and mean absolute error (MAE,
Formula (9)). R2 measures the fit between the predicted and observed values, while RMSE,
RRMSE, and MAE calculate the estimation error of the model. A larger R2 indicates a better
fit between the observed and predicted values. Smaller RMSE, RRMSE, and MAE values
indicate smaller estimation errors in the model. The calculation formulas are as follows:
∧
∑in=1 (yi − yi )2
R2 = 1 − 2
(6)
∑in=1 (yi − yi )
s
∧
∑in=1 (yi − yi )2
RMSE = (7)
n
RMSE
RRMSE = × 100 (8)
y
∧
∑in=1 yi − yi
MAE = (9)
n
∧
In the formulas, n represents the number of samples; yi and yi are the observed and
predicted AGC values, respectively, for the i-th sample; and yi is the average AGC value of
the i-th sample.

3. Results
3.1. Extraction of Forest Structural Parameters Based on Remote Sensing Data
First, we conducted sampling at the field site, where each tree within the plot was
scanned using backpack LiDAR. The data were then exported and processed in LiDAR360
 3.
3. Results
Results
3.1.
3.1. Extraction
Extraction of
of Forest
Forest Structural
Structural Parameters
Parameters Based
Based on
on Remote
Remote Sensing
Sensing Data
Data
Remote Sens. 2024, 16, 3927 11 of 23
First,
First, we
we conducted
conducted sampling
sampling at at the
the field
field site,
site, where
where each
each tree
tree within
within the the plot
plot was
was
scanned
scanned using
using backpack
backpack LiDAR.
LiDAR. The The data
data were
were then
then exported
exported andand processed
processed in in LiDAR360
LiDAR360
software,
software, where the point cloud data underwent cropping, filtering, ground point
where the point cloud data underwent cropping, filtering, ground point classi-
classi-
software, where the point
fication, cloud data underwent cropping, filtering, ground pointpro- classifi-
fication, normalization,
normalization, and and individual
individual tree
tree segmentation.
segmentation. This This process
process ultimately
ultimately pro-
cation,the
vided normalization,
DBH and H of and
trees individual
with treecoordinates.
specific segmentation. Thisthe
Finally, process
DBH ultimately
and H provided
measured
vided the DBH and H of trees with specific coordinates. Finally, the DBH and H measured
the DBH
in and Hfitted
of trees with specific coordinates. Finally, thefrom
DBH and H measured in the
in the
the field
field were
were fitted and and cross-validated
cross-validated withwith those
those extracted
extracted from the the backpack
backpack LiDAR
LiDAR
field were fitted and cross-validated with those extracted from the backpack LiDAR data.
data.
data.
Figure7a
Figure
Figure 7ashows
7a showsthe
shows thethecomparison
comparison
comparison between
between
between thethe
the measured
measured
measured DBH
DBH DBHandand
and the the DBH
the DBH
DBH extracted
extracted
extracted
from the Backpack LiDAR for the Larix gmelinii forest plot 2(R 2 = 0.9966, RMSE = 0.25, and
from the Backpack LiDAR for the Larix gmelinii forest plot (R
from the Backpack LiDAR for the Larix gmelinii forest plot (R2 = 0.9966, RMSE = 0.25, and
= 0.9966, RMSE = 0.25, and
MAE
MAE = 0.17). Figure 7b presents the comparison between the
MAE = 0.17). Figure 7b presents the comparison between the measured H and the H ex-
= 0.17). Figure 7b presents the comparison between the measured
measured H andH and
the H the
ex-H ex-
tracted 22 == 0.4507,
tracted from
from the Backpack LiDAR for the Larix
Larix gmelinii
gmelinii forest
forest plot
plot (R
tracted from the Backpack LiDAR for the Larix gmelinii forest plot (R = 0.4507, RMSE === 1.48,
(R 2 0.4507, RMSE
RMSE
1.48, and
and MAE
and MAE
1.48, == 3.15).
= 3.15).
MAE 3.15).

(a)
(a) (b)
(b)
Figure 7. (a) Comparison
Figure7.7.(a)(a)
Figure of
Comparison
Comparison measured DBH
of measured
of measured with
DBHDBH DBH
with with extracted
DBH DBH by
by backpack
extracted
extracted LiDAR
LiDAR in
by backpack
backpack Larix;
LiDAR
in (b)
Larix; in
(b)Larix;
Comparison
Comparison of
of measured
measured DBH
DBH with
with DBH
DBH extracted
extracted by
by backpack
backpack LiDAR
LiDAR in
in Larix.
Larix.
(b) Comparison of measured DBH with DBH extracted by backpack LiDAR in Larix.

Figure
Figure8a
Figure 8ashows
8a showsthe
shows thecomparison
the comparison
comparison between
between
between the measured
thethe measured
measured DBH
DBH and
DBH
andand the
the DBH
the DBH
DBH extracted
extracted
extracted
from
from the
from the Backpack
the Backpack
BackpackLiDARLiDAR
LiDARforfor the
forthe Betula
theBetula
Betulaplatyphylla
platyphylla forest
forest
platyphylla plot (R
plotplot
forest
2 = 0.984, RMSE = 0.51,
(R2 (R 2 = 0.984,
= 0.984, RMSE = 0.51,
RMSE = 0.51,
and
and MAE
MAE == 0.37).
0.37). Figure
Figure 8b
8b presents
presents the
the comparison
comparison between
between the
the measured
measured H
H and
and the
the H
H
and MAE = 0.37). Figure 8b presents the comparison between the measured H and the
extracted
extracted from
from the
the Backpack
Backpack LiDAR
LiDAR for
for the Betula
thefor
Betula platyphylla forest
forest plot (R
(R2 == 0.6227,
2 RMSE
2 = 0.6227,
H extracted from the Backpack LiDAR theplatyphylla
Betula platyphyllaplot forest 0.6227,
plot (RRMSE
== 1.51,
1.51, and
and MAE
MAE == 1.61).
1.61).
RMSE = 1.51, and MAE = 1.61).

(a)
(a) (b)
(b)
Figure
Figure 8. (a) Comparison of measured DBH with DBH extracted by backpack LiDAR
LiDAR in Betula;
Betula;in(b)
Figure8.8.(a)(a)
Comparison
Comparisonof measured DBHDBH
of measured with with
DBH DBH
extracted by backpack
extracted by backpack in
LiDAR (b)
Betula;
Comparison
Comparison of
of measured
measured DBH
DBH with
with DBH
DBH extracted
extracted by
by backpack
backpack LiDAR
LiDAR in
in Betula.
Betula.
(b) Comparison of measured DBH with DBH extracted by backpack LiDAR in Betula.

3.2. Aboveground Carbon Stock Estimation in Forests at the Sample Plot Scale Based on LiDAR
Remote Sensing Data
3.2.1. Aboveground Carbon Stock Estimation in Forests Based on LiDAR Remote Sensing
Using Multiple Stepwise Linear Regression Methods
The modeling approach used a simple regression fitting method to model LiDAR-
estimated DBH and H with forest AGC. We used model accuracy evaluation metrics (such
as R2 and RMSE) to test the significance of the model to eliminate variables that have
an insignificant impact on the dependent variable. Finally, we established a multivariate
Remote Sens. 2024, 16, 3927 12 of 23

linear regression equation containing all important variables to explain the variation in the
dependent variable. Therefore, in this study, using IBM SPSS Statistics 27 statistical analysis
software, we selected the multivariate linear stepwise regression method based on the
correlation between the biomasses of different types of tree species and modeling factors.
We retained factors with strong significance and eliminated those with weak significance
until we formed the optimal model equation for estimating aboveground carbon storage
for different tree species. The DBH and H of each tree extracted from the backpack LiDAR
were used as independent variables (a Pearson correlation analysis showed a significant
correlation between DBH and H, with R = 0.7, p < 0.05), and the measured aboveground
carbon storage as the dependent variable. Various linear and nonlinear fitting models were
applied using Excel to explore the relationship between forest structural variables (such as
DBH and H) and predicted AGC. The best-fitting model was then selected from among
these models (Tables 4 and 5).

Table 4. Multivariate stepwise linear regression modeling for LiDAR prediction of AGC in Larix.

Variable Model R2 RMSE

DBHLiDAR AGCLiDAR = 0.0962 × DBHLiDAR 2.1023 0.937 4.19
DBHLiDAR AGCLiDAR = 3.7302 × DBHLiDAR − 25.612 0.882 5.94
HLiDAR AGCLiDAR = 2.9865 × HLiDAR 1.0256 0.244 14.28
HLiDAR AGCLiDAR = 5.6161 × HLiDAR − 0.9979 0.230 14.63
DBHLiDAR × HLiDAR AGCLiDAR = 0.1272 (DBHLiDAR × HLiDAR )1.1291 0.770 11.20
DBHLiDAR × HLiDAR AGCLiDAR = 0.2634 (DBHLiDAR × HLiDAR ) − 2.2132 0.737 11.97
DBHLiDAR + HLiDAR AGCLiDAR = 0.0124 (DBHLiDAR + HLiDAR )2.4667 0.904 7.60
DBHLiDAR + HLiDAR AGCLiDAR = 2.7801 (DBHLiDAR + HLiDAR ) − 33.604 0.847 11.39

Table 5. Multivariate stepwise linear regression modeling for LiDAR prediction of AGC in Betula.

Variable Model R2 RMSE

DBHLiDAR AGCLiDAR = 0.1724 × DBHLiDAR 1.9684 0.948 3.71
DBHLiDAR AGCLiDAR = 4.4728 × DBHLiDAR − 28.31 0.939 4.56
HLiDAR AGCLiDAR = 0.0391 × HLiDAR 2.534 0.411 17.26
HLiDAR AGCLiDAR = 5.3764 × HLiDAR − 39.113 0.281 17.00
DBHLiDAR × HLiDAR AGCLiDAR = 0.0151 (DBHLiDAR × HLiDAR )1.4633 0.942 5.74
DBHLiDAR × HLiDAR AGCLiDAR = 0.2659 (DBHLiDAR × HLiDAR ) − 15.195 0.909 6.65
DBHLiDAR + HLiDAR AGCLiDAR = 0.002 (DBHLiDAR + HLiDAR )2.9239 0.938 5.47
DBHLiDAR + HLiDAR AGCLiDAR = 3.3565 (DBHLiDAR + HLiDAR ) − 56.464 0.880 7.41

In this study, simple stepwise regression fitting methods were used to establish various
linear and nonlinear models that relate LiDAR-estimated forest structural variables such as
tree height (H) and diameter at breast height (DBH) with the measured forest AGC. The
model with the highest correlation coefficient was selected from multiple fitted models. As
shown in Tables 4 and 5, among all the trained multivariate linear and nonlinear models,
the multivariate power model exhibited higher accuracy than the multivariate linear model.
In L. gmelinii forests, the LiDAR-estimated DBH provided the best fit for predicting L.
gmelinii AGC (R2 = 0.9371). Similarly, in B. platyphylla forests, the LiDAR-estimated DBH
was the best predictor for B. platyphylla AGC (R2 = 0.9482). Therefore, in this study, a
power function regression model was used, with LiDAR-estimated DBH as the predictor, to
simulate and predict AGC in L. gmelinii and B. platyphylla forests, respectively. The optimal
prediction regression models are shown in Equations (10) and (11):

AGCLarix_LiDAR = 0.0962 × DBHLiDAR 2.1023 (10)

AGCBetula_LiDAR = 0.1724 × DBHLiDAR 1.9684 (11)

Using the aforementioned formulas, the AGC of L. gmelinii and B. platyphylla forests
was predicted using LiDAR data. Figure 9 shows the results of the accuracy validation of
 2.1023
AGCLarix_ LiDAR = 0.0962× DBHLiDAR (10)

1.9684
AGCBetula_ LiDAR = 0.1724× DBHLiDAR (11)
Remote Sens. 2024, 16, 3927 13 of 23
Using the aforementioned formulas, the AGC of L. gmelinii and B. platyphylla forests
was predicted using LiDAR data. Figure 9 shows the results of the accuracy validation of
the AGC predicted by the LiDAR MSLR model compared with the AGC estimated in the
field in
in L.
L. gmelinii
gmeliniiand
andB.
B.platyphylla
platyphyllaforests.
forests.As
Asdemonstrated,
demonstrated,the
theAGC
AGC ofof
L. L. gmelinii
gmelinii andand
B.
B. platyphylla
platyphylla predicted
predicted byby LiDAR
LiDAR usingthe
using theMSLR
MSLRmodel
modelwas
wassignificantly
significantly correlated
correlated with
measured AGC, with
the measured with R22 values of 0.91 and 0.94, RMSE values of 6.21 and 5.20, RRMSE
respectively. Among the
values of 19.36% and 18.37%, and MAE values of 3.83 and 3.32, respectively.
different tree species, the R22, RMSE, RRMSE, and MAE values were greatest for the B.
two different
platyphylla forests.

Remote Sens. 2024, 16, x FOR PEER REVIEW 15 of 25

Figure 9. Forest AGC(a) (Tg) of (a) Larix and (b) Betula measured via (b)
LiDAR using the MSLR model
versus the predicted forest AGC. (Blue dots indicate data sample points for Larix; red dots indicate
Figure 9. Forest
data sample AGC
points for(Tg) of (a) Larix and (b) Betula measured via LiDAR using the MSLR model
Betula).
versus the predicted forest AGC. (Blue dots indicate data sample points for Larix; red dots indicate
data
3.2.2.sample points for Carbon
Aboveground Betula). Stock Estimation in Forests Based on LiDAR Remote
Sensing
3.2.2. Using the Random
Aboveground Carbon Forest RegressioninMethod
Stock Estimation Forests Based on LiDAR Remote Sensing
UsingInthe Random
this Forest
study, 186 Regression
L. gmelinii treesMethod
and 164 B. platyphylla trees were randomly selected
for RF modeling.
In this Figure
study, 186 10 shows
L. gmelinii treesthe
and results
164 B.of the accuracy
platyphylla trees validation
were randomly of theselected
AGC pre-
for
dicted by the LiDAR RF model compared with the AGC estimated in the
RF modeling. Figure 10 shows the results of the accuracy validation of the AGC predicted field in L. gmelinii
and
by B. LiDAR
the platyphylla
RF forests. As illustrated,
model compared withthetheAGCs
AGC of L. gmelinii
estimated inand
the B. platyphylla
field predicted
in L. gmelinii and
byplatyphylla
B. LiDAR viaforests.
the RF As
model were significantly
illustrated, the AGCs of correlated
L. gmeliniiwith
andtheB. measured
platyphyllaAGC, with
predicted
R2 LiDAR
by values of 0.9501
via the RFand 0.9618,
model wereRMSE values of correlated
significantly 4.4023 andwith3.54,the
RRMSE values
measured of 13.61%
AGC, with
2 values
Rand 14.03%, and MAE
of 0.9501 values of
and 0.9618, 3.2041
RMSE and of
values 2.79, respectively.
4.4023 and 3.54, Among
RRMSE the twoofdifferent
values 13.61%
tree 14.03%,
and species,and
the MAE
R2, RMSE,
valuesRRMSE,
of 3.2041andandMAE values were Among
2.79, respectively. greatestthe fortwo
the different
B. platyphylla
tree
forests. the
species, R2 , RMSE,
Overall, the fitting
RRMSE, models basedvalues
and MAE on RFwere
performed
greatest better
for thethan those based
B. platyphylla on
forests.
MSLR. the fitting models based on RF performed better than those based on MSLR.
Overall,

(a) (b)
Figure10.
Figure 10.Forest
ForestAGC
AGC(Tg)
(Tg)
of of
(a) (a) Larix
Larix andand (b) Betula
(b) Betula measured
measured by LiDAR
by LiDAR using
using the RF the RF versus
model model
versus the predicted forest AGC. (Blue dots indicate data sample points for Larix; red dots indicate
the predicted forest AGC. (Blue dots indicate data sample points for Larix; red dots indicate data
data sample points for Betula).
sample points for Betula).

3.3. Estimation of Aboveground Carbon Stocks in Forests at the Sample Site Scale Based on
Multisource Remote Sensing Data
3.3.1. Aboveground Carbon Stock Estimation in Forests Based on Multisource Remote
Sensing Using Multiple Stepwise Linear Regression Methods
Considering the shortcomings of LiDAR data, which are spatially discrete and do not
have imaging capability, multispectral information was combined to improve the accu-
racy of forest AGC estimation on the basis of obtaining LiDAR tree height and diameter
Remote Sens. 2024, 16, 3927 14 of 23

3.3. Estimation of Aboveground Carbon Stocks in Forests at the Sample Site Scale Based on
Multisource Remote Sensing Data
3.3.1. Aboveground Carbon Stock Estimation in Forests Based on Multisource Remote
Sensing Using Multiple Stepwise Linear Regression Methods
Considering the shortcomings of LiDAR data, which are spatially discrete and do not
have imaging capability, multispectral information was combined to improve the accuracy
of forest AGC estimation on the basis of obtaining LiDAR tree height and diameter at breast
height (DBH). In this study, based on previous estimates of forest AGC using LiDAR data,
the inversion model was optimized by combining multispectral imagery to complement the
spectral information of vegetation. First, the UAV multispectral orthorectified image was
opened, and the required individual band information (red, near-infrared, blue, green, etc.)
was extracted in ArcGIS 10.7. Second, seven vegetation indices, such as the NDVI, DVI,
EVI, RVI, NDRE, GNDVI, and SAVI, were calculated using the above band information,
and the results of each index were saved as separate GeoTIFF files. Finally, a Pearson
correlation analysis was conducted using IBM SPSS Statistics 27 software to assess the
relationship between the seven vegetation indices and the measured AGC (as shown in
Table 6). This analysis was used to identify the optimal vegetation indices that could
enhance the accuracy of AGC estimation. As known from Table 3, the seven selected
vegetation indices were weakly correlated with the measured AGC. The vegetation index
with the highest correlation with forest AGC in larch forests was RVI, while the index with
the highest correlation with forest AGC in birch forests was GNDVI.

Table 6. Vegetation index and measured AGC—Pearson correlation.

Correlation
Vegetation Index
Larix Betula
NDVI 0.0397 0.1273
DVI 0.0417 0.1162
EVI 0.0407 0.0903
RVI 0.0426 0.1328
NDRE 0.0413 0.1329
GNDVI 0.0367 0.1614
SAVI 0.0235 0.0861

Through the analysis of the correlation between AGC and modeling factors in the
sample data of larch and birch forests, we found a weak correlation between various
modeling factors and AGC. We hypothesize that incorporating these vegetation indices
could improve the accuracy of AGC predictions. Therefore, in this study, we combined
multispectral indices with LiDAR parameters to develop models. For L. gmelinii, DBH and
H extracted from LiDAR data, along with the vegetation index RVI calculated from UAV
multispectral imagery, were used as independent variables, with the measured AGC as
the dependent variable. For B. platyphylla, DBH, H, and the vegetation index GNDVI were
used as independent variables, with the measured AGC as the dependent variable. Various
linear and nonlinear fitting models were applied using Excel to explore the relationship
between these forest structure variables and the predicted AGC. The best-fitting models
were selected from multiple models, and predictive AGC models under multi-source
remote sensing were established using various linear and nonlinear equations, with the
optimal model chosen for AGC prediction (Tables 7 and 8). As indicated in the following
two tables, similar to the regression models predicting AGC with LiDAR, multivariate
power models exhibit higher accuracy than multivariate linear models among all the
models. The use of the RVI alone for predicting AGC had poor effectiveness (R2 = 0.043,
RMSE = 29.02). However, incorporating LiDAR parameters such as DBH and tree height
significantly improves the prediction performance and modeling accuracy. In L. gmelinii
forests, the AGC prediction model with the highest accuracy was constructed via simple
power function fitting of the RVI combined with the LiDAR DBH (R2 = 0.939, RMSE = 5.06)
Remote Sens. 2024, 16, 3927 15 of 23

(Table 7). Similarly, in B. platyphylla forests, the highest accuracy was achieved using a
simple power function fitting of the GNDVI combined with the LiDAR DBH (R2 = 0.942,
RMSE = 3.67) (Table 8). These results are more accurate than those of models fitted with
only LiDAR data, significantly enhancing the precision of AGC fitting. This outcome
reflects that while vegetation indices can provide some information about the horizontal
structure of vegetation cover, they lack information on the vertical structure of vegetation
height and are prone to saturation. However, incorporating LiDAR parameters not only
overcomes saturation issues but also enhances the accuracy of AGC estimation.

Table 7. Multivariate stepwise linear regression modeling for multisource remote sensing prediction
of AGC in Larix.

Variable Model R2 RMSE

RVI AGCRVI = 13.802 × RVI − 12.697 0.043 29.02
RVI + HLiDAR AGCRVI+H(LiDAR) = 3.5266 × (RVI + HLiDAR ) − 11.54 0.394 27.21
RVI × HLiDAR AGCRVI×H(LiDAR) = 1.0189 × (RVI × HLiDAR ) + 1.6993 0.386 14.48
RVI × DBHLiDAR AGCRVI×DBH(LiDAR) = 1.268 × (RVI × DBHLiDAR ) − 30.04 0.889 6.16
RVI × DBHLiDAR AGCRVI×DBH(LiDAR) = 0.0106 × (RVI × DBHLiDAR )2.039 0.924 5.74
RVI + DBHLiDAR AGCRVI+DBH(LiDAR) = 4.3185 × (RVI + DBHLiDAR ) − 46.748 0.913 5.37
RVI + DBHLiDAR AGCRVI+DBH(LiDAR) = 0.0144 (RVI + DBHLiDAR )2.6197 0.939 5.06
RVI + (DBH + H) LiDAR AGCRVI+(DBH+H) LiDAR = 0.0102 [RVI + (DBH + H) LiDAR ]2.4015 0.863 6.80
RVI × (DBH + H) LiDAR AGCRVI×(DBH+H) LiDAR = 0.006 [RVI × (DBH + H) LiDAR ]1.9538 0.818 8.55
RVI + (DBH × H) LiDAR AGCRVI+(DBH×H) LiDAR = 0.2316 [RVI + (DBH × H) LiDAR ]0.9215 0.774 8.37
RVI × (DBH × H) LiDAR AGCRVI×(DBH×H) LiDAR = 0.1075 [RVI × (DBH × H) LiDAR ]2.0231 0.759 8.94

Table 8. Multivariate stepwise linear regression modeling for multisource remote sensing prediction
of AGC in Betula.

Variable Model R2 RMSE

GNDVI + HLiDAR AGCGNDVI+H(LiDAR) = 3.6502 × (GNDVI + HLiDAR ) − 21.84 0.191 16.41
GNDVI × HLiDAR AGCGNDVI×H(LiDAR) = 5.9133 × (GNDVI × HLiDAR ) − 11.63 0.150 16.82
GNDVI × DBHLiDAR AGCGNDVI×DBH(LiDAR) = 8.4868 × (GNDVI × DBHLiDAR ) − 26.025 0.903 5.67
GNDVI × DBHLiDAR AGCGNDVI×DBH(LiDAR) = 0.7576 × (GNDVI × DBHLiDAR )1.9001 0.919 4.95
GNDVI + DBHLiDAR AGCGNDVI+DBH(LiDAR) = 4.4075 × (GNDVI + DBHLiDAR ) − 29.202 0.933 4.49
GNDVI + DBHLiDAR AGCGNDVI+DBH(LiDAR) = 0.1424 (GNDVI + DBHLiDAR )2.0231 0.942 3.67
GNDVI + (DBH + H) LiDAR AGCGNDVI+(DBH+H) LiDAR = 0.0017 [GNDVI +(DBH + H) LiDAR ]2.9292 0.911 5.32
GNDVI × (DBH + H) LiDAR AGCGNDVI×(DBH+H) LiDAR = 0.0248 [GNDVI ×(DBH + H) LiDAR ]2.6788 0.845 0.76
GNDVI + (DBH × H) LiDAR AGCGNDVI+(DBH×H) LiDAR = 0.0166 [GNDVI + (DBH × H) LiDAR ]1.4349 0.923 5.84
GNDVI × (DBH × H) LiDAR AGCGNDVI×(DBH×H) LiDAR = 0.0495 [GNDVI × (DBH × H) LiDAR ]1.4068 0.905 6.80

This study employs a combination of the multispectral vegetation indices RVI and
GNDVI with LiDAR data for modeling, constructs models using various linear and nonlin-
ear equations, and selects the best models for predicting forest AGCs. In L. gmelinii forests,
the combination of multisource remote sensing-estimated DBH + RVI (power function)
yielded the best fit for predicting AGCs, with an R2 value of 0.939. In B. platyphylla forests,
the combination of DBH + GNDVI (power function) estimated through multisource remote
sensing provides the best fit for predicting AGCs, with an R2 value of 0.942.

AGCLarix_LiDARδmultispectral = 0.0144 × ( DBHLiDAR + RV Imultispectral )2.6197 (12)

AGCBetula_LiDARδmultispectral = 0.1424 × ( DBHLiDAR + GNDV Imultispectral )2.0231 (13)

Using Equations (12) and (13), the AGC of L. gmelinii and B. platyphylla forests is
predicted through multisource remote sensing. Figure 11 shows the results of the accuracy
validation of the AGC predicted by the multisource remote sensing MSLR model compared
with the AGC estimated in the field in L. gmelinii and B. platyphylla forests. As demonstrated,
 AGC Betula _ LiDAR δmultispect ral = 0 .1424 × ( DBH LiDAR + GNDVI multispect ral ) 2.0231 (13)

Using Equations (12) and (13), the AGC of L. gmelinii and B. platyphylla forests is pre-
dicted through multisource remote sensing. Figure 11 shows the results of the accuracy
Remote Sens. 2024, 16, 3927 validation of the AGC predicted by the multisource remote sensing MSLR model16com- of 23
pared with the AGC estimated in the field in L. gmelinii and B. platyphylla forests. As
demonstrated, the AGC of L. gmelinii and B. platyphylla predicted by multisource remote
the
sensing of L. gmelinii
AGCusing the MSLR B. platyphylla
andmodel predicted correlated
was significantly by multisource
with remote sensingAGC,
the measured usingwith
the
MSLR model
R2 values was and
of 0.92 significantly
0.96, RMSEcorrelated
values with theand
of 6.15 measured AGC, with
3.57, RRMSE R2 of
values values
19.06%of 0.92
and
and 0.96,and
12.44%, RMSE MAE values of of
values 6.15 and
4.41 3.57,
and RRMSE
2.70, valuesAmong
respectively. of 19.06%theand
two12.44%,
differentand
treeMAE
spe-
values ofR4.41
cies, the and 2.70,
2, RMSE, respectively.
RRMSE, and MAE Among
valuesthe two
were different
greatest fortree species, the
B. platyphylla R2 , RMSE,
forests. Com-
RRMSE,
pared toand MAE predicted
the AGC values were greatest
solely with for B. platyphylla
LiDAR, forests.
the accuracy Compared
improved to the
by 0.01 andAGC0.03,
predicted solely
respectively. with LiDAR, the accuracy improved by 0.01 and 0.03, respectively.

Remote Sens. 2024, 16, x FOR PEER REVIEW 18 of 25

(a) (b)
Figure 11. Forest AGC (Tg) of (a) Larix and (b) Betula measured by multisource remote sensing
Figure 11. Forest AGC (Tg) of (a) Larix and (b) Betula measured by multisource remote sensing using
using the MSLR model versus the predicted forest AGC. (Blue dots indicate data sample points for
the MSLR
Larix; red model versus data
dots indicate the predicted forestfor
sample points AGC. (Blue dots indicate data sample points for Larix;
Betula).
red dots indicate data sample points for Betula).
3.3.2. Aboveground
3.3.2. Aboveground Carbon Carbon Stock
Stock Estimation
Estimation in in Forests
Forests Based
Based ononMultisource
MultisourceRemote
Remote
Sensing Using a Random Forest Regression
Sensing Using a Random Forest Regression Approach Approach
Figure 12
Figure 12 shows
shows thethe results
results ofof the
theaccuracy
accuracy validation
validation ofof the
the AGC
AGC predicted
predicted by by the
the
multisourceremote
multisource remotesensing
sensingRF RFmodel
model compared
comparedwith with the
the AGC
AGC estimated
estimated in
in the
the field
field in
in L.
L.
gmeliniiand
gmelinii andB. B. platyphylla
platyphylla forests.
forests. The
The results
results indicate
indicate aa significant
significant correlation
correlation between
between the the
predictedAGC
predicted AGC viavia multisource
multisource remote
remote sensing
sensing via
via the
the RF
RF model
model andand the
the measured
measuredAGC AGC
for both tree species.
for both tree species. The R The R22 values for L. gmelinii and B. platyphylla were both 0.95, the
L. gmelinii and B. platyphylla were both 0.95, the
RMSEvalues
RMSE valueswere
were3.993.99 and
and 3.45,
3.45, thethe
RRMSERRMSE values
values werewere 12.49%
12.49% and 13.83%,
and 13.83%, and the and
MAEthe
MAE values
values were
were 3.10 and3.10 and
2.78. 2.78. Among
Among these two these two different
different tree the
tree species, 2
species, the R RRMSE,
R , RMSE, 2 , RMSE,
RRMSE,
and MAEand MAE
values values
were wereingreatest
greatest in B. platyphylla
B. platyphylla forests. Moreover,
forests. Moreover, comparedcompared
to the AGC to
the AGC predicted
predicted solely withsolely
LiDAR,with LiDAR,
there was anthere was an improvement
improvement in accuracyinofaccuracy of gmelinii
0.02 for L. 0.02 for
L. gmelinii
and 0.01 forand 0.01 for B. platyphylla.
B. platyphylla.

(a) (b)
Figure12.
Figure 12.Forest
ForestAGC
AGC(Tg)
(Tg)
of of
(a)(a) Larix
Larix andand (b) Betula
(b) Betula measured
measured by multisource
by multisource remoteremote
sensingsensing
using
using the RF model versus the predicted forest AGC. (Blue dots indicate data sample
the RF model versus the predicted forest AGC. (Blue dots indicate data sample points for Larix;pointsred
for
Larix; red dots indicate data sample points
dots indicate data sample points for Betula). for Betula).

3.4. Comparative Analysis of Multiple Stepwise Linear Regression and Random Forest
Regression Models Based on Different Remote Sensing Methods
This study utilized various remote sensing data and employed both MSLR and RF
regression models to predict the AGC of L. gmelinii and B. platyphylla forests. Figures 13
and 14 display comparisons between the prediction results of these two models under the
different remote sensing data and the actual observation results for AGC. Overall, the R2
values show an increasing trend from left to right, approaching 1, indicating progressively
 Remote Sens. 2024, 16, 3927 17 of 23

3.4. Comparative Analysis of Multiple Stepwise Linear Regression and Random Forest Regression
Models Based on Different Remote Sensing Methods
This study utilized various remote sensing data and employed both MSLR and RF re-
gression models to predict the AGC of L. gmelinii and B. platyphylla forests. Figures 13 and 14
display comparisons between the prediction results of these two models under the different
remote sensing data and the actual observation results for AGC. Overall, the R2 values
show an increasing trend from left to right, approaching 1, indicating progressively better
model fitting. The RMSE, RRMSE, and MAE values display a decreasing trend from left
to right, suggesting that the smaller these values are, the less the discrepancy between
the model predictions and observed values, and the better the predictive power of the
model. This overall trend indicates the feasibility of the models. Specifically: (1) Among
both the parametric and nonparametric models, the RF model constructed from machine
learning algorithms demonstrated greater accuracy in estimating AGC for L. gmelinii and
B. platyphylla forests than the MSLR model. (2) When combining vegetation indices from
UAV multispectral images with LiDAR remote sensing data in the MSLR and RF models
for estimating forest AGC, the accuracy surpasses that of using only LiDAR remote sensing
Remote Sens. 2024, 16, x FOR PEER REVIEW 19 of 25
data. (3) In both L. gmelinii and B. platyphylla forests, regardless of whether the MSLR
Remote Sens. 2024, 16, x FOR PEER REVIEW
or RF
19 of 25
model was used, the accuracy of estimating forest AGC was greater for B. platyphylla than
for L. gmelinii.

Figure
Figure13.13.Comparison
Comparison ofofthe
theprediction
predictionresults
resultsofoftwo
twomodels
modelsbased
basedonondifferent
differentremote
remotesensing
sensing
Figure
data 13. Comparison of the prediction results of two models based on different remote sensing
dataononthe
theAGC
AGCofofLarix
Larixwith
withthe
theactual
actualobservation
observationresults.
results.
data on the AGC of Larix with the actual observation results.

Figure 14. Comparison of the prediction results of two models based on different remote sensing
Figure 14. Comparison of the prediction results of two models based on different remote sensing
data on
Figure the AGC of Betula with the actual observation results.
data on14.
theComparison of the
AGC of Betula withprediction
the actualresults of tworesults.
observation models based on different remote sensing
data on the AGC of Betula with the actual observation results.
4. Discussion
4. Discussion
4.1. Potential of LiDAR Combined with Multispectral Imagery for Estimating AGC in Forests
4.1. Potential of LiDAR Combined with Multispectral Imagery for Estimating AGC in Forests
Compared to manual measurements, backpack LiDAR offers precise scanning and
Compared to manual measurements, backpack LiDAR offers precise scanning and
Remote Sens. 2024, 16, 3927 18 of 23

4. Discussion
4.1. Potential of LiDAR Combined with Multispectral Imagery for Estimating AGC in Forests
Compared to manual measurements, backpack LiDAR offers precise scanning and
real-time data integration while in motion, providing a more flexible and efficient method
for forest inventory collection [60]. In the data collection process, backpack LiDAR requires
only one surveyor to carry the equipment across the measurement site, significantly reduc-
ing time and costs and improving efficiency [61]. As shown in Table 9, when collecting
point cloud data for a 10 m × 40 m sample, traditional measurement methods require
3–4 people to complete the data collection, whereas backpack LiDAR needs only one per-
son. While traditional manual measurements take approximately 36 min to measure a plot,
backpack LiDAR takes only approximately 5 min. Preprocessing the collected data via
traditional methods took approximately 14 min, while preprocessing the backpack LiDAR
point cloud data took approximately 10 min. The internal data processing time depends on
the size of the dataset and the computer configuration. Overall, compared with traditional
methods, backpack LiDAR saves approximately 30 min per plot, illustrating its time effi-
ciency. In terms of optical data, acquiring airborne multispectral images under favorable
weather conditions enhances efficiency and reduces costs to a certain extent. Therefore, the
combined use of optical imagery and LiDAR further reduces the cost of assessing forest
emission reductions. This combination enables the mapping of large areas near real-time
carbon stocks [62]. The findings of this study underscore the high precision and potential
of LiDAR technology for estimating AGC, offering significant value for enhancing forest
management practices and informing sustainable ecosystem management strategies [63].
However, scaling up this approach to a broader level may present significant challenges,
particularly in low-income countries where limited financial and technical resources could
hinder its widespread implementation and reduce its overall effectiveness [64].

Table 9. Timing comparison between traditional and Backpack LiDAR measurement methods.

Time Consumption (min)

Measurement Method Personnel Sample Site (m2 )
Data Collection Data Processing Total
Traditional measurement 3–4 10 × 40 30:16 14:16 44:32
Backpack LiDAR 1 10 × 40 5:42 10:04 15:46

Optical images have been applied in earlier studies to estimate forest AGB and AGCs,
but the results showed that optical signals are weakly penetrating. Spectral images mainly
record the horizontal structure of the forest and cannot record the vertical structure in-
formation of the forest. However, LiDAR can penetrate the forest canopy and record
vertical structure information. This approach compensates for the shortcomings of op-
tical images. In this study, there are two main reasons for the small improvement after
adding multispectral information. The first reason may be that when the visible light of
multispectral data is saturated in dense forest areas [65], the accuracy is lower in complex
forest structures, resulting in the deviation of the AGC estimated by the NDVI from the
measured AGC. Another reason for this difference may be that the LiDAR forest structure
attributes themselves have a strong correlation with AGC, and the addition of multispectral
information did not result in much improvement. Overall, although these improvements
are not significant, novel multisensor earth observation methods that involve the combi-
nation of satellite-borne LiDAR data with optical data using machine learning techniques
enable accurate measurements of carbon stocks and provide effective data support for
forest emission reduction. For example, Jiao et al. [66] proposed a practical framework
for assessing forest emission reductions via the fusion of optical satellite imagery and
spaceborne LiDAR data. Shen et al. mapped subtropical forest AGB data by combining
Landsat TM/ETM+ and ALOS l-band SAR imagery from Guangdong Province, and the
results demonstrated that multisensor imagery-based AGBs had a good correlation [67,68].
Remote Sens. 2024, 16, 3927 19 of 23

Our results further suggest that combining LiDAR and multispectral data is essential for
improving the accuracy of AGB and AGC estimation.

4.2. Main Challenges and Uncertainty Analyses for Estimating Forest AGCs
In response to the challenges in estimating vegetation biomass and carbon storage
(specifically, whether obtaining large-scale forest structure and spectral information im-
proves biomass and carbon stock estimations [69]), this study integrates forest structural
attributes and spectral data to estimate forest AGCs at the plot level. Despite the difficulty
in accurately capturing AGC changes in forests with complex structures using only struc-
tural and spectral information, the heterogeneity of canopy spectral information provided
by multispectral images has enhanced the accuracy of our multisource data integration
modeling approach, increasing the AGC estimation accuracy from 90.29% to 90.6%. Ad-
ditionally, we utilized various multiple linear regression and power regression models to
select the best-fitting models for AGC estimation. Compared to multiple linear regression
models, power regression models exhibited greater accuracy in AGC estimation. This
indicates that the dominant tree species in our study area, L. gmelinii, conforms to a power
allometric relationship and that using this relationship can improve the accuracy of forest
AGC estimates. Therefore, the power allometric relationship based on forest structural
attributes and spectral information represents a new method for enhancing AGC estima-
tion. This method can be used to explore the relationship between tree metabolism and
biomass [70], and such relationships may be more stable in similar landscapes [71].
However, there are still uncertainties in this study. First, the laser beams of backpack
LiDAR cannot penetrate the lower canopy layers in dense forest structures; second, due
to obstruction from the understory, backpack LiDAR faces challenges in capturing the
treetops of the upper canopy, resulting in notable differences between the LiDAR-estimated
and actual measured heights. The results and conclusions of this study are currently valid
only for coniferous forests with relatively simple stand structures, and further validation
is needed for broadleaf forests, mixed forests, or other forest types with more complex
structures based on additional forest plots and remote sensing data. In addition, this study
utilized ultra-high-resolution UAV imagery with a spatial resolution of 0.02 m. While such
high spatial detail enables capturing fine-scale variations, it may also introduce significant
spatial variability, particularly in areas with heterogeneous vegetation distribution. This
level of granularity can result in weak correlations between vegetation indices (VI) and
AGC, ultimately impacting the model’s predictive accuracy. Despite the observed low
correlation in our findings, VI still holds considerable promise for capturing ecosystem
dynamics and monitoring environmental changes [72].

4.3. Research and Perspectives on Estimating Late-Season Forest AGCs

This study revolves around the theme of estimating regional-scale forest AGCs by
integrating multispectral imagery and LiDAR data; encompassing a comprehensive and
systematic exploration from field data collection to preprocessing steps such as atmospheric,
radiometric, and geometric correction of multispectral imagery and cropping; resampling,
denoising, filtering, ground classification, and normalization of LiDAR data; constructing
forest AGC estimation models suitable for complex terrain conditions; and then spatially
extrapolating regional-scale forest AGCs. However, due to the scarcity of field measurement
data and the complexity of mountainous terrain, the accuracy of regional forest AGC
estimation combined with multisource remote sensing data is still not precise enough,
warranting further research.
At the current stage, calibration and validation still require high-quality field measure-
ment data. Due to the complex terrain of mountainous areas, more accessible sites were
chosen for field inventory collection, resulting in spatial discontinuity and discreteness in
the regional forest AGC density spatial distribution map. Future research should aim to
select spatially continuous plots for data collection. Limited by time, the collected samples
Remote Sens. 2024, 16, 3927 20 of 23

were insufficient, suitable only for single-tree or regional forest AGC estimation, and not
representative of the entire forest AGC storage in the Dulaer forest.
The backpack LiDAR data collection is affected by poor GPS signals, directly impacting
the quality of trajectory files and leading to failures in point cloud resolution or significant
errors in absolute coordinates. Moreover, obtaining high-precision absolute coordinate
point cloud data is crucial for determining individual tree locations within sample areas.
Therefore, efficiently and accurately collecting absolute geographical reference point cloud
data in dense forests without GPS signals will be a focus of future research. Additionally,
in L. gmelinii plots with dense branches, it was necessary to cut branches in advance
along the designed route to ensure the safe operation of the backpack LiDAR, which
affects the data collection time and quality. Thus, the accuracy of image data collection
via backpack LiDAR needs further verification in more operational environments. The
multispectral data used in this study had limited spectral bands, resulting in less correlation
between the calculated vegetation spectral indices and forest AGC. Future research should
explore the capability of regional forest spectral inversion via hyperspectral imaging via
UAVs at different flight altitudes in conjunction with LiDAR data. To further mitigate
the impact of spatial variability, future research could explore the use of Object-Based
Image Analysis (OBIA) and texture features. These advanced methodologies offer more
stable and structured variables by grouping adjacent pixels into cohesive objects based on
their spectral and morphological similarities, thereby minimizing the variability inherent
in high-resolution data. Additionally, texture features can capture the intricate spatial
patterns and distribution characteristics of vegetation, providing a richer representation of
the landscape and ultimately enhancing the precision of AGC estimation [71,72].
In summary, combining LiDAR data with traditional remote sensing data can comple-
ment data sources better, facilitating the acquisition and classification of ground information
and improving the accuracy of ecological parameter estimation, ecological monitoring, and
simulation. Effectively integrating multisource remote sensing data for ecological research
is currently a trending topic.

5. Conclusions
In this study, LiDAR and multispectral data were effectively integrated to estimate
the AGC of Xing’an larch and white birch forests. The findings highlight the strong
correlation between LiDAR-derived forest structure attributes and the AGC, underscoring
the critical role of LiDAR in carbon monitoring and assessment. Although the relationship
between vegetation indices (VI) and the AGC was comparatively weaker, their potential
value in ecological monitoring and assessment should not be overlooked and warrants
further investigation. This research serves as a valuable reference for future applications of
multi-source remote sensing technologies in forest carbon stock estimation, particularly in
assessing their feasibility and effectiveness under diverse environmental conditions.

Author Contributions: R.S.: Data curation, Formal analysis, Writing—original draft, Writing—review
& editing author; W.D.: Methodology, Project administration, Supervision, Writing—review & editing
author; H.Y.: Investigation, Supervision; Y.S.: Conceptualization; Funding acquisition author; W.R.:
Resources, Supervision; R.L.: Software, Validation. All authors have read and agreed to the published
version of the manuscript.
Funding: This research was supported by several key funding initiatives, including the Science and
Technology Programme of Inner Mongolia Autonomous Region (2024KJHZ0002 and 2022YFSH0027),
the Key Special Project of Inner Mongolia’s “Science and Technology for the Development of Mongo-
lia” Action Plan (2020ZD0028), and the Project for Introducing High-Level Talents of Inner Mongolia
Autonomous Region (2022NMRC003). Additional support was provided by the National Natural
Science Foundation of China (42201374), the Natural Science Foundation of Inner Mongolia Au-
tonomous Region (2022LHQN04001), and the “Integrated Demonstration of Ecological Protection
and Comprehensive Utilization of Resource Technology in Aershan” under the Central Guidance
of Local Science and Technology Development Funds. We also acknowledge the “Introduction of
Remote Sens. 2024, 16, 3927 21 of 23

High-Level Talents Project” and the Master’s Degree Research and Innovation Fund (CXJJS23065) of
Inner Mongolia Normal University (2020YJRC050).
Data Availability Statement: The original contributions presented in the study are included in the
article, further inquiries can be directed to the corresponding author.
Acknowledgments: We are grateful for the fieldwork support from the lnner Mongolia Key Labora-
tory of Remote Sensing and Geographic Information Systems and for the support of the ArshanForest
and Grassland Disaster Prevention and Mitigation Field Scientific Observation and Research Station
of the Inner Mongolia Autonomous Region.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Wani, A.A.; Joshi, P.K.; Singh, O. Estimating Biomass and Carbon Mitigation of Temperate Coniferous Forests Using Spectral
Modeling and Field Inventory Data. Ecol. Inform. 2015, 25, 63–70. [CrossRef]
2. Su, R.; Du, W.; Ying, H.; Shan, Y.; Liu, Y. Estimation of Aboveground Carbon Stocks in Forests Based on LiDAR and Multispectral
Images: A Case Study of Duraer Coniferous Forests. Forests 2023, 14, 992. [CrossRef]
3. Luderer, G.; Vrontisi, Z.; Bertram, C.; Edelenbosch, O.Y.; Pietzcker, R.C.; Rogelj, J.; De Boer, H.S.; Drouet, L.; Emmerling, J.; Fricko,
O.; et al. Residual Fossil CO2 Emissions in 1.5–2 ◦ C Pathways. Nat. Clim. Chang. 2018, 8, 626–633. [CrossRef]
4. Poorazimy, M.; Shataee, S.; McRoberts, R.E.; Mohammadi, J. Integrating Airborne Laser Scanning Data, Space-Borne Radar
Data and Digital Aerial Imagery to Estimate Aboveground Carbon Stock in Hyrcanian Forests, Iran. Remote Sens. Environ. 2020,
240, 111669. [CrossRef]
5. Chen, Y.; Li, L.; Lu, D.; Li, D. Exploring Bamboo Forest Aboveground Biomass Estimation Using Sentinel-2 Data. Remote Sens.
2018, 11, 7. [CrossRef]
6. Nandy, S.; Srinet, R.; Padalia, H. Mapping Forest Height and Aboveground Biomass by Integrating ICESat-2, Sentinel-1
and Sentinel-2 Data Using Random Forest Algorithm in Northwest Himalayan Foothills of India. Geophys. Res. Lett. 2021,
48, e2021GL093799. [CrossRef]
7. Li, W.; Niu, Z.; Shang, R.; Qin, Y.; Wang, L.; Chen, H. High-Resolution Mapping of Forest Canopy Height Using Machine Learning
by Coupling ICESat-2 LiDAR with Sentinel-1, Sentinel-2 and Landsat-8 Data. Int. J. Appl. Earth Obs. Geoinf. 2020, 92, 102163.
[CrossRef]
8. Li, M.; Im, J.; Quackenbush, L.J.; Liu, T. Forest Biomass and Carbon Stock Quantification Using Airborne LiDAR Data: A Case
Study Over Huntington Wildlife Forest in the Adirondack Park. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2014, 7, 3143–3156.
[CrossRef]
9. Luo, S.; Wang, C.; Xi, X.; Pan, F.; Peng, D.; Zou, J.; Nie, S.; Qin, H. Fusion of Airborne LiDAR Data and Hyperspectral Imagery for
Aboveground and Belowground Forest Biomass Estimation. Ecol. Indic. 2017, 73, 378–387. [CrossRef]
10. Chen, Q.; McRoberts, R. Statewide Mapping and Estimation of Vegetation Aboveground Biomass Using Airborne Lidar. In
Proceedings of the 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Beijing, China, 10–15 July
2016; IEEE: Beijing, China, 2016; pp. 4442–4444.
11. Tao, S.; Wu, F.; Guo, Q.; Wang, Y.; Li, W.; Xue, B.; Hu, X.; Li, P.; Tian, D.; Li, C.; et al. Segmenting Tree Crowns from Terrestrial and
Mobile LiDAR Data by Exploring Ecological Theories. ISPRS J. Photogramm. Remote Sens. 2015, 110, 66–76. [CrossRef]
12. Holmgren, J.; Tulldahl, H.M.; Nordlöf, J.; Nyström, M.; Olofsson, K.; Rydell, J.; Willén, E. Estimation of tree position and stem
diameter using simultaneous localization and mapping with data from a backpack-mounted laser scanner. Int. Arch. Photogramm.
Remote Sens. Spat. Inf. Sci. 2017, XLII-3/W3, 59–63. [CrossRef]
13. Oveland, I.; Hauglin, M.; Giannetti, F.; Schipper Kjørsvik, N.; Gobakken, T. Comparing Three Different Ground Based Laser
Scanning Methods for Tree Stem Detection. Remote Sens. 2018, 10, 538. [CrossRef]
14. Su, Y.; Guo, Q.; Fry, D.L.; Collins, B.M.; Kelly, M.; Flanagan, J.P.; Battles, J.J. A Vegetation Mapping Strategy for Conifer Forests by
Combining Airborne LiDAR Data and Aerial Imagery. Can. J. Remote Sens. 2016, 42, 1–15. [CrossRef]
15. Cutler, M.E.J.; Boyd, D.S.; Foody, G.M.; Vetrivel, A. Estimating Tropical Forest Biomass with a Combination of SAR Image Texture
and Landsat TM Data: An Assessment of Predictions between Regions. ISPRS J. Photogramm. Remote Sens. 2012, 70, 66–77.
[CrossRef]
16. Nichol, J.E.; Sarker, M.d.L.R. Improved Biomass Estimation Using the Texture Parameters of Two High-Resolution Optical
Sensors. IEEE Trans. Geosci. Remote Sens. 2011, 49, 930–948. [CrossRef]
17. Reddersen, B.; Fricke, T.; Wachendorf, M. A Multi-Sensor Approach for Predicting Biomass of Extensively Managed Grassland.
Comput. Electron. Agric. 2014, 109, 247–260. [CrossRef]
18. Tilly, N.; Aasen, H.; Bareth, G. Fusion of Plant Height and Vegetation Indices for the Estimation of Barley Biomass. Remote Sens.
2015, 7, 11449–11480. [CrossRef]
19. Ghosh, S.M.; Behera, M.D. Aboveground Biomass Estimation Using Multi-Sensor Data Synergy and Machine Learning Algorithms
in a Dense Tropical Forest. Appl. Geogr. 2018, 96, 29–40. [CrossRef]
Remote Sens. 2024, 16, 3927 22 of 23

20. Verrelst, J.; Rivera, J.P.; Veroustraete, F.; Muñoz-Marí, J.; Clevers, J.G.P.W.; Camps-Valls, G.; Moreno, J. Experimental Sentinel-2
LAI Estimation Using Parametric, Non-Parametric and Physical Retrieval Methods—A Comparison. ISPRS J. Photogramm. Remote
Sens. 2015, 108, 260–272. [CrossRef]
21. Silva, C.; Hudak, A.; Vierling, L.; Klauberg, C.; Garcia, M.; Ferraz, A.; Keller, M.; Eitel, J.; Saatchi, S. Impacts of Airborne Lidar
Pulse Density on Estimating Biomass Stocks and Changes in a Selectively Logged Tropical Forest. Remote Sens. 2017, 9, 1068.
[CrossRef]
22. Brown, S.; Narine, L.L.; Gilbert, J. Using Airborne Lidar, Multispectral Imagery, and Field Inventory Data to Estimate Basal Area,
Volume, and Aboveground Biomass in Heterogeneous Mixed Species Forests: A Case Study in Southern Alabama. Remote Sens.
2022, 14, 2708. [CrossRef]
23. Kim, S.-R.; Kwak, D.-A.; oLee, W.-K.; Son, Y.; Bae, S.-W.; Kim, C.; Yoo, S. Estimation of Carbon Storage Based on Individual Tree
Detection in Pinus Densiflora Stands Using a Fusion of Aerial Photography and LiDAR Data. Sci. China Life Sci. 2010, 53, 885–897.
[CrossRef] [PubMed]
24. Guisan, A.; Edwards, T.C.; Hastie, T. Generalized Linear and Generalized Additive Models in Studies of Species Distributions:
Setting the Scene. Ecol. Model. 2002, 157, 89–100. [CrossRef]
25. Ahmadi, K.; Kalantar, B.; Saeidi, V.; Harandi, E.K.G.; Janizadeh, S.; Ueda, N. Comparison of Machine Learning Methods for
Mapping the Stand Characteristics of Temperate Forests Using Multi-Spectral Sentinel-2 Data. Remote Sens. 2020, 12, 3019.
[CrossRef]
26. Ali, I.; Greifeneder, F.; Stamenkovic, J.; Neumann, M.; Notarnicola, C. Review of Machine Learning Approaches for Biomass and
Soil Moisture Retrievals from Remote Sensing Data. Remote Sens. 2015, 7, 16398–16421. [CrossRef]
27. Mu, C.; Lu, H.; Wang, B.; Cui, W. Short-Term Effects of Harvesting on Carbon Storage of Boreal Larix Gmelinii–Carex Schmidtii
Forested Wetlands in Daxing’anling, Northeast China. For. Ecol. Manag. 2013, 293, 140–148. [CrossRef]
28. Liu, Y. Carbon Density in Boreal Forests Responds Non-Linearly to Temperature: An Example from the Greater Khingan
Mountains, Northeast China. Agric. For. Meteorol. 2023, 338, 109519. [CrossRef]
29. Xu, Y. Current Situation of Forest Resources and Management Countermeasures in Dural Forest. Master’s Thesis, Inner Mongolia
Agricultural University, Hohhot, China, 2022.
30. Fang, C.L.; Chen, Q.; Ren, Y.; Wang, Y.J. Modelling of subtropical forest biomass estimation based on airborne LiDAR. For. Surv.
Plan. 2021, 46, 1–8.
31. Yang, H.; Hu, C.M.; Zhang, L.M.; Li, S.K. Progress in characterising forest carbon sinks in Inner Mongolia. J. Appl. Ecol. 2014, 25,
3366–3372. [CrossRef]
32. Daniel, I.; Rollet, B. Phytomasse Aerienne et Production Primaire Dans La Mangrove Du Grand Cul-De-Sac Marin (Guadeloupe,
Antillas Francaises). Bull. Ecol. 1989, 20, 27–39.
33. Fromard, F.; Puig, H.; Mougin, E.; Marty, G.; Betoulle, J.L.; Cadamuro, L. Structure, above-Ground Biomass and Dynamics of
Mangrove Ecosystems: New Data from French Guiana. Oecologia 1998, 115, 39–53. [CrossRef] [PubMed]
34. Han, A. Forest Biomass and Carbon Stock Remote Sensing Methods. Ph.D. Thesis, Beijing Forestry University, Beijing, China,
2009.
35. He, H. Carbon Sink Capacity of Xing’an Larch Primary Forest and Post-Harvest Restoration Stand. Master’s Thesis, Inner
Mongolia Agricultural University, Hohhot, China, 2009.
36. Ma, Z.; Wang, S.; Wang, C.; Zhao, B.; Zhao, J.; Wu, F. A study on carbon layer delineation of natural Xing’an larch and birch
forests in the DaXing’an Mountains. J. Cent. South For. Univ. Sci. Technol. 2017, 37, 112–117. [CrossRef]
37. Paustian, K.; Ravindranath, N.H.; Amstel, A.V. 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Int. Panel Clim.
Chang. 2006, 2, 335–346. [CrossRef]
38. Lu, D.; Chen, Q.; Wang, G.; Liu, L.; Li, G.; Moran, E. A Survey of Remote Sensing-Based Aboveground Biomass Estimation
Methods in Forest Ecosystems. Int. J. Digit. Earth 2016, 9, 63–105. [CrossRef]
39. Gleason, C.J.; Im, J. A Review of Remote Sensing of Forest Biomass and Biofuel: Options for Small-Area Applications. GIScience
Remote Sens. 2011, 48, 141–170. [CrossRef]
40. Jiang, F.; Deng, M.; Tang, J.; Fu, L.; Sun, H. Integrating Spaceborne LiDAR and Sentinel-2 Images to Estimate Forest Aboveground
Biomass in Northern China. Carbon Balance Manag. 2022, 17, 12. [CrossRef]
41. Bannari, A.; Morin, D.; Bonn, F.; Huete, A.R. A Review of Vegetation Indices. Remote Sens. Rev. 1995, 13, 95–120. [CrossRef]
42. Crippen, R.E. Calculating the Vegetation Index Faster. Remote Sens. Environ. 1990, 34, 71–73. [CrossRef]
43. Huete, A.R. A Soil-Adjusted Vegetation Index (SAVI). Remote Sens. Environ. 1988, 25, 295–309. [CrossRef]
44. Jordan, C.F. Derivation of Leaf-Area Index from Quality of Light on the Forest Floor. Ecology 1969, 50, 663–666. [CrossRef]
45. Tucker, C.J. Red and Photographic Infrared Linear Combinations for Monitoring Vegetation. Remote Sens. Environ. 1979, 8, 127–150.
[CrossRef]
46. Huete, A.; Didan, K.; Miura, T.; Rodriguez, E.P.; Gao, X.; Ferreira, L.G. Overview of the Radiometric and Biophysical Performance
of the MODIS Vegetation Indices. Remote Sens. Environ. 2002, 83, 195–213. [CrossRef]
47. Buschmann, C.; Nagel, E. In Vivo Spectroscopy and Internal Optics of Leaves as Basis for Remote Sensing of Vegetation. Int. J.
Remote Sens. 1993, 14, 711–722. [CrossRef]
48. Li, X.; Xiang, F.; Wu, S.; Liu, X.; Tian, Y.; Zhu, Y.; Cao, Q. A diagnostic method for nitrogen nutrition in winter wheat based on the
time-series dynamics of vegetation index. J. Wheat Crop. 2022, 42, 109–119.
Remote Sens. 2024, 16, 3927 23 of 23

49. Straub, C.; Weinacker, H.; Koch, B. A Comparison of Different Methods for Forest Resource Estimation Using Information from
Airborne Laser Scanning and CIR Orthophotos. Eur. J. For. Res 2010, 129, 1069–1080. [CrossRef]
50. Ku, N.-W.; Popescu, S.C. A Comparison of Multiple Methods for Mapping Local-Scale Mesquite Tree Aboveground Biomass with
Remotely Sensed Data. Biomass Bioenergy 2019, 122, 270–279. [CrossRef]
51. Kronseder, K.; Ballhorn, U.; Böhm, V.; Siegert, F. Above Ground Biomass Estimation across Forest Types at Different Degradation
Levels in Central Kalimantan Using LiDAR Data. Int. J. Appl. Earth Obs. Geoinf. 2012, 18, 37–48. [CrossRef]
52. Morin, D.; Planells, M.; Guyon, D.; Villard, L.; Mermoz, S.; Bouvet, A.; Thevenon, H.; Dejoux, J.-F.; Le Toan, T.; Dedieu, G.
Estimation and Mapping of Forest Structure Parameters from Open Access Satellite Images: Development of a Generic Method
with a Study Case on Coniferous Plantation. Remote Sens. 2019, 11, 1275. [CrossRef]
53. Fassnacht, F.E.; Hartig, F.; Latifi, H.; Berger, C.; Hernández, J.; Corvalán, P.; Koch, B. Importance of Sample Size, Data Type and
Prediction Method for Remote Sensing-Based Estimations of Aboveground Forest Biomass. Remote Sens. Environ. 2014, 154,
102–114. [CrossRef]
54. Hudak, A.T.; Strand, E.K.; Vierling, L.A.; Byrne, J.C.; Eitel, J.U.H.; Martinuzzi, S.; Falkowski, M.J. Quantifying Aboveground
Forest Carbon Pools and Fluxes from Repeat LiDAR Surveys. Remote Sens. Environ. 2012, 123, 25–40. [CrossRef]
55. Tanase, M.A.; Panciera, R.; Lowell, K.; Tian, S.; Hacker, J.M.; Walker, J.P. Airborne Multi-Temporal L-Band Polarimetric SAR Data
for Biomass Estimation in Semi-Arid Forests. Remote Sens. Environ. 2014, 145, 93–104. [CrossRef]
56. Karlson, M.; Ostwald, M.; Reese, H.; Sanou, J.; Tankoano, B.; Mattsson, E. Mapping Tree Canopy Cover and Aboveground
Biomass in Sudano-Sahelian Woodlands Using Landsat 8 and Random Forest. Remote Sens. 2015, 7, 10017–10041. [CrossRef]
57. Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [CrossRef]
58. Adam, E.; Mutanga, O.; Abdel-Rahman, E.M.; Ismail, R. Estimating Standing Biomass in Papyrus (Cyperus papyrus L.) Swamp:
Exploratory of in Situ Hyperspectral Indices and Random Forest Regression. Int. J. Remote Sens. 2014, 35, 693–714. [CrossRef]
59. Powell, S.L.; Cohen, W.B.; Healey, S.P.; Kennedy, R.E.; Moisen, G.G.; Pierce, K.B.; Ohmann, J.L. Quantification of Live Aboveground
Forest Biomass Dynamics with Landsat Time-Series and Field Inventory Data: A Comparison of Empirical Modeling Approaches.
Remote Sens. Environ. 2010, 114, 1053–1068. [CrossRef]
60. Hyyppä, E.; Yu, X.; Kaartinen, H.; Hakala, T.; Kukko, A.; Vastaranta, M.; Hyyppä, J. Comparison of Backpack, Handheld,
Under-Canopy UAV, and Above-Canopy UAV Laser Scanning for Field Reference Data Collection in Boreal Forests. Remote Sens.
2020, 12, 3327. [CrossRef]
61. Ruhan, A.; Du, W.; Ying, H.; Wei, B.; Shan, Y.; Dai, H. Estimation of Aboveground Biomass of Individual Trees by Backpack
LiDAR Based on Parameter-Optimized Quantitative Structural Models (AdQSM). Forests 2023, 14, 475. [CrossRef]
62. Jiao, Y.; Wang, D.; Yao, X.; Wang, S.; Chi, T.; Meng, Y. Forest Emissions Reduction Assessment Using Optical Satellite Imagery and
Space LiDAR Fusion for Carbon Stock Estimation. Remote Sens. 2023, 15, 1410. [CrossRef]
63. Schmidt, M.; Lucas, R.; Bunting, P.; Verbesselt, J.; Armston, J. Multi-Resolution Time Series Imagery for Forest Disturbance and
Regrowth Monitoring in Queensland, Australia. Remote Sens. Environ. 2015, 158, 156–168. [CrossRef]
64. Rodríguez-Fernández, N.; Al Bitar, A.; Colliander, A.; Zhao, T. Soil Moisture Remote Sensing across Scales. Remote Sens. 2019,
11, 190. [CrossRef]
65. Wang, C.; Feng, M.-C.; Yang, W.-D.; Ding, G.-W.; Sun, H.; Liang, Z.-Y.; Xie, Y.-K.; Qiao, X.-X. Impact of Spectral Saturation on Leaf
Area Index and Aboveground Biomass Estimation of Winter Wheat. Spectrosc. Lett. 2016, 49, 241–248. [CrossRef]
66. Jin, S.; Sun, X.; Wu, F.; Su, Y.; Li, Y.; Song, S.; Xu, K.; Ma, Q.; Baret, F.; Jiang, D.; et al. Lidar Sheds New Light on Plant Phenomics for
Plant Breeding and Management: Recent Advances and Future Prospects. ISPRS J. Photogramm. Remote Sens. 2021, 171, 202–223.
[CrossRef]
67. Coops, N.C.; Tompalski, P.; Goodbody, T.R.H.; Queinnec, M.; Luther, J.E.; Bolton, D.K.; White, J.C.; Wulder, M.A.; van Lier, O.R.;
Hermosilla, T. Modelling Lidar-Derived Estimates of Forest Attributes over Space and Time: A Review of Approaches and Future
Trends. Remote Sens. Environ. 2021, 260, 112477. [CrossRef]
68. Schlund, M.; Scipal, K.; Quegan, S. Assessment of a Power Law Relationship Between P-Band SAR Backscatter and Aboveground
Biomass and Its Implications for BIOMASS Mission Performance. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2018, 11, 3538–3547.
[CrossRef]
69. Yang, Q.; Su, Y.; Hu, T.; Jin, S.; Liu, X.; Niu, C.; Liu, Z.; Kelly, M.; Wei, J.; Guo, Q. Allometry-Based Estimation of Forest
Aboveground Biomass Combining LiDAR Canopy Height Attributes and Optical Spectral Indexes. For. Ecosyst. 2022, 9, 100059.
[CrossRef]
70. Rius, M.; Darling, J.A. How Important Is Intraspecific Genetic Admixture to the Success of Colonising Populations? Trends Ecol.
Evol. 2014, 29, 233–242. [CrossRef]
71. Blaschke, T. Object Based Image Analysis for Remote Sensing. ISPRS J. Photogramm. Remote Sens. 2010, 65, 2–16. [CrossRef]
72. Duro, D.C.; Franklin, S.E.; Dubé, M.G. A Comparison of Pixel-Based and Object-Based Image Analysis with Selected Machine
Learning Algorithms for the Classification of Agricultural Landscapes Using SPOT-5 HRG Imagery. Remote Sens. Environ. 2012,
118, 259–272. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.