Access to this full-text is provided by MDPI.
Content available from Forests
This content is subject to copyright.
Citation: Tang, Y.; Liu, X.; Lian, J.;
Cheng, X.; Wang, G.G.; Zhang, J. Soil
Depth Can Modify the Contribution
of Root System Architecture to the
Root Decomposition Rate. Forests
2023,14, 1092. https://doi.org/
10.3390/f14061092
Academic Editor: Antonio
Montagnoli
Received: 14 March 2023
Revised: 18 May 2023
Accepted: 22 May 2023
Published: 25 May 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Soil Depth Can Modify the Contribution of Root System
Architecture to the Root Decomposition Rate
Yingzhou Tang 1, Xin Liu 1, Jingwei Lian 2, Xuefei Cheng 1, G. Geoff Wang 3and Jinchi Zhang 1,*
1Co-Innovation Center for Sustainable Forestry in Southern China, Jiangsu Province Key Laboratory of Soil
and Water Conservation and Ecological Restoration, Nanjing Forestry University, Nanjing 210037, China
2Jiangsu Academy of Forestry, Dongshanqiao, Jiangning, Nanjing 211153, China
3Department of Forestry and Environmental Conservation, Clemson University, Clemson, SC 29634, USA
*Correspondence: zhang8811@njfu.edu.cn
Abstract:
Aims: Changes in root system architecture (RSA) and soil depth affect the root decomposi-
tion rate. However, due to soil opacity, many variables of RSA have not been well studied or even
measured. Methods: To investigate the effects of soil depth and the characteristics of RSA on the
root decomposition rate, soil samples (Soil cores were collected in October 2020 from Cunninghamia
lanceolata and Pinus taeda plantations, which were 40 years old) were obtained using a soil auger and
had a diameter of 10 cm and a length of 60 cm. Samples were taken from six different soil depths,
ranging from 0 to 60 cm with a 10 cm interval between each depth. The RSA in the in-situ soil cores
was analyzed using computed tomography scans and Avizo. Results: Root volume and the number
of root throats were significantly higher at the 0–10 cm soil depth than at the 10–60 cm soil depth,
but root length was significantly lower at the 50–60 cm soil depth (p< 0.05). Structural equation
modeling showed that different stand types influenced root biomass and thus the root decomposition
rate directly or indirectly through the characteristics of the stand types. RSA, i.e., root thickness
and breadth, affected root biomass indirectly and then affected the root decomposition rate. Root
biomass contributed the most to the root decomposition rate in the Cunninghamia lanceolata (20.19%)
and Pinus taeda (32.26%) plantations. The contribution of the RSA variables to the root decomposition
rate exceeded 50% at the 20–30 cm and 40–50 cm soil depths. Conclusions: Our findings suggested
that the influence of the RSA variables on the root decomposition rate varies with soil depth. This
deserves more consideration in our future studies on root decomposition and RSA.
Keywords: soil depth; CT scanning; traditional RSA; CT-based RSA; connectivity of roots; BRT
1. Introduction
Roots play a vital role in trees’ ability to acquire nutrients and water from the soil [
1
,
2
],
making them crucial for tree productivity. The biomass of roots constitutes a considerable
portion, ranging from 10% to 65%, of the overall biomass of trees [
3
–
5
]. This substan-
tial root biomass greatly influences the carbon dynamics and storage capacity of forest
ecosystems [6–8]
. At the same time, roots are affected by a combination of factors, including
the soil environment in which the plant is located and the tree species itself [
5
,
8
]. Despite
their importance, many aspects of roots remain relatively unknown. This is mainly because
the roots are underground and the opaque nature of soils is a major impediment to root
studies [
9
]. Therefore, the number of studies on roots is small compared to studies on
the above-ground parts. Among these studies on the roots of forest trees, studies on root
system architecture (RSA) are often neglected.
RSA describes the spatial arrangement pattern of different types of roots distributed
in the soil [
10
], i.e., the size and definite location of each specific point of the roots [
11
]. RSA
has important effects on nutrient acquisition and storage and carbon allocation in forest
trees. Different types of RSA create different growth habits in forest trees. At the same time,
Forests 2023,14, 1092. https://doi.org/10.3390/f14061092 https://www.mdpi.com/journal/forests
Forests 2023,14, 1092 2 of 22
forest trees also adjust their RSA to adapt to the changing external environment [
12
,
13
].
When roots encounter a nutrient-rich medium, trees increase [
14
] or decrease [
15
] lateral
root growth near nutrient-rich areas. When nutrient-poor areas are present, the tree will
suppress root growth in these areas. It has been shown that drought causes a decrease
in root length density and an increase in root diameter [
16
]. Trees can also adjust their
roots to utilize nutrients at different soil depths [
17
]. When nutrients in the upper soil are
not sufficient, the tree will adjust its roots to grow deeper into the soil to obtain sufficient
nutrients. When there are sufficient nutrients in the upper soil, the roots will adjust again
to make more use of the upper soil nutrients [
18
,
19
]. The variation in RSA at soil depth
largely reflects the distribution of soil nutrients at soil depth. The dynamic nature of
the root structure, as it adapts to the environment and its own requirements, plays a
crucial role in optimizing the growth and development of forest trees [
20
,
21
]. Conducting
a comprehensive and detailed investigation of RSA has significant implications for the
understanding of soil–plant systems. The RSA serves as an indicator of root growth and
soil nutrient distribution, directly influencing the rate of root decomposition [
22
]. Typically,
the quantification of RSA involves variables such as root diameter, root length, root surface
area, and root volume, all of which impact the root decomposition rate [
23
]. However,
these variables only provide a two-dimensional representation of the distribution pattern
of the RSA [
11
]. Since roots exist in soil as a three-dimensional structure with significant
variability, analyzing the RSA solely from a two-dimensional perspective might overlook
crucial information [
24
]. This disregarded information could hold great importance for the
rate of root decomposition.
To improve the accuracy of characterizing RSA, a large number of research methods
have been applied, including excavation methods, minirhizotrons, and artificial transparent
growth media to observe roots [
25
,
26
]. However, all of these methods have certain draw-
backs. For example, the manual method involves the destructive excavation of the roots by
hand and cleaning the roots to measure the root diameter, length, angle, and depth [
27
]. The
coordinates of the root surface are recorded, and the results of the manual measurements are
inputted into software to reconstruct the entire root system in three-dimensional (3D). This
method is time-consuming and labor-intensive, and if not handled properly, this destructive
method can bring external disturbances to the inter-root environment, leading to inaccurate
repeated measurements over time [
28
]. Further, because of the need to manually clean the
roots, some fine roots will inevitably be broken and washed away in the process, causing
damage to the original structure of the roots and leading to errors in the final results [
29
].
Although the minirhizotron method can achieve in situ observations, it can only observe a
small volume of soil surrounding the transparent minirhizotron observation tubes [
30
]. The
problem with the artificial medium method is that while it simulates the natural habitat,
due to the diversity and uncertainty of the field environment, the method cannot fully
restore the real growth condition of the forest roots in the field [
31
]. In summary, the
opacity of the soil medium is a major problem limiting the study of roots [
9
]. Computed
tomography (CT) has been applied as a non-destructive method to investigate the RSA
with the advancement of technology and research
tools [32–34]
. The three-dimensional
reconstruction of the roots can be achieved by CT, and the reconstructed RSA can be quan-
tified using analysis software. With CT nondestructive techniques, we can analyze more
structural characteristics of the roots, and we can further investigate the influence of the
real structural characteristics of the roots in soil on the root decomposition rate.
To better investigate the distribution characteristics of forest RSA at different soil
depths and to study its influence on the root decomposition rate, we designed a 12-month
field experiment within Cunninghamia lanceolata and Pinus taeda plantations. The in situ soil
core method was innovatively improved by scanning the in situ soil cores using medical
CT to obtain the RSA distribution characteristics in a non-destructive manner. Our study
intends to address the following main questions: (1) what is the distribution pattern of the
RSA at each soil depth?, (2) How do RSA variables affect the root decomposition rate, and
Forests 2023,14, 1092 3 of 22
(3) Does the contribution of each RSA variable to the root decomposition rate differ by tree
species and soil depth?
2. Materials and Methods
2.1. Study Site
Our study site was located at Xiashu Forest Farm, Jurong City, Jiangsu Province,
China (119
◦
22
0
46
00
E, 32
◦
12
0
57
00
N), at an elevation of 100 m, which was established by
Nanjing Forestry University (Figure 1). According to the Harmonized World Soil Database
(HWSD) and World Reference Base for soil resources (WRB), the soil type in this area is
yellow–brown soil, mainly eluvial, slightly acidic, with a high sediment content, and a
depth of about 1.0 m. The soil pH is 5.22. The natural forest at the Xiashu Forest Farm was
dominated by secondary forest [35].
Forests 2023, 14, x FOR PEER REVIEW 3 of 23
study intends to address the following main questions: (1) what is the distribution paern
of the RSA at each soil depth?, (2) How do RSA variables affect the root decomposition
rate, and (3) Does the contribution of each RSA variable to the root decomposition rate
differ by tree species and soil depth?
2. Materials and Methods
2.1. Study Site
Our study site was located at Xiashu Forest Farm, Jurong City, Jiangsu Province,
China (119°22′46″ E, 32°12′57″ N), at an elevation of 100 m, which was established by Nan-
jing Forestry University (Figure 1). According to the Harmonized World Soil Database
(HWSD) and World Reference Base for soil resources (WRB), the soil type in this area is
yellow–brown soil, mainly eluvial, slightly acidic, with a high sediment content, and a
depth of about 1.0 m. The soil pH is 5.22. The natural forest at the Xiashu Forest Farm was
dominated by secondary forest [35].
Figure 1. Location map: (a) thumbnail of a map of China; (b) Location map of the Xia Shu Forestry
Field; (c) The sample site was located in the Cunninghamia lanceolata and Pinus taeda L. plantations
in the old mill area of Xia Shu Forestry.
The region has a northern subtropical monsoon climate. The average annual temper-
ature is 15.2 °C. The lowest monthly average temperature is −0.8°C, in January. The high-
est average temperature is 31.6 °C, in July. The average frost-free period is 233 days. The
seasonal fluctuation of rainfall is large. The average annual precipitation is 1055.6 mm,
and the average annual relative air humidity is as high as 79%. The total rainfall during
Figure 1.
Location map: (
a
) thumbnail of a map of China; (
b
) Location map of the Xia Shu Forestry
Field; (
c
) The sample site was located in the Cunninghamia lanceolata and Pinus taeda L. plantations in
the old mill area of Xia Shu Forestry.
The region has a northern subtropical monsoon climate. The average annual tem-
perature is 15.2
◦
C. The lowest monthly average temperature is
−
0.8
◦
C, in January. The
highest average temperature is 31.6
◦
C, in July. The average frost-free period is 233 days.
The seasonal fluctuation of rainfall is large. The average annual precipitation is 1055.6 mm,
and the average annual relative air humidity is as high as 79%. The total rainfall during the
experimental period from October 2020 to October 2021 was 1167.4 mm, of which precipi-
tation in May August accounted for more than 60% of the total precipitation (Figure A1)
(Appendix A).
Forests 2023,14, 1092 4 of 22
2.2. Experimental Design
In this study, Cunninghamia lanceolata and Pinus taeda plantations in good growing
condition were selected, and three plots (20 m
×
20 m) were randomly set up in each
plantation type (The age of the Cunninghamia lanceolata plantation was 43 years, the average
tree height was 13.86 m, and the average diameter at breast height was 23.88 cm. The age
of the Pinus taeda plantation was 39 years, the average tree height was 14.08 m, and the
average diameter at breast height was 29.11 cm) (Table A1). In October 2020, sample strips
were set up at least 1 m from the nearest tree within the selected Cunninghamia lanceolata
and Pinus taeda plantations to ensure that root biomass was as uniform as possible within
each soil core [
36
,
37
]. Ten sample strips were set up in each sample plot. Soil cores were
taken using a special soil drill with an inner diameter of 10 cm and a length of 60 cm,
which we designed and made. The drill was made of stainless steel and was driven to a
predetermined depth using an impact pick. The drill was then removed with the core intact
using a High Lift Jack. Four soil cores were drilled in each sample strip, resulting in 40 soil
cores from each sample plot. We carefully placed the soil core from the drill in a 74
µ
m
nylon bag and returned it to the soil from which the core was extracted. On days 0, 120,
240, and 360 after the completion of sampling, 30 soil cores were randomly removed from
each sample plot. The roots in the soil core were cleaned, dried, and weighed to calculate
the root decomposition rate.
At each sampling session, the in situ soil cores that were incubated were removed
along with the nylon mesh bags. The removed soil cores were protected with a 10 cm
diameter and 60 cm height polyvinyl chloride (PVC) pipe, which was sealed with plastic
wrap and protected from shocks. On days 0 after the completion of sampling, the cores
were scanned using CT equipment within 12 h after sampling and were then cut into
6 sections (0–10 cm, 10–20 cm, 20–30 cm, 30–40 cm, 40–50 cm, and 50–60 cm) at 10 cm
intervals. Detailed measurements of RSA at each soil depth were then made using Avizo.
On days 120, 240, and 360 after the completion of sampling, the soil cores were cut directly
after sampling, and the roots were removed to calculate the root decomposition rate.
2.3. RSA
The intact soil cores were scanned using a United Imaging uCT510 medical scanner
(United Imaging Intelligence, Shanghai, China). Scanning was done at 1 mm intervals
using 120 kV and 335 mA current. The scanned images were reconstructed using Avizo
2019.1 (Thermo Fisher Scientific Avizo, Waltham, MA, USA), and the oblique CT image cor-
rection was performed using VG Studio Max 2022.1 (Volume Graphics GmbH, Heidelberg,
Germany). The images were thresholded for segmentation using Avizo’s intensity his-
togram, and the roots were reconstructed and then analyzed using Avizo’s computational
module [38].
The traditional RSA (RSA variables that can be obtained by traditional means) and the
CT-based RSA (RSA variables that cannot be obtained by traditional means or are difficult
to obtain but can be easily obtained by CT) were calculated directly by the Avizo computing
module (Table A2). The connectivity of roots was calculated by fitting roots to spheres and
throats between roots to synthetic sticks.
2.4. Root Decomposition Rate
In this study, the constant K in the classical decomposition dynamics describing the
exponential loss of mass over time [
39
], that is, the mass loss ratio per unit time, was taken
as the root decomposition rate. The specific calculation method is as follows:
For each soil depth for each removed soil core, the roots were removed, cleaned, and
dried at 65
◦
C to a constant weight. We calculated the roots’ mass loss rate (k) by applying
the first-order exponential decay model [40]:
Xt/Xt−1= e−k∆t(1)
Forests 2023,14, 1092 5 of 22
where X
t
(g) is the root mass obtained by sampling, X
t−1
(g) is the root mass obtained on the
previous sampling date, and
∆
t is the time (in years) that elapsed between the collection of
the two samples (Table 1).
Table 1.
Root mass loss rate in different decomposition periods and the root decomposition rate in
the test period (Note: uppercase letters indicate significant differences between different soil layers,
lowercase letters indicate significant differences between decomposition times (p≤0.05)).
Forest Type Soil Layer 4 Months 8 Months 12 Months Root Decomposition
Rate (Year−1)
Cunninghamia
lanceolata
0–10 cm 8.17% (±8.19%) Aa 5.78% (±3.80%) Ba 4.93% (±4.37%) Aa 0.058 (±0.06) A
10–20 cm 11.00% (±8.16%) Ab 4.66% (±3.73%) Bab 4.23% (±3.22%) Aa 0.035 (±0.028) A
20–30 cm 8.10% (±8.16%) Aa 3.83% (±3.73%) ABa 4.33% (±3.22%) Aa 0.051 (±0.047) A
30–40 cm 8.37% (±6.55%) Ab 3.33% (±1.44%) ABa 2.11% (±1.41%) Aa 0.038 (±0.033) A
40–50 cm 8.32% (±8.47%) Aa 1.25% (±1.36%) Aa 4.45% (±7.29%) Aa 0.029 (±0.024) A
50–60 cm 8.91% (±8.95%) Ab 1.65% (±1.13%) Aa 1.05% (±1.47%) Aa 0.033 (±0.021) A
Pinus
taeda
0–10 cm 14.79% (±9.28%) Ab 4.8% (±2.66%) ABa 9.92% (±6.04%) Bab 0.067 (±0.046) AB
10–20 cm 12.00% (±8.73%) Aa 8.84% (±5.91%) Ba 5.74% (±2.75%) ABa 0.083 (±0.055) AB
20–30 cm 11.31% (±10.92%) Aa 7.91% (±7.53%) Ba 4.41% (±3.99%) ABa 0.051 (±0.047) AB
30–40 cm 7.18% (±9.77%) Aa 3.97% (±3.24%) ABa 2.07% (±2.64%) Aa 0.086 (±0.063) B
40–50 cm 4.40% (±3.54%) Aa 1.65% (±1.71%) Aa 2.76% (±1.98%) Aa 0.066 (±0.045) B
50–60 cm 5.82% (±6.34%) Aa 5.87% (±6.21%) ABa 4.14% (±9.37%) ABa 0.034 (±0.026) A
2.5. Statistical Analysis
All statistical analyses were performed using SPSS 26.0 (SPSS Inc., Chicago, IL, USA).
To facilitate comparison among different variables, we standardized all variables using the
z-score (Figure A2). We used the average approach [
41
] to divide 24 variables into three
indicators. To detect differences in the root decomposition rate, root biomass, and RSA
between different soil depths and different stand types, we first analyzed the interaction
between soil depth and tree species using two-way analysis of variance. With one exception,
we detected a significant interaction between the soil depth and stand type. Therefore, we
conducted and reported one-way ANOVAs for stand type separately for each soil depth
and for soil depth for each stand type. If the ANOVA results were significant (
p< 0.05
),
the least significant difference (LSD) test was used to determine significant differences
between soil depths and tree species (p< 0.05). We used Pearson’s correlation coefficient
to quantify the strength and significance of the relationship between RSA, soil depth, tree
species, and the root decomposition rate coefficient. Structural equation modeling (SEM)
analysis (Amos, 24.0, Chicago, IL, USA) was used to estimate the direct and indirect effects
of RSA on the root decomposition rates. Several metrics (
χ2
,p-value, GFI, and RMSEM)
were used to test the goodness-of-fit of the model:
p-value > 0.05
,
GFI > 0.9
, and RMSEM
< 0.08. Augmented regression tree (BRT) analysis (R 4.2.1) was conducted to estimate the
contribution of tree species, soil depth, and RSA to the root decomposition rate [
42
]. Graphs
were plotted using Origin (Origin, version 2021, Northampton, MA, USA) and GraphPad
Prism 9 (GraphPad Software, Inc., La Jolla, CA, USA).
3. Results
3.1. The Vertical Distribution of RSA
Except for OrientationTheta (Figure 2j), soil depth significantly affected traditional
RSA, but none of the effects of tree species on measured RSA was significant. Except for
Length (Figure 2b), each traditional RSA variable gradually decreased with soil depth,
where the Root volume fraction, Volume, Area, and VoxelFaceArea were significantly
higher in the topsoil (0–10 cm) than at other soil depths (Figure 2a,f,g,h).
Forests 2023,14, 1092 6 of 22
Forests 2023, 14, x FOR PEER REVIEW 6 of 23
3. Results
3.1. The Vertical Distribution of RSA
Except for OrientationTheta (Figure 2j), soil depth significantly affected traditional
RSA, but none of the effects of tree species on measured RSA was significant. Except for
Length (Figure 2b), each traditional RSA variable gradually decreased with soil depth,
where the Root volume fraction, Volume, Area, and VoxelFaceArea were significantly
higher in the topsoil (0–10 cm) than at other soil depths (Figure 2a,f,g,h).
Figure 2.
Traditional RSA: (
a
) Volume fraction of roots; (
b
) Length of roots; (
c
) Width of roots;
(
d
) Breadth of roots; (
e
) Thickness of roots; (
f
) Volume of roots; (
g
) Surface area of roots; (
h
) Voxel
Surface area of roots; (i) OrientationPhi angle of roots; (j) OrientationTheta angle of roots (Note: up-
percase letters indicate significant differences between different soil layers (p
≤
0.05); the light blue
columns indicate the Cunninghamia lanceolata plantation; dark blue columns indicate the Pinus taeda L.
plantation.
Forests 2023,14, 1092 7 of 22
There was no significant effect of tree species on CT-based RSA, and EqDiameter
(Figure 3d) and Flatness (Figure 3e) did not differ significantly between soil depths. Shape
(Figure 3a) and Dimension (Figure 3c) decreased with soil depth, and Euler was significantly
higher in the topsoil (0–10 cm) than at other soil depths. Tortuosity and Euler, on the other
hand, increased with soil depth, and Euler was significantly lower in the topsoil (0–10 cm)
than at the other soil depths.
Forests 2023, 14, x FOR PEER REVIEW 7 of 23
Figure 2. Traditional RSA: (a) Volume fraction of roots; (b) Length of roots; (c) Width of roots; (d)
Breadth of roots; (e) Thickness of roots; (f) Volume of roots; (g) Surface area of roots; (h) Voxel Sur-
face area of roots; (i) OrientationPhi angle of roots; (j) OrientationTheta angle of roots (Note: upper-
case leers indicate significant differences between different soil layers (p ≤ 0.05); the light blue col-
umns indicate the Cunninghamia lanceolata plantation; dark blue columns indicate the Pinus taeda L.
plantation.
There was no significant effect of tree species on CT-based RSA, and EqDiameter
(Figure 3d) and Flatness (Figure 3e) did not differ significantly between soil depths. Shape
(Figure 3a) and Dimension (Figure 3c) decreased with soil depth, and Euler was signifi-
cantly higher in the topsoil (0–10 cm) than at other soil depths. Tortuosity and Euler, on
the other hand, increased with soil depth, and Euler was significantly lower in the topsoil
(0–10 cm) than at the other soil depths.
Figure 3.
CT-based RSA: (
a
) Shape factor of roots; (
b
) Tortuosity of roots; (
c
) Fractal dimension of
roots; (
d
) EqDiameter of roots; (
e
) Flatness of roots; (
f
) Euler of roots. (Note: uppercase letters indicate
significant differences between different soil layers (p
≤
0.05); the light blue columns indicate the
Cunninghamia lanceolata plantation; dark blue columns indicate the Pinus taeda L. plantation.
Root connectivity was significantly different among soil depths, but the effect of tree
species was not significant. Except for Root number, which increased and then decreased
with soil depth, all other variables decreased gradually with soil depth, among which the
Forests 2023,14, 1092 8 of 22
coordination number and throat number were significantly higher in the topsoil (0–10 cm)
than at other soil depths (Figure 4).
Forests 2023, 14, x FOR PEER REVIEW 8 of 23
Figure 3. CT-based RSA: (a) Shape factor of roots; (b) Tortuosity of roots; (c) Fractal dimension of
roots; (d) EqDiameter of roots; (e) Flatness of roots; (f) Euler of roots. (Note: uppercase leers indi-
cate significant differences between different soil layers (p ≤ 0.05); the light blue columns indicate
the Cunninghamia lanceolata plantation; dark blue columns indicate the Pinus taeda L. plantation.
Root connectivity was significantly different among soil depths, but the effect of tree
species was not significant. Except for Root number, which increased and then decreased
with soil depth, all other variables decreased gradually with soil depth, among which the
coordination number and throat number were significantly higher in the topsoil (0–10 cm)
than at other soil depths (Figure 4).
Figure 4.
Connectivity of roots: (
a
) Number of roots; (
b
) Number of root connections; (
c
) Number of
root throats; (
d
) Surface area of root throats; (
e
) Length of root throats; (
f
) Throat EqDiameter of roots.
(Note: uppercase letters indicate significant differences between different soil layers (p
≤
0.05); the
light blue columns indicate the Cunninghamia lanceolata plantation; dark blue columns indicate the
Pinus taeda L. plantation.
3.2. Root Decomposition Rate and Root Biomass
Tree species had a significant effect on the root decomposition rate, but there was
no significant difference in the root decomposition rate among different soil depths in
the Cunninghamia lanceolata plantation. In contrast, the root decomposition rate was sig-
nificantly higher at the 30–40 cm soil depth than at the 50–60 cm soil depth in the Pinus
taeda plantation (Figure 5a). Root biomass gradually decreased with increasing soil depth,
Forests 2023,14, 1092 9 of 22
but there was no significant difference between the different tree species and among the
different soil depths (Figure 5b).
Forests 2023, 14, x FOR PEER REVIEW 9 of 23
Figure 4. Connectivity of roots: (a) Number of roots; (b) Number of root connections; (c) Number of
root throats; (d) Surface area of root throats; (e) Length of root throats; (f) Throat EqDiameter of
roots. (Note: uppercase leers indicate significant differences between different soil layers (p ≤ 0.05);
the light blue columns indicate the Cunninghamia lanceolata plantation; dark blue columns indicate
the Pinus taeda L. plantation.
3.2. Root Decomposition Rate and Root Biomass
Tree species had a significant effect on the root decomposition rate, but there was no
significant difference in the root decomposition rate among different soil depths in the
Cunninghamia lanceolata plantation. In contrast, the root decomposition rate was signifi-
cantly higher at the 30–40 cm soil depth than at the 50–60 cm soil depth in the Pinus taeda
plantation (Figure 5a). Root biomass gradually decreased with increasing soil depth, but
there was no significant difference between the different tree species and among the dif-
ferent soil depths (Figure 5b).
Figure 5. (a) Root decomposition rate; (b) Root biomass; (Note: uppercase leers indicate significant
differences between different soil layers (p ≤ 0.05); the light blue columns indicate the Cunninghamia
lanceolata plantation; dark blue columns indicate the Pinus taeda L. plantation.
The decomposition rate of roots in the 0–10 cm, 10–20 cm, 30–40 cm, and 50–60 cm
soil layers of the Cunninghamia lanceolata plantation and the 10–20 cm, 20–30 cm, and 30–
40 cm soil layers of the Pinus taeda plantation decreased gradually with increasing decom-
position time (Table 1; p < 0.05). During the first four-month cycle of the experiment, the
root mass loss rate in the Cunninghamia lanceolata and Pinus taeda plantations did not show
significant differences among soil layers (Table 1; p < 0.05). However, during the second 4-
month cycle of the experiment, the root mass loss rate in the shallow soil layer of the Cun-
ninghamia lanceolata plantation and Pinus taeda plantation was significantly higher than
that in the deep soil layer (Table 1; p < 0.05). In contrast, when the experiment proceeded
to the last four-month cycle, the root mass loss rate in the Cunninghamia lanceolata planta-
tion did not differ significantly among soil layers. However, in the Pinus taeda plantation,
the root mass loss rate remained significantly higher in the shallow soil layer than in the
deep soil layer based on Table 1 (p < 0.05).
Figure 5.
(
a
) Root decomposition rate; (
b
) Root biomass; (Note: uppercase letters indicate significant
differences between different soil layers (p
≤
0.05); the light blue columns indicate the Cunninghamia
lanceolata plantation; dark blue columns indicate the Pinus taeda L. plantation.
The decomposition rate of roots in the 0–10 cm, 10–20 cm, 30–40 cm, and 50–60 cm soil
layers of the Cunninghamia lanceolata plantation and the 10–20 cm, 20–30 cm, and 30–40 cm
soil layers of the Pinus taeda plantation decreased gradually with increasing decomposition
time (Table 1;p< 0.05). During the first four-month cycle of the experiment, the root mass
loss rate in the Cunninghamia lanceolata and Pinus taeda plantations did not show significant
differences among soil layers (Table 1;p< 0.05). However, during the second 4-month cycle
of the experiment, the root mass loss rate in the shallow soil layer of the Cunninghamia
lanceolata plantation and Pinus taeda plantation was significantly higher than that in the
deep soil layer (Table 1;p< 0.05). In contrast, when the experiment proceeded to the last
four-month cycle, the root mass loss rate in the Cunninghamia lanceolata plantation did not
differ significantly among soil layers. However, in the Pinus taeda plantation, the root mass
loss rate remained significantly higher in the shallow soil layer than in the deep soil layer
based on Table 1(p< 0.05).
3.3. Major Factors Affecting the Root Decomposition Rate in Different Stand Types and Soil Depths
Tortuosity, Euler, and Length were significantly and positively correlated with soil
depth (p< 0.05). The root decomposition rate was significantly correlated only with stand
type, root biomass, EqDiameter, Width, Breadth, and thickness (p< 0.05) (Figure 6).
The enhanced regression tree (BRT) model analyzed the contribution of RSA variables
to the root decomposition rate in different stand types and at different soil depths. The
results showed that the most significant contributor to the root decomposition rate was root
biomass in both Cunninghamia lanceolata (Figure 7a) and Pinus taeda plantations (Figure 7b),
with values of 20.19% and 32.26%, respectively.
Forests 2023,14, 1092 10 of 22
Forests 2023, 14, x FOR PEER REVIEW 10 of 23
3.3. Major Factors Affecting the Root Decomposition Rate in Different Stand Types and
Soil Depths
Tortuosity, Euler, and Length were significantly and positively correlated with soil
depth (p < 0.05). The root decomposition rate was significantly correlated only with stand
type, root biomass, EqDiameter, Width, Breadth, and thickness (p < 0.05) (Figure 6).
Figure 6. Pearson correlation between forest type, soil layer, the root decomposition rate, traditional
RSA, CT-based RSA, and connectivity of roots, *** indicates a significant correlation at p ≤ 0.001, **
indicates a significant correlation at p ≤ 0.01, * indicates a significant correlation at p ≤ 0.05.
The enhanced regression tree (BRT) model analyzed the contribution of RSA varia-
bles to the root decomposition rate in different stand types and at different soil depths.
The results showed that the most significant contributor to the root decomposition rate
was root biomass in both Cunninghamia lanceolata (Figure 7a) and Pinus taeda plantations
(Figure 7b), with values of 20.19% and 32.26%, respectively.
Figure 6.
Pearson correlation between forest type, soil layer, the root decomposition rate, traditional
RSA, CT-based RSA, and connectivity of roots, *** indicates a significant correlation at p
≤
0.001,
** indicates a significant correlation at p≤0.01, * indicates a significant correlation at p≤0.05.
The analysis of the main factors affecting the root decomposition rate at different soil
depths revealed that only Root Biomass contributed more than 10% to the root decom-
position rate at the 0–10 cm soil depth and 50–60 cm soil depth, i.e., 66.32% (Figure 7c)
and 47.95% (Figure 7h), respectively. With increasing soil depth, the contribution of RSA
variables such as the root number, shape, and width to the root decomposition rate first
increased and then decreased, and the contribution of root biomass first decreased and
then increased (Figure 7c–h).
Forests 2023,14, 1092 11 of 22
Forests 2023, 14, x FOR PEER REVIEW 11 of 23
Figure 7.
The contribution of soil layers and the root decomposition rate under different forest
types and different soil layers: (
a
)Cunninghamia lanceolata; (
b
)Pinus taeda; (
c
) 0–10 cm soil layer;
(
d
)
10–20 cm
soil layer; (
e
) 20–30 cm soil layer; (
f
) 30–40 cm soil layer; (
g
) 40–50 cm soil layer;
(h) 50–60 cm soil layer.
Forests 2023,14, 1092 12 of 22
3.4. SEM Analysis
Root biomass (path coefficient = 0.473; p< 0.001) and tree species (
path coefficient = 0.187
;
p< 0.05) had a direct positive effect on the root decomposition rate, while tree species also in-
directly affected the root decomposition rate by affecting root biomass (path
coefficient = 0.197
;
p< 0.05) (Figure 8a).
Forests 2023, 14, x FOR PEER REVIEW 12 of 23
Figure 7. The contribution of soil layers and the root decomposition rate under different forest types
and different soil layers: (a) Cunninghamia lanceolata; (b) Pinus taeda; (c) 0–10 cm soil layer; (d) 10–20
cm soil layer; (e) 20–30 cm soil layer; (f) 30–40 cm soil layer; (g) 40–50 cm soil layer; (h) 50–60 cm soil
layer.
The analysis of the main factors affecting the root decomposition rate at different soil
depths revealed that only Root Biomass contributed more than 10% to the root decompo-
sition rate at the 0–10 cm soil depth and 50–60 cm soil depth, i.e., 66.32% (Figure 7c) and
47.95% (Figure 7h), respectively. With increasing soil depth, the contribution of RSA var-
iables such as the root number, shape, and width to the root decomposition rate first in-
creased and then decreased, and the contribution of root biomass first decreased and then
increased (Figure 7c–h).
3.4. SEM Analysis
Root biomass (path coefficient = 0.473; p < 0.001) and tree species (path coefficient =
0.187; p < 0.05) had a direct positive effect on the root decomposition rate, while tree spe-
cies also indirectly affected the root decomposition rate by affecting root biomass (path
coefficient = 0.197; p < 0.05) (Figure 8a).
Figure 8. (a) Structural equation model (SEM) analysis estimating the direct and indirect effects of
the CT scan data of the root on the root decomposition rate. Boxes show variables included in the
model. Test results of the goodness-of-model fit Chi-square (2) = 1.063, p-value = 0.588 > 0.05, good-
ness-of-fit index (GFI) = 0.997 > 0.9, and root square mean error of approximation (RMSEA) = 0.000
< 0.08. Numbers on arrows are standardized path coefficients. The widths of the arrows represent
the strength of the relationships. Blue arrows indicate positive relationships, and yellow arrows in-
dicate negative relationships. Solid arrows indicate significance (p < 0.05) and dashed arrows repre-
sent non-significance (p > 0.05) (the p-value was calculated in terms of the nonnormalized path). (b)
Standardized total effects (direct plus indirect effects) derived from the structural equation models
depicted above.
To further assess the main factors affecting the root decomposition rate, the stand-
ardized total effects of different parameters were analyzed. The results showed that root
biomass had the largest positive effect on the root decomposition rate, followed by Thick-
ness, EqDiameter, and stand type. Root width had the largest negative effect on the root
decomposition rate, followed by root breadth (Figure 8b).
4. Discussion
4.1. Vertical Distribution of the RSA
We found that root volume and area were significantly greater in the 0–10 cm soil
depth than at the other depths and accounted for nearly 50% of the total volume of the in
situ soil core. The topsoil contains more nutrients that promote root growth in upper soils,
Figure 8.
(
a
) Structural equation model (SEM) analysis estimating the direct and indirect effects of the
CT scan data of the root on the root decomposition rate. Boxes show variables included in the model.
Test results of the goodness-of-model fit Chi-square (2) = 1.063,
p-value = 0.588 > 0.05
, goodness-of-fit
index (GFI) = 0.997 > 0.9, and root square mean error of approximation
(RMSEA) = 0.000 < 0.08
. Num-
bers on arrows are standardized path coefficients. The widths of the arrows represent the strength of
the relationships. Blue arrows indicate positive relationships, and yellow arrows indicate negative re-
lationships. Solid arrows indicate significance (p< 0.05) and dashed arrows represent non-significance
(p> 0.05) (the p-value was calculated in terms of the nonnormalized path). (
b
) Standardized total
effects (direct plus indirect effects) derived from the structural equation models depicted above.
To further assess the main factors affecting the root decomposition rate, the stan-
dardized total effects of different parameters were analyzed. The results showed that
root biomass had the largest positive effect on the root decomposition rate, followed by
Thickness, EqDiameter, and stand type. Root width had the largest negative effect on the
root decomposition rate, followed by root breadth (Figure 8b).
4. Discussion
4.1. Vertical Distribution of the RSA
We found that root volume and area were significantly greater in the 0–10 cm soil
depth than at the other depths and accounted for nearly 50% of the total volume of the in
situ soil core. The topsoil contains more nutrients that promote root growth in upper soils,
and as the soil depth increases, nutrients gradually decrease, limiting root development.
Root length, however, is the opposite of the above, as upper soils provide roots with easy
access to nutrients. However, with increasing soil depth, nutrient acquisition becomes
difficult and roots have to grow toward the nutrient-rich areas by increasing their length to
obtain nutrients [
43
]. The thickness of the root was also significantly greater in 0–10 cm soils
than at other soil depths. The reason for this is that coarse roots have a greater thickness
and are rarely found in deeper soils. Roots in deeper soils are mainly in the form of fine
roots [44].
The analysis of all RSA variables analyzed by the Avizo model showed that all vari-
ables except EqDiameter and Flatness were significantly correlated with soil depth. Tortu-
osity and Euler were significantly and positively correlated with depth soil depth, while
shape and dimension were significantly and negatively correlated with soil depth. Tortuos-
ity reflects the variation of the nutrient content in the soil. Root Tortuosity results from a
specific growth response, which depends mainly on soil matrix properties. Deeper soils
Forests 2023,14, 1092 13 of 22
have lower and unevenly distributed water and nutrient concentrations, forcing the roots to
grow and meander with the location of the water and nutrients. As a result, a significantly
higher tortuosity was found in deeper soils [
45
]. At the same time, the mechanical strength
of deeper soils is greater, and roots have to grow by constantly meandering around dense
soil areas to obtain nutrients. It has been shown that the fractal dimension of the roots is
consistent with the reinforcement capacity of the roots [
46
]. The coarse roots in upper soils
have a greater ability to hold the soil, while the fractal dimension of the roots decreases
with soil depth, decreasing the ability of the roots to maintain regional soil and water and
soil shear strength.
Using CT with a computer model construction of the 3D structure of the roots, we
obtained data on the natural growth state of the roots. Based on our innovative measure-
ment of root characteristics, we proposed a root connectivity index. We found that all
six variables used to derive the root connectivity index were significantly and negatively
correlated with soil depth. Our analysis suggests that the number of roots was higher in
upper soils, and roots were interspersed with each other, so there were more channels and
connections between roots. As the soil depth increased, the number of roots decreased
significantly. At the same time, due to the increase in the density and mechanical strength
of the deeper soil depths, the channels of connection between the roots disappeared. The
roots grew and developed more as independent individuals.
4.2. Vertical Distribution of the Root Decomposition Rate
During the first 4 months of the root decomposition rate, the amount of root mass
lost did not decrease significantly with increasing soil depth, supporting the previous
findings [
47
] (Table 1). This may be due to favorable conditions for the root decomposition
rate at the initial stage when both the roots and soil nutrient content were adequate. Over
time, the rate of root decomposition began to slow down with increasing soil depth, sug-
gesting the conditions for the root decomposition rate were being depleted. Higher levels
of SM, CEC, and SOM were present in the surface soils, indicating a greater decomposition
potential than in the deeper soils. High levels of SM, CEC, and SOM increase enzyme
activity and thus stimulate the root decomposition rate and promote microbial uptake
of unstable C and N [
48
]. With changes in soil depth, bulk density, temperature, and
moisture [
49
], nutrient availability and microbial communities are also changing, affecting
the root decomposition rate [
50
–
53
]. Furthermore, the lack of soil animal activity can also
result in a lower root decomposition rate in deeper soils than in surface soils [47].
The mass loss rate of the root decomposition rate did not decrease linearly with in-
creasing soil depth but tended to be much faster at the 10–20 cm depth [
54
]. This is related
to the complexity of factors influencing the root decomposition rate. The root decomposi-
tion rate is influenced by both the structures and chemical characteristics of roots and the
environment [
55
]. As a result, the root decomposition rates may vary considerably even for
the same species in different environments [
56
,
57
]. Site-specific conditions (e.g., tempera-
ture, precipitation, different forest floor types, and different nutrient interactions between
plants and belowground communities) can also affect the root decomposition rate [
58
,
59
].
Although our studied stands were coniferous and occurred at the same location, their soil
temperature and humidity were found to differ significantly. Changes in temperature and
humidity cause a series of chain reactions in nutrient distribution and microbial community
composition and distribution, which result in differences in the root decomposition rate
in time and space. At the same time, we believe that this may be the result of a water
deficit in mid-summer in the 0–10 cm layer, leading to decrease in the decomposition rate
of dead roots.
4.3. Major Factors Affecting the Root Decomposition Rate in Different Stand Types and at
Different Soil Depths
Throughout the study period, there was a significant positive correlation between
the root decomposition rate and indicators such as root thickness, which is reflected by
Forests 2023,14, 1092 14 of 22
root diameter. We found that the root decomposition rate increased with the root diameter
and length (Figure A3b). The effects of root length on the decomposition rate supported
previous studies [
60
,
61
]. Carbon in thicker roots may be present as carbohydrates, amino
acids, and other substances that are more easily decomposed than the finer diameter roots,
which causes the larger diameter roots to be more easily decomposed [
62
]. The slowing
effect of fine roots on the nutrient supply also leads to a slower decomposition of fine
roots than coarse roots [
63
]. Although larger diameter roots contain higher levels of N and
stimulate the formation of N-lignin complexes to retard decomposition [
64
], our study was
conducted on coniferous species, which usually have lower N and P contents [
65
,
66
]. It
has also been shown that the root decomposition rate capacity of roots varies with the root
morphology [
67
]. However, we found that RSA variables reflected the nutrient acquisition
capacity of root growth and, to some extent, soil nutrient distribution [
52
,
68
]. A significant
linear correlation was found between the root decomposition rate and root connectivity
in the Cunninghamia lanceolata plantation (Figure A3f). We did not find any correlation
between RSA characteristics (tortuosity, etc.) and the root decomposition rate in the Pinus
taeda plantation (Figure A4). The SEM results also confirmed our conjecture that root
biomass and stand type directly affect the root decomposition rate. Root Breadth indirectly
but negatively changed root biomass and thus the root decomposition rate. Roots with
smaller breadth contributed to lower root biomass, and the decomposition was therefore
also slower.
Through the analysis of RSA and the root decomposition rate, we observed that
the effects were inconsistent across stand type and soil depth. Therefore, we needed to
investigate the influential factors that affect the root decomposition rate of each stand
type at each soil depth. In both Cunninghamia lanceolata and Pinus taeda plantations, root
biomass contributed the most to the root decomposition rate, which reached 20.19% and
32.26%, respectively. This result further validates the SEM results that root biomass directly
influences the root decomposition rate while other variables indirectly influence root
biomass and thus the root decomposition rate. Further analysis revealed that differences
between Cunninghamia lanceolata and Pinus taeda plantations, the second most important
variable, were likely due to the difference in RSA and root element contents between
species.
Various conditions in upper soils are more favorable to the root decomposition
rate. Roots in upper soils are also more susceptible to external environmental influences
(e.g., trampling by forest animals, gnawing by soil animals, rainfall erosion, etc.) [
63
].
Therefore, the effect of RSA on the root decomposition rate was overshadowed by other
factors, and root biomass became the decisive factor affecting the root decomposition rate in
upper soils (66.32%). As soil depth increased, the contribution of RSA variables to the root
decomposition rate gradually increased. However, when the soil depth reached 50–60 cm,
the largest contributor to the root decomposition rate was once again the root biomass,
which accounted for nearly half (47.95%) of the total contribution. In deeper soils, the
number of roots was lower and the conditions were less favorable to the root decomposition
rate. Consequently, root biomass plays a decisive role in the root decomposition rate [
60
,
69
].
5. Conclusions
RSA variables varied with soil depth but did not differ significantly between the
two stand types. Root volume, root surface area, and root thickness were greater and
inter-root connectivity was higher in upper soils. Root length gradually increased with
depth, and root tortuosity also increased with increasing soil depth. Roots of different stand
types directly influenced the root decomposition rate through tree species’ composition.
Structural variables such as root thickness positively affected the root decomposition
rate indirectly by changing the root biomass. Root biomass was the main contributor
affecting the root decomposition rate. Finally, the contribution of RSA variables to the root
decomposition rate gradually increased with increasing soil depth. However, when the soil
Forests 2023,14, 1092 15 of 22
depth reached 50–60 cm, root biomass was again the largest contributor affecting the root
decomposition rate.
Author Contributions:
Conception and design of the research: Y.T.; Acquisition of data: Y.T., J.L.,
X.C. and X.L.; Analysis and interpretation of data: Y.T. and J.L.; Statistical analysis: Y.T. and J.L.;
Drafting the manuscript: Y.T.; Revision of manuscript for important intellectual content: Y.T., X.L.,
G.G.W. and J.Z. All authors have read and agreed to the published version of the manuscript.
Funding:
This research was funded by the Jiangsu Science and Technology Plan Project [BE2022420];
Innovation and Promotion of Forestry Science and Technology Program of Jiangsu Province [LYKJ
(2021) 30] Scientific Research Project of Baishanzu National Park [2021ZDLY01]; Jiangsu Province
Science Foundation for Youths (BK20200785); and Priority Academic Program Development of
Jiangsu Higher Education Institutions [PAPD].
Data Availability Statement:
The data presented in this study are available on request from the
corresponding author.
Acknowledgments:
Yingzhou Tang would especially like to thank Jingwei Lian for her patience,
care, support, and company over the years. Let’s get married!
Conflicts of Interest: The authors declare no conflict of interests.
Forests 2023,14, 1092 16 of 22
Appendix A
Forests 2023, 14, x FOR PEER REVIEW 16 of 23
Appendix A
Figure A1. Meteorological data for the study area: (a) Air Temperature; (b) Relative Humidity; (c)
Rainfall; (d) Wind Speed; (e) Atmospheric Pressure; (f) Solar Irradiance.
Figure A1.
Meteorological data for the study area: (
a
) Air Temperature; (
b
) Relative Humidity;
(c) Rainfall; (d) Wind Speed; (e) Atmospheric Pressure; (f) Solar Irradiance.
Forests 2023,14, 1092 17 of 22
Forests 2023, 14, x FOR PEER REVIEW 17 of 23
Figure A2. Heat map showing the change in root factors. The coefficient of variation (CV) indicated
the sensitivity of root factors.
Figure A2.
Heat map showing the change in root factors. The coefficient of variation (CV) indicated
the sensitivity of root factors.
Forests 2023, 14, x FOR PEER REVIEW 17 of 23
Figure A2. Heat map showing the change in root factors. The coefficient of variation (CV) indicated
the sensitivity of root factors.
Figure A3.
RSA in the Cunninghamia lanceolata plantation: (
a
) Traditional RSA; (
b
) Relationship
between traditional RSA and the root decomposition rate; (
c
) CT-based RSA; (
d
) Relationship between
CT-based RSA and the root decomposition rate; (
e
) Connectivity of roots; (
f
) Relationship between
the connectivity of roots and the root decomposition rate; (Note: uppercase letters indicate significant
differences between different soil layers (p≤0.05)).
Forests 2023,14, 1092 18 of 22
Forests 2023, 14, x FOR PEER REVIEW 18 of 23
Figure A3. RSA in the Cunninghamia lanceolata plantation: (a) Traditional RSA; (b) Relationship be-
tween traditional RSA and the root decomposition rate; (c) CT-based RSA; (d) Relationship between
CT-based RSA and the root decomposition rate; (e) Connectivity of roots; (f) Relationship between
the connectivity of roots and the root decomposition rate; (Note: uppercase leers indicate signifi-
cant differences between different soil layers (p ≤ 0.05)).
Figure A4. RSA in the Pinus taeda L. plantation: (a) Traditional RSA; (b) Relationship between tradi-
tional RSA and the root decomposition rate; (c) CT-based RSA; (d) Relationship between CT-based
RSA and the root decomposition rate; (e) Connectivity of roots; (f) Relationship between the con-
nectivity of roots and the root decomposition rate; (Note: uppercase leers indicate significant dif-
ferences between different soil layers (p ≤ 0.05)).
Figure A4.
RSA in the Pinus taeda L. plantation: (
a
) Traditional RSA; (
b
) Relationship between
traditional RSA and the root decomposition rate; (
c
) CT-based RSA; (
d
) Relationship between CT-
based RSA and the root decomposition rate; (
e
) Connectivity of roots; (
f
) Relationship between the
connectivity of roots and the root decomposition rate; (Note: uppercase letters indicate significant
differences between different soil layers (p≤0.05)).
Table A1. Basic information about the sample site.
Sample
Type
Age
(Years) Place Aspect Slope ASL
(m)
Density
(Trees
ha−2)
Mean
DBH
(cm)
Mean
TH
(m)
Mean
CD
(m)
Crown
Density
Intrusion
Ratio
C. lanceolata
C1 43 32◦0701700 N
119◦1300300 E347◦N 14◦100 975 22.19
(4.83) 12.77
(4.33) 6.12
(2.16) 0.82 8%
C. lanceolata
C2 43 32◦0701700 N
119◦1300200 E350◦N 15◦100 875 25.38
(6.72) 14.39
(4.80) 8.11
(3.26) 0.89 12%
C. lanceolata
C3 43 32◦0701500 N
119◦1300300 E214◦SW 16◦100 750 24.06
(4.75) 14.43
(3.19) 6.93
(2.20) 0.93 20%
P. taeda L.
P1 39 32◦0701500 N
119◦1301600 E203◦SW 13◦110 675 28.80
(5.62) 15.11
(3.95) 7.93
(2.72) 0.54 9%
P. taeda L.
P2 39 32◦0701700 N
119◦1301800 E207◦SW 10◦110 500 29.91
(6.52) 13.13
(3.15) 7.75
(2.22) 0.61 17%
P. taeda L.
P3 39 32◦0701700 N
119◦1301700 E
330
◦
NW
7◦110 525 28.61
(4.98) 14.00
(2.66) 9.62
(1.83) 0.63 13%
Forests 2023,14, 1092 19 of 22
Table A2. The characteristics of the root system.
Parameters Explanatory Notes
The volume fraction of roots The volume of roots/volume of undisturbed soil core (The range is 0 to 1)
Length Maximum Feret diameter
The Feret diameter is defined as the distance between two tangent planes of a particle in a given
direction
Breadth Largest distance between two parallel lines touching the object without intersecting it and lying
in a plane orthogonal to the maximum Feret diameter
Width Minimum Feret Diameter
Thickness
The largest segment that touches the object by its endpoints and lying in a plane orthogonal to the
maximum Feret diameter and orthogonal to the breadth diameter
OrientationPhi
Phi orientation of the particle in degrees [0, +90], computed with the inertia moments. It defines,
with OrientationTheta, the eigenvector of the largest eigenvalue of the covariance matrix
OrientationTheta Theta orientation of the particle in degrees [−180, 180], computed with the inertia moments. It
defines, with OrientationPhi, the eigenvector of the largest eigenvalue of the covariance matrix
Shape The shape factor is defined as shape = area3/(36 ×π×volume2)
Tortuosity
Tortuosity is defined as the ratio between the length of the path and the distance between its ends
along the z-axis. In our case, the distance between the ends of the curve is given by the number of
planes along the z-axis
Fractal dimension
A fractal dimension is a number greater than 2 and strictly lower than 3. The result is 2 in the case
of standard geometric surfaces. Applied to 3D images, the fractal dimension is an effective
indicator to measure and compare the roughness of a surface. It is also a good indicator to
evaluate how the curve fills the space. The less smooth the surface, the greater the fractal
dimension. It can also be interpreted as a quantification of how complex the surface is and how it
fills the space
Flatness The ratio of the smallest to the medium eigenvalue of the covariance matrix
EqDiameter The equivalent diameter is the diameter of the sphere of the same volume
VoxelFaceArea This value is the sum of voxel surfaces that are on the outside of each connected component
Euler Connectedness indicator. It is an indicator of the connectivity of a 3D complex structure
References
1.
Teramoto, S.; Takayasu, S.; Kitomi, Y.; Arai-Sanoh, Y.; Tanabata, T.; Uga, Y. High-Throughput Three-Dimensional Visualization of
Root System Architecture of Rice Using X-ray Computed Tomography. Plant Methods 2020,16, 66. [CrossRef]
2.
Gregory, P.J. Plant Roots: Growth, Activity and Interactions with the Soil; John Wiley & Sons: Hoboken, NJ, USA, 2008; ISBN
978-1-4051-7308-7.
3.
Barton, C.V.M.; Montagu, K.D. Detection of Tree Roots and Determination of Root Diameters by Ground Penetrating Radar under
Optimal Conditions. Tree Physiol. 2004,24, 1323–1331. [CrossRef]
4.
Bert, D.; Danjon, F. Carbon Concentration Variations in the Roots, Stem and Crown of Mature Pinus pinaster (Ait.). For. Ecol.
Manag. 2006,222, 279–295. [CrossRef]
5.
Montagnoli, A.; Terzaghi, M.; Miali, A.; Chiatante, D.; Dumroese, R.K. Unusual Late-Fall Wildfire in a Pre-Alpine Fagus Sylvatica
Forest Reduced Fine Roots in the Shallower Soil Layer and Shifted Very Fine-Root Growth to Deeper Soil Depth. Sci. Rep.
2023
,
13, 6380. [CrossRef] [PubMed]
6. Litton, C.M.; Raich, J.W.; Ryan, M.G. Carbon Allocation in Forest Ecosystems. Glob. Chang. Biol. 2007,13, 2089–2109. [CrossRef]
7.
Montagnoli, A.; Dumroese, R.K.; Terzaghi, M.; Onelli, E.; Scippa, G.S.; Chiatante, D. Seasonality of Fine Root Dynamics and
Activity of Root and Shoot Vascular Cambium in a Quercus ilex L. Forest (Italy). For. Ecol. Manag. 2019,431, 26–34. [CrossRef]
8.
Kondori, A.A.; Vajari, K.A.; Feizian, M.; Montagnoli, A.; Di Iorio, A. Gap Size in Hyrcanian Forest Affects the Lignin and N
Concentrations of the Oriental Beech (Fagus orientalis Lipsky) Fine Roots but Does Not Change Their Morphological Traits in the
Medium Term. Forests 2021,12, 137. [CrossRef]
9.
Bontpart, T.; Concha, C.; Giuffrida, M.V.; Robertson, I.; Admkie, K.; Degefu, T.; Girma, N.; Tesfaye, K.; Haileselassie, T.;
Fikre, A.; et al
. Affordable and Robust Phenotyping Framework to Analyse Root System Architecture of Soil-Grown Plants. Plant
J. 2020,103, 2330–2343. [CrossRef]
10.
Rogers, E.D.; Monaenkova, D.; Mijar, M.; Nori, A.; Goldman, D.I.; Benfey, P.N. X-ray Computed Tomography Reveals the
Response of Root System Architecture to Soil Texture. Plant Physiol. 2016,171, 2028–2040. [CrossRef]
Forests 2023,14, 1092 20 of 22
11.
Fang, S.; Yan, X.; Liao, H. 3D Reconstruction and Dynamic Modeling of Root Architecture in Situ and Its Application to Crop
Phosphorus Research. Plant J. 2009,60, 1096–1108. [CrossRef]
12.
Bauerle, T.L.; Centinari, M. Assessment of Root System Development among Four Ornamental Tree Species through Time via
X-ray Computed Tomography. HortScience 2014,49, 44–50. [CrossRef]
13.
Pang, Y.; Tian, J.; Yang, H.; Zhang, K.; Wang, D. Responses of Fine Roots at Different Soil Depths to Different Thinning Intensities
in a Secondary Forest in the Qinling Mountains, China. Biology 2022,11, 351. [CrossRef]
14. Hodge, A. The Plastic Plant: Root Responses to Heterogeneous Supplies of Nutrients. New Phytol. 2004,162, 9–24. [CrossRef]
15.
Montagnoli, A.; Lasserre, B.; Terzaghi, M.; Byambadorj, S.-O.; Nyam-Osor, B.; Scippa, G.S.; Chiatante, D. Fertilization Reduces
Root Architecture Plasticity in Ulmus pumila Used for Afforesting Mongolian Semi-Arid Steppe. Front. Plant Sci.
2022
,13, 878299.
[CrossRef]
16.
Zhou, G.; Zhou, X.; Nie, Y.; Bai, S.H.; Zhou, L.; Shao, J.; Cheng, W.; Wang, J.; Hu, F.; Fu, Y. Drought-Induced Changes in Root
Biomass Largely Result from Altered Root Morphological Traits: Evidence from a Synthesis of Global Field Trials. Plant Cell
Environ. 2018,41, 2589–2599. [CrossRef] [PubMed]
17.
Ma, Z.; Chen, H.Y.H. Positive Species Mixture Effects on Fine Root Turnover and Mortality in Natural Boreal Forests. Soil Biol.
Biochem. 2018,121, 130–137. [CrossRef]
18.
Nyam-Osor, B.; Byambadorj, S.-O.; Park, B.B.; Terzaghi, M.; Scippa, G.S.; Stanturf, J.A.; Chiatante, D.; Montagnoli, A. Root
Biomass Distribution of Populus sibirica and Ulmus pumila Afforestation Stands Is Affected by Watering Regimes and Fertilization
in the Mongolian Semi-Arid Steppe. Front. Plant Sci. 2021,12, 638828. [CrossRef] [PubMed]
19.
Montagnoli, A.; Chiatante, D.; Godbold, D.L.; Koike, T.; Rewald, B.; Dumroese, R.K. Editorial: Modulation of Growth and
Development of Tree Roots in Forest Ecosystems. Front. Plant Sci. 2022,13, 850163. [CrossRef]
20.
Kaestner, A.; Schneebeli, M.; Graf, F. Visualizing Three-Dimensional Root Networks Using Computed Tomography. Geoderma
2006,136, 459–469. [CrossRef]
21.
Peng, S.; Chen, H.Y.H. Global Responses of Fine Root Biomass and Traits to Plant Species Mixtures in Terrestrial Ecosystems.
Glob. Ecol. Biogeogr. 2021,30, 289–304. [CrossRef]
22.
Pierret, A.; Doussan, C.; Capowiez, Y.; Bastardie, F.; Pagès, L. Root Functional Architecture: A Framework for Modeling the
Interplay between Roots and Soil. Vadose Zone J. 2007,6, 269–281. [CrossRef]
23.
de Graaff, M.-A.; Six, J.; Jastrow, J.D.; Schadt, C.W.; Wullschleger, S.D. Variation in Root Architecture among Switchgrass Cultivars
Impacts Root Decomposition Rates. Soil Biol. Biochem. 2013,58, 198–206. [CrossRef]
24.
Lontoc-Roy, M.; Dutilleul, P.; Prasher, S.O.; Han, L.; Brouillet, T.; Smith, D.L. Advances in the Acquisition and Analysis of CT
Scan Data to Isolate a Crop Root System from the Soil Medium and Quantify Root System Complexity in 3-D Space. Geoderma
2006,137, 231–241. [CrossRef]
25.
Ding, Y.; Leppälammi-Kujansuu, J.; Helmisaari, H.-S. Fine Root Longevity and Below- and Aboveground Litter Production in a
Boreal Betula Pendula Forest. For. Ecol. Manag. 2019,431, 17–25. [CrossRef]
26.
Li, A.; Zhu, L.; Xu, W.; Liu, L.; Teng, G. Recent Advances in Methods for in Situ Root Phenotyping. PeerJ
2022
,10, e13638.
[CrossRef]
27.
Dupuy, L.X.; Fourcaud, T.; Lac, P.; Stokes, A. A Generic 3D Finite Element Model of Tree Anchorage Integrating Soil Mechanics
and Real Root System Architecture. Am. J. Bot. 2007,94, 1506–1514. [CrossRef] [PubMed]
28.
Guo, L.; Chen, J.; Cui, X.; Fan, B.; Lin, H. Application of Ground Penetrating Radar for Coarse Root Detection and Quantification:
A Review. Plant Soil 2013,362, 1–23. [CrossRef]
29.
Smith, A.; Astrup, R.; Raumonen, P.; Liski, J.; Krooks, A.; Kaasalainen, S.; Åkerblom, M.; Kaasalainen, M. Tree Root System
Characterization and Volume Estimation by Terrestrial Laser Scanning and Quantitative Structure Modeling. Forests
2014
,5,
3274–3294. [CrossRef]
30.
Jose, S.; Gillespie, A.R.; Seifert, J.R.; Pope, P.E. Comparison of Minirhizotron and Soil Core Methods for Quantifying Root Biomass
in a Temperate Alley Cropping System. Agrofor. Syst. 2001,52, 161–168. [CrossRef]
31.
Saltveit, M.E.; Young, E. A Method for Studying the Three-Dimensional Distribution of Roots Grown in an Artificial Medium. J.
Am. Soc. Horticult. Sci. 1983,108, 1023–1025. [CrossRef]
32.
Burr-Hersey, J.E.; Mooney, S.J.; Bengough, A.G.; Mairhofer, S.; Ritz, K. Developmental Morphology of Cover Crop Species Exhibit
Contrasting Behaviour to Changes in Soil Bulk Density, Revealed by X-ray Computed Tomography. PLoS ONE
2017
,12, e0181872.
[CrossRef]
33.
Hu, X.; Li, X.-Y.; Zhao, Y.-D.; Cheng, Y.-Q.; Gao, Z.; Yang, Z.-G. Identification of Water Flow through Non-Root Soil Macropores
and along Roots in Shrub-Encroached Grassland. Eur. J. Soil Sci. 2022,73, e13260. [CrossRef]
34.
Suresh, N.; Stephens, S.A.; Adams, L.; Beck, A.N.; McKinney, A.L.; Varga, T. Extracting Metrics for Three-Dimensional Root
Systems: Volume and Surface Analysis from In-Soil X-ray Computed Tomography Data. J. Vis. Exp.
2016
,110, 53788. [CrossRef]
35.
FAO. World Reference Base for Soil Resources 2014: International Soil Classification System for Naming Soils and Creating Legends for Soil
Maps; FAO: Rome, Italy, 2014; ISBN 978-92-5-108369-7.
36.
Dornbush, M.E.; Isenhart, T.M.; Raich, J.W. Quantifying Fine-Root Decomposition: An Alternative to Buried Litterbags. Ecology
2002,83, 2985–2990. [CrossRef]
37.
Li, X.; Zheng, X.; Zhou, Q.; McNulty, S.; King, J.S. Measurements of Fine Root Decomposition Rate: Method Matters. Soil Biol.
Biochem. 2022,164, 108482. [CrossRef]
Forests 2023,14, 1092 21 of 22
38.
Katuwal, S.; Norgaard, T.; Moldrup, P.; Lamandé, M.; Wildenschild, D.; de Jonge, L.W. Linking Air and Water Transport in Intact
Soils to Macropore Characteristics Inferred from X-ray Computed Tomography. Geoderma 2015,237–238, 9–20. [CrossRef]
39.
Olson, J.S. Energy Storage and the Balance of Producers and Decomposers in Ecological Systems. Ecology
1963
,44, 322–331.
[CrossRef]
40.
Aulen, M.; Shipley, B.; Bradley, R. Prediction of in Situ Root Decomposition Rates in an Interspecific Context from Chemical and
Morphological Traits. Ann. Bot. 2012,109, 287–297. [CrossRef] [PubMed]
41.
Han, S.; Delgado-Baquerizo, M.; Luo, X.; Liu, Y.; Van Nostrand, J.D.; Chen, W.; Zhou, J.; Huang, Q. Soil Aggregate Size-Dependent
Relationships between Microbial Functional Diversity and Multifunctionality. Soil Biol. Biochem. 2021,154, 108143. [CrossRef]
42.
Li, C.; Yu, Z.; Lin, J.; Meng, M.; Zhao, Y.; Jia, Z.; Peng, X.; Liu, X.; Zhang, J. Forest Conversion and Soil Depth Can Modify the
Contributions of Organic and Inorganic Colloids to the Stability of Soil Aggregates. Forests 2022,13, 546. [CrossRef]
43.
Chen, W.; Chen, Y.; Siddique, K.H.M.; Li, S. Root Penetration Ability and Plant Growth in Agroecosystems. Plant Physiol. Biochem.
2022,183, 160–168. [CrossRef]
44.
Materechera, S.A.; Alston, A.M.; Kirby, J.M.; Dexter, A.R. Influence of Root Diameter on the Penetration of Seminal Roots into a
Compacted Subsoil. Plant Soil 1992,144, 297–303. [CrossRef]
45.
Popova, L.; van Dusschoten, D.; Nagel, K.A.; Fiorani, F.; Mazzolai, B. Plant Root Tortuosity: An Indicator of Root Path Formation
in Soil with Different Composition and Density. Ann. Bot. 2016,118, 685–698. [CrossRef] [PubMed]
46.
Zhou, Z.; Wang, Y.; An, Z.; Li, R.; Xu, Y.; Zhang, P.; Yang, Y.; Wang, T. Deep Root Information “Hidden in the Dark”: A Case
Study on the 21-m Soil Profile of Robinia pseudoacacia in the Critical Zone of the Chinese Loess Plateau. CATENA
2022
,213, 106121.
[CrossRef]
47.
Hicks Pries, C.E.; Sulman, B.N.; West, C.; O’Neill, C.; Poppleton, E.; Porras, R.C.; Castanha, C.; Zhu, B.; Wiedemeier, D.B.; Torn,
M.S. Root Litter Decomposition Slows with Soil Depth. Soil Biol. Biochem. 2018,125, 103–114. [CrossRef]
48.
Han, S.H.; Kim, S.; Chang, H.; Kim, H.-J.; Khamzina, A.; Son, Y. Soil Depth- and Root Diameter-Related Variations Affect Root
Decomposition in Temperate Pine and Oak Forests. J. Plant Ecol. 2019,12, 871–881. [CrossRef]
49.
Chen, Y.; Zhang, Y.; Cao, J.; Fu, S.; Wang, J.; Lambers, H.; Liu, Z. Divergent Responses of Fine Root Decomposition to Removal of
Understory Plants and Overstory Trees in Subtropical Eucalyptus Urophylla Plantations. Plant Soil
2022
,476, 639–652. [CrossRef]
50.
Pregitzer, K.S.; Laskowski, M.J.; Burton, A.J.; Lessard, V.C.; Zak, D.R. Variation in Sugar Maple Root Respiration with Root
Diameter and Soil Depth. Tree Physiol. 1998,18, 665–670. [CrossRef] [PubMed]
51.
Birouste, M.; Kazakou, E.; Blanchard, A.; Roumet, C. Plant Traits and Decomposition: Are the Relationships for Roots Comparable
to Those for Leaves? Ann. Bot. 2012,109, 463–472. [CrossRef]
52.
Donovan, L.A.; Mason, C.M.; Bowsher, A.W.; Goolsby, E.W.; Ishibashi, C.D.A. Ecological and Evolutionary Lability of Plant Traits
Affecting Carbon and Nutrient Cycling. J. Ecol. 2014,102, 302–314. [CrossRef]
53.
Smith, S.W.; Woodin, S.J.; Pakeman, R.J.; Johnson, D.; van der Wal, R. Root Traits Predict Decomposition across a Landscape-Scale
Grazing Experiment. New Phytol. 2014,203, 851–862. [CrossRef]
54.
Wang, W.; Zhang, X.; Tao, N.; Ao, D.; Zeng, W.; Qian, Y.; Zeng, H. Effects of Litter Types, Microsite and Root Diameters on Litter
Decomposition in Pinus sylvestris Plantations of Northern China. Plant Soil 2014,374, 677–688. [CrossRef]
55.
Djukic, I.; Kepfer-Rojas, S.; Schmidt, I.K.; Larsen, K.S.; Beier, C.; Berg, B.; Verheyen, K.; Caliman, A.; Paquette, A.;
Gutiérrez-Girón, A.; et al
. Early Stage Litter Decomposition across Biomes. Sci. Total Environ.
2018
,628–629, 1369–1394.
[CrossRef] [PubMed]
56.
Harmon, M.E.; Silver, W.L.; Fasth, B.; Chen, H.; Burke, I.C.; Parton, W.J.; Hart, S.C.; Currie, W.S. Lidet Long-Term Patterns of Mass
Loss during the Decomposition of Leaf and Fine Root Litter: An Intersite Comparison. Glob. Chang. Biol.
2009
,15, 1320–1338.
[CrossRef]
57.
Hobbie, S.E.; Oleksyn, J.; Eissenstat, D.M.; Reich, P.B. Fine Root Decomposition Rates Do Not Mirror Those of Leaf Litter among
Temperate Tree Species. Oecologia 2010,162, 505–513. [CrossRef] [PubMed]
58.
Liu, H.; Lin, L.; Wang, H.; Zhang, Z.; Shangguan, Z.; Feng, X.; He, J.-S. Simulating Warmer and Drier Climate Increases Root
Production but Decreases Root Decomposition in an Alpine Grassland on the Tibetan Plateau. Plant Soil
2021
,458, 59–73.
[CrossRef]
59.
Santonja, M.; Milcu, A.; Fromin, N.; Rancon, A.; Shihan, A.; Fernandez, C.; Baldy, V.; Hättenschwiler, S. Temporal Shifts in Plant
Diversity Effects on Carbon and Nitrogen Dynamics During Litter Decomposition in a Mediterranean Shrubland Exposed to
Reduced Precipitation. Ecosystems 2019,22, 939–954. [CrossRef]
60.
Guerrero-Ramírez, N.R.; Craven, D.; Messier, C.; Potvin, C.; Turner, B.L.; Handa, I.T. Root Quality and Decomposition Environ-
ment, but Not Tree Species Richness, Drive Root Decomposition in Tropical Forests. Plant Soil 2016,404, 125–139. [CrossRef]
61.
Prieto, I.; Stokes, A.; Roumet, C. Root Functional Parameters Predict Fine Root Decomposability at the Community Level. J. Ecol.
2016,104, 725–733. [CrossRef]
62.
Chapin, F.S., III; Zavaleta, E.S.; Eviner, V.T.; Naylor, R.L.; Vitousek, P.M.; Reynolds, H.L.; Hooper, D.U.; Lavorel, S.; Sala, O.E.;
Hobbie, S.E.; et al. Consequences of Changing Biodiversity. Nature 2000,405, 234–242. [CrossRef]
63.
Ohashi, M.; Makita, N.; Katayama, A.; Kume, T.; Matsumoto, K.; Kumagai, T.; Endo, I.; Kho, L.K. Characteristics of Root
Decomposition Based on in Situ Experiments in a Tropical Rainforest in Sarawak, Malaysia: Impacts of Root Diameter and Soil
Biota. Plant Soil 2019,436, 439–448. [CrossRef]
Forests 2023,14, 1092 22 of 22
64.
Camiré, C.; Côté, B.; Brulotte, S. Decomposition of Roots of Black Alder and Hybrid Poplar in Short-Rotation Plantings: Nitrogen
and Lignin Control. Plant Soil 1991,138, 123–132. [CrossRef]
65.
Zhuang, L.; Yang, W.; Wu, F.; Tan, B.; Zhang, L.; Yang, K.; He, R.; Li, Z.; Xu, Z. Diameter-Related Variations in Root Decomposition
of Three Common Subalpine Tree Species in Southwestern China. Geoderma 2018,311, 1–8. [CrossRef]
66.
Sun, T.; Mao, Z.; Han, Y. Slow Decomposition of Very Fine Roots and Some Factors Controlling the Process: A 4-Year Experiment
in Four Temperate Tree Species. Plant Soil 2013,372, 445–458. [CrossRef]
67.
Wang, P.; Guo, J.; Xu, X.; Yan, X.; Zhang, K.; Qiu, Y.; Zhao, Q.; Huang, K.; Luo, X.; Yang, F.; et al. Soil Acidification Alters Root
Morphology, Increases Root Biomass but Reduces Root Decomposition in an Alpine Grassland. Environ. Pollut.
2020
,265, 115016.
[CrossRef] [PubMed]
68. Reich, P.B. The World-Wide ‘Fast–Slow’ Plant Economics Spectrum: A Traits Manifesto. J. Ecol. 2014,102, 275–301. [CrossRef]
69.
Sun, T.; Dong, L.; Zhang, L.; Wu, Z.; Wang, Q.; Li, Y.; Zhang, H.; Wang, Z. Early Stage Fine-Root Decomposition and Its
Relationship with Root Order and Soil Depth in a Larix Gmelinii Plantation. Forests 2016,7, 234. [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.
Available via license: CC BY 4.0
Content may be subject to copyright.
