ArticlePDF Available

Consequences of tropical rainforest conversion to tree plantations on fine root dynamics and functional traits

Authors:

Abstract and Figures

Despite the crucial role of fine roots for water and nutrient uptake, soil biological activity and ecosystem carbon cycling, the response of root dynamics to rapidly advancing land-use change in the tropics is still poorly understood. To address this uncertainty, we investigated the consequences of tropical forest conversion to intensively managed tree plantations for a suite of functional fine root traits. We analysed fine root morphology (diameter, specific root length (SRL), tissue density) and chemistry, as well as root lifespans in four prevalent land-use systems in the lowlands of Sumatra (Indonesia), namely natural forest, jungle rubber, rubber and oil palm monocultures. Fine root production was estimated using three independent approaches (sequential coring, ingrowth cores, mini-rhizotrons). Contradicting the expected trend from more conservative to more acquisitive fine root traits with increasing land-use intensity, we found that SRL and tissue density were significantly higher in forest trees, while fine root diameter was largest in rubber trees and root N content lowest in the oil palm system. Median fine root longevity in the top soil was 11% higher in rubber plantations (238 days) than in jungle rubber (211 days), and more than 50% greater than in the forest (140 days) and oil palm plantations (125 days). Fine root production was higher in the forest and oil palm plantations (ranging between 2 and 9 Mg ha −1 year −1) than the rubber stands, but annual totals varied depending on the methodological approach. Conversion of tropical lowland forest to agricultural systems significantly altered community-level fine root morphology, dynamics and longevity, with likely consequences for soil carbon cycling and soil biological activity. However, land-use intensification did not consistently lead to more acquisitive fine root systems; rather, differences in root morphology and dynamics were driven by species-specific root trait syndromes especially of rubber and oil palm.
Content may be subject to copyright.
www.oikosjournal.org
OIKOS
Oikos
1
© 2022 e Authors. Oikos published by John Wiley & Sons Ltd on behalf of Nordic Society Oikos.
is is an open access article under the terms of the Creative Commons
Attribution License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited.
Subject Editor: Luke McCormack
Editor-in-Chief:
Gerlinde B. De Deyn
Accepted 15 May 2022
doi: 10.1111/oik.08898
1–19
2022: e08898
Despite the crucial role of fine roots for water and nutrient uptake, soil biological activ-
ity and ecosystem carbon cycling, the response of root dynamics to rapidly advancing
land-use change in the tropics is still poorly understood. To address this uncertainty,
we investigated the consequences of tropical forest conversion to intensively man-
aged tree plantations for a suite of functional fine root traits. We analysed fine root
morphology (diameter, specific root length (SRL), tissue density) and chemistry, as
well as root lifespans in four prevalent land-use systems in the lowlands of Sumatra
(Indonesia), namely natural forest, jungle rubber, rubber and oil palm monocultures.
Fine root production was estimated using three independent approaches (sequential
coring, ingrowth cores, mini-rhizotrons). Contradicting the expected trend from more
conservative to more acquisitive fine root traits with increasing land-use intensity, we
found that SRL and tissue density were significantly higher in forest trees, while fine
root diameter was largest in rubber trees and root N content lowest in the oil palm
system. Median fine root longevity in the top soil was 11% higher in rubber planta-
tions (238 days) than in jungle rubber (211 days), and more than 50% greater than
in the forest (140 days) and oil palm plantations (125 days). Fine root production was
higher in the forest and oil palm plantations (ranging between 2 and 9 Mg ha1 year1)
than the rubber stands, but annual totals varied depending on the methodological
approach. Conversion of tropical lowland forest to agricultural systems significantly
altered community-level fine root morphology, dynamics and longevity, with likely
consequences for soil carbon cycling and soil biological activity. However, land-use
intensification did not consistently lead to more acquisitive fine root systems; rather,
differences in root morphology and dynamics were driven by species-specific root trait
syndromes especially of rubber and oil palm.
Keywords: fine root dynamics, fine root lifespan, fine root production, ingrowth
cores, land-use change, mini-rhizotron, root economics spectrum, root morphology,
sequential coring
Consequences of tropical rainforest conversion to tree
plantations on fine root dynamics and functional traits
Martyna M. Kotowska, Sasya Samhita, Dietrich Hertel, Triadiati Triadiati, Friderike Beyer, Kara Allen,
Roman M. Link and Christoph Leuschner
M. M. Kotowska (https://orcid.org/0000-0002-2283-5979) (mkotows@gwdg.de), S. Samhita, D. Hertel and C. Leuschner, Dept of Plant Ecology and
Ecosystems Research, Albrecht-von-Haller Inst. for Plant Sciences, Univ. of Goettingen, Göttingen, Germany. – T. Triadiati, Dept of Biology, Faculty of
Mathematics and Natural Sciences, IPB Univ., Bogor, Indonesia. – F. Beyer, Chair of Silviculture, Faculty of Environment and Natural Resources, Univ. of
Freiburg, Freiburg, Germany. – K. Allen, Manaaki Whenua-Landcare Research, Lincoln, New Zealand. – R. M. Link, Chair of Ecophysiology and
Vegetation Ecology, Julius von Sachs Inst. of Biological Sciences, Univ. of Würzburg, Würzburg, Germany.
Research
2
Introduction
Fine roots fulfil vital functions in plant metabolism and
growth, as they are responsible for water and nutrient uptake,
anchorage, contribute to carbohydrate storage and enable
beneficial interactions with soil microorganisms (de Kroon
and Visser 2003, Mommer et al. 2016). Penetrating deep into
the ground, they are controlling soil organic matter dynam-
ics and nutrient transfer to the soil through root exudates,
rhizodeposition, mutual matter exchange with symbiotic
mycorrhizal fungi and tissue decomposition, thereby influ-
encing soil fertility (Jackson et al. 1997, Rasse et al. 2005,
Dijkstra et al. 2021). In addition, the mechanical properties
of fine roots contribute to erosion control and slope stabi-
lization (Reubens et al. 2007). eir anatomical, morpho-
logical and chemical characteristics vary substantially across
species and along environmental gradients and can be con-
siderably altered by anthropogenic management and chang-
ing vegetation composition (Eissenstat 1991, Hodge 2004,
Leuschner et al. 2009, Bardgett et al. 2014, Addo-Danso et al.
2020, Pierick et al. 2020, Weemstra et al. 2020). Hence, a
better understanding of fine root functional traits and their
variation with vegetation change is critical for improving
our capacity to predict ecosystem functioning with land-use
intensification (Erktan et al. 2018, Freschet et al. 2021).
Macroscopic root morphological and chemical character-
istics may serve as indicators of root functioning, and suites
of root traits have been used to characterize species’ below-
ground resource acquisition strategies. Within root systems,
the acquisitive potential and resource uptake activity generally
decrease from root tips towards the root basis with increas-
ing root age and advancing suberisation (Wells and Eissenstat
2002, Hishi 2007). is advocates for assigning different root
sections along the longitudinal axis to broadly defined root
functional types in order to facilitate comparison among spe-
cies. Traditionally, diameter thresholds (such < 2 mm or <
1 mm) have been used, but an architectural approach based
on root orders has increasingly been employed (Pregitzer et al.
2002, McCormack et al. 2015). In analogy to the leaf econom-
ics spectrum (Wright et al. 2004), a root economics spectrum
(RES) has been proposed which predicts a growth-survival
tradeoff for fine roots (Kong et al. 2014). Accordingly, root
form and function are optimised either for quick growth and
high resource acquisition rates (‘acquisitive strategy’), or for
long lifespans and lower resource uptake rates (‘conservative
strategy’) (Freschet et al. 2010, Reich 2014, Weemstra et al.
2016). In this framework, high tissue density (RTD), larger
fine root diameters (FRD), lower specific root length (SRL)
and lower N concentrations are considered conservative, while
thinner diameters, lower RTD, higher SRL and N concentra-
tions are indicative of an acquisitive root strategy (Reich 2014,
Weemstra et al. 2016, Pierick et al. 2020). Although some
evidence suggests that root trait syndromes of different species
can indeed be arranged one-dimensionally along the acquis-
itive-conservative axis (de la Riva et al. 2018), other results
point towards more complex, multi-dimensional relationships
among root traits (Kramer-Walter et al. 2016, McCormack
and Iversen 2019, Sierra Cornejo et al. 2020). Many studies
have found fine root traits to vary with species turnover along
environmental gradients, and specifically soil fertility gradi-
ents, with abundant resources promoting acquisitive strate-
gies and resource limitation favouring conservative strategies
(Reich 2014, Addo-Danso et al. 2020). Less is known about
possible changes in root strategies along gradients of land-use
intensity, e.g. with the conversion of natural forest to inten-
sively managed agricultural systems (Leuschner et al. 2009,
Prieto et al. 2015, Pransiska et al. 2016). is is especially true
for dynamic fine root functions, such as growth and mortal-
ity rates, and fine root longevity (Han and Zhu 2021). Root
longevity (or lifespan) is a key property which largely deter-
mines the root systems carbohydrate demand and also influ-
ences the C transfer from root systems to the soil upon root
death. Moreover, fine root lifespans influence the age com-
position of root populations, thereby indirectly controlling
the resource uptake capacity of the root system, as younger
roots are usually more active than older ones (Eissenstat and
Yanai 1997, McCormack et al. 2012). However, studies on
fine root lifespan have mostly been conducted in temperate or
boreal forests or grasslands with strong seasonality in tempera-
ture, where frequently one or two root growth peaks per year
have been observed (Norby et al. 2004, Noguchi et al. 2005,
Withington et al. 2021). Much less is known about fine root
dynamics in hot and moist tropical climates, where root turn-
over is thought to be much higher (Lauenroth and Gill 2003).
Human population growth and increased resource
demand drive rapid land-use intensification in many tropi-
cal regions, causing large-scale forest conversion to agricul-
tural systems (Lambin et al. 2003, Lambin and Meyfroidt
2011, Davis et al. 2020) with far-reaching consequences
for biodiversity, ecosystem carbon and nutrient stores, and
regional climate and hydrological systems (Tilman et al.
2001, Fitzherbert et al. 2008, Findell et al. 2017, Houghton
and Nassikas 2017). In the lowlands of Indonesia and other
Southeast Asian countries, the rapid expansion especially
of tree plantations such as oil palm and rubber monocul-
tures has caused significant deforestation in the last decades
(Ziegler et al. 2012, Margono et al. 2014, Curtis et al. 2018,
Austin et al. 2019). Tropical forests differ from the agricul-
tural replacement systems in many functional aspects, nota-
bly aboveground productivity, water consumption, soil C
sequestration, decomposition rate, soil fertility and ecosys-
tem resilience (Allen et al. 2015, Kotowska et al. 2015, van
Straaten et al. 2015, Guillaume et al. 2018, Krashevska et al.
2018, Röll et al. 2019). e consequences of a transition
between land use systems or profound plant community
change on root system dynamics are not well studied, largely
due to the methodological challenges associated with this
task (Vogt et al. 1996, Clark et al. 2001).
With the aim to quantify fine root production, mortal-
ity and lifespan across a transformation landscape in the
lowlands of central Sumatra (Indonesia), we applied three
different techniques (ingrowth cores, sequential coring,
mini-rhizotrons) to measure root production synchronously
in stands of natural forest, jungle rubber (rubber trees under
3
natural forest tree cover), rubber and oil palm monoculture.
We further analysed fine root morphology in these systems
to characterize changes in fine root trait syndromes with
land-use change from tropical forest to intensively man-
aged agricultural systems. Tropical lowland forest productiv-
ity has frequently been found to be limited by phosphorus
(and nitrogen) deficiency (Vitousek and Sanford 1986,
Turner et al. 2018). In fact, late-successional forest species
are reported to show rather conservative shoot and root traits
(Caplan et al. 2019, Hogan et al. 2020), while higher soil fer-
tility in managed systems is expected to favor acquisitive root
strategies. erefore, we expected that land-use change from
natural forest to oil palm or rubber plantations 1) shifts the
community-level means of fine root functional traits from
more conservative to more acquisitive strategies, with roots in
intensively managed tree plantations exhibiting smaller fine
root diameters, higher specific root lengths and nitrogen con-
centrations and lower tissue density. We further postulated
that, as a consequence, 2) fine root production increases, and
average root longevity decreases in tree plantations, 3) lead-
ing to a more pronounced temporal variability of fine root
growth and mortality, as root functional diversity is reduced
in comparison to the forest.
Material and methods
Study area
e study was conducted in the lowlands of Jambi Province,
Sumatra (Indonesia). Once covered by dipterocarp-rich
tropical rainforest, over the last decades, large areas of the
province have been transformed into agricultural systems.
We investigated four major land-use types, namely natural
forest, extensively-farmed mixed rubber plantations (rubber
trees planted under natural forest tree cover, hereafter called
‘jungle rubber’), rubber Hevea brasiliensis and oil palm Elaeis
guineensis monoculture plantations. e research was con-
ducted within the framework of the collaborative research
project ‘EFForTS’ (Ecological and Socioeconomic Functions
of Tropical Lowland Transformation Systems, Sumatra,
Indonesia; Drescher et al. 2016).
irty-two 50 × 50 m plots, i.e. four replicates of each
land-use system in each landscape, were established in
the Harapan forest region (01°.47'24'S, 103°16'48''E –
02°11'24''S, 103°20'24''E) and the Bukit Duabelas region
(01°56'24''S, 102°34'48''E – 02°08'24''S, 102°50'60''E),
ranging in elevation from 40 to 100 m a.s.l. Mean annual
rainfall during the study period (2013–2014) was 2609 mm
at the Jambi city climate station, while the 30-year long-term
average of the two investigated landscapes ranged from 2567
mm year1 in the Harapan region to 2902 mm year1 in the
Bukit Duabelas region (Worldclim database: <www.world-
clim.org>). Even though precipitation patterns vary between
years, May to August usually receive lower precipitation than
the season between October and April. Air temperature is
relatively constant throughout the year with an annual aver-
age of 26.7°C.
e plots were selected to be representative in terms of
stand age and management of the respective land-use systems
and were comparable with respect to soil (Allen et al. 2015),
topographic and climatic conditions (Table 1). Natural rain-
forest plots with closed canopy cover and at least 200 m dis-
tance to visible disturbances were chosen as reference sites. e
natural forest plots in both investigated landscapes are most
likely remnants of the primary forest that received protection
Table 1. Stand structural properties and soil variables from 0 to 50 cm soil depth in the four land-use systems studied (natural forest, jungle
rubber, rubber monocultures and oil palm plantations) (mean ± SD, n = 8 plots per system). Canopy height is the 95% quantile of trees with
dbh > 10 cm. Average daily maximum soil temperature was measured in 5 cm soil depth. *Oil palm trunk diameter measured including
frond stumps.
Units Forest Jungle rubber Rubber Oil palm
Stand structure
Stem density n ha1571 ± 110 710 ± 178 499 ± 155 137 ± 11
Basal area m2 ha129.7 ± 2.5 18.6 ± 3.1 10.9 ± 2.7 65.8 ± 11.6*
Stem diameter cm 22.0 ± 1.4 17.0 ± 1.0 16.5 ± 2.3 76.5 ± 4.6*
Canopy height m 32.3 ± 2.9 18.2 ± 1.2 16.3 ± 0.9 5.7 ± 1.2
AGB Mg ha1316.2 ± 55.0 112.2 ± 11.9 59.8 ± 17.9 78.6 ± 14.6
NPPwood Mg ha1 year19.3 ± 2.1 6.9 ± 1.4 5.7 ± 1.6 7.4 ± 2.1
NPPlitter Mg ha1 year19.0 ± 2.1 7.7 ± 1.1 3.8 ± 0.6 6.3 ± 1.6
Soil variables
Clay content % 29 ± 7 39 ± 17 40 ± 12 47 ± 16
Bulk density g cm31.14 ± 0.16 1.08 ± 0.15 1.24 ± 0.15 1.21 ± 0.18
Soil temperature °C 26.3 ± 0.9 27.0 ± 1.0 27.5 ± 0.9 27.6 ± 1.1
pH 1:4 H2O 4.32 ± 0.08 4.57 ± 0.10 4.46 ± 0.10 4.48 ± 0.06
Soil organic C kg C m27.83 ± 2.77 10.85 ± 3.93 7.49 ± 2.90 8.16 ± 2.64
Total N g N m2676.14 ± 379.98 867.13 ± 315.00 640.51 ± 237.24 662.82 ± 197.21
Soil C:N g g112.00 ± 2.19 11.95 ± 1.86 11.51 ± 1.63 12.12 ± 1.37
δ15N 6.32 ± 0.57 6.54 ± 0.39 6.43 ± 0.80 6.95 ± 0.51
Extractable P g P m21.85 ± 0.82 1.91 ± 1.67 0.97 ± 0.45 2.37 ± 1.43
ECEC mmolc kg1194.22 ± 187.98 259.69 ± 215.37 181.83 ± 98.27 165.93 ± 73.47
Base saturation % 10.90 ± 3.76 10.53 ± 4.43 9.76 ± 2.71 16.12 ± 2.93
4
status in 2000 (Bukit Duabelas) and 2007 (Harapan), but
signs of selective logging activities and extraction of non-
timber forest products are visible. In the plantations, the
oil palms had an age of 8–15 years at study onset, while the
rubber trees were 7–16 years old. Rubber trees in the jungle
rubber system were between 15 and 40 years old. e rubber
trees in our study were tapped regularly and latex harvested
every 6–10 days except for the very dry months July and early
August 2013. e management of the smallholder planta-
tions was conducted with intensities typical for the respec-
tive transformation system and included the application of
herbicides (often Gramoxone) and the addition of 100–200
kg ha1 mineral NPK complete fertilizer, potassium chloride
(KCl) and/or urea (CO(NH2)2) once a year in the rubber
monocultures and of 150–300 kg ha1 in the oil palm plan-
tations twice a year (Allen et al. 2015, Euler et al. 2016). In
case of oil palm plantations, the application of both fertilizer
and herbicides typically is restricted to circular areas in close
proximity of the palm stems. Additionally, the stacking of cut
palm fronds in lines along the palm rows increases spatial het-
erogeneity in soil conditions across the oil palm plantations
(Formaglio et al. 2021). e jungle rubber plots were in most
cases neither fertilized nor systematically treated with herbi-
cides. In both study landscapes, Acrisols were the prevailing
soil types. In the Harapan region, the dominant soil type
according to the WRB classification (2014) was sandy loamy
Acrisol, whereas in Bukit Duabelas the major soil type was
clayey Acrisol. Soil physical and chemical properties down to
50 cm depth (presented in Table 1) were collected across the
32 plots between June 2013 and December 2013 and anal-
ysed following commonly used protocols, described in detail
in Allen et al. (2015, 2016) and Kurniawan et al. (2018). In
general, the soil nutrient status was comparable across the
four land-use systems, with levels of soil N, extractable P and
exchangeable bases in the plantation systems augmented by
the management practices (i.e. fertilization, liming, biomass
burning, etc.) applied in these systems (Allen et al. 2016).
Fine root biomass, morphology and chemistry
Root sampling for fine root morphology and biomass was
conducted from October 2013 to March 2014 in all 32
plots of the four land-use systems at 16 coring locations per
plot using a random grid map. Soil cores were taken with
a split-tube steel corer 3.5 cm in diameter in 0–10, 10–30
and 30–50 cm depth. e soil samples were stored in plastic
bags at 5°C and processed within six weeks in the laboratory
of the University of Jambi. For determining fine root mass
(defined as roots with a diameter < 2 mm), the samples were
soaked in water and cleaned from soil residues by washing the
root material twice with sieves of 1 and 0.2 mm mesh size.
Fine root fragments of woody species longer than 1 cm were
extracted manually with tweezers and separated under a ste-
reomicroscope into live (biomass) and dead (necromass) frac-
tions (Kotowska et al. 2015). Roots of herbaceous plants were
separated in all systems from those of woody plants and omit-
ted from further analysis. We were able to reliably distinguish
rubber tree roots from the roots of other tree species using
visual criteria such as periderm colour and branching pat-
terns. is was relevant in the jungle rubber system, where
rubber grew in mixture with forest trees. e high tree species
diversity in our jungle rubber and natural forest plots ren-
dered it impractical to separate the forest tree roots by species.
All living fine roots from each soil core were then subjected
to morphological analysis. e fine root samples were placed
in a flat transparent container filled with purified water and
scanned for the analysis of root segment length, projected sur-
face area and root diameter using a modified flat-bed scanner.
e scans were then analyzed with WinRHIZO 2005c soft-
ware (Régent Instruments Inc., Québec, QC, Canada). From
the obtained data and root dry weight, specific root length
(SRL, m g1), specific root area (SRA, cm2 g1), root tissue
density (RTD, g cm3) and mean root diameter (FRD, mm)
of the fine root sample were calculated. After morphological
investigation, the living root material was dried at 70°C for
48 h and ground with a ball mill. e concentrations of C
and N were determined by gas chromatography with a C/N-
elemental analyzer.
Fine root longevity and growth dynamics
In situ observation of fine root longevity and growth dynam-
ics has recently been improved by the introduction of visual
processing tools, automated root detection algorithms and
machine learning (Johnson et al. 2001, Zeng et al. 2008,
Yu et al. 2020). In October 2012, we installed a total of
64 mini-rhizotrons (16 tubes in one plot of each land-use
system) vertically in the soil down to 35 cm. Transparent
butyrate tubes with an inner diameter of 6 cm and a wall
thickness of 0.5 cm were used. Image acquisition started ten
months later in August 2013. e top part of the tubes was
covered with black adhesive foil and sealed with a waterproof
removable lid. To ensure that water does not flow along the
tube and create preferential conditions for root growth at the
tube surface, a 5 cm-wide foil manchette was placed around
each tube on the soil surface. e tubes were scanned once a
month with a mobile full rotation in-situ scanner system for
12 consecutive months until July 2014.
e images were analyzed using the WinRHIZO Tron MF
2012a software ). From each of the full 360-degree images,
two 10 × 10 cm squares were framed at exactly the same
location for each consecutive scan to obtain separate data for
0–10 and 10–20 cm soil depth. Each visible fine root seg-
ment was marked with its length and diameter and classified
as alive, dead or disappeared with values adjusted for each ses-
sion in case changes occurred. For calculating total fine root
growth and mortality, the length of all newly appeared or
elongated root segments (= growth) and disappeared or dead
root segments (= mortality) in each observation interval were
summed up. In addition, we computed relative root growth
by relating measured values to the initial root length visible in
the mini-rhizotron window. While roots of non-woody plants
such as grasses were excluded from further rhizotron analysis,
it was not feasible to distinguish between the roots of rubber
5
trees and other woody species in the images. Live and dead
fine roots were separated on the basis of color and root peri-
derm morphology. e color of living oil palm and rubber
fine roots ranges from white to beige, while dead roots turn to
darker brownish and blackish shades. In the forest and jungle
rubber systems, a variety of tree species follows similar color
changes and a slow disintegration and decomposition of roots
is typically observable. Herbivory was assumed when roots
disappeared from the alive root class between two subsequent
sessions and no changes were observed in the surrounding
soil. Affected roots were included in the dead fine root class.
If a fine root was still alive at the end of the data collection
period, it was marked as right-censored. Roots present at the
first scan were excluded from the survival analysis, as their
age was unknown (left-censored). Individual root lifespan
was expressed in days by calculating the period between the
assumed date of death and the date of first observation; the
dates of first observation and death were assumed to be the
mid-point between two subsequent scanning dates.
Fine root production
To determine fine root production per ground area and year,
we used data from the mini-rhizotron observations as well
as calculations from sequential coring and an ingrowth-core
approach. e latter approach quantifies the regrowth of fine
roots into site-specific soil from which all root mass has been
removed (Powell and Day 1991, Majdi et al. 2005). In our
study sites, 16 ingrowth cores per plot were installed down
to 30 cm depth in March and April 2013, following a ran-
dom grid map at least 40 cm distant to other installed equip-
ment such as litter traps. Initially, soil cores were extracted
with a soil corer (3.5 cm in diameter), and the soil material
cleaned by hand from all visible live and dead rootlets (>
1 cm length). To mark the location of the core and facili-
tate precise re-sampling, a rubberized mesh-wire tube with
1-cm spacing was inserted. Afterwards, the root-free soil was
refilled into the hole taking care to maintain its structure and
bulk density as well as possible. Re-sampling of the cores was
carried out after 234–269 days. e extracted soil cores were
divided into the 0–10 and 10–30 cm layers, and processed
in the same manner as done in the fine root inventory in the
laboratory. e fine root biomass production in the cores was
expressed in g dry mass produced per m2 soil surface area and
year to obtain data on annual fine root production, following
Vogt et al. (1996). We assumed that fine root re-colonization
started immediately after the installation of the ingrowth
cores based on observations from earlier ingrowth core and
mini-rhizotron studies in tropical forests and assuming that
root mass losses due to root death and subsequent decompo-
sition during the experiment were minimal (Harteveld et al.
2007, Leuschner et al. 2013).
e sampling campaigns for the sequential coring
approach took place in April 2013, July 2013, October 2013
and January 2014 and followed the approach described by
Vogt and Persson (1991). Due to the large number of sam-
ples and long processing time, only one representative plot
per system was sampled for this methodology, which was the
same plot set used for investigating the fine root dynamics
using mini-rhizotrons. Per plot and campaign, 16 vertical
soil cores (3.5 cm in diameter, 0–10, 10–30 and 30–50 cm
depth) were sampled at 40 cm distance to ingrowth cores to
avoid disturbance. Repeated coring took place in 20 cm dis-
tance to the previous coring, yielding a quadrangular sam-
pling scheme with four cores. e soil cores were then stored
and fine roots processed in the same manner as done for the
fine root inventory (Kotowska et al. 2016).
Annual fine root production was then estimated using
these three different methodologies and several calculation
approaches. For data obtained via sequential coring, i.e. a
series of consecutive measurements of fine root biomass and
necromass, the decision matrix (Fairley and Alexander 1985)
and the ‘minimum–maximum method’ (McClaugherty et al.
1982) were applied. Using the simplified decision matrix
(Supporting information) developed by Yuan and Chen
(2013), we considered all temporal differences in root bio-
mass and necromass between sampling dates, while in the
minimum–maximum method, only the difference between
the highest and lowest biomass per sampling point was taken
into account. Ingrowth cores were used to obtain an estimate
of annual fine root production as the amount of roots growing
into root-free soil after disturbance (Kotowska et al. 2015).
Finally, we used mini-rhizotron observation data to calculate
production rates with two different approaches, 1) based on
fine root lifespan by multiplying turnover (1/median lifes-
pan) with maximal fine root biomass per sampling location,
and 2) based on relative root growth at each tube (cm cm1
year1) divided by SRL and multiplied by average fine root
lengths per hectare in the investigated soil horizons.
Statistical analysis
e influence of land-use system on fine root morphological
traits and root tissue C:N ratio was tested separately for each
variable using linear mixed effects models based on R package
lme4 (ver. 1.1-23, Bates et al. 2015) with system as a fixed
and plot as a random effect. Subsequently, multiple compari-
son tests between group means were analyzed post hoc with
Tukey HSD tests using the multcomp package with the glht
function (ver. 1.4-13, Hothorn et al. 2008). Normality and
homoscedasticity of residuals were assessed based on residual
diagnostic plots. A significance level of p < 0.05 was used
throughout all statistical tests.
A principal components analysis (PCA) was performed
with log-transformed values of fine root biomass, the four
root traits (RTD, FRD, SRL, SLA), and C and N content
using the function prcomp. e input variables were cen-
tered and scaled. Root lifespan data were not included in
the PCA due to the different resolution of trait data from
fine root inventories and data from visual mini-rhizotron
observations. Furthermore, the statistical significance of mul-
tivariate differences between the root trait combinations of
different land-use types was assessed using simulation-based
analyses of similarity based on Euclidean distances between
6
scaled, centered and natural log-transformed traits apply-
ing the anosim function from the R package vegan ver. 2.5-7
(Oksanen et al. 2020).
Fine root lifespans were calculated based on median lon-
gevity using the survfit function from the survival package
(ver. 3.1-8, erneau 2020), and survival curves were plotted
according to the Kaplan–Meier method (Kaplan and Meier
1958, Andersson and Majdi 2005, Ding et al. 2019) using
the survminer package (ver. 0.4.6, Kassambara et al. 2019).
Concurrently, we performed Cox hazard regression analysis
to predict the mortality risk in dependence of system, soil
depth and root length using the coxph function from the sur-
vival package (ver. 3.1-8, erneau 2020). For the fine root
growth and mortality model, we described root growth (i.e.
root length increment per unit time) by a two-part hurdle
model with a probability ϕ of observing a length of zero and
a log-linear mixed effects model component for the non-zero
observations, as the root growth data contained a large num-
ber of zeros that were driven by limitations to the accuracy
of the measurement process. e non-zero observations were
modelled as a function of soil depth and land-use system,
while permitting random variation between tubes and both
overall and plot-specific random time effects. e hurdle
component of the model was fit with a generalized linear
model using the base R function glm, while the component
for the non-zero observations was fit with a mixed effects
model based on the R package lme4 (ver. 1.1-23, Bates et al.
2015). We calculated R2 values using the r.squaredGLMM
function in the MuMIn package (Barton 2020) as: mar-
ginal R2 (R2
mar) which describes the proportion of variance
explained by the fixed factor alone, and conditional R2
(R2
cond) which includes both the fixed and random factors.
We computed the contribution of the different model com-
ponents to the total variance explained by the LME compo-
nent of the hurdle model based on Nakagawa and Schielzeth
(2013). We further tested the influence of distance to the
proximal trees on fine root growth building linear models
using the lm function. All analyses were conducted using R
ver. 3.6.3 (<www.r-project.org>).
Results
Fine root functional traits
Across the four land-use systems, we found similar soil depth-
dependent changes in fine root morphology and chemistry
in forest tree, rubber and oil palm roots between 10 and 50
cm depth (Fig. 1A–D). Yet, fine root morphology differed
considerably between tree types in these land-use systems
(Supporting information). Average fine root diameter (FRD)
per plot ranged from 0.38 ± 0.12 to 0.84 ± 0.14 mm (mean
± SE) in the different systems and tree types, with signifi-
cantly thicker roots in rubber (linear mixed effects model
with Tukey’s contrasts; p < 0.001) than in the other species
(Fig. 1A). Average root tissue density (RTD) ranged between
0.24 ± 0.06 and 0.61 ± 0.12 g m3 in the four systems with
significantly higher values in the trees of the natural forest
and in the jungle rubber plots (mean ± SE: 0.5 ± 0.03 g m3)
as compared to the rubber trees at all soil depths.
Average specific root length (SRL) per plot ranged from
6.55 ± 0.54 m g1 in rubber trees to 22.43 ± 4.1 m g1 in
the forest trees with particularly high (and variable) values in
the forest trees of the jungle rubber system. SRL increased
with soil depth especially in the forest trees, while it remained
relatively constant in rubber trees. Root C:N ratio increased
with depth by 8–10% from 10 to 50 cm in all species. Oil
palm roots had a significantly higher C:N ratio especially in
50 cm depth (65.7 ± 2.17) than the tree roots in the other
systems, while the pattern is reversed in leaves (Supporting
information).
In a principal components analysis (PCA) of community-
level means of morphological and chemical fine root traits of
all land-use systems (Supporting information), we observed
that rubber tree roots clustered prominently along the first
principal component, which was associated with fine root
diameter and SRL and accounted for 36% of variation, while
oil palm roots clustered at the second principal component,
which was primarily explained by total fine root biomass,
RTD and SRA, accounting for 26% of variation. Fine root
nitrogen content (N) and fine root carbon content (C) were
only weakly correlated with the first two principal compo-
nents (Supporting information). We found significant mul-
tivariate differences in the root trait combinations between
the land-use systems (ANOSIM statistic R = 0.505; p <
0.001). Looking at the trait dispersion of only the forest trees
community in jungle rubber and natural forest (Supporting
information), we observed a trend with fine roots of forest
trees growing in jungle rubber plantations showing higher
SRL and SRA values than natural forest trees. However, we
found no significant multivariate differences between the
root trait combinations of forest trees between these two sys-
tems (ANOSIM statistic R: 0.040; Significance: 0.0743).
Fine root lifespan
We constructed Kaplan–Meier fine root survival curves for
the four land-use systems and two soil depths from the mini-
rhizotron data obtained for a twelve-month period (Fig. 2).
Of all 6700 traced roots, 40.2% were identified in oil palm
plantations, 16.4% in monoculture rubber, 22.5% in jungle
rubber and 20.9% in the forest plots. Overall, 29% of the
initially present roots were still alive at the end of the 12-mo
period, while 52% of all traced roots died during the study
period. In 15% of the dead roots, mortality and disintegra-
tion could be observed in the mini-rhizotrons before root
disappearance. Median fine root longevity per plot was 12%
higher in rubber (238 days) than in jungle rubber (211 days),
and 51% greater than in the forest (140 days). With a median
lifespan of 125 days, oil palm exhibited the shortest lifespan
of all systems at 10 cm depth, while lifespan at 20 cm was
shortest in the forest (Table 2). Similarly, at 20 cm depth,
rubber had a longer median lifespan (> 296 days) than the
jungle rubber system (265 days) and oil palm (231 days), and
7
a roughly twice as long lifespan as forest tree roots (154 days).
Fine root longevity increased significantly from 10 to 20 cm
soil depth in all land-use systems (log rank test, p < 0.001).
e Cox proportional hazards analysis showed that land-
use system, soil depth and fine root length significantly
affected the longevity of the fine roots in our sample (Table
3). e mortality hazard ratio (HR) of roots in jungle rub-
ber (HR = 0.59) and rubber plantations (HR = 0.50) was
significantly lower than in natural forest, while roots in oil
palm plantations (HR = 0.93) had a similar hazard ratio.
is implies that single fine roots in jungle rubber and rub-
ber over the study period had on average half the hazard
ratio of roots in the natural forest and oil palm plantations.
Root length and soil depth were positively related to fine
root longevity. Deeper and longer roots tended to show a
lower hazard ratio and thus had a longer lifespan. On aver-
age, every additional cm in root length was associated with
a 19.5% lower hazard ratio. Roots at 20 cm depth had on
average 5.3% lower instantaneous mortality risk than roots
at 10 cm depth.
Figure 1.Box–whisker plots (bold lines: median; boxes: 25th–75th percentile; whiskers: range of the data or single outlier values if exceeding
1.5 interquartile ranges). (A) Average fine root diameter, (B) fine root tissue density, (C) fine root specific root length (SRL), and (D) C:N
ratio of community-level mean values in four tropical land-use systems, namely: forest (F), jungle rubber (J), rubber (R) and oil palm (O)
at three different soil depths (10, 30, 50 cm). Different coloration of the boxes denotes the different tree types – forest trees, rubber trees
and oil palms. Note that in the jungle rubber both rubber and forest trees were present which results in two separate bars in the jungle rub-
ber system. Pairwise significant differences between tree type/land-use system within the depths (p < 0.05) are indicated by different low-
ercase letters (n = 8 plots per system, total of 32 plots).
8
Monthly fine root growth and mortality
Fine root dynamics were observed as monthly root growth
and mortality in the mini-rhizotrons revealing distinct tem-
poral and spatial patterns across land-use systems (Fig. 3A–B).
Considering forest as reference with an average monthly root
growth of 9.93 ± 3.40 cm month1 per 200 cm2 tube sur-
face from 0–20 cm soil depth, oil palm had a significantly
higher growth with 19.84 ± 6.13 cm month1, while jun-
gle rubber and rubber roots did not differ from forest with
11.82 ± 5.50 cm month1 and 8.70 ± 9.66 cm month1,
respectively (Table 4). Soil depth also had an influence on
fine root growth (Table 4). Average fine root growth rate
was about twice as high in 0–10 cm compared to 10–20 cm
depth. Land-use system and depth as fixed effects explained
19.5% of overall variance in the log-transformed growth
data (R2
mar = 0.195, R2
cond = 0.369), while time explained 6%
and the seasonal interaction within systems 10.2% of overall
variance. Root growth tended to be higher during months
with higher precipitation (October–April) than in drier
months (May–August) in all land-use systems (Fig. 3C).
Furthermore, we observed increased fine root growth in the
oil palm plantation after a fertilization event that happened at
the end of April 2014. is effect was not as clearly visible in
the rubber monocultures, where fertilization was conducted
at the beginning of January 2014. In contrast to root growth,
fine root mortality (Fig. 3B) was shifted to the drier months
with peaks occurring from February to April 2014 and again
in May and June 2014, when less than 200 mm and < 100
mm of rain fell, respectively (Fig. 3C). Lower mortality and
disappearance of root segments were recorded in the rubber
plantations and jungle rubber stands compared to the for-
est and oil palm plantations (Table 4). Similar to growth,
mortality was significantly higher in the upper than in the
deeper soil layer, decreasing by 66% from 0–10 to 10–20 cm.
In our model, land-use system and depth explained 22.2%
of overall variance in log-transformed fine root mortality
(R2
mar = 0.2219, R2
cond = 0.3039), while 1.5% of the variance
were related to time and 3.4% to the seasonal interaction
within systems. Mini-rhizotron tube identity, i.e. within-
plot variability, explained only 1.1% of the variance in root
growth and 3.2% of the variance in mortality (Table 4).
e distance to the nearest tree neither influenced initial
fine root length in the images (lm, F-value (3,57) = 6.148;
Figure 2. Plots of Kaplan–Meier estimates of survival probabilities with 95% confidence intervals of fine roots observed in 16 mini-rhizo-
trons within each land-use system (total of 64) namely natural forest, jungle rubber, rubber monocultures and oil palm plantations (see
colour legend) separately for 10 and 20 cm soil depth (left and right panel, respectively) monitored over a period of one year.
Table 2. Median lifespan and 95% confidence intervals and mean lifespan ± SE of fine roots in the four land-use systems for 10 and 20 cm
soil depth. Data were obtained from mini-rhizotron observation of 16 tubes per system.
System Soil depth (cm) Number of roots (n) Death events (n) Median lifespan (95% CI) (days) Mean lifespan± SE (days)
Forest 10 1034 550 140 (122–150) 153 ± 3.45
20 373 175 154 (140–178) 172 ± 5.97
Jungle rubber 10 1098 451 211 (205–239) 192 ± 3.55
20 413 141 265 (239 to >296) 214 ± 4.90
Rubber 10 761 384 238 (212–243) 204 ± 3.56
20 343 112 > 296 238 ± 4.94
Oil palm 10 1627 931 125 (123–146) 148 ± 2.50
20 1074 420 231 (213–243) 204 ± 3.31
9
t-value = 0.460; p > 0.05) nor average root length growth
(F-value (3,57) = 8.119; t-value = 0.892; p > 0.001).
Fine root production estimated with different
methods
Annual fine root production estimates varied markedly
between the land-use systems, but the magnitude of values
and observed temporal and spatial patterns strongly depended
on the technique and calculation method applied (Table 5,
Supporting information). In general, the results from the
mini-rhizotron technique yielded the highest estimates. is
was true for both calculation methods, i.e. the lifespan-based
calculation (all land-use systems) and the approach based
on relative root growth (particularly for the oil palm plan-
tations). Compared to the mini-rhizotron approach, fine
root production derived from sequential coring tended to be
18–52% lower and that from ingrowth cores 40–62% lower
in the land-use systems except for oil palm plantations. In
oil palm plantations, we found overall the largest variance
of production estimates with values differing more than ten-
fold (0.76–23.47 Mg ha1 year1). Consequently, the relative
ranking of systems according to root production along the
transformation systems differed with the methodology. e
scheme based on root growth applied to the mini-rhizotron
data gave the ranking ‘oil palm > rubber > jungle rubber >
forest’, whereas the ingrowth cores suggested a different rank-
ing with ‘jungle rubber > forest > rubber > oil palm’. e
former method gave a similar pattern as the sequential coring
approach, whereas the latter was close to the results of the
lifespan calculation. While the different methods gave largely
different production values for the oil palm plantations, the
general agreement was better for the other land-use systems.
For the other methods, we found fine root production differ-
ing up to fourfold within a given system. Production values
obtained with the decision matrix and the root growth based
on mini-rhizotrons were comparable (average difference
per sampling site: 6.5 ± 15%). Averaging over all methods
yielded highest average fine root production in the oil palm
plantations (8.28 ± 8.9 Mg ha1 year1; mean ± SD) fol-
lowed by jungle rubber (4.72 ± 1.9 Mg ha1 year1), natural
forest (4.54 ± 2.8 Mg ha1 year1) and finally rubber planta-
tions (4.0 ± 2.0 Mg ha1 year1).
Discussion
Consequences of forest conversion for fine root
morphology and chemistry
Conversion of tropical lowland forests to extensively-farmed
tree plantations and intensive monocultures in Sumatra
causes major changes in fine root morphology and below-
ground dynamics of the dominant woody species. Contrary
to our expectation and our first hypothesis, community-level
means of fine root functional traits did not uniformly shift
towards more acquisitive strategies with land-use intensifica-
tion. Limited by light and soil resources, natural forests are
typically characterized by an abundance of slower-growing
species with relatively conservative root traits (Reich 2014,
Addo-Danso et al. 2020), whereas managed systems and sec-
ondary forests tend to be dominated by faster growing spe-
cies that display more acquisitive traits (Wright et al. 2004,
Poorter et al. 2006, Lebrija-Trejos et al. 2010, Carreno-
Rocabado et al. 2016, Hogan et al 2020). We expected that
intensive fertilization with > 100 kg NPK ha1 year1 in the
investigated plantation systems would favor acquisitive root
strategies, while the tree root systems in the natural forest
would exhibit more conservative characteristics, as tropical
lowland forest productivity is frequently limited by phos-
phorus deficiency (Vitousek and Sanford 1986, Turner et al.
2018). Indeed, high N:P ratios in leaf litter and high nutri-
ent use efficiency were observed at our study sites, indicat-
ing strong P limitation in forest and jungle rubber systems
(Kotowska et al. 2016), while δ15N as indicator for N avail-
ability and extractable phosphorus in the soil were increased
in the fertilized monocultures, especially the oil palm plan-
tations, as compared to the forest ecosystems (Allen et al.
2015). However, community-level fine root traits in our for-
est plots were not clearly aligning along the expected acquis-
itive-conservative spectrum. Forest trees had relatively high
tissue densities, but also high specific root length with low
diameters particularly when compared to rubber tree roots.
Recent studies suggest that the traits of tropical forest tree
roots may not always mirror the tradeoff between acquisi-
tive and conservative strategies seen in leaf traits. Rather, their
trait expressions are likely multi-dimensional and at least
partially independent of observed above-ground patterns
(Kramer-Walter et al. 2016, Weemstra et al. 2016, Sierra
Cornejo et al. 2020), potentially with a distinct axis of trait
variation reflecting a gradient of collaboration with mycor-
rhizal fungi (Kong et al. 2019, Bergmann et al. 2020).
e indistinct expression of conservative traits in our for-
est sites could be due to the high species diversity with mul-
tiple contrasting growth strategies, ranging from tall-growing
species of the upper canopy to obligate smaller-sized shade-
tolerant trees of the understory (Rembold et al. 2017). It is
possible that the much higher species diversity in the forests is
Table 3. Results of Cox proportional hazards regression analyses for
fine-root survival time in dependence of land-use systems (jungle rub-
ber (J), oil palm (O), rubber (R) and natural forest (F) as the reference),
soil depth (10, 20 cm) and root length in cm (continuous variable).
Factors
Parameter
estimate SE z p-value
Hazard
ratio
System
Forest (ref.)
Jungle
rubber
0.523 0.055 9.51 < 0.001 0.592
Oil palm 0.072 0.047 1.55 0.168 0.930
Rubber 0.692 0.059 11.69 < 0.001 0.501
Soil depth
10 cm (ref.)
20 cm 0.055 0.004 13.49 < 0.001 0.947
Root length 0.217 0.021 10.25 < 0.001 0.805
10
allowing for complementary use of N compounds in the soil,
thereby increasing N availability in the system (Weigelt et al.
2005). Additionally, selective tree species removal might have
caused changes in the tree species composition. Lowland
forests in Sumatra are typically characterized by their abun-
dance of dipterocarp species (Laumonier et al. 2010), which
form associations with ectomycorrhizal fungi (Moyersoen
2006, Peay et al. 2010). Yet, only 7% of tree individuals in
Figure 3. (A) Monthly fine root growth and (B) monthly fine root mortality in natural forest, jungle rubber, rubber and oil palm plantations
observed within a 10 × 10 cm mini-rhizotron window in two soil depths over a period of one year. Shown are mean values ± SD (n = 16
tubes per system). (C) Monthly rainfall at weather stations located close to the Bukit Duabelas forest and the plantation systems (jungle
rubber, rubber, oil palm) over one year starting from September 2013 to August 2014. October to April were the wetter months, May to
July the drier months.
11
our plots were found to be dipterocarps, which may result
from selective logging in the past, as these species are valuable
timber species. Other target species of selective logging are
usually also characterized by high wood densities and are in
general slow-growing and thus may develop more conserva-
tive root traits which would be missing in the community
after extraction (Bunker et al. 2005, Carreño-Rocabado et al.
2012). On the other hand, when comparing the multivariate
trait space differences of forest trees growing in the natural
forest and in the extensively-managed jungle rubber systems,
where many more fast-growing pioneer species occur, the
trend towards acquisitive strategies was also only marginally
significant.
Accordingly, roots of rubber trees in our study showed
more conservative features with relatively long lifespans, large
diameters and low branching, but low C:N ratios. is is in
line with other studies on rubber roots (Eissenstat et al. 2000,
Withington et al. 2006, Prieto et al. 2015, Pransiska et al.
2016). One likely explanation is that some fine root traits,
particularly those associated with root diameter, are pri-
marily determined by phylogeny, while the environmental
influence on these properties is secondary (Liese et al. 2017,
Valverde-Barrantes et al. 2017, Pierick et al. 2020). Hevea
brasiliensis belongs to the Euphorbiaceae, the species of which
are reported to have relatively large root diameters (Valverde-
Barrantes et al. 2017) as well as low RTD (Pierick et al.
2022). Further, it is likely that cultivated rubber trees are
exposed to additional stress through intensive latex tapping,
which could force the trees to display more conservative
root traits than expected for a tree system well supplied with
Table 4. The effects of land-use type and soil depth on fine root growth (upper part of table) and mortality (lower part of table) in the systems
jungle rubber (J), oil palm (O) and rubber (R) analyzed with log-normal hurdle models with a GLM component for observations equal to
zero and a mixed effects model component for the non-zero observations. Shown are the estimates and likelihood profile-based 95% con-
fidence intervals for the probability of observing a value of zero (ϕ), and the results of the mixed models with system as a fixed and time and
location of mini-rhizotron as random effects. Fixed effects are expressed as differences to ‘forest’ as a baseline; bold face marks fixed effects
coefficients whose 95% confidence intervals exclude zero. Data were obtained from mini-rhizotron observations.
Growth Fixed effects Estimate SE Statistic 2.5% CI 97.5% CI
Intercept 2.11 0.176 12.00 1.77 2.45
J 0.13 0.179 0.75 0.21 0.48
O0.82 0.178 4.61 0.48 1.16
R0.34 0.180 1.90 0.69 0.00
Depth 0.07 0.005 12.20 0.08 0.06
Random effects
Location 0.13 0.05 0.22
Time 0.29 0.10 0.52
System:Time 0.38 0.27 0.49
Residual 0.94 0.91 0.98
Φ0.11 0.09 0.12
Mortality Fixed effects Estimate SE Statistic 2.5% CI 97.5% CI
Intercept 2.67 0.134 19.80 2.41 2.93
J0.27 0.119 2.31 0.50 0.05
O 0.14 0.119 1.20 0.09 0.37
R0.53 0.120 4.44 0.77 0.30
Depth 0.10 0.006 16.50 0.11 0.08
Random effects
Location 0.20 0.12 0.32
Time 0.14 0.00 0.29
System:Time 0.21 0.10 0.29
Residual 0.94 0.90 0.98
Φ0.21 0.19 0.23
Table 5. Annual fine root production estimates (Mg ha1 year1) using data obtained from sequential coring, ingrowth cores and mini-rhizo-
tron observation in natural forest, jungle rubber, rubber monocultures and oil palm plantations for a soil depth of 0–30 cm. Shown are means
± SD of 16 sample locations per system.
Fine root production (Mg ha1 year1) Based on Forest Jungle rubber Rubber Oil palm
Sequential coring
Min–max 2.09 ± 1.08 2.80 ± 1.48 1.66 ± 0.64 3.01 ± 2.13
Decision matrix 4.37 ± 2.04 4.61 ± 1.88 4.04 ± 1.56 6.56 ± 2.74
Ingrowth cores 2.63 ± 1.26 3.16 ± 1.99 2.57 ± 1.15 0.76 ± 0.67
Mini-rhizotrons
Root growth 4.44 ± 2.92 5.49 ± 4.00 6.85 ± 3.69 23.47 ± 19.81
Lifespan 9.18 ± 3.84 7.59 ± 3.74 4.93 ± 1.37 7.64 ± 4.23
12
nutrients. In our study, also oil palm fine roots revealed an
unexpected trait combination with high C:N ratio, dense
tissue and relatively low mean diameters. e fine roots of
the natural forest had a twice as high N concentration as the
oil palm roots (Sahner et al. 2015), which may relate to the
typically higher N content of dicot roots than monocot roots
(Salpagarova et al. 2013). Pena et al. (2013), Pena and Polle
(2014) and Edy et al. (2020) explain the low N content of
oil palm roots with a more rapid plant-internal N translo-
cation to sinks, notably in palm fronds and fruits, and not
with a lower N uptake capacity. In addition, the microbial
community in the intensively managed oil palm plantations
of Sumatra has been found to contain a larger proportion of
pathogenic fungi, while beneficial fungi dominate in the for-
est soils (Brinkmann et al. 2019). is might hamper mineral
nitrogen supply to the oil palm roots in comparison to the
forest with a more abundant and stable microbial community
(Brundrett and Tedersoo 2018).
While higher root tissue density is usually associated with
lower SRL, as greater surface development is linked to young
root segments of low order with lower tissue density and also
shorter lifespan (Weemstra et al. 2020), this relationship is
not supported by our data. Fine roots with higher specific
root length possess a greater absorptive area per unit biomass
invested, enabling faster turnover rates, as thin roots with high
SRL are associated with faster growth and shorter lifespans
(Ostonen et al. 2007, Metcalfe et al. 2008). On the other
hand, specific root length was found to be either unrelated to
root tissue density or positively associated (Holdaway et al.
2011, Kong et al. 2014, Kramer-Walter et al. 2016).
Furthermore, higher SRL is usually also related with higher
water and nutrient uptake capability per root mass (Eissenstat
1992, Reich 2002, Hodge 2004, Prieto et al. 2015), which
suggests that forest conversion to tree plantations in Sumatra
does not increase the specific absorption capacity of the fine
root system. Earlier studies comparing submontane or low-
land forests in Indonesia with cacao, rubber and oil palm
plantations indicated an increase in average fine root diam-
eter with increasing management intensity, but a decrease in
root tissue density (Leuschner et al. 2009, Pransiska et al.
2016). ese observations together with our results add evi-
dence against the assumption that transformation systems are
generally dominated by trees with more acquisitive root traits
and suggest that the trajectory of root trait changes with land-
use intensification in tropical landscapes reflects the species-
specific characteristics in fine root morphology and chemistry
of the tree crop to which the system is converted to.
Consequences of forest conversion for fine root
dynamics
Fine root growth and longevity are thought to be largely under
species-specific endogenous control, i.e. intrinsically deter-
mined by the plant, but they also depend on environmen-
tal changes and resource-availability (Dickmann et al. 1996,
Torreano and Morris 1998, Tierney et al. 2003, Ponti et al.
2004, McCormack et al. 2012, Mao et al. 2013). Our results
show a clear land-use system influence on fine root growth
and mortality patterns as well as root longevity, suggesting
effects of both land-use intensity and species identity on fine
root dynamics. In contradiction to our hypothesis, root lon-
gevity was relatively low in the forest as well as the upper soil
of oil palm plantations and highest in the rubber and jungle
rubber systems. With median lifespans between 125 and 295
days, the fine roots in our systems were shorter-lived than e.g.
temperate deciduous forests (180–900 days, Withington et al.
2006, 220–335 days, McCormack et al. 2013).
According to the Cox hazard regression analysis, rubber
fine roots had a considerably lower instantaneous mortality
risk than oil palm and forest tree roots. e resulting higher
average lifespan of the rubber roots may partly be explained
by the greater average fine root diameter of this species. is
would be in accordance with the prediction of the carbon
optimization theory that the lifespan of an organ has to be
greater when the cost of construction is high in relation to
the cost of maintenance and nutrient uptake (Eissenstat
1992, Eissenstat et al. 2000). In correspondence, the rela-
tively thin oil palm fine roots were rather short-lived in the
upper soil (0–10 cm); yet, they had a much higher tissue
density than the longer-lived rubber roots, suggesting that
other factors than tissue density are decisive for root lifespan
in oil palm. A similar depth dependence of fine root lon-
gevity has also been found in other studies (Gu et al. 2011,
Chen and Brassard 2013, Mao et al. 2013). However, the
opposite, i.e. longer-lived topsoil fine roots as compared to
deeper roots, has also been reported for example in Hevea
brasiliensis (Maegh et al. 2015), which was explained by the
occurrence of thicker roots in the upper soil layers. e upper
soil layers in the palm plantations of our study dry out more
strongly than the deeper soil, as daytime temperatures in the
topsoil are higher (Sabajo et al. 2017, Meijide et al. 2018).
is might increase root respiration and possibly shorten fine
root lifespan (Hendrick and Pregitzer 1993, Pregitzer et al.
2000, Leppälammi-Kujansuu et al. 2014, McCormack and
Guo 2014). On the other hand, a faster fine-root turnover
increases the uptake capacity of fine roots and thus may
enable a more dynamic plant response to changes in nutri-
ent availability (Hodge et al. 2009). is might be beneficial
in spatially heterogeneous soil environments, where nutrients
are quickly depleted in the rhizosphere or nutrient supply
rates vary with soil moisture fluctuation.
In fact, tree root longevity has been found to depend
more on species-specific root anatomical features such as the
existence of thick-walled exodermal cells (Withington et al.
2006) than on root diameter or tissue density, but these fea-
tures have not been studied here. e observed shorter lifes-
pan of the fine roots in the forest as compared to the oil palms
may also relate to the higher root nitrogen content, as organs
higher in nitrogen usually are metabolically more active with
higher respiration costs (Reich et al. 1998, Eissenstat et al.
2000, Green et al. 2005). On the other hand, this factor can-
not explain the high longevity of the relatively N-rich rubber
fine roots. We did not observe a distinct short-term effect of
fertilization on root growth and mortality in the oil palm and
13
rubber plantations, where fertilizer was added once or twice
a year. Such an effect on fine root growth has been observed,
for example, in a fertilized coffee agroforest system in Costa
Rica (van Kanten et al. 2005).
Various climatic and edaphic factors are known to affect
fluctuations in fine root dynamics (Pregitzer et al. 1993,
Tierney et al. 2003), with water availability being a key fac-
tor influencing root growth, longevity and mortality (Deans
1979, Espeleta and Eissenstat 1998, Yavitt and Wright 2001,
Comas et al. 2005). In woody plants of the tropical regions,
root growth usually culminates in the wet season, while root
mortality often peaks during or immediately after the onset
of the dry season, when rainfall decreases (Muñoz and Beer
2001, Green et al. 2005, Kho et al. 2013, Maeght et al.
2015). However, there are also reports that root growth was
enhanced during the dry season (Lima et al. 2010), that both
root growth and mortality culminated during the wet season
(Cordeiro et al. 2020), or that growth was independent of
monthly rainfall (Endo et al. 2019). Our results for all four
land-use systems indicated a root growth stimulation during
the wetter months (October–April) and increased mortality
during the drier season from May to August, in line with
the majority of studies in tropical ecosystems. A rhizotron
study in a rubber plantation in ailand demonstrated a
strong effect of rainfall and associated soil moisture on fine
root elongation with more than 80% of fine roots stopping
growth in the dry season (Chairungsee et al. 2013). e root
growth of rubber trees may also depend on the timing of tap-
ping, i.e. the stimulation of latex flow by removing a thin
slice of bark by the farmer, as carbohydrate allocation to the
root system likely is affected by this disturbance (Silpi et al.
2007, Chairungsee et al. 2013).
Besides environmental influence, fine root growth and
mortality depend on endogenous factors and aboveground
growth, which in turn influences photosynthetic activity and
thus carbohydrate supply to the roots. Studies on rubber
tree seedlings have shown that apical diameter and elonga-
tion rate of roots were depressed during the period of shoot
growth (aler and Pages 1996). Indeed, we have observed
that fine root production peaked just after the rubber trees
had shed most of their leaves and before starting to flush new
leaves, while root mortality increased simultaneously with
leaf shedding during the sampling period. In the forest and
jungle rubber systems, in contrast, leaf litterfall and stem
growth were less variable throughout the year than in the
rubber monoculture, and synchronicity between root growth
and mortality and aboveground phenology was less obvious
(Kotowska et al. 2016). Other studies in tropical forests in
Panama also found root growth to be unrelated to canopy
phenology (Yavitt and Wright 2001).
Despite the importance of fine root dynamics for soil car-
bon input, fine root production in an ecosystem is not exclu-
sively defining the velocity of carbon transfer from dead fine
root mass to the soil, as fine root decomposition rates can
differ. In our study, the natural forest had a lower fine root
production than the oil palm plantations, but the root decom-
position rate was higher (Violita et al. 2015, Krashevska et al.
2018), most likely due to increased palatability and the higher
nutritive value of the relatively N-rich fine roots of the forest
trees. Moreover, the soil microflora has been found to be more
diverse and more active in the forest soil as compared to the
oil palm plantation soil (Krashevska et al. 2015). erefore,
the consequences of forest conversion for long-term carbon
storage in an ecosystem will be impacted not only by root
longevity and production, but most likely by an interplay of
these processes together with soil microbial activities.
Challenges of fine root production estimates
Our results confirm the earlier reported circumstance that
belowground production estimates are significantly influ-
enced by the methodology applied for calculating fine root
production per area (Hertel and Leuschner 2002, Majdi et al.
2005, Jiménez et al. 2009, Moser et al. 2010, Finér et al.
2011), especially when structurally different vegetation types
such as forests and palm stands are compared. We obtained
fine root production figures for the oil palm stands that dif-
fered more than twentyfold between ingrowth core and mini-
rhizotron image approaches when upscaled to the plot level.
e sequential coring technique is usually considered a reli-
able approach to determine the standing fine root biomass
and necromass, but it does not account for simultaneously
occurring root growth, death and decomposition processes
(Vogt et al. 1996, Majdi et al. 2005, Li and Lange 2015).
erefore, sequential coring-based production figures may
underestimate fine root production especially in systems with
high fine root turnover, where growth and mortality peaks
between sampling dates may have been overlooked. Mini-
rhizotrons, in contrast, allow continuous observation of the
growth and death of individual fine roots, while minimiz-
ing soil disturbance (Vogt et al. 1996, Majdi et al. 2005,
Hendricks et al. 2006). However, the measurements are
only reflecting the fine root dynamics on the visible mini-
rhizotron image and not per unit soil volume (Johnson et al.
2001, Li et al. 2020). Upscaling these observations to the
stand area may introduce considerable bias, as it is not
known whether root growth at or near the tube surface is
promoted or impeded. is may at least partly explain the
large deviation in estimated fine root production especially
for the oil palm plantations. Projections of root longevity
and root carbon residence time are based on the assumption
that root dynamics are in equilibrium (Tierney and Fahey
2001, Milchunas et al. 2005), which is often not the case in
agricultural systems such as the oil palm plantations. When
interpreting mini-rhizotron data of fine root mortality, it has
to be kept in mind to consider, over which timespan root
observation takes place, as well as whether roots that have
disappeared are classified as dead or censored (Leppälammi-
Kujansuu et al. 2014). It has been reported that root longev-
ity estimates increased by up to 50% with increases in study
duration (Strand et al. 2008). Moreover, assessing the exact
timing of root death from the images is in most cases prob-
lematic (Comas et al. 2000, Mainiero and Kazda 2006). Fine
root senescence may proceed more or less continuously, and
14
root surface structures and colour are often not indicative of
root functionality and the progress of dieback and decompo-
sition. is makes it likely that roots often are dead before
death is detected by visual inspection in the images.
Oil palm is a highly productive crop which invests
much of its carbon gain in the growth of fronds and fruits
(Kotowska et al. 2015, Röll et al. 2019). Among the four
studied land-use systems, the palm plantations are the sys-
tems with most intense human intervention in terms of her-
bicide application, fertilisation, weeding and frond pruning
(Darras et al. 2019). Oil palm plantations therefore are char-
acterized by a particularly high temporal and spatial hetero-
geneity of the soil, which must influence fine root production
and its temporal variability, and also the reliability of pro-
duction estimates recorded with the different methods. Such
intensively managed, highly productive tree crop systems
belong to the vegetation types for which reliable root pro-
duction data are most difficult to obtain. e presented root
production figures especially of the oil palm plantations must
therefore be regarded with great care, as many steady-state
assumptions underlying root production measurement may
not apply. is suggests that using different methods of fine
root production measurement to identify likely upper and
lower limits of production is a possible option, even though
a ‘gold standard’ for fine root production measurement does
not exist.
Implications of land-use change on belowground
ecosystem functions
Tropical forest conversion to rubber and oil palm plantations
is accompanied by marked alterations in fine root morphol-
ogy, fine root system size and fine root dynamics, with likely
consequences for the root-borne flux of carbon to the soil,
soil biological activity and soil organic matter pools. Our
root-trait analysis does only partially confirm our hypoth-
esis of a shift from conservative to more acquisitive root
traits along the land-use intensity gradient, as community-
weighted root trait means in the forest were in general not
fully matching the expectations of a conservative root trait
spectrum. In addition, the comparison of trees from the nat-
ural forest to those in the extensively managed jungle rubber
system did reveal only a weak shift towards more acquisitive
traits in the latter. Some of the unexpected trends may be
explained by the specific biology of the tree crops involved,
i.e. the rubber tree as a member of Euphorbiaceae and oil
palm as a monocot, with the latter being characterized as a
species with mainly conservative plant traits despite its high
productivity (Rocabado et al. 2016). Hence, forest conver-
sion to intensively managed tree plantations is not necessarily
associated with a shift to a more acquisitive trait combina-
tion, when species with largely different growth strategies
are involved. As fine root turnover and rhizodeposition are
major sources of soil carbon (Rasse et al. 2005), assessing
the consequences for carbon storage and cycling of forest
transformation requires reliable carbon budget data also for
the belowground dynamics of forest and plantation systems
(Kumar et al. 2006, Brunner et al. 2013). is information is
also crucial for ecosystem modelling approaches and carbon
management decisions, in particular as soil carbon dynamics
take place over longer time scales and typically only respond
to land-use changes with some delay (Post and Kwon 2000,
Guo and Gifford 2002). With our study, we contribute to a
better understanding of changes in fine root dynamics with
tropical forest conversion, while also pointing out method-
ological difficulties to reliably assess fine root productivities
in dynamic tropical conversion systems.
Acknowledgements – We thank our local assistants for support with
fieldwork and the village leaders, local plot owners, PT REKI, the
authorities of the Bukit Duabelas National Park, and the Indonesian
Research Foundation (LIPI) for the research permissions. We also
thank Christian Stiegler for providing precipitation data from the
study region. e samples analysed in this study were collected with
the Collection Permit no. 2704/IPH.1/KS.02/X1/2012 issued by
the FRP-Kemenristek.
Funding – is study was financed by the German Research
Foundation (DFG) in the framework of the research platform EFForTS
(CRC990). FB acknowledges funding through the Margarete-von-
Wrangell Fellowship of the Ministry of Science, Research and the Arts
Baden-Württemberg and European Social Fund.
Author contributions
Martyna M. Kotowska: Conceptualization (equal); Data
curation (lead); Formal analysis (lead); Investigation (lead);
Methodology (equal); Project administration (support-
ing); Software (equal); Validation (equal); Visualization
(equal); Writing – original draft (lead); Writing – review
and editing (equal). Sasya Samhita: Data curation (support-
ing); Formal analysis (equal); Methodology (supporting);
Software (equal); Visualization (equal); Writing – original
draft (equal); Writing – review and editing (equal). Dietrich
Hertel: Conceptualization (equal); Funding acquisition
(equal); Methodology (supporting); Project administra-
tion (equal); Resources (equal); Supervision (lead); Writing
– original draft (supporting); Writing – review and edit-
ing (supporting). Triadiati Triadiati: Conceptualization
(supporting); Funding acquisition (equal); Investigation
(supporting); Project administration (equal); Resources
(equal); Supervision (supporting); Writing – original draft
(supporting); Writing – review and editing (supporting).
Friderike Beyer: Data curation (supporting); Formal anal-
ysis (supporting); Methodology (supporting); Validation
(supporting); Writing – original draft (supporting); Writing
– review and editing (supporting). Kara Allen: Data cura-
tion (equal); Formal analysis (supporting); Investigation
(equal); Methodology (supporting); Validation (supporting);
Visualization (supporting); Writing – original draft (support-
ing); Writing – review and editing (supporting). Roman M.
Link: Data curation (supporting); Formal analysis (support-
ing); Methodology (supporting); Software (equal); Validation
(equal); Visualization (equal); Writing – original draft (sup-
porting); Writing – review and editing (equal). Christoph
15
Leuschner: Conceptualization (equal); Funding acquisition
(equal); Project administration (equal); Resources (lead);
Supervision (equal); Writing – original draft (supporting);
Writing – review and editing (equal).
Data availability statement
Data are available from GöttingenResearchOnline Data:
<https://doi.org/10.25625/8PFVPS> (Kotowska et al. 2022).
Supporting information
e Supporting information associated with this article is
available with the online version.
References
Addo-Danso, S. D. et al. 2020. Fine-root morphological trait vari-
ation in tropical forest ecosystems: an evidence synthesis. –
Plant Ecol. 221: 1–13.
Allen, K. et al. 2015. Soil nitrogen-cycling responses to conversion
of lowland forests to oil palm and rubber plantations in Suma-
tra, Indonesia. – PLoS One 10: e0133325.
Allen, K. et al. 2016. Spatial variability surpasses land-use change
effects on soil biochemical properties of converted lowland
landscapes in Sumatra, Indonesia. – Geoderma 284: 42–50.
Andersson, P. and Majdi, H. 2005. Estimating root longevity at sites
with long periods of low root mortality. – Plant Soil 276: 9–14.
Austin, K. G. et al. 2019. What causes deforestation in Indonesia?
– Environ. Res. Lett. 14: 024007.
Bardgett, R. D. et al. 2014. Going underground: root traits as
drivers of ecosystem processes. – Trends Ecol. Evol. 29:
692–699.
Barton, K. 2020. MuMIn: multi-model inference. – R package ver.
1.43.17, <https://CRAN.R-project.org/package=MuMIn>.
Bates, D. et al. 2015. Fitting linear mixed-effects models using
lme4. – J. Stat. Softw. 67: 1–48.
Bergmann J. et al. 2020 e fungal collaboration gradient domi-
nates the root economics space in plants. – Sci. Adv. 6: eaba3756.
Brinkmann, N. et al. 2019. Intensive tropical land use massively
shifts soil fungal communities. – Sci. Rep. 9: 3403.
Brundrett, M. C. and Tedersoo, L. 2018. Evolutionary history of
mycorrhizal symbioses and global host plant diversity. – New
Phytol. 220: 1108–1115.
Brunner, I. et al. 2013. Fine-root turnover rates of European forests
revisited: an analysis of data from sequential coring and ingrowth
cores. – Plant Soil 362: 357–372.
Bunker, D. E. et al. 2005. Species loss and aboveground carbon
storage in a tropical forest. – Science 310: 1029–1031.
Caplan, J. S. et al. 2019. Fine-root traits are linked to species dynam-
ics in a successional plant community. – Ecology 100: e02588.
Carreno-Rocabado, G. et al. 2016. Land-use intensification effects
on functional properties in tropical plant communities. – Ecol.
Appl. 26: 174–189.
Carreño-Rocabado, G. et al. 2012. Effects of disturbance intensity
on species and functional diversity in a tropical forest. – J. Ecol.
100: 1453–1463.
Chairungsee, N. et al. 2013. Impact of tapping and soil water sta-
tus on fine root dynamics in a rubber tree plantation in ai-
land. – Front. Plant Sci. 4: 538.
Chen, H. Y. H. and Brassard, B. W. 2013. Intrinsic and extrinsic
controls of fine root life span. – CRC Crit. Rev. Plant Sci. 32:
151–161.
Clark, D. A. et al. 2001. Net primary production in tropical forests:
an evaluation and synthesis of existing field data. – Ecol. Appl.
11: 371–384.
Comas, L. H. et al. 2000. Assessing root death and root system
dynamics in a study of grape canopy pruning. – New Phytol.
147: 171–178.
Comas, L. H. et al. 2005. Canopy and environmental control of
root dynamics in a long-term study of Concord grape. – New
Phytol. 167: 829–840.
Cordeiro, A. L. et al. 2020. Fine-root dynamics vary with soil depth
and precipitation in a low-nutrient tropical forest in the central
Amazonia. – J. Plant Interact. 1: 3–16.
Curtis, P. G. et al. 2018. Classifying drivers of global forest loss.
– Science 361: 1108–1111.
Darras, K. F. A. et al. 2019. Reducing fertilizer and avoiding her-
bicides in oil palm plantations-ecological and economic valua-
tions. – Front. For. Global Change 2: 65.
Davis, K. F. et al. 2020. Tropical forest loss enhanced by large-scale
land acquisitions. – Nat. Geosci. 13: 482–488.
de Kroon, H. and Visser, E. J. (eds) 2003. Root ecology. – Springer
Science & Business Media, p. 397.
de la Riva et al. 2018. Root traits across environmental gradients
in Mediterranean woody communities: are they aligned along
the root economics spectrum? – Plant Soil 424: 35–48.
Deans, J. D. 1979. Fluctuations of the soil environment and fine
root growth in a young Sitka spruce plantation. – Plant Soil 52:
195–208.
Dickmann, D. I. et al. 1996. Effects of irrigation and coppicing on
above-ground growth, physiology and fine-root dynamics of
two field-grown hybrid poplar clones. – For. Ecol. Manage. 80:
163–174.
Dijkstra, F. A. et al. 2021. Root effects on soil organic carbon: a
double-edged sword. – New Phytol. 230: 60–65.
Ding, Y. et al. 2019. Fine root longevity and below- and above-
ground litter production in a boreal Betula pendula forest. – For.
Ecol. Manage. 431: 17–25.
Drescher, J. et al. 2016. Ecological and socio–economic functions
across tropical land use systems after rainforest conversion. –
Phil. Trans. R. Soc. B 371: 20150275.
Edy, N. et al. 2020. Differences in root nitrogen uptake between
tropical lowland rainforests and oil palm plantations. – Front.
Plant Sci. 11: 92.
Eissenstat, D. M. 1991. On the relationship between specific root
length and the rate of root proliferation: a field study using
citrus rootstocks. – New Phytol. 118: 63–68.
Eissenstat, D. M. 1992. Costs and benefits of constructing roots of
small diameter. – J. Plant Nutr. 15: 763–782.
Eissenstat, D. M. and Yanai, R. D. 1997. e ecology of root
lifespan. – Adv. Ecol. Res. 27: 1–60.
Eissenstat, D. M. et al. 2000. Building roots in a changing environ-
ment: implications for root longevity. – New Phytol. 147: 33–42.
Endo, I. et al. 2019. Spatial and temporal patterns of root dynam-
ics in a Bornean tropical rainforest monitored using the root
scanner method. – Plant Soil 443: 323–335.
Erktan, A. et al. 2018. Frontiers in root ecology: recent advances
and future challenges. – Plant Soil 424: 1–9.
Espeleta, J. F. and Eissenstat, D. M. 1998. Responses of citrus fine
roots to localized soil drying: a comparison of seedlings with
adult fruiting trees. – Tree Physiol. 18: 113–119.
16
Euler, M. et al. 2016. Exploring yield gaps in smallholder oil palm
production systems in eastern Sumatra, Indonesia. – Agric. Syst.
146: 111–119.
Fairley, R. and Alexander, I. 1985. Methods of calculating fine root
pro-duction in forests. – In: Fitter, A. H. et al. (eds), Ecological
interactions in soil: plants, microbes and animals. Blackwell, pp.
37–4.
Findell, K. L. et al. 2017. e impact of anthropogenic land use
and land cover change on regional climate extremes. – Nat.
Commun. 8: 989.
Finér, L. et al. 2011. Fine root production and turnover in forest
ecosystems in relation to stand and environmental characteris-
tics. – For. Ecol. Manage. 262: 2008–2023.
Fitzherbert, E. B. et al. 2008. How will oil palm expansion affect
biodiversity? – Trends Ecol. Evol. 23: 538–545.
Formaglio, G. et al. 2021. Mulching with pruned fronds promotes
the internal soil N cycling and soil fertility in a large-scale oil
palm plantation. – Biogeochemistry 154: 63–80.
Freschet, G. T. et al. 2010. Evidence of the ‘plant economics spec-
trum’ in a subarctic flora. – J. Ecol. 98: 362–373.
Freschet, G. T. et al. 2021. Root traits as drivers of plant and eco-
system functioning: current understanding, pitfalls and future
research needs. – New Phytol. 232: 1123–1158.
Green, J. J. et al. 2005. Fine root dynamics in a tropical rain forest
is influenced by rainfall. – Plant Soil 276: 23–32.
Gu, J. et al. 2011. Influence of root structure on root survivorship:
an analysis of 18 tree species using a minirhizotron method. –
Ecol. Res. 26: 755–762.
Guillaume, T. et al. 2018. Carbon costs and benefits of Indonesian
rainforest conversion to plantations. – Nat. Commun. 9: 2388.
Guo, L. B. and Gifford, R. M. 2002. Soil carbon stocks and land
use change: a meta analysis. – Global Change Biol. 8: 345–360.
Han, M. and Zhu, B. 2021. Linking root respiration to chemistry and
morphology across species. – Global Change Biol. 27: 190–201.
Harteveld, M. et al. 2007 Spatial and temporal variability of fine
root abundance and growth in tropical moist forests and agro-
forestry systems (Sulawesi, Indonesia). – Ecotropica 13: 111–120.
Hendrick, R. L. and Pregitzer, K. S. 1993. e dynamics of fine
root length, biomass and nitrogen content in two northern
hardwood ecosystems. – Can. J. For. Res. 23: 2507–520.
Hendricks, J. J. et al. 2006. Assessing the patterns and controls of
fine root dynamics: an empirical test and methodological
review. – J. Ecol. 94: 40–57.
Hertel, D. and Leuschner, C. 2002. A comparison of four different
fine root production estimates with ecosystem carbon balance
data in a Fagus-Quercus mixed forest. – Plant Soil 239: 237–251.
Hishi, T. 2007. Heterogeneity of individual roots within the fine
root architecture: causal links between physiological and eco-
system functions. – J. For. Res. 12: 126–133.
Hodge, A. 2004. e plastic plant: root responses to heterogeneous
supplies of nutrients. – New Phytol. 162: 9–24.
Hodge, A. et al. 2009. Plant root growth, architecture and function.
– Plant Soil 321: 153–187.
Hogan, J. A. et al. 2020. Morphological variation of fine root sys-
tems and leaves in primary and secondary tropical forests of
Hainan Island, China. – Ann. For. Sci. 77: 1–21.
Holdaway, R. J. et al. 2011. Species- and community-level patterns
in fine root traits along a 120 000-year soil chronosequence in
temperate rain forest: root trait shifts along soil chronosequence.
– J. Ecol. 99: 954–963.
Hothorn, T. et al. 2008. Simultaneous inference in general para-
metric models. – Biometrical J. 50: 346–363.
Houghton, R. A. and Nassikas, A. A. 2017. Global and regional
fluxes of carbon from land use and land cover change
1850–2015: carbon emissions from land use. – Global Biogeo-
chem. Cycles 31: 456–472.
Jackson, R. B. et al. 1997. A global budget for fine root biomass,
surface area and nutrient contents. – Proc. Natl Acad. Sci. USA
94: 7362–7366.
Jiménez, E. M. et al. 2009. Fine root dynamics for forests on con-
trasting soils in the Colombian Amazon. – Biogeosciences 6:
2809–2827.
Johnson, M. G. et al. 2001. Advancing fine root research with
minirhizotrons. – Environ. Exp. Bot. 45: 263–289.
Kaplan, E. L. and Meier, P. 1958. Nonparametric estimation from
incomplete observations. – J. Am. Stat. Assoc. 53: 457–481.
Kassambara, A. et al. 2019. survminer: drawing survival curves
using ‘ggplot2’. – R package ver. 0.4.6, <https://CRAN.R–pro-
ject.org/package=survminer>.
Kho, L. K. et al. 2013. Annual budget and seasonal variation of
aboveground and belowground net primary productivity in a
lowland dipterocarp forest in Borneo: productivity in a Bornean
forest. – J. Geophys. Res. Biogeosci. 118: 1282–1296.
Kong, D. et al. 2014. Leading dimensions in absorptive root trait
variation across 96 subtropical forest species. – New Phytol.
203: 863–872.
Kong, D. et al. 2019. Nonlinearity of root trait relationships and
the root economics spectrum. – Nat. Commun. 10: 2203.
Kotowska, M. M. et al. 2015. Quantifying above- and below-
ground biomass carbon loss with forest conversion in tropical
lowlands of Sumatra (Indonesia). – Global Change Biol. 21:
3620–3634.
Kotowska, M. M. et al. 2016. Conversion of tropical lowland for-
est reduces nutrient return through litterfall and alters nutrient
use efficiency and seasonality of net primary production. –
Oecologia 180: 601–618.
Kotowska, M. M. et al. 2022. Data from: Consequences of tropical
rainforest conversion to tree plantations on fine root dynamics
and functional traits. – GöttingenResearchOnline Datay,
<https://doi.org/10.25625/8PFVPS>.
Kramer-Walter, K. R. et al. 2016. Root traits are multidimensional:
specific root length is independent from root tissue density and
the plant economic spectrum. – J. Ecol. 104: 1299–1310.
Krashevska, V. et al. 2015. Impact of tropical lowland rainforest
conversion into rubber and oil palm plantations on soil micro-
bial communities. – Biol. Fertil. 51: 697–705.
Krashevska, V. et al. 2018. Micro-decomposer communities and
decomposition processes in tropical lowlands as affected by land
use and litter type. – Oecologia 187: 255–266.
Kumar, R. et al. 2006. Plant roots and carbon sequestration. – Curr.
Sci. 91. 885–890.
Kurniawan, S. et al. 2018. Conversion of tropical forests to small-
holder rubber and oil palm plantations impacts nutrient leach-
ing losses and nutrient retention efficiency in highly weathered
soils. – Biogeosciences 15: 5131–5154.
Lambin, E. F. and Meyfroidt, P. 2011. Global land use change,
economic globalization and the looming land scarcity. – Proc.
Natl Acad. Sci. USA 108: 3465–3472.
Lambin, E. F. et al. 2003. Dynamics of land-use and land-cover
change in tropical regions. – Annu. Rev. Environ. Resour. 28:
205–241.
Lauenroth, W. K. and Gill, R. 2003. Turnover of root systems. – In:
de Kroon, H. and Visser, E. J. (eds), Root ecology. Springer,
pp. 61–89.
17
Laumonier, Y. et al. 2010. Landscape-scale variation in the structure
and biomass of the hill dipterocarp forest of Sumatra: implica-
tions for carbon stock assessments. – For. Ecol. Manage. 259:
505–513.
Lebrija-Trejos, E. et al. 2010. Functional traits and environmental
filtering drive community assembly in a species-rich tropical
system. – Ecology 91: 386–398.
Leppälammi-Kujansuu, J. et al. 2014. Fine root turnover and lit-
ter production of Norway spruce in a long-term temperature
and nutrient manipulation experiment. – Plant Soil 374:
73–88.
Leuschner, C. et al. 2009. Consequences of increasing forest use
intensity for biomass, morphology and growth of fine roots in
a tropical moist forest on Sulawesi, Indonesia. – Agric. Ecosyst.
Environ. 129: 474–481.
Leuschner, C. et al. 2013. Conversion of tropical moist forest into
cacao agroforest: consequences for carbon pools and annual C
sequestration. – Agrofor. Syst. 87: 1173–1187.
Li, X. and Lange, H. 2015. A modified soil coring method for
measuring fine root production, mortality and decomposition
in forests. – Soil Biol. Biochem. 91: 192–199.
Li, X. et al. 2020. An improved method for quantifying total fine
root decomposition in plantation forests combining measure-
ments of soil coring and minirhizotrons with a mass balance
model. – Tree Physiol. 40: 1466–1473.
Liese, R. et al. 2017. Root branching is a leading root trait of the
plant economics spectrum in temperate trees. – Front. Plant
Sci. 8: 315.
Lima, T. T. S. et al. 2010. Effects of water and nutrient availability
on fine root growth in eastern Amazonian forest regrowth, Bra-
zil. – New Phytol. 187: 622–630.
Maeght, J.-L. et al. 2015. Seasonal patterns of fine root production
and turnover in a mature rubber tree (Hevea brasiliensis Müll.
Arg.) stand – differentiation with soil depth and implications
for soil carbon stocks. – Front. Plant Sci. 6: 1022.
Mainiero, R. and Kazda, M. 2006. Depth-related fine root dynam-
ics of Fagus sylvatica during exceptional drought. – For. Ecol.
Manage. 237: 135–142.
Majdi, H. et al. 2005. Measuring fine root turnover in forest eco-
systems. – Plant Soil 276: 1–8.
Mao, Z. et al. 2013. Modelling root demography in heterogeneous
mountain forests and applications for slope stability analysis.
– Plant Soil 363: 357–382.
Margono, B. A. et al. 2014. Primary forest cover loss in Indonesia
over 2000–2012. – Nat. Clim. Change 4: 730–735.
McClaugherty, C. A. et al. 1982. e role of fine roots in the
organic matter and nitrogen budgets of two forested ecosys-
tems. – Ecology 63: 1481–1490.
McCormack, M. et al. 2013. Regional scale patterns of fine root
lifespan and turnover under current and future climate. –
Global Change Biol. 19: 1697–1708.
McCormack, M. L. and Guo, D. 2014. Impacts of environmental
factors on fine root lifespan. – Front. Plant Sci. 5: 205.
McCormack, M. L. and Iversen, C. M. 2019. Physical and func-
tional constraints on viable belowground acquisition strategies.
– Front. Plant Sci. 10: 1215.
McCormack, M. L. et al. 2012. Predicting fine root lifespan from
plant functional traits in temperate trees. – New Phytol. 195:
823–831.
McCormack, M. L. et al. 2015. Redefining fine roots improves
understanding of below-ground contributions to terrestrial bio-
sphere processes. – New Phytol. 207: 505–518.
Meijide, A. et al. 2018. Impact of forest conversion to oil palm and
rubber plantations on microclimate and the role of the 2015
ENSO event. – Agric. For. Meteorol. 252: 208–219.
Metcalfe, D. B. et al. 2008. e effects of water availability on root
growth and morphology in an Amazon rainforest. – Plant Soil
311: 189–199.
Milchunas, D. G. et al. 2005. Root dynamics and demography in
shortgrass steppe under elevated CO2, and comments on minirhi-
zotron methodology. – Global Change Biol. 11: 1837–1855.
Mommer, L. et al. 2016. Root–root interactions: towards a rhizo-
sphere framework. – Trends Plant Sci. 21: 209–217.
Moser, G. et al. 2010. Biomass and productivity of fine and coarse
roots in five tropical mountain forests stands along an altitudi-
nal transect in southern Ecuador. – Plant Ecol. Divers. 3:
151–164.
Moyersoen, B. 2006. Pakaraimaea dipterocarpacea is ectomycorrhi-
zal, indicating an ancient Gondwanaland origin for the ecto-
mycorrhizal habit in Dipterocarpaceae. – New Phytol. 172:
753–762.
Muñoz, F. and Beer, J. 2001. Fine root dynamics of shaded cacao
plantations in Costa Rica. – Agrofor. Syst. 51: 119–130.
Nakagawa, S. and Schielzeth, H. 2013. A general and simple
method for obtaining R2 from generalized linear mixed-effects
models. – Methods Ecol. Evol. 4: 133–142.
Noguchi, K. et al. 2005. Estimating the production and mortality
of fine roots in a Japanese cedar (Cryptomeria japonica D. Don)
plantation using a minirhizotron technique. – J. For. Res. 10:
435–441.
Norby, R. J. et al. 2004. Fine-root production dominates response
of a deciduous forest to atmospheric CO2 enrichment. – Proc.
Natl Acad. Sci. USA 101: 9689–9693.
Oksanen, J. F. et al. 2020. vegan: community ecology package. R pack-
age ver. 2.5-7. – <https://CRAN.R-project.org/package=vegan>.
Ostonen, I. et al. 2007. Specific root length as an indicator of
environmental change. – Plant Biosyst. 141: 426–442.
Peay, K. G. et al. 2010. Potential link between plant and fungal
distributions in a dipterocarp rainforest: community and phy-
logenetic structure of tropical ectomycorrhizal fungi across a
plant and soil ecotone. – New Phytol. 185: 529–542.
Pena, R. and Polle, A. 2014. Attributing functions to ectomycor-
rhizal fungal identities in assemblages for nitrogen acquisition
under stress. – ISME J. 8: 321–330.
Pena, R. et al. 2013. Interspecific temporal and spatial differences
in the acquisition of litter-derived nitrogen by ectomycorrhizal
fungal assemblages. – New Phytol. 199: 520–528.
Pierick, K. et al. 2020. Topography as a factor driving small-scale
variation in tree fine root traits and root functional diversity in
a species-rich tropical montane forest. – New Phytol. 230:
129–138.
Ponti, F. et al. 2004. Fine root dynamics of pedunculate oak and
narrow-leaved ash in a mixed-hardwood plantation in clay soils.
– Plant Soil 259: 39–49.
Poorter, L. et al. 2006. Architecture of 54 moist-forest tree species:
traits, tradeoffs and functional groups. – Ecology 87: 1289–1301.
Post, W. M. and Kwon, K. C. 2000. Soil carbon sequestration and
land-use change: processes and potential: soil carbon sequestra-
tion and land-use change. – Global Change Biol. 6: 317–327.
Powell, S. W. and Day, F. P. 1991. Root production in four com-
munities in the great dismal swamp. – J. Bot. 78: 288.
Pransiska, Y. et al. 2016. Forest conversion impacts on the fine and
coarse root system, and soil organic matter in tropical lowlands
of Sumatera (Indonesia). – For. Ecol. Manage. 379: 288–298.
18
Pregitzer, K. S. et al. 1993. e demography of fine roots in response
to patches of water and nitrogen. – New Phytol. 125: 575–580.
Pregitzer, K. S. et al. 2000. Responses of tree fine roots to tem-
perature: review responses of fine roots to temperature. – New
Phytol. 147: 105–115.
Pregitzer, K. S. et al. 2002. Fine root architecture of nine North
American trees. – Ecol. Monogr. 72: 293–309.
Prieto, I. et al. 2015. Root functional parameters along a land-use
gradient: evidence of a community-level economics spectrum.
– J. Ecol. 103: 361–373.
Rasse, D. et al. 2005. Is soil carbon mostly root carbon? Mecha-
nisms for specific stabilization. – Plant Soil 269: 341–356.
Reich, P. 2002. Root–shoot relations: optimality in acclimation and
adaptation or the ‘Emperor’s new clothes’? – In: Waisel, R. et al.
(eds), Plant roots – the hidden half. Marcel Dekker Inc.,
pp. 314–333.
Reich, P. B. 2014. e world-wide ‘fast–slow’ plant economics spec-
trum: a traits manifesto. – J. Ecol. 102: 275–301.
Reich, P. B. et al. 1998. Close association of RGR, leaf and root
morphology, seed mass and shade tolerance in seedlings of nine
boreal tree species grown in high and low light: RGR and tissue
morphology in boreal trees. – Funct. Ecol. 12: 327–338.
Rembold, K. et al. 2017. Plant diversity, forest dependency and
alien plant invasions in tropical agricultural landscapes. – Biol.
Conserv. 213: 234–242.
Reubens, B. et al. 2007. e role of fine and coarse roots in shallow
slope stability and soil erosion control with a focus on root
system architecture: a review. – Trees 21: 385–402.
Röll, A. et al. 2019. Transpiration on the rebound in lowland Suma-
tra. – Agric. For. Meteorol. 274: 160–171.
Sabajo, C. R. et al. 2017. Expansion of oil palm and other cash
crops causes an increase of the land surface temperature in the
Jambi province in Indonesia. – Biogeosciences 14: 4619–4635.
Sahner, J. et al. 2015. Degradation of root community traits as
indicator for transformation of tropical lowland rain forests into
oil palm and rubber plantations. – PLoS One 10: e0138077.
Salpagarova, F. S. et al. 2013. Fine root nitrogen contents and morpho-
logical adaptations of alpine plants. – Zh. Obshch. Biol. 74: 190–200.
Sierra Cornejo, N. et al. 2020. Biomass, morphology and dynamics
of the fine root system across a 3000-m elevation gradient on
Mt. Kilimanjaro. – Front. Plant Sci. 11: 13.
Silpi, U. et al. 2007. Carbohydrate reserves as a competing sink:
evidence from tapping rubber trees. – Tree Physiol. 27: 881–889.
Strand, A. E. et al. 2008. Irreconcilable differences: fine-root life
spans and soil carbon persistence. – Science 319: 456–458.
aler, P. and Pages, L. 1996. Periodicity in the development of the
root system of young rubber trees (Hevea brasiliensis Muell.
Arg.): relationship with shoot development. – Plant Cell Envi-
ron. 19: 56–64.
erneau, T. M. et al. 2020. A package for survival analysis in R.
– R package ver. 3.1–12, <https://CRAN.R–project.org/
package=survival>.
Tierney, G. and Fahey, T. 2001. Evaluating minirhizotron estimates
of fine root longevity and production in the forest floor of a
temperate broadleaf forest. – Plant Soil 229: 167–176.
Tierney, G. et al. 2003. Environmental control of fine root dynam-
ics in a northern hardwood forest. – Global Change Biol. 9:
670–679.
Tilman, D. et al. 2001. Forecasting agriculturally driven global
environmental change. – Science 292: 281–284.
Torreano, S. J. and Morris, L. A. 1998. Loblolly pine root growth
and distribution under water stress. – Soil Sci. Soc. J. 62:
818–827.
Turner, B. L. et al. 2018. Pervasive phosphorus limitation of tree
species but not communities in tropical forests. – Nature 555:
367–370.
Valverde-Barrantes, O. J. et al. 2017. A worldview of root traits:
the influence of ancestry, growth form, climate and mycorrhizal
association on the functional trait variation of fine-root tissues
in seed plants. – New Phytol. 215: 1562–1573.
van Kanten, R. et al. 2005. Fine-root dynamics of coffee in asso-
ciation with two shade trees in Costa Rica. – Agrofor. Syst. 63:
247–261.
van Straaten, O. et al. 2015. Conversion of lowland tropical forests
to tree cash crop plantations loses up to one-half of stored soil
organic carbon. – Proc. Natl Acad. Sci. USA 112: 9956–9960.
Violita, V. et al. 2015. Transformation of lowland rainforest into
oil palm plantations results in changes of leaf litter production
and decomposition in Sumatra, Indonesia. – J. Biodivers. Envi-
ron. Sci. 6: 546–556.
Vitousek, P. M. and Sanford Jr., R. L. 1986. Nutrient cycling in
moist tropical forest. – Annu. Rev. Ecol. Syst. 17: 137–167.
Vogt, K. A. et al. 1996. Review of root dynamics in forest ecosystem
grouped by climate, climatic forest type and species. – Plant
Soil 187: 159–219.
Weemstra, M. et al. 2016. Towards a multidimensional root trait
framework: a tree root review. – New Phytol. 211: 1159–1169.
Weemstra, M. et al. 2020. e role of fine-root mass, specific root
length and life span in tree performance: a whole-tree explora-
tion. – Funct. Ecol. 34: 575–585.
Weigelt, A. et al. 2005. Preferential uptake of soil nitrogen forms
by grassland plant species. – Oecologia 142: 627–635.
Wells, C. E. and Eissenstat, D. M. 2002. Beyond the roots of young
seedlings: the influence of age and order on fine root physiology.
– J. Plant Growth Regul. 21: 324–334.
Withington, J. M. et al. 2006. Comparisons of structure and life
span in roots and leaves among temperate trees. – Ecol. Mon-
ogr. 76: 381–397.
Withington, J. M. et al. 2021. Remarkable similarity in timing of
absorptive fine-root production across 11 diverse temperate tree
species in a common garden. – Front. Plant Sci. 11: 623722.
Wright, I. J. et al. 2004. e worldwide leaf economics spectrum.
– Nature 428: 821–827.
Yavitt, J. B. and Wright, S. J. 2001. Drought and irrigation effects
on fine root dynamics in a tropical moist forest, Panama. –
Biotropica 33: 421–434.
Yu, G. et al. 2020. Root identification in minirhizotron imagery
with multiple instance learning. – Mach. Vis. Appl. 31: 43.
Yuan, Z. Y. and Chen, H. Y. 2013. Simplifying the decision matrix
for estimating fine root production by the sequential soil coring
approach. – Acta Oecol. 48: 54–61.
Zeng, G. et al. 2008. Automatic discrimination of fine roots in
minirhizotron images. – New Phytol. 177: 549–557.
Ziegler, A. D. et al. 2012. Carbon outcomes of major land-cover
transitions in SE Asia: great uncertainties and REDD + policy
implications. – Global Change Biol. 18: 3087–3099.
... The mature rubber plantation in this study (33 years) was likely to create more litterfall than the young rubber plantations in Sumatra (7-16 years) (Kotowska et al., 2016) and in Côte d'Ivoire (7-25 years) (N'Dri et al., 2018). Moreover, the higher stem density in the present RM (600 ha -1 ) may produce more litterfall than in rubber plantation (469 ha -1 ) in Indonesia (Kotowska et al., 2016(Kotowska et al., , 2023. In addition, the variation pattern of litterfall production among the studied forests coincided with canopy cover and leaf area index, indicating that a more diverse species composition and canopy structure was likely to increase litterfall. ...
... Similarly, the current rubber agroforestry systems exhibited 1.3-1.6 times higher litterfall production compared to RM (Table 2), which is consistent with findings in the rubber plantations in southern Thailand (Tongkaemkaew et al., 2018). In addition, the litter productions in jungle rubber (multi-species rubber plantations) in Indonesia (7.7 Mg ha -1 yr -1 ) (Kotowska et al., 2016(Kotowska et al., , 2023, in southern China (9.2 Mg ha -1 yr -1 ) (Tang et al., 2010), and in central Amazonia (6.4-7.6 Mg ha -1 yr -1 ) (Martius et al., 2004) were also higher than that in the RM, which has been attributed to the more diverse species composition and canopy structure of jungle rubber (Martius et al., 2004;Tang et al., 2010). However, litterfall amounts of jungle rubber were slightly lower than those in the rubber agroforestry systems from our study, which may be Table 3 Annual mean stand litter (Mg ha -1 yr -1 ) and the proportion (%) of its components from 2017 to 2019 in the four plantation types. ...
Article
Increasing global demand for natural rubber has resulted in rapid expansion of rubber plantations and subsequent environmental issues in the tropics. Diversified rubber agroforestry (i.e. mixed planting of rubber with various crops) are proposed to counteract negative effects induced by rubber monoculture (RM). However, little is known about the changes in ecosystem function and services associated with biogeochemical cycles following RM conversion to rubber agroforestry. A three-year study was conducted to investigate litterfall biomass, stand litter and nutrient return in RM and three rubber agroforestry systems, i.e., rubber-tea (RTAs), rubber-Flemingia macrophylla (RFAs), and rubber-cacao (RCAs). Annual litterfall production was the highest in RCAs (8.92 Mg ha-1 yr-1), followed by RTAs (8.25 Mg ha-1 yr-1), RFAs (7.30 Mg ha-1 yr-1), and the lowest in RM (5.42 Mg ha-1 yr-1). Temperature, relative humidity, and solar radiation were dominant drivers of the distinct seasonality of litterfall production, with most of production (64-71%) occurring in the dry and cold season. Average stand litter showed the decreasing order of RTAs > RCAs > RFAs > RM. Litter turnover rate was generally lower in RM than in rubber agroforestry systems. Carbon and nutrient returns via litterfall exhibited 1.3-1.6 times higher in rubber agroforestry than in RM. RM had the greatest P use efficiency due to low P availability in the soil. Such strong P limitation was mitigated by introducing understory intercrops. RTAs exhibited the most efficient Ca and Mg cycling, limiting their losses through litterfall. These results demonstrates that intercropping perennial cash crops with rubber could effectively increase biomass productivity, accumulation, turnover rate, and accompanying nutrient deposition. Constructing multispecies rubber agroforestry is an environmentally friendly management strategy to improve the ecosystem nutrient cycling flexibility and soil fertility restoration and thus ecosystem services of rubber plantations.
... RTL mortality that fine root production usually peaks in summer (Wang et al. 2020;Huo et al. 2022), but mortality often increases in late summer or autumn (Brassard et al. 2009). By contrast, other studies recorded that relatively lower fine root production occurs in summer Joslin et al. 2001;Kotowska et al. 2023). Such confliction suggests that the environmental control of fine root dynamics may vary geographically (Tierney et al. 2003). ...
Article
Full-text available
Root tips are the main components of absorptive fine roots, but their seasonal dynamics and relationship to environmental factors remain unclear due to the difficulties in methodology. In this study, we explored the temporal patterns of root-tip production and mortality in monoculture plantations of five temperate tree species at a common site in northeastern China, and identified the general environmental controls on such processes. We made monthly in-situ assessments of root tip length (RTL) production and mortality in two hardwood and three coniferous species with a minirhizotron (MR) method during the growing seasons of 2008 and 2009. Air temperature, rainfall, soil temperature and water content at 10 cm depth were determined concurrently. RTL production in all species exhibited consistent peaks in summer (June–August) in two growing seasons. RTL mortality showed substantial interannual and interspecific variability, with peaks in autumn and winter in 2008, but various patterns in 2009. RTL production positively correlated with monthly soil and air temperature across all species, and with monthly rainfall in three coniferous species. However, there was no significant correlation between RTL production and soil water content. By contrast, RTL mortality was weakly related to environmental factors, showing positive correlations with soil temperature in Korean spruce, and with rainfall in Korean pine and Korean spruce. Our findings suggest that the seasonal patterns of RTL production are convergent across the five temperate tree species due to the overlapped distribution of heat and rainfall, which can conduce roots to maximizing the acquisition of nutrient resources in the soil.
... In their case, although there were weak trends towards more conservative root trait syndromes on high altitudes with deteriorating soil conditions, all tropical forest communities showed high within-community variance in root traits at each altitude, confirming the importance of multiple strategies for resource capture in tropical forests. In addition, Kotowska et al. (2022) present compelling evidence that expected ecological trends in trait syndromes, like the dominance of "acquisitive traits" in secondary succession, were not supported after comparing mature and secondary forests in Indonesia, despite the large productivity of the secondary forests. Finally, Erktan and colleagues (2022) provide an in-depth view of root diameter distributions, where most of the temperate plant communities (including woody and non-woody species) they studied, showed a multimodal distribution of root diameter reflecting the distinct belowground strategies to exploit soil resources that exist within plant communities. ...
Article
Full-text available
We investigated the variation in tree fine root traits and their functional diversity along a local topographic gradient in a Neotropical montane forest to test if fine root trait variation along the gradient is consistent with the predictions of the root economics spectrum on a shift from acquisitive to conservative traits with decreasing resource supply. We measured five fine root functional traits in 179 randomly selected tree individuals of 100 species and analysed the variation of single traits (using Bayesian phylogenetic multilevel models) and of functional trait diversity with small-scale topography. Fine roots exhibited more conservative traits (thicker diameters, lower specific root length and nitrogen concentration) at upper slope compared with lower slope positions, but the largest proportion of variation (40–80%) was explained by species identity and phylogeny. Fine root functional diversity decreased towards the upper slopes. Our results suggest that local topography and the related soil fertility and moisture gradients cause considerable small-scale variation in fine root traits and functional diversity along tropical mountain slopes, with conservative root traits and greater trait convergence being associated with less favourable soil conditions due to environmental filtering. We provide evidence of a high degree of phylogenetic conservation in fine root traits.
Article
Full-text available
Intensive management practices in large-scale oil palm plantations can slow down nutrient cycling and alter other soil functions. Thus, there is a need to reduce management intensity without sacrificing productivity. The aim of our study was to investigate the effect of management practices on gross rates of soil N cycling and soil fertility. In Jambi province, Indonesia, we established a management experiment in a large-scale oil palm plantation to compare conventional practices (i.e. high fertilization rates and herbicide weeding) with reduced management intensity (i.e. reduced fertilization rates and mechanical weeding). Also, we compared the typical management zones characterizing large-scale plantations: palm circle, inter-row and frond-stacked area. After 1.5 years of this experiment, reduced and conventional management showed comparable gross soil N cycling rates; however, there were stark differences among management zones. The frond-stacked area had higher soil N cycling rates and soil fertility (high microbial biomass, extractable C, soil organic C, extractable organic N, total N and low bulk density) than inter-row and palm circle (all p ≤ 0.05). Microbial biomass was the main driver of the soil N cycle, attested by its high correlation with gross N-cycling rates (r = 0.93–0.95, p < 0.01). The correlations of microbial N with extractable C, extractable organic N, soil organic C and total N (r = 0.76–0.89, p < 0.01) suggest that microbial biomass was mainly regulated by the availability of organic matter. Mulching with senesced fronds enhanced soil microbial biomass, which promoted nutrient recycling and thereby can decrease dependency on chemical fertilizers.
Article
Full-text available
Long-term minirhizotron observations of absorptive fine roots provide insights into seasonal patterns of belowground root production and carbon dynamics. Our objective was to compare root dynamics over time across mature individuals of 11 temperate trees species: five evergreen and six deciduous. We analyzed the timing and growth on 1st-and 2nd-order roots in minirhizotron images down to a vertical depth of 35 cm, as well as monthly and total annual length production. Production patterns were related to total annual precipitation of the actual and previous year of root production over 6 years. The main or largest peak of annual fine-root production occurred between June and September for almost all species and years. In most years, when peaks occurred, the timing of peak root production was synchronized across all species. A linear mixed model revealed significant differences in monthly fine-root length production across species in certain years (species x year, P < 0.0001), which was strongly influenced by three tree species. Total annual root production was much higher in 2000–2002, when there was above-average rainfall in the previous year, compared with production in 2005–2007, which followed years of lower-than-average rainfall (2003–2006). Compared to the wetter period all species experienced a decline of at least 75% in annual production in the drier years. Total annual root length production was more strongly associated with previous year’s (P < 0.001) compared with the actual year’s precipitation (P = 0.003). Remarkably similar timing of monthly absorptive fine-root growth can occur across multiple species of diverse phylogeny and leaf habit in a given year, suggesting a strong influence of extrinsic factors on absorptive fine-root growth. The influence of previous year precipitation on annual absorptive fine-root growth underscores the importance of legacy effects in biological responses and suggests that a growth response of temperate trees to extreme precipitation or drought events can be exacerbated across years.
Article
Full-text available
From recent developments on how roots affect soil organic carbon (SOC) an apparent paradox has emerged where roots drive SOC stabilization causing SOC accrual, but also SOC destabilization causing SOC loss. We synthesize current results and propose the new Rhizo‐Engine framework consisting of two linked components: microbial turnover and the soil physicochemical matrix. The Rhizo‐Engine is driven by rhizodeposition, root turnover, and plant uptake of nutrients and water, thereby accelerating SOC turnover through both stabilization and destabilization mechanisms. This Rhizo‐Engine framework emphasizes the need for a more holistic approach to study root‐driven SOC dynamics. This framework would provide better understanding of plant root effects on soil carbon sequestration and the sensitivity of SOC stocks to climate and land‐use changes.
Article
Full-text available
Root respiration is a critical physiological trait involved in root resource acquisition strategies, yet it is less represented in root traits syndrome. Here we compiled a large dataset of root respiration associated with root chemical and morphological traits from 245 plant species. Our results demonstrated that root respiration correlated positively with root nitrogen concentration (RNC) and negatively with root tissue density (RTD) across and within woody and non‐woody species. However, the relationships between root respiration and specific root length (SRL) and root diameter (RD) were weak or even insignificant. Such root respiration‐traits relationships were not completely in line with predictions by the root economics spectrum. Furthermore, the principal component analysis showed that root traits syndrome was multi‐dimensional. Root respiration was associated more strongly to the RNC‐RTD axis (the classical root economics spectrum) than to the orthogonal SRL‐RD axis for woody species, but not for non‐woody species. Collectively, the linkages of root physiological, chemical and morphological traits provide a better understanding of root traits covariation and root resource acquisition strategies.
Article
Full-text available
The effects of plants on the biosphere, atmosphere and geosphere are key determinants of terrestrial ecosystem functioning. However, despite substantial progress made regarding plant belowground components, we are still only beginning to explore the complex relationships between root traits and functions. Drawing on the literature in plant physiology, ecophysiology, ecology, agronomy and soil science, we reviewed 24 aspects of plant and ecosystem functioning and their relationships with a number of root system traits, including aspects of architecture, physiology, morphology, anatomy, chemistry, biomechanics and biotic interactions. Based on this assessment, we critically evaluated the current strengths and gaps in our knowledge, and identify future research challenges in the field of root ecology. Most importantly, we found that belowground traits with the broadest importance in plant and ecosystem functioning are not those most commonly measured. Also, the estimation of trait relative importance for functioning requires us to consider a more comprehensive range of functionally relevant traits from a diverse range of species, across environments and over time series. We also advocate that establishing causal hierarchical links among root traits will provide a hypothesis‐based framework to identify the most parsimonious sets of traits with the strongest links on functions, and to link genotypes to plant and ecosystem functioning.
Article
Full-text available
--Key message-- In older, unlogged rainforest of Hainan Island, China, leaves of saplings were larger, and fine root systems of saplings were thicker with fewer root tips than in historically logged areas. These results were consistent among 15 Angiosperm lineages, even though families differed widely in their leaf and root traits. --Context-- How plant organ morphologies vary with environment is key for inferring plant functional strategies. --Aims-- We were interested in quantifying any changes in fine root and leaf morphology of saplings with local-scale environmental variation in tropical forest, and if any variation in organ morphologies differed with plant lineage. --Methods-- We measured functional traits of fine root systems and leaves of saplings from 15 families in historically logged and unlogged Chinese tropical forest, where soil fertility and texture slightly decreased with greater forest age. --Results-- Root morphological traits were more conservative, while leaf morphologies were more acquisitive in primary forest than in secondary forest. From secondary to primary forests, mean root system diameter increased 0.4 mm, mean specific root length decreased 0.35 m g−1, and mean root system branching intensity decreased by 0.3 tips cm−1. Similarly, from secondary to primary forests, average leaf area increased 7 cm2 and specific leaf area decreased 0.8 m2 kg−1. Leaf thickness and root tissue density were not different. Among the selected plant families, root and leaf morphological differences between forest types were consistent. --Conclusion-- Within lineage (i.e., intraspecific) root and leaf morphological variation showed contrasting patterns. Local-scale variation in soil phosphorus and base saturation affected intraspecific variation in root diameter and specific root length. Please note: a correction has been published for this article (see link in comments)
Article
Full-text available
Plant economics run on carbon and nutrients instead of money. Leaf strategies aboveground span an economic spectrum from “live fast and die young” to “slow and steady,” but the economy defined by root strategies belowground remains unclear. Here, we take a holistic view of the belowground economy and show that root-mycorrhizal collaboration can short circuit a one-dimensional economic spectrum, providing an entire space of economic possibilities. Root trait data from 1810 species across the globe confirm a classical fast-slow “conservation” gradient but show that most variation is explained by an orthogonal “collaboration” gradient, ranging from “do-it-yourself” resource uptake to “outsourcing” of resource uptake to mycorrhizal fungi. This broadened “root economics space” provides a solid foundation for predictive understanding of belowground responses to changing environmental conditions.