Content uploaded by Jin-Hua Li
Author content
All content in this area was uploaded by Jin-Hua Li on May 31, 2023
Content may be subject to copyright.
Journal of Ecology. 2023;00:1–12.
|
1wileyonlinelibrary.com/journal/jec
Received: 10 November 2022
|
Accepted: 30 April 2023
DOI : 10.1111/136 5-2745.14133
RESEARCH ARTICLE
Biomass temporal stability increases at two spatial scales
during secondary succession
Wenjin Li1 | Xi Zhou1 | Zhiqiang Xiang1 | Jinhua Li1 | Shaopeng Wang2 |
Michel Loreau3 | Lin Jiang4
© 2023 The Authors. Journal of Ecology © 2023 British Ecological Society.
Wenjin Li and Xi Zho u authors contr ibuted equally to this wo rk as co– first a uthors.
1State Key Laboratory of Herbage
Improvement and Grassland Agro-
ecosystems, Gansu Gannan Grassland
Ecosystem National Observation and
Research Station, College of Ecology,
Lanzhou University, Lanzhou, China
2Key Laboratory for Ear th Sur face
Processes of the Ministr y of Education,
Institute of Ecology, College of Urban
and Environmental Sciences, Peking
University, Beijing, China
3Centre for Biodiversit y Theory and
Modelling, Theoretical and Experimental
Ecology Station, CNRS and Paul Sabatier
University, Moulis, France
4School of Biological Sciences, Georgia
Institute of Technology, Atlanta, Georgia,
USA
Correspondence
Wenji n Li
Email: liwj@lzu.edu.cn
Funding information
the TULIP Laboratory of Excellence,
Grant/Award Number: ANR- 10- LABX- 41;
the US National Science Foundation,
Grant/Award Number: CBET- 1833988
and DEB- 1856318; National Natural
Science Foundation of China, Grant/
Award Number: 31470480 and 32271761;
TULIP Laboratory of E xcellence, Grant/
Award Number: ANR- 10- LABX- 41; US
National Science Foundation, Grant/
Award Number: CBET- 1833988 and DEB-
1856318
Handling Editor: Eric Lamb
Abstract
1. Ecological stability has long been considered to change over succession, but how
secondary succession influences the relationship between diversity and temporal
stability of biomass production at different spatial scales is poorly understood.
2. We studied changes in plant diversity, functional temporal stability (biomass
production) and compositional temporal stability (the latter two are hereafter
referred to as functional stability and compositional stability) and explored the
stabilizing roles of plant diversity at two spatial scales (small plots of 0.25 m2 and
large transects of 1.25 m2) during secondary succession in a subalpine meadow
from 2003 to 2010.
3. Our results showed that both plant diversity and functional and compositional
stability increased at the small plot scale and large transect scale during secondary
succession. As secondary succession proceeded, higher average alpha diversity
(i.e. species diversity at the plot scale) led to higher functional and compositional
stability at the plot scale by mainly species stability, predominantly contribut-
ing to higher functional and compositional stability at the large transect scale.
In addition, Simpson- based beta diversity (i.e. compositional dissimilarity among
communities within the same transect), while unaffected by succession, contrib-
uted to functional stability at the large transect scale by promoting asynchronous
dynamics among communities.
4. Synthesis. Our study highlights the stabilizing effects of plant diversity across the
two spatial scales during secondary succession. Our findings provide the first em-
pirical evidence that biodiversity- mediated effects on ecosystem temporal stabil-
ity strengthen over successional time, suggesting that the stabilizing effects of
biodiversity should be considered across spatial and temporal scales in the face
of global changes and biodiversity loss.
KEYWORDS
beta diversity, ecosystem stability, old fields, spatial asynchrony, spatial scale, succession,
temporal scale
2
|
Journal of Ecology
LI et al.
1 | INTRODUC TION
Ecological stability is a multidimensional concept that can be quan-
tified using various metrics, often including resistance, resilience,
recovery and temporal stability (Donohue et al., 2013; Grimm &
Wissel, 1997; Pimm, 198 4; White et al., 2020). Temporal stability,
which can be quantified using the inverse of the coefficient of vari-
ation, is the most common aspect among stability metrics (Donohue
et al., 2016). Furthermore, most studies have focused on the tem-
poral stability of biomass production (Donohue et al., 2016; Tilman
et al., 2006; Wagg et al., 2022), which is more formally defined as
the degree of biomass fluctuation over time (Tilman, 1996; Wagg
et al., 2022) and calculated as the ratio of the mean of community
biomass to its standard deviation in a given ecosystem (Tilman, 1996).
The temporal stability of biomass production is also the measure of
stability adopted in the present study because (i) it is an integra-
tive measure of stability (Loreau, 2022), frequently used in ecology
(Donohue et al., 2016; Tilman et al., 2006); (ii) it describes the com-
bined effects of resistance and resilience on community dynamics
over time (Clark et al., 2021) and (iii) it is also very useful when ex-
amining the stability of a community to sustain temporally stable
biomass production across multiple years (Lehman & Tilman, 2000).
In the past two decades, many studies have emphasized the
effects of global change, including nutrient addition, precipitation
change, elevated CO2 and warming (Su et al., 2022), and diver-
sity change (Tilman et al., 2006), on the local temporal stability of
biomass. However, predicting the relationships between biodi-
versity and ecological stability at different temporal and spatial
scales remains challenging in ecology (Clark et al., 2021; Wang &
Loreau, 2016). Theoretical and experimental research over the past
decade has begun to explore ecosystem stability at larger spatial
scales (Hautier et al., 2020; Liang et al., 2022; Wang et al., 2021; Wang
& Loreau, 2014; Wilcox et al., 2017) and has revealed that ecosys-
tem stability often increases with the spatial scale (Qiao et al., 2022;
Wang et al., 2017). The theory predicts that ecosystem stability (γ
stability) at a large spatial scale is driven by local community sta-
bility and asynchronous dynamics among local communities (spatial
asynchrony; Wang et al., 2019; Wang & Loreau, 2014, 2016). For ex-
ample, Wilcox et al. (2017 ) revealed that spatial asynchrony among
local communities was an important predictor of stability at large
spatial scales in global grasslands. Clark et al. (2021) showed that
the joint stabilizing effects of both plant α and β diversity contribute
to the stability of the grassland ecosystem at a large scale in North
America and Europe. However, these empirical studies covered rela-
tively short periods and mainly addressed the effects of spatial scale
on ecosystem stability (Hautier et al., 2020; Liang et al., 2022). The
studies did not examine the effects of temporal scale (e.g. duration of
experiments), especially the successional stage (e.g. ecosystem de-
velopment and land- use change), on ecosystem stabilit y, which limits
our understanding to longer- term, large- scale ecosystem manage-
ment and conservation (Qiu & Cardinale, 2020). Indeed, ecosystem
stability changes with the duration of experiments or observations
(Luo et al., 2021; Pimm & Redfearn, 198 8; Wagg et al., 2022). The
duration of experiments can also change biodiversity– ecosystem
functioning relationships (Meyer et al., 2016; Wagg et al., 2022). The
plant diversity– productivity relationship often strengthens with the
duration of artificially assembled biodiversity experiments (Meyer
et al., 2016; Qiu & Cardinale, 2020). Species complementarity and
asynchrony can take more than 10 years to play a strong role in
stabilizing the effects of biodiversity on productivity in plant com-
munities (Wagg et al., 2022). Thus, these results suggest that the
duration of experiments or observations could modulate the effects
of biodiversity on ecosystem stability (Wagg et al., 2022). However,
to the best of our knowledge, the effects of successional change on
ecosystem temporal stability via biodiversity across temporal and
spatial scales remain largely unexplored.
The study of ecological succession is often regarded as a prom-
ising approach for addressing the temporal dynamics of ecosys-
tem structure and functioning (Foster & Tilman, 2000; Prach &
Wal ke r, 2011; Walker & Wardle, 2014). Classic ecological succession
studies indicate that population fluctuations and species turnover
during succession depend on scales of time and space (Connell &
Slatyer, 1977 ). The earliest synchronic chronosequence studies
(Cowles, 1899; Oosting, 1942) and diachronic permanent plot stud-
ies (Foster & Tilman, 2000) have claimed that the rate of commu-
nity change decreased during primary or secondary succession
(Anderson, 20 07; Li et al., 2016). Succession can also alter biodiver-
sity and ecosystem functioning relationships (Lasky et al., 2014; Mori
et al., 2017). One of the longest- running biodiversity experiments,
conducted in Jena, Germany, has shown that the temporal stabilizing
effect of species richness on plant productivity increased with com-
munity age at the local scale (Wagg et al., 2022). In recent years, with
growing awareness of the spatial scale dependence of biodiversity
and ecosystem functioning (Gonzalez et al., 2020; Isbell et al., 2018),
β diversity has attracted increasing attention (Mori et al., 2018; Reu
et al., 2022), and it has been determined that β diversity contrib-
utes to increasing ecosystem functioning (Mori et al., 2018) and
ecosystem stability (Clark et al., 2021; Mellin et al., 2014) at large
spatial scales. However, it is not clear whether β diversity consis-
tently influences ecosystem stability at large spatial scales through
spatially asynchronous dynamics among local communities during
succession.
In our study, a chronosequence of five old fields (1, 3, 5, 15 and
30 years since abandonment) and one natural meadow (without agri-
cultural land use for approximately 100 years) in a subalpine meadow
was employed to assess how secondary succession influences the
relationship between the diversity and stability of biomass produc-
tion at two spatial scales and the underlying mechanisms. We quan-
tified temporal stability from functional (biomass) and structural
(composition) perspectives to reflect ecosystem and community dy-
namics. The following hypotheses were specifically tested:
(i) Plant diversity and functional and compositional stability at the
plot and transect scales increase over succession because com-
munity change rates generally decline over the course of succes-
sion (Anderson, 2007).
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
|
3
Journal of Ecology
LI et al.
(ii) This increase in functional and compositional stability at two
spatial scales is driven by increasing plant diversity over suc-
cession. Because increased average α plant diversity at the plot
scale during succession (Li et al., 2017) induces greater α func-
tional and compositional stability at the plot scale and a decline
in β diversity over succession, resulting in lower asynchronous
dynamics among local communities (Wang & Loreau, 20 14,
2016), the relative contribution of α diversity to functional and
compositional stability at the large transect scale is greater than
that of β diversity.
2 | MATERIALS AND METHODS
2.1 | Study sites and experimental design
The study site was located at the Gannan Grassland Ecosystem
National Observation and Research Station and established on a
subalpine meadow on the eastern Qinghai- Tibetan Plateau, China
(N34°55′, E102°53′). Using a space- for- time substitution approach,
we selected different stages of secondary succession consisting
of five old fields (1, 3, 5, 15 and 30 years since abandonment) and
one natural meadow (without agricultural land use for approxi-
mately 100 years) by referring to the historical data of local land use
and interviewing the local herdsmen. We also considered that the
early- stage plant communities were dominated by annual or bien-
nial weeds and herbs, such as Aconitum gymnandrum, Poa annua
and Potentilla sp., and that the later stage plant communities were
dominated by perennial species, such as Elymus nutan, Kobresia
humilis and Roegneria nutans (Li et al., 2009). All sites share similar
substrate, topographic position and historic cultivation conditions
(Li et al., 2017). The only agricultural practice in this alpine region
is rotational cultivation of oat (Avena sativa)— fallow— and rapeseed
(Brassica napus) in the last century (Li et al., 2009). After the cessa-
tion of cultivation, these abandoned meadows were grazed by live-
stock until 2003. After ward, large herbivores were excluded using a
wire fence. The area of these study meadows is small, ranging from
0.10 to 0.20 ha, located within an area of 10 km2 at a simil ar el evation
(from 2926 to 3000 m above sea level) and at least 300 m apart. The
annual mean temperature is 3.2°C, ranging from 9.9°C in January to
12.8°C in July. The mean annual rainfall was 540 mm from 2000 to
2010, with 86% of the precipitation concentrated during the grow-
ing season. The natural vegetation in this region is typical subalpine
meadow, which is dominated by Agrostis hugoniana Rendle, Stipa
aliena Keng, and Kobresia humilis (C. A. Mey.) Serg and Polygonum
viviparum L. In some patches, a shrub, sea buckthorn (Hippophae
rhamnoides), is also present, but not in our study sites.
In July 2003, two parallel transects (A and B) were established
at each old field site and at the control meadow, and 10 permanent
50 × 50 cm sampling plots were established along the two transects
inside each field. The distance between the two parallel transects
ranged from 5 to 8 m; both transects were located at least 5 m from
the edge. The interval between two adjacent plots within each
transect was 3.5 m. The plots in each old field were sampled annu-
ally every August from 2003 to 2010 (with the exception of 2005,
when no sampling was conducted). All above- ground biomass was
estimated by clipping the above- ground plant parts at the 1 cm soil
surface level. All clipped plants were sorted into individual species
and litter and then dried to a constant mass at 60°C. Species diver-
sity is quantified as species richness (i.e. number of species in the
plot or transect), and plant biomass is quantified as the weight of
aboveground dry materials per m2.
2.2 | Alpha, beta and gamma diversity
Species diversity and temporal stability of biomass were consid-
ered at both the plot scale (0.25 m2 area) and transect scale (1.25 m2
area). Each 0.25 m2 plot was treated as the small plot scale, and
the combination of the five replicated plots along a transect was
classified as the large transect scale. The Simpson- based diversity
index can best explain ecosystem stability at different spatial scales
(Wang & Loreau, 2016). Therefore, we calculated the Simpson index
Dk
=∑
S
i
p
ik2
, where
pik
is the relative biomass of species i and S is
the number of species within community
k
. Thus, we applied the in-
verse of the Simpson index as α diversity (αsimp). The γ diversity was
calculated as
𝛾simp
=1∕∑
S
i
p
iM2
, where
piM
is the relative biomass of
species i and S is the number of species at the transect scale. As
alpha diversity is measured at the plot scale and gamma diversity is
measured at the transect scale, multiplicative beta diversity at the
transect scale is the ratio of gamma diversity to mean alpha diversity.
Multiplication- based β diversity (β
simp) was calculated as the ratio of
γsimp to αsimp (Wang & Loreau, 2016), which represents the turnover
of species among local communities.
To quantify different aspects of the change in plant composition
at the plot level in the 7- year sampling, we used the ‘species exchange
ratio’ (SER) approach (Hillebrand, Blasius, et al., 2018) to measure
the proportional exchange of species or relative abundances of spe-
cies between an earlier sampling and later sampling in a time se-
ries at the plot level (the caption of Figure S1 in the Supplementary
Material provides detailed information on the calculation of SER).
We found that the richness- based species exchange ratio (SERr) and
abundance- based species exchange ratio (SERa) at most sites (with
the exception of one or two sites) did not significantly change with
sampling year. Thus, we quantified average species diversity per plot
by averaging species diversity in the same plot over a 7- year period
to examine plant diversity effects on temporal stability across two
spatial scales (Tilman, 1996; Wagg et al., 2022).
2.3 | Stability and asynchrony across two
spatial scales
Following previous work (Wang et al., 2019), species temporal sta-
bility (hereafter referred to as species stability) was defined as the
weighted average of local population stability across species and
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4
|
Journal of Ecology
LI et al.
plots. We utilized total above- ground biomass to quantify functional
temporal stability. Following previous work (Tilman, 1999), func-
tional temporal stability was defined as μT/σT, where μT and σT are
the interannual mean and standard deviation, respectively, of com-
munity biomass over the 7 years (2003– 2010, with the exception
for 2005). α functional stability at the plot scale was defined as the
biomass- weighted average of the temporal stability of each plot, and
γ functional stability at the transect scale was defined as the tem-
poral stability of the total community biomass of the 5 plots along a
transect (Wang et al., 2019). The α compositional stabilit y at the plot
scale was quantified by one minus the mean Euclidean distance from
each plot in each year to its centroid over 7 years (2003– 2010, with
the exception of 2005), with the distance calculated based on the
Bray– Curtis dissimilarity between two communities of the same plot
in each year and then averaged at the transect level (Xu et al., 2022).
γ compositional stability at the transect scale was quantified by one
minus the mean Euclidean distance from each transect in each year
to its 7- year (2003– 2010, with the exception of 2005) transect cen-
troid, with distance calculated based on the Bray– Curtis dissimilarity
among communities of the same transect.
In addition, functional stability at two spatial scales was also
determined after detrending to eliminate potentially confounding
effects of directional biomass change on temporal stability (Lepš
et al., 2019; Valencia et al., 2020). Specifically, we replaced σ in the
stability definition as the standard deviation of the residuals of the
linear regression between aboveground biomass and the sampling
year (Tilman et al., 2006). Here, the detrended functional α and γ
stability was applied.
The previous framework for partitioning ecosystem stability (see,
e.g. Wang et al., 2019; Wang & Loreau, 2014, 2016) demonstrated
that local α stability can be partitioned into species stability and
species synchrony (see also Thibaut & Connolly, 2013). Therefore,
species asynchrony indicates that the dynamics of asynchronous
species within local communities respond to environmental fluctu-
ations, which is defined as the ratio of α stability to species stability
(Wang et al., 2021). Similarly, spatial asynchrony is calculated as the
ratio of γ stability to α stability, which indicates the asynchronous
community dynamics among local communities in response to envi-
ronmental fluctuations. More details on the asynchrony index equa-
tions are provided in Wang et al. (2019).
2.4 | Statistical analysis
To improve normality and linear relationships, diversity, stability, and
asynchrony measures and successional year (as a continuous vari-
able) were log10 transformed before analyses. First, linear mixed-
effects models (LMMs) were performed using the R package nlme
(Pinheiro et al., 2021) to assess the effects of successional time on
diversity, stability, and asynchrony at two different spatial scales,
where the sites were treated as random factors. Similarly, we em-
ployed LMMs to test the diversity– asynchrony– stability relation-
ships at the two spatial scales during succession. We calculated the
marginal R2 (R2
m) and conditional R2 (R2
c) using the package mumIn
to evaluate the model performance (Nakagawa & Schielzeth, 2013).
Specifically, the marginal R2 (R2
m) and conditional R2 (R2
c)2 corre-
spond to ‘fixed effects’ and ‘fixed + random effects’, respectively.
We used the R package ‘piecewise structural equation model
(SEM)’ (Lefcheck, 2016) to conduct a stepwise selection of a piece-
wise structural equation model to quantify the relative contribution
of α and β diversity to γ stability during succession. An initial model
based on theory (see, e.g. Wang et al., 2019; Wang & Loreau, 2014,
2016) was establ ished (Table S1 and Figure S2), with site as a random
factor. In the SEM, all diversity, stability and asynchrony metrics
were log10 transformed, and Shipley's d- separation test was con-
ducted to ensure all possible paths. Next, the non- significant paths
were iteratively removed. Lastly, the final model that had the most
simplified path and the lowest AIC was chosen. All analyses were
performed using R 3.6.3 (R Core Team, 2019).
3 | RESULTS
3.1 | Changes in diversity, stability and asynchrony
across two spatial scales during succession
We found that the annual α diversity did not strongly vary with the
sampling year, with the exception of the control and 5- year sites
(p < 0.05; Figure 1a), and that the annual β diversity did not change
with sampling year, with the exception of the 1- year and 5- year sites
(p < 0.05; Figure 1b). During succession, both the average α diversity
at the plot scale (Simpson- based, Rm
2 = 0.69, p < 0.05; richness based,
Rm
2 = 0.79, p < 0.01) and γ diversity at the transect scale (Simpson
based, Rm
2 = 0.61, p < 0.05; richness based, R2 = 0.60, p < 0.05) con-
sistently increased over time (Figure 1c and Figure S3c, Tables S2
and S3). Simpson- based β diversity did not show a significant trend
with succession time (Rm
2 = 0.53, p > 0.05; Figure 1c and Table S2),
but richness- based β diver si ty (Rm
2 = 0.71, p < 0.05) decreased during
succession (Figure S3c and Table S3). Likewise, there were consist-
ent increases in species stability (Rm
2 = 0.60, p = 0.04), α functional
stability (Rm
2 = 0.67, p = 0.03) and γ functional stability (Rm
2 = 0.57,
p = 0.05) over successional time (Figure 2a and Table S2). Both α
compositional stability (Rm
2 = 0.73, p = 0.01) at the plot scale and γ
compositional stability (Rm
2 = 0.68, p = 0.02) at the transect scale
also increased with successional time (Figure 2b and Table S2). In
contrast, neither species asynchrony (Rm
2 = 0.002, p > 0.05) nor spa-
tial asynchrony (Rm
2 = 0.012, p > 0.05) significantly changed with
successional time (Figure S4a,b and Table S3).
3.2 | Biodiversity- mediated effects of successional
time on asynchrony and stability across two
spatial scales
During succession, the average α diversity was consistently positively
associated with α functional stability (Simpson based, Rm
2 = 0.88,
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
|
5
Journal of Ecology
LI et al.
p < 0.001; richness based, Rm
2 = 0.69, p < 0.05) and γ fu n ctio n al sta bil-
ity (Simpson based, Rm
2 = 0.83, p < 0.01; richness based, Rm
2 = 0.63,
p < 0.05; Figure 3a,c, Figure S5a,c and Table S2). The α and γ compo-
sitional st abil ities at the two sp atial scal es were also positi vely co rre-
lated with the average α diversity (compositional α stability, Simpson
based, Rm
2 = 0.39, p < 0.1; richness based, Rm
2 = 0.58, p < 0.05 and
compositional γ stability, Simpson based, Rm
2 = 0.37, p < 0.1; richness
based, Rm
2 = 0.62, p < 0.05; Figure 3b,d, Figure S5b,d and Table S2).
The average α diversity was not significantly positively correlated
with species stability (Simpson based, Rm
2 = 0.14, p > 0.05; richness
based, Rm
2 = 0.11, p > 0.05) or species asynchrony (Simpson based,
Rm
2 = 0.04, p > 0.05; richness based, Rm
2 = 0.01, p > 0.05; Figure S6
and Table S4). Simpson- based β diversity was positively associ-
ated with spatial asynchrony (Rm
2 = 0.78, p < 0.01; Figure 3g and
Table S2) and γ functional stability (Rm
2 = 0.03, p < 0.1; Figure 3e and
Table S2), but Simpson- based β diversity was not associated with γ
compositional stability (Rm
2 < 0.01, p = 0.77; Figure 3f and Table S2).
Richness- based β diversity was not correlated with spatial asyn-
chrony (Rm
2 = 0.05, p = 0.55; Figure S5g and Table S3) but was posi-
tively related to γ functional stability (Rm
2 = 0.05, p < 0.1; Figure S5e
and Table S3) and negatively related to γ compositional stability
(Rm
2 = 0.43, p = 0.05; Figure S5f and Table S3).
FIGURE 1 Temporal changes in annual α diversity (Simpson) at the plot scale (a) and annual β diversity (Simpson) at the transect scale
(b) at each old field site with sampling year. (c) Changes in the average 7- year diversity (Simpson) at both the plot and transect scales during
secondary succession. These diversity metrics are based on biomass. Solid lines represent the significant relationships from the linear
models (a and b) and the linear mixed- effects model (c) at p < 0.05, and the dashed lines represent the non- significant relationships at
p > 0.05. The marginal (R2
m) r- squared represents ‘fixed ef fects’ explanation. Details of the models can be found in Table S2.
FIGURE 2 Changes in functional
(a) and compositional (b) stability at
both plot and transect scales during
secondary succession. These functional
stability metrics are based on biomass.
The solid lines represent the significant
relationships from linear mixed- effects
models at p ≤ 0.05. The marginal (R2
m)
r- squared represents ‘fixed ef fects’
explanation. The details of the models can
be found in Table S2.
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6
|
Journal of Ecology
LI et al.
The SEM demonstrated that successional time positively af-
fected functional stability and compositional stability at the
two spatial scales mainly through increased average α diversity
(Figure 4a,b). During succession, the average α diversity signifi-
cantly increased with time (direct effect: 0.87) and further en-
hanced γ functional and compositional stability at the transect
scale by increasing α functional and compositional stability at
the plot scale, respectively (Figure 4a,b). Average α diversity had
stronger positive impacts on α functional stability than on α com-
positional stability. β diversity was not affected by succession but
significantly increased spatial asynchrony (direct effect: 0.88),
thus also increasing γ functional stability (indirect effect of β di-
versity: 0.88*0.23 = 0.20; Figure 4). However, β diversity did not
increase γ compositional stability via increased spatial asynchrony
(Figure 4b). We obtained qualitatively similar results when analys-
ing the relationships of α and β diver sity based on species richn ess
with functional stability and compositional stability at the two
spatial scales (Figure S7a,b). Although β diversity was negatively
affected by succession, it did not significantly increase spatial
asynchrony. Therefore, plant diversity at the two spatial scales
promoted γ functional and compositional stability at the tran-
sect scale during succession, and there were stronger stabilizing
FIGURE 3 The diversity–
synchrony– stability relationships at
both plot and transect scales during
secondary succession. These diversity,
asynchrony and stability metrics are
based on biomass. Species diversity is
measured using Simpson- based metrics.
Relationships between alpha diversity and
functional stability at plot (a) and larger
transect (c) scales; relationships between
alpha diversity and compositional stability
at plot (b) and larger transect (d) scales;
relationships between beta diversity and
gamma (e: functional; f: compositional)
stability; relationships between beta
diversity and spatial asynchrony (g). The
colour of the dots represents the different
old field sites. Solid lines represent
the overall significant relationships
from a linear mixed effects model at
p < 0.1, and the dashed line represents
a nonsignificant relationship at p > 0.1.
The shaded areas are the error bands and
denote 95% confidence inter vals. The
marginal (R2
m) r- squared represents ‘fixed
effects’ explanation. The details of the
models can be found in Table S2.
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
|
7
Journal of Ecology
LI et al.
effects of average α diversity on γ functional stability and com-
positional stability at the transect scale than on β diversity during
succession.
4 | DISCUSSION
To the best of our knowledge, our study provided the first assess-
ment of how succession influences functional and compositional
stability and their relationships with plant diversity at two different
spatial scales. Our results were consistent with the classic notion
that plant diver sity and ecosystem stability at the small plot scale in-
creased over the course of succession (Odum, 1969). Our study also
extended plant diversity and ecosystem stability from the small plot
scale to the large transect spatial scale during succession, supporting
our hypothesis that there is a stronger stabilizing effect of average
α diversity than β diversity at the large transect scale during succes-
sion. We also discovered that β diversity could improve functional
stability at the large transect scale by increasing spatial asynchrony
among local communities without altering compositional stability at
the large transect scale during natural succession. Our results high-
light that multiple spatial scales should be considered to fully under-
stand the stabilizing effects of biodiversity during succession.
4.1 | α diversity and α stability during succession
We found that although succession did not affect species asyn-
chrony, species stability significantly increased over successional
time (Figure 2a and Table S2). Both species stability and species syn-
chrony strongly positively affected α fu n ct io n al st ab il it y (Figure S8a,b
and Table S4), while only species stability strongly enhanced α com-
positional stability (Figure S8c and Table S4). This result is consist-
ent with previous studies that show that local ecosystem stability
can be driven by both species stability and species synchrony (Li
et al., 2022; Wilcox et al., 2017; Xu et al., 2021). In our study, species
stability had a relatively stronger effect on functional and composi-
tional stability than species asynchrony at the plot scale (Figure S8
and Table S4). However, during forest succession, the direct effect
of species asynchrony on local ecosystem stability increased with
FIGURE 4 Structural equation models (SEM) describing the relative effects of alpha and beta diversity on gamma stability (a: functional;
b: compositional), through alpha stability (a: functional; b: compositional) and spatial synchrony at two spatial scales. In both SEMs, the
average species diversity is measured by Simpson- based metrics. (a) Fisher's C = 24.495, p = 0.079, df = 16, AIC = 64.495. (b) Fisher's
C = 16.343, p = 0.293, df = 14, AIC = 58. 343. The solid arrows represent positive relationships and the light- grey (bidirectional) arrows
represent a correlation between variables. All the lines marked are significant effects (the significance levels of each predictor are *p < 0.05,
**p < 0.01, ***p < 0.001, a: 0.05 < p < 0.1). For each response variable, marginal (R2
m) values showing the variance explained by fixed effects
and conditional (R2
c) values indicating the variance explained by the whole model are provided. The widths of each arrows are relative to the
standardized path coefficients.
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8
|
Journal of Ecology
LI et al.
time. Other findings have also reported the important role of spe-
cies asynchrony in stabilizing local community dynamics (Wilcox
et al., 2017; Xu et al., 2021). Th e incre ase in sp ecies asynchr ony com-
pensates for the decline in species stability (Xu et al., 2021). During
succession, increasing α diversity over time weakly increased spe-
cies stability and species asynchrony (Figure S6), but the increase in
species stability surpasses the increase in species asynchrony, both
resulting in enhanced local ecosystem stability.
Although diversity and stability at the two spatial scales demon-
strated an overall significant increase over secondary succession, we
also observed that alpha and gamma diversity and functional stabil-
ity decreased at the early stage of succession (between 1 year and
5 years) and then increased at the late stage of succession (between
5 years and 100 years). Explanations for this trend are presented as
follows: (1) a confounding site- specific effect because there was no
replication of sites with the same successional time in our exper-
iment and (2) the disappearance of pioneer weed species such as
Aconitum gymnandrum and Galium aparine prior to the es tablishment
of late- successional species such as Elymus nutans, Roegneria nutans
and Kobresia humilis (Li et al., 2009).
4.2 | α stability and γ stability during succession
Metacommunity stability theory has clarified the close link between
stability at a small spatial scale and stability at a large spatial scale
(Wang et al., 2019; Wang & Loreau, 2014, 2016). Using the SEM,
we discovered that increased α diversity over time promoted α
functional and compositional stability at the plot scale, which pre-
dominantly contributed to functional and compositional γ stability at
the large transect scale during succession. This finding also showed
agreement with a recent finding of Wang et al. (2021), who also re-
ported stronger effects of α diversity on γ stability than of β diver-
sity on γ stability in long- term grassland observation studies. The
stronger stabilizing effect of α diversity during succession emerged
presumably because increasing α diversity over time enhanced local
α stability by a combination of species stability and species asyn-
chrony, supporting previous findings that α diversity is a key driv-
ing mechanism for local ecosystem stability (Tilman et al., 2006; Xu
et al., 2021). Our study also provides evidence that α diversity is a
primary contributor to ecosystem stability via local α stability at a
larger spatial scale.
4.3 | Stabilizing effects of β diversity
during succession
Theoretical studies (Wang et al., 2019; Wang & Loreau, 2014 ,
2016) and several experimental studies (Clark et al., 2021; Patrick
et al., 2021) suggest that β diversity can increase ecosystem sta-
bility via spatial asynchrony among communities at larger spatial
scales due to an increase in dissimilarity in the local community
structure leading to greater differences in community dynamics
(Wang & Loreau, 2014, 2016). However, to the best of our knowl-
edge, few studies have examined the stabilizing effects of β diver-
sity by spatial asynchrony during succession. Previous succession
studies have shown that β diversity declines with successional time
(Anderson, 2007; Li et al., 2016), implying that the magnitude of the
stabilizing effect of β diversity via spatial asynchrony may decline
over time (Mori et al., 2018). However, our SEM showed that β diver-
sity did not change with successional time but significantly increased
spatial asynchrony, thus increasing functional stability at the large
transect scale (Figure 3). Our results support the prediction that
spatial asynchrony among localities was enhanced by β diversity, al-
though β diversity contributed to a smaller stabilizing effect than α
diversity. In contrast, some studies have also shown that β diversity
is not related to spatial asynchrony (Wilcox et al., 2017), probably
because the relatively small sampling size and homogeneous envi-
ronmental conditions in these studies could weaken the stabilizing
effect of β diversity. Two recent studies, conducted with a larger
spatial extent, provided robust evidence for the stronger contribu-
tion of β diversity than α diversity to ecosystem stability at large
spatial scales (Liang et al., 2022; Qiao et al., 2022). The responses of
spatial asynchrony among communities to environmental variations
provide a spatial insurance effect to stabilize ecosystem function at
large spatial scales (Loreau et al., 2003; Wang & Loreau, 2014).
4.4 | Caveats
One caveat is that we utilized a space- for- time approach to analysing
the natural succession of plant communities (Foster & Tilman, 2000;
Johnson & Miyanishi, 2008; Walker et al., 2010). Although this ap-
proach is often criticized due to reaching false ecological patterns
and temporal dynamics (Johnson & Miyanishi, 2008), it can still be a
useful tool for analysing successional trajectories if it is judiciously
utilized (Walker et al., 2010). In this study, the combination of the
chronosequence approach with the long- term study of perma-
nent plots could better predict successional trajectories (Foster &
Tilman, 2000, Johnson & Miyanishi, 2008, Walker et al., 2010) and
ecosystem stability. Furthermore, we acknowledge that pseudorep-
lication (i.e. no replication in our study sites of the same successional
age) could affect the outcomes and generalizability of our study, al-
though site identity as a random effect was included in LMMs. In
this region, natural grasslands are dominant, and old field sites are
rare. It is difficult to identify multiple sites that share similar succes-
sional stages. Another caveat is that our experiment was conducted
in a relatively small spatial area at each field site. Therefore, our re-
sult must be queried to determine whether it can be generalized to
large spatial scales where β diversity may contribute to ecosystem
stability at regional scales and spatial asynchrony among local com-
munities (Liang et al., 2022; Wang & Loreau, 2016). In our study, we
focused on the temporal stability of both ecosystem function and
community composition. We discovered that functional stability was
significantly positively correlated with compositional stability at the
pl ot scal e but not at th e trans ect sc a le (Figure S9). However, no single
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
|
9
Journal of Ecology
LI et al.
facet of stability can sufficiently reflect the overall ecosystem sta-
bility (Donohue et al., 2013; Hillebrand, Langenheder, et al., 2018),
thus preventing generalizable conclusions regarding overall stability
across ecosystems. Therefore, in the future, it is necessary to con-
duct more successional experiments on different ecosystem types
to verify how succe ssion influences the rel at io nships bet ween diver-
sity and multidimensional ecosystem stability at large spatial scales.
5 | CONCLUSIONS
Our study demonstrates that ecological succession increases plant
diversity and functional and compositional stability at spatiotempo-
ral scales. α diversity and β diversity provide stabilizing effects for
a large spatial scale by local α stability and spatial asynchrony, re-
spectively, during succession. The effect of α stability on γ stability
is greater than that of spatial asynchrony on γ stability during suc-
cession, regardless of functional and compositional stability. These
findings have the following important implications: (i) the positive
effects of biodiversity (α, γ and β) on ecosystem stability in natural
systems are spatially and temporally dependent. The effects of bio-
diversity are not only important on short term, small local scales but
also become even stronger in long term, large- scale landscapes. (ii)
Findings on biodiversity stabilizing mechanisms at local spatial scales
from shor t- term studies (e.g. Hautier et al., 2020; Wang et al., 2021;
Wilcox et al., 2017; Zhang et al., 2019) are likely to underestimate the
temporal impacts of biodiversity change in real- world ecosystems
(Qiu & Cardinale, 2020), and their effects could strengthen over
longer periods (Wagg et al., 2022), particularly across successional
stages. (iii) In the context of ongoing biodiversit y loss, it is vit al to re-
store multiple components of biodiversity at multiple spatial scales
(such as local, regional and habitat) to stabilize ecosystem functions,
especially macrosystem stability (Patrick et al., 2021). Temporal or
successional changes should be considered, if possible, for a more
comprehensive understanding of the effects of biodiversity on eco-
system functioning in a given ecological system (Lasky et al., 2014;
Wagg et al., 2022).
AUTHOR CONTRIBUTIONS
Wenjin Li and Xi Zhou developed and framed research questions and
drafted the manuscript, Jinhua Li designed the experiment and con-
tributed to paper writing, Shaopeng Wang, Michel Loreau and Lin
Jiang revised the manuscript, Xi Zhou and Zhiqiang Xiang contrib-
uted to data analyses, Wenjin Li and Jinhua Li carried out the field-
work and collected the data. All authors contributed substantially to
manuscript writing and revisions. Wenjin Li and Xi Zhou contributed
equally to this work.
ACKNO WLE DGE MENTS
We appreciated the staff of Gannan Grassland Ecosystem National
Observation and Research Station and the interns for their help in
the field work. This research was supported by funding from National
Natural Science Foundation of China (32271761; 31470480) to W.L.,
the TULIP Laboratory of Excellence (ANR- 10- LABX- 41) to M.L. and
the US National Science Foundation (DEB- 1856318 and CBET-
1833988) to L.J. We also thank Yann Hautier, François Gillet and
two anonymous reviewers for their valuable comments.
CONFLICT OF INTEREST STATEMENT
The authors declare no competing interests.
PEER REVIEW
The peer review history for this article is available at h t tps://
www.webof scien ce.com/api/gatew ay/wos/peer- revie w/10.1111/
1365- 2745.14133.
DATA AVA ILAB ILITY STATE MEN T
The data are archived with Dryad https://doi.org/10.5061/dryad.
qv9s4 mwkq. (Li et al., 2023).
ORCID
Wenjin Li https://orcid.org/0000-0002-6426-4852
Xi Zhou https://orcid.org/0000-0002-7315-2086
Zhiqiang Xiang https://orcid.org/0000-0002-9879-9089
Jinhua Li https://orcid.org/0000-0003-3338-2925
Shaopeng Wang https://orcid.org/0000-0002-9430-8879
Michel Loreau https://orcid.org/0000-0002-0122-495X
Lin Jiang https://orcid.org/0000-0002-7114-0794
REFERENCES
Anderson, K. J. (2007). Temporal patterns in rates of communit y change
during succession. The American Naturalist, 16 9, 780– 793. https://
doi.org/10.1086/516653
Clark, A. T., Arnoldi, J.- F., Zelnik, Y. R., Barabas, G., Hodapp, D., Karakoç,
C., König, S., Radchuk, V., Donohue, I., Huth, A., Jacquet, C., de
Mazancourt, C., Mentges, A., Nothaaß, D., Shoemaker, L. G.,
Taube rt, F., Wiega nd, T., Wan g, S. , Ch ase , J. M., … Harp ole, S. (2021).
General statistical scaling laws for stability in ecological systems.
Ecology Letters, 24, 1474– 1486. https://doi.org/10.1111/el e.13760
Connell, J. H., & Slatyer, R. O. (1977). Mechanisms of succession in nat-
ural communities and their role in community stability and orga-
nization. The A merican Naturalist, 111, 1119– 1144. ht tps:// d o i .
org/10.1086/283241
Cow les, H. C. (1899). The ecologic al rel at io ns of the vegetation on the sand
dunes of Lake Michigan. Part I.- geographical relations of the dune
floras. Botanical Gazette, 27, 95– 117. https://doi.org/10.10 86/327796
Donohue, I., Hillebrand, H., Montoya, J. M., Petchey, O. L., Pimm, S.
L., Fowler, M. S., Healy, K., Jackson, A. L ., Lurgi, M., McClean, D.,
O'Connor, N. E., O'Gorman, E. J., & Yang, Q. (2016). Navigating the
complexity of ecological stability. Ecology Letters, 19, 1172– 1185.
https://doi.or g/10.1111/ele.126 48
Donohue, I., Petchey, O. L., Montoya, J. M., Jackson, A. L., McNally, L .,
Viana, M., Healy, K., Lurgi, M., O'Connor, N. E., & Emmerson, M. C.
(2013). On the dimensionality of ecological stability. Ecology Letters,
16, 421– 429. https://doi.org/10.1111/ele .12086
Foster, B. L., & Tilman, D. (2000). Dynamic and static views of succes-
sion: Testing the descriptive power of the chronosequence ap-
proach. Plant Ecology, 146 , 1– 10. https://doi.org/10.1023/A:10098
9510 3017
Gonzalez, A., Germain, R. M., Srivastava, D. S., Filotas, E., Dee, L. E.,
Gravel, D., Thompson, P. L., Isbell, F., Wang, S. P., Kefi, S., Montoya,
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10
|
Journal of Ecology
LI et al.
J., Zelnik, Y. R., & Loreau, M. (2020). Scaling- up biodiversity-
ecosystem functioning research. Ecology Letters, 23, 757– 776.
https://doi.or g/10.1111/ele.13456
Grimm, V., & Wissel, C. (1997). Babel, or the ecological stability dis-
cussions: an inventory and analysis of terminology and a guide
for avoiding confusion. Oecologia, 109, 323– 334. h t t ps : //doi.
org/10.1007/s0044 20050090
Haut ie r, Y., Zh an g, P., Lo re au, M., Wi lcox, K. R., Seabl oom, E. W., Bo re r,
E. T., Byrnes, J. E. K., Koerner, S. E., Komatsu, K. J., Lefcheck, J.
S., Hector, A., Adler, P. B., Alberti, J., Arnillas, C. A., Bakker, J.
D., Brudvig, L. A ., Bugalho, M. N., Cadotte, M., Caldeira, M. C.,
… Wang, S. (2020). General destabilizing effects of eutrophica-
tion on grassland productivity at multiple spatial scales. Nature
Communications, 11, 5375. https://doi.org/10.1038/s4146 7- 020-
1 9 2 5 2 - 4
Hillebrand, H., Blasius, B., Borer, E. T., Chase, J. M., Downing, J. A.,
Eriksson, B. K., Filstrup, C. T., Harpole, W. S., Hodapp, D., Larsen,
S., Lewandowska, A. M., Seabloom, E. W., Van de Waal, D. B., &
Ryabov, A. B. (2018). Biodiversity change is uncoupled from
species richness trends: Consequences for conservation and
monitoring. Journal of Applied Ecology, 55, 169– 184. ht t p s : //d o i .
org /10.1111/1365- 2664.12959
Hillebrand, H., L angenheder, S., Lebret, K., Lindström, E., Östman, Ö., &
Striebel, M. (2018). Decomposing multiple dimensions of stability in
global change experiments. Ecology Letters, 21, 21– 30. htt p s://d o i .
org /10.1111/ele.12867
Isbell, F., Cowles, J., Dee, L. E., Loreau, M., Reich, P. B., Gonzalez, A.,
Hector, A., & Schmid, B. (2018). Quantifying effects of biodiversity
on ecosystem functioning across times and places. Ecology Letters,
21, 763– 778. ht tps://doi.org/10.1111/ele.12928
Johnson, E. A., & Miyanishi, K. (2008). Testing the assumptions of chro-
nosequences in succession. Ecology Letters, 11, 419– 431. ht t p s : //
doi.org/10.1111/j.1461- 0248 .2 008.01173.x
Lasky, J. R., Uriarte, M., Boukili, V. K., Erickson, D. L., John Kress, W.,
& Chazdon, R. L. (2014). The relationship between tree biodiver-
sity and biomass dynamics changes with tropical forest succession.
Ecology Letters, 17, 1158– 1167. htt ps://doi.or g/10.1111/ele.12322
Lefcheck, J. S. (2016). piecewiseSEM: Piecewise structural equa-
tion modelling in R for ecology, evolution, and systemat-
ics. Methods in Ecolog y and Evolution, 7, 573– 579. h t t p s : //doi.
org /10.1111/2041- 210X .12512
Lehman, C. L., & Tilman, D. (200 0). Biodiversity, stability, and productiv-
ity in competitive communities. The American Naturalist, 156 , 534–
552. https://doi.org/10.1086/303402
Lepš, J., Götzenberger, L., Valencia, E., & de Bello, F. (2019). Accounting
for long- term directional trends on year- to- year synchrony in
species fluctuations. Ecography, 42, 1728– 1741. h t t p s : //doi.
org /10.1111/ecog.04528
Li, S., Cadotte, M. W., Meiners, S. J., Pu, Z., Fukami, T., & Jiang, L. (2016).
Convergence and divergence in a long- term old- field succession:
the importance of spatial scale and species abundance. Ecology
Letter s, 19, 1101– 1109. https://doi.org/10.1111/el e.12647
Li, W., Li, J., Knops, J. M., Wang, G., Jia, J., & Qin, Y. (2009). Plant com-
munities, soil carbon, and soil nitrogen properties in a successional
gradient of sub- alpine meadows on the eastern Tibetan plateau
of China. Environmental Management, 44, 755– 765. h t tps://doi.
o r g / 1 0 . 1 0 0 7 / s 0 0 2 6 7 - 0 0 9 - 9 3 6 1- 1
Li, W., Li, J., Liu, S., Zhang, R., Qi, W., Zhang, R., Knops, J. M. H., & Lu, J.
(2017). Magnitude of species diversity effect on aboveground plant
biomass increases through successional time of abandoned farm-
lands on the Eastern Tibetan Plateau of China. Land Degradation &
Development, 28, 370– 378. https://doi.org/10.1002/ldr.2607
Li, W., Luo, S., Wang, J., Zheng, X., Zhou, X., Xiang, Z., Liu, X., & Fang, X.
(2022). Nitrogen deposition magnifies destabilizing effects of plant
functional group loss. Science of the Total Environment, 835, 155419.
https://doi.org/10.1016/j.scito tenv.2022.155419
Li, W., Zhou, X., Xiang, Z., Li, J., Wang, S., Loreau, M., & Jiang, L. (2023).
Data from: Biomass temporal stability increases at two spatial
scales during secondary succession. Journal of Ecology https://d o i .
org/10.5061/dryad.qv9s4 mwkq
Liang, M., Baiser, B., Hallett, L. M., Hautier, Y., Jiang, L., Loreau, M.,
Record, S., Sokol, E. R., Zarnetske, P. L., & Wang, S. (2022).
Consistent stabilizing effects of plant diversity across spatial scales
and climatic gradients. Nature Ecology & Evolution, 6, 1669– 1675.
h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / s 4 1 5 5 9 - 0 2 2 - 0 1 8 6 8 - y
Loreau, M. (2022). Biodiversity and ecosystem stability: New theoretical
insights. In The ecological and societal consequences of biodiversity
loss (pp. 145– 166). John Wiley & Sons.
Loreau, M., Mouquet, N., & Gonzalez, A. (2003). Biodiversity as spatial
insurance in heterogeneous landscapes. Proceedings of the National
Academy of Sciences of the United States of America, 100, 12765–
12770. https://doi.org /10.1073/pnas .22354 65100
Luo, M., Reuman, D. C., Hallett, L. M., Shoemaker, L., Zhao, L., Castorani,
M. C. N., Dudney, J. C., Gherardi, L. A., Rypel, A . L., Sheppard, L.
W., Walter, J. A., & Wang, S. (2021). The effec ts of dispersal on sp a-
tial synchrony in metapopulations differ by timescale. Oikos, 130,
1762– 1772. https://doi.or g/10.1111/oik.08298
Mellin, C., Bradshaw, C. J. A., Fordham, D. A., & Caley, M. J. (2014).
Strong but opposing β diversity- stability relationships in
coral reef fish communities. Proceedings of the Royal Society B:
Biological Sciences, 281, 20131993. https://doi.org/10.1098/
rspb. 2013.1993
Meyer, S. T., Ebeling, A., Eisenhauer, N., Hertzog, L., Hillebrand, H., Milcu,
A., Pompe, S., Abbas, M., Bessler, H., Buchmann, N., De Luca, E.,
Engels, C., Fischer, M., Gleixner, G., Hudewenz, A., Klein, A.- M., de
Kroon, H., Leimer, S., Loranger, H., … Weisser, W. W. (2016). Effects
of biodiversity strengthen over time as ecosystem functioning
declines at low and increases at high biodiversity. Ecosphere, 7,
e0 1619. ht tps://doi.org /10.10 02/ecs2.1619
Mori, A. S., Isbell, F., & Seidl, R. (2018). β- Diversity, community assembly,
and ecosystem functioning. Trends in Ecology & Evolution, 33, 549–
564. https://doi.org/10.1016/j.tree.2018.04.012
Mori, A. S., Osono, T., Cornelissen, J. H. C., Craine, J., & Uchida, M. (2017).
Biodiversity– ecosystem function relationships change through pri-
mary succession. Oikos, 126, 1637– 1649. htt ps://doi.org/10 .1111/
oik.0 4345
Nakagawa, S., & Schielzeth, H. (2013). A general and simple method
for obtaining R2 from generalized linear mixed- effects mod-
els. Methods in Ecology and Evolution, 4, 133– 142. h t t p s : //doi.
org /10.1111/j .2 041- 210 x.2012.0 0261.x
Odum, E. P. (1969). The strategy of ecosystem development. Science,
164, 262– 270. https://doi.org/10.1126/scien ce.164.3877.262
Oosting, H. J. (1942). An ecological analysis of the plant communities of
piedmont, North Carolina. American Midland Naturalist, 28, 1.
Patrick, C. J., McCluney, K. E., Ruhi, A., Gregor y, A., Sabo, J., & Thorp,
J. H. (2021). Multi- scale biodiversity drives temporal variability in
macrosystems. Frontiers in Ecology and the Environment, 19, 47– 56.
https://doi.org/10.1002/fee.2297
Pimm, S., & Redfearn, A. (1988). The variability of population densities.
Nature, 334, 613– 614. https://doi.org/10.1038/334613a0
Pimm, S. L. (1984). The complexity and stability of ecosystems. Nature,
307, 321– 326. htt ps://doi.org/10.1038/307321a0
Pinheiro, J., Bates, D., DebRoy, S., & Sarkar, D. (2021). nlme: linear and
nonlinear mixed effects models, R package version 3.1- 153.
Pr ach , K., & Walker, L. R. (2011 ). Four oppor tunit ies for st udies of ecolo g-
ical succession. Trends in Ecolog y & Evolution, 26, 119– 123. h t t p s : //
doi.org/10.1016/j.tree.2010.12.007
Qiao, X., Geng, Y., Zhang, C., Han, Z., Zhang, Z., Zhao, X., & von Gadow,
K. (2022). Spatial asynchrony matters more than alpha stability in
stabilizing ecosystem productivity in a large temperate forest re-
gion. Global Ecology and Biogeogra phy, 31, 1133– 1146. htt p s://d o i .
org /10.1111/geb.1348 8
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
|
11
Journal of Ecology
LI et al.
Qiu , J., & Cardinale, B. J. (2020 ). Scaling up biod iversity– ecosystem fu nc-
tion relationships across space and over time. Ecology, 101, e03166.
https://doi.org/10.1002/ecy.3166
R Core Team. (2019). R: A language and environment for statistical comput-
ing. R Foundation for Statistical Computing.
Reu, J. C., Catano, C. P., Spasojevic, M. J., & Myers, J. A. (2022). Beta
diversity as a driver of forest biomass across spatial scales. Ecology,
103, e3774. https://doi.org/10.1002/ecy.3774
Su, J., Zhao, Y., Xu, F., & Bai, Y. (2022). Multiple global changes drive grass-
land productivit y and stability: A meta- analysis. Journal of Ecology,
110, 2850– 2869. https://doi.org/10.1111/1365- 2745 .13983
Thibaut, L. M., & Connolly, S. R. (2013). Understanding diversity– stability
relationships: Towards a unified model of portfolio ef fects. Ecology
Letter s, 16, 140– 150. htt ps://doi.org/10 .1111/ele.12019
Tilman, D. (1996). Biodiversity: population versus ecosystem stability.
Ecology, 77, 350– 363. https://doi.org/10.2307/2265614
Tilman, D. (1999). The ecological consequences of changes in biodiver-
sity: A search for general principles. Ecology, 80, 1455– 1474. https://
doi.org/10.1890/0012-9658(1999)080[1455:TECOC I]2.0.CO;2
Tilman, D., Reich, P. B., & Kn ops, J. M. H. (20 06). Bi od iv er si ty and ecos ys-
tem stability in a decade- long grassland experiment. Nature, 441,
6 2 9 – 6 3 2 . h ttp s://doi.or g/10 .103 8/n atur e04742
Valencia, E., de Bello, F., Lepš, J., Galland, T., E- Vojtkó, A., Conti, L.,
Danihelka, J., Dengler, J., Eldridge, D. J., Estiarte, M., García-
González, R., Garnier, E., Gómez, D., Harrison, S., Herben, T., Ibáñez,
R., Jentsch, A., Juergens, N., Kertész, M., … Götzenberger, L. (2020).
Directional trends in species composition over time can lead to a
widespread overemphasis of year- to- year asynchrony. Journal of
Vegetation Science, 31, 792– 802. https://doi.or g/10.1111/jvs.12916
Wagg, C., Roscher, C., Weigelt, A., Vogel, A ., Ebeling, A., de Luca,
E., Roeder, A., Kleinspehn, C., Temperton, V. M., Meyer, S. T.,
Scherer- Lorenzen, M., Buchmann, N., Fischer, M., Weisser, W. W.,
Eisenhauer, N., & Schmid, B. (2022). Biodiversity– stability relation-
ships strengthen over time in a long- term grassland experiment.
Nature Communications, 13, 7752. https://doi.org/10.1038/s4146
7 - 0 2 2 - 3 5 1 8 9 - 2
Walker, L. R., & Wardle, D. A. (2014). Plant succession as an integrator of
contrasting ecological time scales. Trends in Ecology & Evolution, 29,
50 4– 510 . https://doi.org/10.1016/j.tree.2014.07.002
Walker, L. R., Wardle, D. A ., Bardgett, R. D., & Clarkson, B. D. (2010).
The use of chronosequences in studies of ecological succession
and soil development. Journal of Ecology, 98, 725– 736. h t tps: //doi.
org/10.1111/j.1365- 2745.2010.01664.x
Wang, S., Lamy, T., Hallett, L. M., & Loreau, M. (2019). Stability and
synchrony across ecological hierarchies in heterogeneous meta-
communities: Linking theory to data. Ecography, 42, 1200– 1211.
https://doi.org/10.1111/ecog.04290
Wang, S., & Loreau, M. (2014). Ecosystem stability in space: α, β and γ
variability. Ecology Letters, 17, 891– 901. http s://doi.o rg /10.1111/
ele.12292
Wang, S., & Loreau, M. (2016). Biodiversity and ecosystem stability
across scales in metacommunities. Ecology Letters, 19, 510– 518.
https://doi.or g/10.1111/ele.12582
Wang, S., Loreau, M., Arnoldi, J.- F., Fang, J., Rahman, K. A., Tao, S., &
de Mazancourt, C . (2017). An invariability- area relationship sheds
new light on the spatial scaling of ecological stability. Nature
Communications, 8, 15211. https://doi.org /10.1038/ncomm s15211
Wang, S., Loreau, M., de Mazancourt, C., Isbell, F., Beierkuhnlein, C.,
Connolly, J., Deutschman, D. H., Doležal, J., Eisenhauer, N., Hector,
A., Jentsch, A., Kreyling, J., Lanta, V., Lepš, J., Polley, H. W., Reich, P.
B., van Ruijven, J., Schmid, B., Tilman, D., … Craven, D. (2021). Biotic
homogenization destabilizes ecosystem functioning by decreasing
spatial asynchrony. Ecology, 102, e03332. https://doi.org/10.10 02/
ecy.3332
White, L., O'Connor, N. E., Yang, Q., Emmerson, M. C., & Donohue, I.
(2020). Individual species provide multifaceted contributions to the
stability of ecosystems. Nature Ecolog y & Evolution, 4, 1594– 1601.
h t t p s : / / d o i . o r g / 1 0 . 1 0 3 8 / s 4 1 5 5 9 - 0 2 0 - 0 1 3 1 5 - w
Wilcox, K. R., Tredennick, A . T., Koerner, S. E., Grman, E., Hallett, L. M.,
Avolio, M. L., La Pierre, K . J., Houseman, G. R., Isbell, F., Johnson,
D. S., Alatalo, J. M., Baldwin, A. H., Bork, E. W., Boughton, E. H.,
Bowman, W. D., Britton, A. J., Cahill, J. F., Jr., Collins, S. L., Du, G.,
… Zhang, Y. (2017). Asynchrony among local communities stabi-
lises ecosystem function of metacommunities. Ecology Letters, 20,
1534– 1545. https://doi.org/10.1111/ele.128 61
Xu, Q., Yang, X., Yan, Y., Wang, S., Loreau, M., & Jiang, L. (2021).
Consistently positive effect of species diversity on ecosystem, but
not population, temporal stability. Ecology Letters, 24, 2256– 2266.
https://doi.or g/10.1111/ele.13777
Xu, Q. N., Yang, X., Song, J., Ru, J. Y., Xia, J. Y., Wang, S. P., Wan, S. Q., &
Jiang, L. (2022). Nitrogen enrichment alters multiple dimensions of
grassland functional stability via changing compositional stability.
Ecology Letters, 25, 2713– 2725. https ://doi.or g/10.1111/ele.14119
Zhang, Y., Feng, J., Loreau, M., He, N., Han, X., & Jiang, L. (2019). Nitrogen
addition does not reduce the role of spatial asynchrony in stabilis-
ing grassland communities. Ecology Letters, 22, 563– 571. h t t p s : //
doi.org/10.1111/ele.13212
SUPPORTING INFORMATION
Additional supporting information can be found online in the
Supporting Information section at the end of this article.
Table S1. Rationales of the priori structural equation model were
developed to test the direct and indirect effects of successional year
through plant diversity on stability at the plot and larger transect
spatial scales.
Table S2. The results of linear mixed- effects models (LMMs) from
Fi gur e 1 to Fi g ure 3; th e mar gin al (R2
m) an d con dit i o nal (R2
c) r- squared
represent “fixed effects” and “fixed + random effects” explanations,
respectively.
Table S3. The results of linear mixed- effects models (LMMs) from
Figure S3 to Figure S5, with “site” as random effect. The marginal
(R2
m) and conditional (R2
c) r- squared represent “fixed effects” and
“fixed + random effects” explanations, respectively.
Table S4. The results of linear mixed- effects models (LMMs) from
Figure S6 to Figure S10 (except for Figure S7), with “site” as random
effect. The marginal (R2
m) and conditional (R2
c) r- squared represent
“fixed effects” and “fixed + random effects” explanations, respectively.
Figure S1. Temporal changes in SERa and SERr at the plot level in
each old field site with sampling year. This richness- based species-
exchange ratio(SERr), is quantified as SERr =
S
imm
+Sext
S
tot
, where Simm is
the number of species immigrating (newly recorded in the later
sample), Sext is the number of species extinct (lost from the previous
sample) and Stot is the total number of species across both samples.
SERa as a measure of turnover by changes in species proportional
abundances, SERa=∑i(pi−p�i)
2
∑pi
2
+∑p�i
2
−∑pip�i
,
pi
and
p′i
represent the
proportional abundances of species i in the first (time 1, i.e. in 2004)
and second (time 2, i.e. in 2005) communities, respectively. The solid
lines represent the significant relationships from linear models at
p < 0.05, and the dash lines represent the non- significant relationships
at p > 0.05.
Figure S2. A init ial structura l equati on modelin g (SEM) for predicting
successional year on ecosystem temporal stability via plant diversity
at two spatial scales, please see the rationales in the Table S1.
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12
|
Journal of Ecology
LI et al.
Figure S3. Temporal changes in α diversity (richness) (a) and β
diversity (richness) (b) in each old field site with sampling year. (c)
Changes in seven- year average diversity (richness) at both plot and
transect scales during secondary succession. These diversity metrics
are based on biomass. The solid lines represent the significant
relationships from linear models (a and b) or linear mixed- effects
model (c) at p < 0.05, and the dash lines represent the non- significant
relationships at p > 0.05. The marginal (R2
m) r- squared represents
“fixed effects” explanation. The details of the models can be found
in Table S3.
Figure S4. Th e ef fe ct s of s uc c e ss i on a l ye a r on sp ec ie s as y nc hr o ny an d
spatial asynchrony. The color of the dots represents the different old
field sites. The dash lines represent the non- significant relationships
from a linear mixed- effects model at p > 0.05. The shaded areas are
the error bands and denote 95% confidence intervals. The marginal
(R2
m) r- squared represents ‘fixed effects’ explanation. The details of
the model can be found in Table S3.
Figure S5. The diversity– asynchrony- stability relationships at
both plot and transect scales during secondary succession.
These diversit y, asynchrony and stability metrics are based on
biomass. Species diversity is measured by richness- based metrics.
Relationships between alpha diversity and functional stability at
plot (a) and larger transect (c) scales; relationships between alpha
diversity and compositional stability at plot (b) and larger transect
(d) scales; relationships between beta diversity and gamma (e:
functional; f: compositional) stability; relationships between beta
diversity and spatial asynchrony (g). The color of the dots represents
the different old field sites. The black lines represent the significant
relationships from a linear mixed- effects model at p < 0.1, and the
dash line represent a non- significant relationship at p > 0.1. The
shaded areas are the error bands and denote 95% confidence
intervals. The marginal (R2
m) r- squared represents “fixed effects”
explanation. The detail of the models can be found in Table S3.
Figure S6. Relationships between alpha diversity and species
stability, species asynchrony. The color of the dots represents the
different old field sites. The dash lines represent non- significant
relationships at p > 0.1. The shaded areas are the error bands and
denote 95% confidence intervals. The marginal (R2
m) r- squared
represents “fixed effects” explanation. The details of the model can
be found in Table S4.
Figure S7. Structural equation model (SEM) describing the relative
effects of alpha and beta diversity on gamma (a: Functional;
b: Compositional) stability, through alpha (a: Functional; b:
Compositional) stability and spatial asynchrony at two spatial scales.
In this SEM, species diversity is measured by richness- based metrics.
(a) Fisher's C = 16.537, p = 0.555, df = 18, AIC = 56.537. (b) Fisher's
C = 16.33, p = 0.43, df = 16, AIC = 58. 33. The black and red arrows
represent positive and negative relationships, respectively. The
light- grey (bidirectional) arrows represent a correlation between
variables. All of the lines marked are significant effects (significance
levels of each predictor are *p < 0.05, **p < 0.01, ***p < 0.001). For
each response variable, marginal (R2
m) values showing the variance
explained by the fixed effects and conditional (R2
c) values indicating
the variance explained by the whole model are provided. The width
of each arrows is relative to the standardized path coefficients.
Figure S8. Relationships between alpha functional stability and
compositional stability and species stability, species asynchrony.
The color of the dots represents the different old field sites. The
black lines represent the overall significant relationships from a
linear mixed- effects model at p < 0.05, and the dash line represent
a non- significant relationship at p > 0.05. The shaded areas are the
error ba nds and denote 95% co nf idenc e inter vals. The marg inal (R2
m)
r- squared represents “fixed ef fects” explanation. The details of the
model can be found in Table S4.
Figure S9. Relationships between compositional stability and
functional stability at two spatial scales. The color of the dots
represents the different old field sites. The black lines represent the
overall significant relationships from a linear mixed- effects model at
p < 0.05, and the dash line represent a non- significant relationship
at p > 0.05. The shaded areas are the error bands and denote 95%
confidence intervals. The marginal (R2
m) r- squared represents “fixed
effects” explanation. The detail of the models can be found in Table
S4.
How to cite this article: Li, W., Zhou, X., Xiang, Z., Li, J.,
Wang, S., Loreau, M., & Jiang, L. (2023). Biomass temporal
stability increases at two spatial scales during secondary
succession. Journal of Ecology, 00, 1–12. htt p s : //d oi.
org /10.1111/1365-2745.14133
13652745, 0, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2745.14133 by Lanzhou University, Wiley Online Library on [28/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
A preview of this full-text is provided by Wiley.
Content available from Journal of Ecology
This content is subject to copyright. Terms and conditions apply.
