Forest soils represent a major stock of organic carbon (C) in the
terrestrial biosphere, but the dynamics of soil organic C (SOC) stock are
poorly quantified, largely due to lack of direct field measurements. In this
study, we investigated the 20-year changes in SOC stocks in eight permanent
forest plots, which represent boreal (1998–2014), temperate (1992–2012),
subtropical (1987–2008), and tropical forest biomes (1992–2012) across
China. SOC contents increased significantly from the 1990s to the 2010s,
mostly in the upper 0–20 cm soil depth, and soil bulk densities do not
change significantly during the same period. As a result, the averaged SOC
stocks increased significantly from 125.2±85.2 Mg C ha-1 in the
1990s to 133.6±83.1 Mg C ha-1 in the 2010s across the forest
plots, with a mean increase of 127.2–907.5 kg C ha-1 yr-1. This
SOC accumulation resulted primarily from increasing leaf litter and fallen
logs, which accounts 3.6 %–16.3 % of above-ground net primary production.
Our findings provided direct evidence that China's forest soils have been
acting as significant C sinks, although their strength varies in forests
with different climates.
Introduction
Terrestrial ecosystems have absorbed approximately 30 % of the carbon
dioxide (CO2) emitted from human activity since the beginning of the
industrial era (IPCC, 2013). Forests have contributed more than half of
these carbon (C) fluxes of terrestrial ecosystems (Pan et al., 2011). Since
soils contain a huge C stock in forest ecosystems, even a slight change in
this stock will induce a considerable feedback to atmospheric CO2
concentrations (Lal, 2004; Luo et al., 2011). Thus, accurate assessment of
the changes in soil organic carbon (SOC) is critical to understanding how
forest soils will respond to global climate change. However, it is difficult
to capture the SOC change with short-term measurements (Smith, 2004) because
the soil C pool typically has a longer turnover time and higher spatial
variability compared to the vegetation C pool (Schrumpf et al., 2011;
Canadell and Schulze, 2014).
Previous efforts have estimated the changes in regional SOC stocks with
indirect approaches, such as regional assessments (Yang et al., 2014) and
model simulations (Todd-Brown et al., 2013). These estimates often involve
large uncertainties due to the inherently high spatial variability of soils
and a lack of direct measurements representing large areas (Sitch et al.,
2015). One reliable approach to reducing the uncertainties is to conduct
long-term monitoring of forest SOC stocks at sites that represent broader
landscapes (Prietzel et al., 2016). Unfortunately, such repeated, accurate
field-based measurements of SOC stocks from which to generate change
estimates are generally lacking and inadequate worldwide (Zhao et al.,
2019).
A few soil resampling studies have explored SOC changes in different
forests, but the results are often contradictory. For instance, Schrumpf et
al. (2014) found that SOC in deciduous broadleaved forests in central
Germany increased, with a change rate of 650.0 kg C ha-1 yr-1 from
2004–2009. In contrast, Prietzel et al. (2016) indicated that SOC stocks
in German forests decreased significantly, with average change rates of
988.2 kg C ha-1 yr-1 in forests in the Alps between 1986 and 2011,
and 441.1 kg C ha-1 yr-1 in the Berchtesgaden region between 1976
and 2011. Kiser et al. (2009) found that the hardwood forest soils in
central Tennessee, USA, exhibited a slight C source and that the relative
change rate ranged from -0.4 % yr-1 to 0.3 % yr-1 between 1976
and 2006. Chen et al. (2015) synthesized global SOC changes and found that
the relative rates of change in forest SOC stocks were contradictory among
long-term experiments (0.2 % yr-1), regional comparisons (0.3 % yr-1), and repeated soil samplings (-0.1 % yr-1). Such
discrepancies can be partly attributed to insufficient observations and
inconsistent methodologies. The different effects of changing environmental
factors and nitrogen inputs on soil C dynamics may also be involved (Norby
and Zak, 2011). In addition, to date these studies have primarily been
conducted in the forests of Europe and the USA, but few have been carried
out in China's forests.
Forests in China cover an area of 156 Mha (Guo et al., 2013) and range from
boreal coniferous forests and deciduous broadleaved forests in the northeast
to tropical rain forests and evergreen broadleaved forests in the south and
southwest. They include almost all major forest biomes of the Northern
Hemisphere (Fang et al., 2012). Such variations in climate and forest types
have provided ideal opportunities to examine the spatial patterns of SOC in
relation to meteorological and biological factors. At the national scale,
the mean annual air temperature of China increased by more than 1 ∘C between 1982 and 2011, which is considerably higher than the
global average (Fang et al., 2018). Since the 1980s, the Chinese Government
has implemented several large-scale national forest protection projects.
These climatic changes and conservation practices in China have
significantly stimulated C uptake into forest ecosystems (Fang et al., 2014,
2018; Feng et al., 2019). Several studies have assessed the temporal
dynamics of SOC stock across China's forests using model simulations (Piao
et al., 2009) or regional assessments (Pan et al., 2011; Tang et al., 2018).
However, these estimates revealed contrasting trends in SOC dynamics and
also lacked direct measurements of SOC change.
Therefore, in this study we measured SOC density (C amount per unit area) of
eight permanent forest plots from tropical, subtropical, temperate, and
boreal forests in China during two periods in the 1990s and 2010s to
quantify their SOC changes. We then analyzed the potential biotic and
climatic drivers in the SOC dynamics across these forests. Finally, we
assessed the changes in SOC stocks in China's forests using the site data
obtained from this study.
Locations and climatic conditions of the sites. (a) Great
Xing'anling, the boreal site; (b) Mt. Dongling, the temperate site; (c) Mt.
Dinghu, the subtropical site; and (d) Jianfengling, the tropical site. The
blue and red lines in climatic diagrams are the monthly mean values of
precipitation and temperature, respectively. The blue areas indicate the
period in the year when the precipitation exceeded 100 mm month-1. MAT is
mean annual temperature, and MAP is mean annual precipitation. Publisher’s note: Please note that the above figure contains disputed territories.
Materials and methodsStudy sites
We investigated eight permanent forest plots in four forest sites (from
north to south: Great Xing'anling, Mt. Dongling, Mt. Dinghu, and
Jianfengling) (Fig. 1). The four sites spanned a wide range from 18.7
to 52.6∘ N in latitude, and belonged to boreal,
temperate, subtropical, and tropical climate zones, respectively, with a
climatic difference of approximately 26 ∘C in mean annual
temperature and 1200 mm in mean annual precipitation. The eight plots
are comprised of a boreal larch forest (Larix gmelinii), two temperate deciduous broadleaved
forests (Betula platyphylla and Quercus wutaishanica), a temperate pine plantation (Pinus tabuliformis), a subtropical evergreen
broadleaved forest, a subtropical pine plantation (P. massoniana), a subtropical pine and
broadleaved mixed forest, and a tropical mountain rainforest (for details,
see Table 1).
Location, forest type, mean annual temperature (MAT), and mean
annual precipitation (MAP) at eight forest plots in four climate zones,
together with forest origin and study periods.
Stand characteristics of all eight plots are summarized in Table 1. The
boreal larch forest was a 100-year-old mature stand at the time of the first
sampling (Wang et al., 2001). Three temperate forest plots (birch, oak, and
pine forests) were located along an elevation gradient on Mt. Dongling,
Beijing. Both birch and oak forest plots were 55-year-old secondary forests
at the time of the first sampling, dominated by B. platyphylla and Q. wutaishanica, respectively. The
temperate pine plantation was 30 years old at the time of the first
sampling and was dominated by P. tabuliformis (Fang et al., 2007). Three subtropical
forest plots were located in Dinghu Biosphere Reserve in Guangdong Province,
southern China (Zhou et al., 2006). The subtropical evergreen broadleaved
forest is an old-growth stand that is more than 400 years old, co-dominated by
Castanopsis chinensis, Canarium pimela, Schima superba, and Engelhardtia roxburghiana. The subtropical pine (P. massoniana) plantation was approximately 40 years
old at the time of the first sampling. The mature mixed pine and broadleaved
forest was approximately 110 years old at the time of the first sampling
and represented the mid-successional stages of monsoon evergreen broadleaved
forest in this region. The tropical mountain rainforest plot was located at
the Jianfengling National Natural Reserve, southwestern Hainan (Zhou et al.,
2013). It had not been disturbed for more than 300 years and was dominated
by species in the families Lauraceae and Fagaceae, such as Mallotus hookerianus, Gironniera subaequalis, Cryptocarya chinensis,
Cyclobalanopsis patelliformis, and Nephelium topengii. For detailed descriptions of these eight plots, see the Supplement.
Soil sampling and calculation of SOC content
The first sampling was conducted between 1987 and 1998 in each of the eight
forests (Table 1). We remeasured the same sample plots in each forest
between 2008 and 2014 using identical sampling protocols.
In each forest plot, two to five pits were dug to collect soil samples for
analyzing the physical and chemical properties during the two sampling
periods (mostly in the 1990s during the first sampling period and in the 2010s
during the second sampling period). The samples were taken at depth
intervals of 10 cm down to the maximum soil depth. In brief, for the boreal
forest, three soil pits were established down to the 40 cm soil depth in
random locations in the growing season in 1998. In August 2014, three soil
pits were again randomly excavated to the same soil depth to allow sampling
for SOC content and bulk density. For the three temperate forests, two soil
profiles (100 cm depth) were dug in each plot to collect soil samples at 10 cm intervals during the summer of 1992. In the summer of 2012, three soil
profiles were dug, and soils were sampled from the same horizons in each
soil profile (Zhu et al., 2015). The first sampling in the three subtropical
forests was conducted in September 1988 in the evergreen and pine plots, and
in 1987 for the mixed plot, both at the end of the rainy season and at the
beginning of the dry season. Five soil pits (60 cm depth) were randomly
excavated to collect samples for the calculation of SOC content and bulk
density. In September 2008, the soil sampling was repeated. For the tropical
forest, five soil profiles (100 cm depth) were established at 10 cm
intervals during summer 1992 and again in summer 2012.
We used consistent sampling and analysis approaches to determine the bulk
density and SOC content between the two sampling times. Three bulk density
samples were obtained for each layer using a standard container that was 100 cm3
in volume. The soil moisture was determined by weighing to the nearest 0.1 g
after 48 h oven-drying at 105 ∘C. The bulk density was calculated
as the ratio of the oven-dried mass to the container volume. Another three
paired samples for C analysis were air-dried. Following this, fine roots were removed by
hand and sieved (2 mm mesh). The SOC content was measured using the wet
oxidation method (Nelson and Sommers, 1982) and was calculated according to
Eq. (1):
SOC=∑i=1nCCi×Bdi×Vi×HFi,
where CCi, Bdi, and Vi are SOC content (%), bulk density (kg m-3), and volume (m3) at the ith soil horizon, respectively.
HFi is calculated as 1-stonevolume+rootvolumeVi and is a dimensionless factor that represents the fine soil
fraction within a certain soil volume.
Calculation of above-ground biomass (AGB) and net primary production
Diameter at breast height (DBH, 1.3 m) and height of all living trees with
DBH >5 cm were both measured in each plot in the 1990s and 2010s. The
AGB of different components (stem, bark, branches, and foliage) was
estimated for all tree species using allometric equations (Table S1 in the Supplement). A
standard factor of 0.5 was used to convert biomass to C (Leith and
Whittaker, 1975). The net increment of AGB (ΔStore) was
calculated for each plot as the difference between the biomass in the 1990s
and the 2010s. The above-ground net primary production (ANPP, kg C ha-1 yr-1) was calculated from Eq. (2):
ANPP=Litterfall+ΔStore+Mortality,
where Litterfall and ΔStore are litter production and above-ground
net biomass increment per year, respectively. Mortality (defined as
above-ground deadwood production) was estimated as the summed production of
fallen logs and standing snags per year.
Litter and fallen log production
Annual litterfall was collected from June 2010 to June 2013 in the tropical
sites, from June 1990 to June 2008 in the subtropical sites, from April to
November 2011–2014 in the temperate sites, and from May to October 2010–2014 in the boreal sites. Litter (leaves, flowers, fruits, and woody
material < 2 cm diameter) was collected monthly from 10 to 15 L
traps (1×1 m2, 1 m above ground) in each plot to calculate
annual litter production. After collection, the samples were taken to the
laboratory, oven-dried at 65 ∘C to a constant mass and weighed.
The 10–15 replicates from each plot were averaged as the monthly mean
value. Annual litter production (kg C ha-1 yr-1) was estimated as
the sum of the monthly production in the year of collection.
Log production represents the mortality (i.e., death of entire trees) per
year. Annual log production was determined from 2010 to 2013 in tropical
sites, from 1989 to 1996 in subtropical sites, from 2011 to 2014 in
temperate sites, and from 2010 to 2014 in boreal sites. Stocks of fallen
logs were harvested and weighed during each investigated year.
Forest area and fossil fuel emission data
To calculate the amount of C sequestration in China's forest soils, we
estimated the changes in the national forest SOC stocks. We used the mean
SOC accumulation rates obtained from this study and the data of forest area
for each forest type documented in the national forest inventory in
1989–1993, which approximates the first sampling period in the present
study (Guo et al., 2013). The changes in national forest SOC stock were
calculated as the product of SOC density, SOC density change rate, and
forest area for major forest types during the period 1989–1993. In
addition, to evaluate the relative importance of forest soil C sequestration
in the national C budget, we obtained the data of fossil fuel emissions
during 1991–2010 from the Carbon Dioxide Information Analysis Center (Zheng
et al., 2016).
ResultsChanges in SOC
SOC stocks were investigated in eight permanent forest plots in four forest
sites from northern to southern China over two periods: the 1990s and 2010s.
The changes in SOC contents, bulk density, and SOC stocks in the top 20 cm
soil layer between the 1990s and the 2010s are shown in Fig. 2 and Figs. S1 and
S2 in the Supplement. The paired t-test analysis indicated that SOC content in the 0–20 cm depth was significantly higher in the 2010s than in the 1990s (3.2±0.7 % vs. 2.9±0.6 %; t=-5.65; P<0.001) (Table 2). The
average rate of increase in SOC content was 0.02 % yr-1 in the top 20 cm depth, ranging from 0.01 % yr-1 to 0.04 % yr-1 across the
study sites. These rates of increase in SOC content in the 0–10 cm horizon
(0.03±0.02 % yr-1) were 3 times larger than those in the
10–20 cm horizon (0.01±0.01 % yr-1) (Table S2). At the same
time, the bulk density of the top 20 cm soil layer decreased in most sites
(6 of 8 sites), with an average rate of decrease of 2.7±3.7 mg cm-3 yr-1 (Table S3). As a result, the SOC stock in the top 20 cm
soil layer was found to have increased significantly in the past 2 decades
(t=-5.85, P<0.001, Table 2), with an average accumulation rate of
332.4±200.2 kg C ha-1 yr-1 (0.7±0.4 % yr-1;
Fig. 2; see Table S3). The temperate pine plantation experienced the
largest increase in SOC stock in the top 20 cm depth (630.8±111.2 kg C ha-1 yr-1). In contrast, the smallest rate of increase was
observed in the subtropical mixed forest (117.3±25.2 kg C ha-1 yr-1). It should be noted that SOC stock in the top 20 cm depth in the
subtropical evergreen old-growth forest increased from 35.6±6.0 Mg C ha-1 in 1988 to 45.6±6.9 Mg C ha-1 in 2008 (increased by
498.3±78.8 kg C ha-1 yr-1), which led to the highest
relative accumulation rate (1.4±0.2 % yr-1) among the study
sites.
Mean soil organic carbon (SOC) content (a), bulk density (b), SOC
stock (c), and their relative change rates (d) within 0–20 cm soil depth in
the 1990s and the 2010s for the four forest sites in China. For more
details, see Table S2 in the Supplement.
Results of the paired-sample t tests for soil organic carbon (SOC)
content, bulk density, and SOC stock at different soil depths in the eight
forest plots between the 1990s and the 2010s.
We further compared SOC stocks of the whole soil profile between 1990s and
2010s at a depth of 0–40 cm in the boreal site, 0–60 cm in the subtropical
site, and 0–100 cm in the temperate and tropical sites (Fig. 3). The SOC
stocks of all sampling sites in the 2010s were higher than those in the
1990s. The paired t-test analysis revealed a significant increase in SOC
stocks for the whole soil profile during the sampling period (t=-4.15, P<0.01; Table 2). The mean SOC stocks of the whole soil profile in the eight
forests increased from 125.2±85.2 Mg C ha-1 in the 1990s to
133.6±83.1 Mg C ha-1 in the 2010s, with an accumulation rate of
421.2±274.4 kg C ha-1 yr-1 and a relative increase rate of
0.6±0.5 % (Fig. 2). The SOC accumulation rates displayed large
variability among different climate zones and forest types. For different
climate zones, the SOC accumulation rates in the subtropical and tropical
sites were relatively higher than those in the boreal and temperate sites
(Fig. 3). The greatest increase in SOC stock occurred in the subtropical
evergreen old-growth forest (907.5±60.1 kg C ha-1 yr-1),
and the least in the temperate deciduous oak forest (127.2±25.3 kg C ha-1 yr-1; Table S3). The relative rates of increase in the
subtropical evergreen old-growth forest (1.3±0.1 % yr-1) and
the subtropical mixed forest (1.5±0.2 % yr-1) were higher than
those in the temperate forests (0.1±0.0 % yr-1 in the oak
forest, 0.1±0.0 % yr-1 in the pine forest, and 0.2±0.0 % yr-1 in the birch forest; Table S3).
In addition, the rates of SOC increase (127.2–907.5 kg C ha-1 yr-1) was equivalent to 3.6 %–16.3 % of ANPP (3340.1–6944.7 kg C ha-1 yr-1), with the highest rate in the subtropical evergreen
forest (16.3±4.2 %) and the lowest in the temperate oak forest
(3.6±3.4 %) (Tables 3 and S4).
Relationships between SOC change rates and biotic and climatic variables
To understand the possible mechanisms for the rates of SOC increase as
described above, we analyzed the driving forces for this significantly
increased SOC stock using measurements of AGB growth rate, above-ground
litter and fallen log production, and ANPP (Table 3). The linear regression
analysis showed that there was no significant correlation between SOC change
rates and AGB growth rate (P>0.05; Fig. 4a). The SOC
accumulation rates were positively and significantly associated with annual
litter (R2=0.66; P=0.01; Fig. 4b) and fallen log production
(R2=0.69; P=0.01; Fig. 4c). The SOC accumulation rates across
these forests were closely associated with the observed ANPP (R2=0.55; P=0.03; Fig. 4d) and also showed an increasing trend with
increasing mean annual temperature and precipitation, despite being insignificant
(both P>0.1; Fig. 4e and f). The multiple regression analysis
indicated the relative effects of biotic factors (AGB growth rate, litter,
and fallen log production) and climatic factors (mean annual temperature and
precipitation) on the rates of SOC increase (Fig. 4g). When the effects of
climatic factors were under control, the biotic factors independently
explained 56.4 % of the variations. By comparison, when the effects of
biotic factors were under control, only 7.5 % of the variations were
explained by the climatic factors.
Measured C stocks and fluxes of the four forest sites in China
during the 1990s and the 2010s.
ParameterBorealTemperateSubtropicalTropicalCarbon pool (Mg C ha-1) AGB91.1±25.089.6±17.4107.0±41.7213.6±41.4Litter4.4±0.03.9±1.32.1±0.71.8±0.2Deadwood1.3±0.54.5±1.27.3±6.75.7±0.8Soil69.4±6.2231.6±14.667.2±19.5102.6±19.9Ecosystem total166.2±31.7329.6±34.5183.7±68.5323.7±62.3Carbon flux (kg C ha-1 yr-1) AGB growth899.4±411.01809.5±521.2798.7±1572.4684.1±145.0litterfall2424.2±283.11946.7±361.23385.4±1444.63970.0±279.8Fallen log13.0±3.7106.1±74.5986.7±967.31034.2±71.6Standing snag3.5±1.8276.7±111.1220.0±135.7803.4±62.4ANPP3340.1±698.84139.0±607.75390.8±1655.36491.6±559.2Soil accumulation243.4±31.1283.6±138.5627.6±370.1397.9±84.2Ratio of soil accumulation to ANPP (%)7.3±7.86.7±2.811.0±5.36.1±3.3(3.6∼9.2)(5.7∼16.3)
Carbon pool of each ecosystem component at the time of the second
sampling (2010s): AGB, above-ground biomass; ANPP, above-ground net primary
production. For details, see Table S1 in the Supplement.
DiscussionSOC accumulation
Previous evidence of forest SOC changes comes mainly from individual
experiments (Prietzel et al., 2006; Kiser et al., 2009; Häkkinen et al.,
2011) or regional comparisons (Lettens et al., 2005; Pan et al., 2011; Ortiz
et al., 2013) in European and American forests. In this study, we performed
a broadscale forest soil resampling to evaluate changes in SOC stock across
eight permanent forest plots in China. Our measurements suggest that SOC
stocks exhibited a significant accumulation in these forests from the 1990s
to the 2010s, at the accumulation rate of 127.2–907.5 kg C ha-1 yr-1. These accumulation rates are comparable to those of other studies
that were primarily conducted in boreal and temperate forests in other
regions (-11.0–812.0 kg C ha-1 yr-1, Fig. 5). In detail, the rate
of SOC accumulation of the boreal forest in the present study was estimated
as 243.4 kg C ha-1 yr-1, which was within the range of boreal
forests in European and American forests (115.6–740.0 kg C ha-1 yr-1) (Prietzel et al., 2006; Dölle and Schmidt, 2009; Häkkinen et al., 2011; Wang et al., 2011; Rantakari et
al., 2012; Chapman et al., 2013; Schrumpf et al., 2014). The rates of SOC
accumulation in the three temperate forests ranged from 127.2 to 390.8 kg C ha-1 yr-1, comparable to the regional comparison data of 200.0 kg C ha-1 yr-1 in the temperate forests of China (Yang et al., 2014).
Evidence from soil inventory-based studies of SOC dynamics also demonstrated
that soil of boreal and temperate forests in European countries is likely to
accumulate C (Berg et al., 2009; Nielsen et al., 2012; Tefs and Gleixner, 2012; Grüneberg et al.,
2014). The mean rate of SOC accumulation in the humus layers of boreal
forests in Sweden was estimated to be 251.0 kg C ha-1 yr-1 during
the period 1961–2002 (Berg et al., 2009). Nielsen et al. (2012) assessed
the rates of SOC change in Denmark's broadleaved deciduous and coniferous
forests using two soil inventories conducted during 1990 and 2005. The
estimated rates of SOC change in the broadleaved and coniferous forests were
90.0 and 310.0 kg C ha-1 yr-1, respectively. Two soil inventories
provided data for analysis of the mineral soils of forests in Germany, which
were found to have sequestrated 410.0 kg C ha-1 yr-1 during the
period of 1987–2008 (Grüneberg et al., 2014). Therefore, evidence from
long-term observations and from the repeated soil sampling in individual
studies and in national soil inventory reports, suggests that soils of
boreal and temperate forests in the Northern Hemisphere have functioned as C
sinks during past decades.
Comparison of soil organic carbon (SOC) stocks in eight forest
plots in China between the 1990s and the 2010s. The SOC stocks in all
forests during the two periods are above the 1:1 line, suggesting that all
these forests have increased their SOC stock during the study period. The
inset graph shows the SOC sink rates by forest biome (i.e., boreal,
temperate, subtropical, and tropical forests), which are categorized from
the eight forest plots. SOC stocks and change rates are presented as means
±1 SD. For details, see Fig. 1 and Tables 1 and S1.
Relationships between rates of increase in soil organic carbon
(SOC) against biotic and climatic factors in eight forests in China. (a) Biomass increment, (b) litter production, (c) log production, (d)
above-ground net primary production (ANPP), (e) mean annual temperature
(MAT), (f) mean annual precipitation (MAP), and (g) the relative effects of
biotic (a, b, c) and climatic (e, f) factors on SOC increase rates (kg C ha-1 yr-1) using partial regression analyses. Solid lines
indicate significant relationships (P<0.05), and dashed lines
represent insignificant trends (P>0.05) between SOC increase
rates and biotic and climatic factors.
In other subtropical and tropical forest ecosystems, direct evidence of SOC
dynamics is relatively scarce (Tang and Li, 2013). However, based on the estimates from regional
comparisons, Pan et al. (2011) showed that global tropical forests were a
source of 1.4 Pg C ha-1 yr-1 from 1990 to 2007. At the global
scale, tropical land-use changes have caused a sharp drop in forest area,
which also led to a large release of C from tropical forest soils. Without
land-use change and deforestation, soils in subtropical and tropical forests
have functioned as a considerable C sink during the past 2 decades
in this study (627.6±370.1 and 397.9±84.2 kg C ha-1 yr-1, respectively; Table 3). Limited forest management (e.g., litter
and deadwood harvest), as well as catastrophic land-use changes, can result
in the loss of C from forest soil. Prietzel et al. (2016) reported a large
loss of SOC in forests in the German Alps, where half of the woody biomass
and deadwood had been harvested over recent decades. On the one hand,
harvesting the forest floor can decrease litter and deadwood inputs into
soils and subsequently lead to the loss of soil C (Davidson and Janssens,
2006). On the other hand, a decrease in the amount of the forest floor may
lead to an increase in soil erosion, especially in mountain forests (Evans
et al., 2013). Additionally, high-elevation ecosystems are expected to be
more sensitive to warming than other regions, with associated changes in
soil freezing and thawing events and in snow cover, which may be another
reason for the SOC losses in forests in the German Alps.
Links between biotic and climatic factors and in SOC accumulation
The forest biomass of China has functioned as a significant C sink over
recent decades (Pan et al., 2011; Fang et al., 2014, 2018). The increase in
C accumulation by vegetation supplied more C inputs into soils, including
inputs of litter, woody debris, and root exudates, and resulted in SOC
accumulation (Zhu et al., 2017). However, the rate of SOC change did not
increase with the rate of biomass change in this study (Table S4). We found
that soil in the subtropical old-growth forest increased at the highest sink
rate of 907.5±60.1 kg C ha-1 yr-1 but that vegetation
functioned as a significant C source (-1000.3±78.2 kg C ha-1 yr-1). This was because the relatively higher annual litterfall and
fallen log production occurred in the old-growth forest, which subsequently
resulted in soil C accumulation (Fig. 4). The positive (but not significant)
trend between climatic factors and SOC dynamics may largely be induced by
the internal correlations between climatic and biotic factors (Fig. 4).
Comparison of the changes in forest soil organic carbon (SOC)
stocks according to repeated soil samplings and/or long-term observation.
Different colors, shapes, and sizes represent different forest biomes, ages,
and soil depths, respectively. The numbers in parentheses indicate the
sampling times and intervals between the two soil samplings.
The heterotrophic respiration of global forest soil has increased
significantly over past decades (Bond-Lamberty et al., 2018), suggesting
that the increment in the rate of soil C input outweighs that of the rate of
soil C output. The increasing heterotrophic respiration of forest soil is
mainly due to ongoing climate change, and especially to increasing
temperature. The increment in forest growth rate is due to increasing
temperature, together with increasing CO2 and nitrogen fertilization
(Norby et al., 2010; Feng et al., 2019). Thus, the sensitivity of forest net
primary production to ongoing climate change should outweigh that of
respiration. We also found that SOC stock increased from 68.4
to 86.6 Mg C ha-1, albeit the biomass C stock decreased significantly
from 1988 to 2008 in the subtropical old-growth plot. The greatest amount of
litter and deadwood production and standing crop occurred in the old-growth
plot, which resulted in relatively higher soil C sequestration in the
old-growth plot compared to other plots (Fig. 4, Table S4). Biotic factors
explained the variation in SOC dynamics better than climatic factors. In
this study, we did not, however, measure root-derived C inputs to SOC,
although below-ground production also makes a significant contribution to
SOC accumulation (Nadelhoffer and Raich, 1992; Majdi, 2001; Pausch and
Kuzyakov, 2018). Above-ground inputs are mineralized from litter and deadwood, and below-ground inputs may benefit from interactions with soils
(Rasse et al., 2005). Even if the effect of climatic factors were controlled
and below-ground biotic factors were not included in the analysis, the
above-ground biotic factors would explain 56.4 % of the variation in the
rate of SOC accumulation.
Regional carbon budget
The rate of SOC accumulation (421.2±274.4 kg C ha-1 yr-1,
Fig. 2 and Table S3) is more than one-half of the vegetation C uptake rate
in China's forests (702.0 kg C ha-1 yr-1) (Guo et al., 2013; Fang
et al., 2018). This result suggests that China's forest soils have
contributed to a negative feedback to climate warming during the past 2 decades, rather than the positive feedback predicted by coupled C–climate
models (Cox et al., 2000; He et al., 2016; Wang et al., 2018).
If we roughly use the inventory-based forest area of 138.8 Mha in China (Guo
et al., 2013) and extend the current SOC sink rates obtained in this study
to all the forests in the country, China's forest soils have sequestered
approximately 1.1±0.5 Pg C during the past 2 decades (57.1±26.5 Tg C yr-1). This C accumulation would be equivalent to
2.4 %–6.8 % of the country's fossil CO2 emissions during the
contemporary period (1991–2010) (Zheng et al., 2016). By comparing forest
SOC data obtained from published literature during the 2000s and a national
soil inventory during the 1980s, Yang et al. (2014) estimated significant C
accumulation in the forest soils of China. Although they did not estimate
the national C budget of these forest soils, we can calculate the national C
sequestration rate of forest soil as 67.2 Tg C yr-1, based on the C
sequestration rates and forest areas of the different forest types in their
study. Our results further confirm the assessment, based on repeated
measurements at eight permanent forest plots, that soils in China's forests
have functioned as a C sink for atmospheric CO2 during the past 2
decades.
According to previous estimates, the C sinks of three C sectors: forest
vegetation biomass (Fang et al., 2014), deadwood, and litter (Zhu et al.,
2017) during the past 2 decades were 70.9, 3.9, and 2.8 Tg C yr-1,
respectively (Table S5). If these previous estimates are incorporated into
the soil C accumulation rate of 57.1±26.5 Tg C yr-1 in the
current study, then China's forests may have sequestered a total of 134.7 Tg C per year between the 1990s and the 2010s. This is equivalent to 14.5 %
of the contemporary fossil CO2 emissions in the country (Zheng et al.,
2016). According to the estimate of Pan et al. (2011), the C sink rate of
forests in the temperate regions of the Northern Hemisphere was 647.1 Tg C yr-1. The C sequestration of China's forests represents 20.8 % of the
total temperate regions. The sequestration rate of China's forests is
slightly higher than the mean value of the total temperate regions, relative
to the forest area of China (i.e., 18.9 % of the forest areas in the
temperate regions). This result indicates that the role of forest soils in
the regional C cycle cannot be ignored, although a large uncertainty about
the national C budget of forest soils remains in our estimates.
Uncertainty analysis
We investigated the SOC stocks in eight permanent plots across four forest
biomes in China. These plots spanned a long-term timescale (approximately 20
years) and a broad spatial scale (approximately 34∘ of latitude).
We also measured several C fluxes (i.e., biomass change rate, production of
litterfall and deadwood) that were relevant to the rate of SOC change. Even
so, the following three factors may introduce uncertainties related to the
estimation of SOC dynamics.
First, the sampling times and intervals between SOC investigations were
different across the sites. The first sampling was performed from 1987 to
1998 and the second was carried out from 2008 to 2014. As a result, the
sampling interval ranged from 16 years in the boreal forest plot to 21 years
in the subtropical mixed forest plot (Table 1). Nonuniform sampling times
and intervals may lead to uncertainties in relation to SOC stocks across the
forest plots.
Second, the depth of soil varied substantially, ranging from 40 cm in the
boreal site to 100 cm in the temperate and tropical sites. In addition,
different numbers (2–5) of soil profiles were dug in different plots during
the first sampling period. To ensure consistency between the two sampling
times, the same number of soil profiles were dug in similar locations
to perform SOC stock investigations during the second sampling period. We
performed continuous observation of litterfall and deadwood production, but
the observation times and durations varied across the plots. Variability in
these items may reduce the comparability of SOC dynamics among plots.
Finally, the rates of SOC change in our study and in inventory-based forest
areas and forest types were used to estimate the C budget of forest soil in
China. However, only eight permanent forest plots were observed in this
study, and this will inevitably lead to uncertainty with respect to national
estimations.
Conclusions
The SOC stocks within the top 20 cm increased by 2.4–12.6 Mg C ha-1
across the forests during the past 2 decades, with an annual accumulation
rate of 332.4±200.2 kg C ha-1. If all soil horizon profiles were
included, the soils may have been found to have sequestered 3.6 %–16.3 % of
the annual net primary production across the investigated sites, and the
averaged accumulated rate (421.2 kg C ha-1 yr-1) may have been
more than one-half of the vegetation C uptake rate (702.0 kg C ha-1 yr-1) in China's forests. These results demonstrate that these forest
soils have functioned as an important C sink over recent decades, although
the phenomenon may not occur uniformly in forests worldwide. Forest soils
store large amounts of C and accumulate it steadily (and often slowly) but
will release it rapidly to the atmosphere once they are disturbed.
Data availability
Allometric equations of above-ground biomass and the data for soil bulk
density, SOC content, stock, and their change rates of the eight permanent
plots are listed as in the Supplement. The remaining data
that support the findings of this study are available from the corresponding
author upon request.
The supplement related to this article is available online at: https://doi.org/10.5194/bg-17-715-2020-supplement.
Author contributions
JF designed the research. JZ and JF designed the data analysis. JZ, JF, ZZ,
LJ, XH, HY, GL, CW, and GZ performed SOC measurements. JF, YL, CJ, and GL
designed sampling and analytical programs and performed data quality
control. JZ, JF, CW, SZ, PL, JZ, ZT, CZ, RAB, and YP contributed to the
writing of the manuscript.
Competing interests
The authors declare that they have no conflict of interest.
Acknowledgements
We thank Qiong Cai and Tianli Zheng for their assistance in the preparation of the manuscript.
Financial support
This research has been supported by the National Key Research and Development Program of China (grant no. 2017YFC0503906), the National Natural Science Foundation of China (grant nos. 31700374 and 31621091), and the International Programs, US Forest Service (grant no. 07-JV-11242300-117).
Review statement
This paper was edited by Yakov Kuzyakov and reviewed by three anonymous referees.
ReferencesBerg, B., Johansson, M. B., Nilsson, Å., Gundersen, P., and Norell, L.:
Sequestration of carbon in the humus layer of Swedish forests – direct
measurements, Can. J. Forest Res., 39, 962–975, 10.1139/X09-022, 2009.Bond-Lamberty, B., Bailey, V. L., Chen, M., Gough, C. M., and Vargas, R.:
Globally rising soil heterotrophic respiration over recent decades, Nature,
560, 80–83, 10.1038/s41586-018-0358-x, 2018.Canadell, J. G. and Schulze, E. D.: Global potential of biospheric carbon
management for climate mitigation, Nat. Commun., 5, 1–12, 10.1038/ncomms6282, 2014.Chapman, S. J., Bell, J. S., Campbell, C. D., Hudson, G., Lilly, A., Nolan,
A. J., Robertson, A. H. J., Potts, J. M., and Towers, W.: Comparison of soil
carbon stocks in Scottish soils between 1978 and 2009, Eur. J. Soil Sci.,
64, 455–465, 10.1111/ejss.12041, 2013.Chen, L., Smith, P., and Yang, Y.: How has soil carbon stock changed over
recent decades?, Glob. Change Biol., 21, 3197–3199, 10.1111/gcb.12992, 2015.Cox, P. M., Betts, R. A., Jones, C. D., Spall, S. A., and Totterdell, I. J.:
Acceleration of global warming due to carbon-cycle feedbacks in a coupled
climate model, Nature, 408, 184–187, 10.1038/35041539, 2000.Davidson, E. A. and Janssens, I. A.: Temperature sensitivity of soil carbon
decomposition and feedbacks to climate change, Nature, 440, 165–173,
10.1038/nature04514, 2006.Dölle, M. and Schmidt, W.: Impact of tree species on nutrient and light
availability: evidence from a permanent plot study of old-field succession,
Plant Ecol., 203, 273–287, 10.1007/s11258-008-9547-2, 2009.Evans, A. M., Perschel, R. T., and Kittler, B. A.: Overview of forest
biomass harvesting guidelines, J. Sustain. Forest., 32, 89–107, 10.1080/10549811.2011.651786, 2013,.Fang, J., Shen, Z., Tang, Z., Wang, X., Wang, Z., Feng, J., Liu, Y., Qiao,
X., Wu, X., and Zheng, C.: Forest community survey and the structural
characteristics of forests in China, Ecography, 35, 1059–1071, 10.1111/j.1600-0587.2013.00161.x, 2012.Fang, J., Guo, Z., Hu, H., Kato, T., Muraoka, H., and Son, Y.: Forest
biomass carbon sinks in East Asia, with special reference to the relative
contributions of forest expansion and forest growth, Glob. Change Biol., 20,
2019–2030, 10.1111/gcb.12512, 2014.Fang, J., Yu, G., Liu, L., Hu, S., and Chapin III, F. S.: Climate change,
human impacts, and carbon sequestration in China, P. Natl. Acad. Sci. USA,
115, 4015–4020, 10.1073/pnas.1700304115, 2018.Fang, J. Y., Liu, G. H., Zhu, B., Wang, X. K., and Liu, S. B.: Carbon
budgets of three temperate forest ecosystems in Dongling Mt., Beijing,
China, Sci. China Earth Sci., 50, 92–101, 10.1007/s11430-007-2031-3, 2007.Feng, Y., Zhu, J., Zhao, X., Tang, Z., Zhu, J., and Fang, J.: Changes in the
trends of vegetation net primary productivity in China between 1982 and
2015, Environ. Res. Lett., 14, 124009, 10.1088/1748-9326/ab4cd8, 2019.Grüneberg, E., Ziche, D., and Wellbrock, N.: Organic carbon stocks and
sequestration rates of forest soils in Germany, Glob. Change Biol., 20,
2644–2662, 10.1111/gcb.12558, 2014.Guo, Z. D., Hu, H. F., Li, P., Li, N. Y., and Fang, J. Y.: Spatio-temporal
changes in biomass carbon sinks in China's forests from 1977 to 2008, Sci.
China Life Sci., 56, 661–671, 10.1007/s11427-013-4492-2, 2013.Häkkinen, M., Heikkinen, J., and Mäkipää, R.: Soil carbon
stock increases in the organic layer of boreal middle-aged stands,
Biogeosciences, 8, 1279–1289, 10.5194/bg-8-1279-2011, 2011.He, Y., Trumbore, S. E., Torn, M. S., Harden, J. W., Vaughn, L. J., Allison,
S. D., and Randerson, J. T.: Radiocarbon constraints imply reduced carbon
uptake by soils during the 21st century, Science, 353, 1419–1424,
10.1126/science.aad4273, 2016.
IPCC: Summary for policymakers, in: Climate Change 2013: The Physical Science Basis, edited by: Stocker, T. F., Qin, D., Plattner, G. K.,
Tignor, M., Allen, S. K., Boschung, J., Nauels, Xia, Y., Bex, V.,
and Midgley, P. M., Contribution
of Working Group I to the Fifth Assessment, Report of the Intergovernmental
Panel on Climate Change, Cambridge, CUP, 1–30, 2013.Kiser, L. C., Kelly, J. M., and Mays, P. A.: Changes in forest soil carbon
and nitrogen after a thirty-year interval, Soil Sci. Soc. Am. J., 73,
647–653, 10.2136/sssaj2008.0102, 2009.Lal, R.: Soil carbon sequestration impacts on global climate change and food
security, Science, 304, 1623–1627, 10.1126/science.1097396, 2004.Leith, H. and Whittaker, R. H.: Primary productivity of the biosphere:
ecological studies, Berlin, Springer, 237–263, 10.1007/978-3-642-80913-2, 1975.Lettens, S., Van Orshoven, J., van Wesemael, B., De Vosc, B., and Muysa, B.:
Stocks and fluxes of soil organic carbon for landscape units in Belgium
derived from heterogeneous data sets for 1990 and 2000, Geoderma, 127,
11–23, 10.1016/j.geoderma.2004.11.001, 2005.Luo, Y., Melillo, J., Niu, S., Beier, C., Clark, J. S., Classen, A. T.,
Davidson, E., Dukes, J. S., Evans, R. D., Field, C. B., Czimczik, C. I.,
Keller, M., Kimball, B. A., Kueppers, L. M., Norby, R. J., Pelini, S. L.,
Pendall, E., Rastetter, E., Six, J., Smith, M., Tjoelker, M. G., and Torn,
M. S.: Coordinated approaches to quantify long-term ecosystem dynamics in
response to global change, Glob. Change Biol., 17, 843–854, 10.1111/j.1365-2486.2010.02265.x, 2011.Majdi, H.: Changes in fine root production and longevity in relation to
water and nutrient availability in a Norway spruce stand in northern Sweden,
Tree Physiol., 21, 1057–1061, 10.1023/A:1011905124393, 2001.Nadelhoffer, K. J. and Raich, J. W.: Fine root production estimates and
belowground carbon allocation in forest ecosystems, Ecology, 73, 1139–1147,
10.2307/1940664, 1992.Nelson, D. W. and Sommers, L. E.: Total carbon, organic carbon, and organic
matter, chap. 29, in: Methods of Soil Analysis, Part 2. Chemical and
Microbiological Properties, 2nd Edn., edited by: Sparks, A. L., American Society of
Agronomy, Inc, Soil Science Society of Agronomy, Inc., 539–579, 10.2136/sssabookser5.3.c34, 1982.Nielsen, O. K., Mikkelsen, M. H., Hoff mann, L., Gyldenkærne, S.,
Winther, M., Nielsen, M., Fauser, P., Thomsen, M., Plejdrup, M. S.,
Albrektsen, R., Hjelgaard, K., Bruun, H. G., Johannsen, V. K., Nord-Larsen,
T., Bastrup-Birk, A., Vesterdal, L., Møller, I. S., Rasmussen, E.,
Arfaoui, K., Baunbæk, L., and Hansen, M. G.: Denmark's National
Inventory Report 2012, Emission Inventories 1990–2010 – Submitted under the
United Nations Framework Convention on Climate Change and the Kyoto
Protocol. Scientific Report from DCE–Danish Centre for Environment and
Energy, 19, http://www.risoe.dtu.dk/rispubl/NEI/NEI-DK-5700.pdf (last access: 3 August 2012);
OSTI as DE01047219, 2012.Norby, R. J. and Zak, D. R.: Ecological lessons from Free-Air CO2
Enrichment (FACE) experiments, Annu. Rev. Ecol. Evol. S., 42, 181–203,
10.1146/annurev-ecolsys-102209-144647, 2011.Norby, R. J., Warren, J. M., Iversen, C. M., Medlyn, B. E., and McMurtrie,
R. E.: CO2 enhancement of forest productivity constrained by limited
nitrogen availability, P. Natl. Acad. Sci. USA, 107, 19368–19373,
10.1073/pnas.1006463107, 2010.Ortiz, C. A., Liski, J., Gärdenäs, A. I., Lehtonen, A., Lundblad,
M., Stendahl, J., Ågren, G. I., and Karltun, E.: Soil organic carbon stock
changes in Swedish forest soils – a comparison of uncertainties and their
sources through a national inventory and two simulation models, Ecol.
Model., 251, 221–231, 10.1016/j.ecolmodel.2012.12.017, 2013.Pan, Y., Birdsey, R. A., Fang, J. Houghton, R., Kauppi, P. E., Kurz, W. A.,
Phillips, O. L., Shvidenko, A., Lewis, S. L., Canadell, J. G., Ciais, P.,
Jackson, R. B., Pacala, S. W., McGuire, A. D., Piao, S., Rautiainen, A.,
Sitch, S., and Hayes, D.: A large and persistent carbon sink in the world's
forests, Science, 333, 988–993, 10.1126/science.1201609, 2011.Pausch, J. and Kuzyakov, Y.: Carbon input by roots into the soil:
quantification of rhizodeposition from root to ecosystem scale, Glob. Change
Biol., 24, 1–12, 10.1111/gcb.13850, 2018.Piao, S., Fang, J., Ciais, P., Peylin, P., Huang, Y., Sitch, S., and Wang,
T.: The carbon balance of terrestrial ecosystems in China, Nature, 458,
1009, 10.1038/nature07944, 2009.Prietzel, J., Stetter, U., Klemmt, H. J., and Rehfuess, K. E.: Recent carbon
and nitrogen accumulation and acidification in soils of two Scots pine
ecosystems in Southern Germany, Plant Soil, 289, 153–170, 10.1007/s11104-006-9120-5, 2006.Prietzel, J., Zimmermann, L., Schubert, A., and Christophel, D.: Organic
matter losses in German Alps forest soils since the 1970s most likely caused
by warming, Nat. Geosci., 9, 543–548, 10.1038/ngeo2732, 2016.Rantakari, M., Lehtonen, A., Linkosalo, T., Tuomi, M., Tamminen, P.,
Heikkinen, J., Liski, J., Mäkipää, R., Ilvesniemi, H.,
and Sievänen, R.: The Yasso07 soil carbon model – Testing against repeated
soil carbon inventory, Forest Ecol. Manag., 286, 137–147, 10.1016/j.foreco.2012.08.041, 2012.Rasse, D. P., Rumpel, C., and Dignac, M. F.: Is soil carbon mostly root
carbon? Mechanisms for a specific stabilisation, Plant Soil, 269, 341–356,
10.1007/s11104-004-0907-y, 2005.Schrumpf, M., Schulze, E. D., Kaiser, K., and Schumacher, J.: How accurately
can soil organic carbon stocks and stock changes be quantified by soil
inventories?, Biogeosciences, 8, 1193–1212, 10.5194/bg-8-1193-2011, 2011.Schrumpf, M., Kaiser, K., and Schulze, E. D.: Soil organic carbon and total
nitrogen gains in an old growth deciduous forest in Germany, PLoS ONE, 9,
e89364, 10.1371/journal.pone.0089364, 2014.Sitch, S., Friedlingstein, P., Gruber, N., Jones, S. D., Murray-Tortarolo, G., Ahlström, A., Doney, S. C., Graven, H., Heinze, C., Huntingford, C., Levis, S., Levy, P. E., Lomas, M., Poulter, B., Viovy, N., Zaehle, S., Zeng, N., Arneth, A., Bonan, G., Bopp, L., Canadell, J. G., Chevallier, F., Ciais, P., Ellis, R., Gloor, M., Peylin, P., Piao, S. L., Le Quéré, C., Smith, B., Zhu, Z., and Myneni, R.: Recent trends and drivers of regional sources and sinks of carbon dioxide, Biogeosciences, 12, 653–679, 10.5194/bg-12-653-2015, 2015.Smith, P.: How long before a change in soil organic carbon can be detected?,
Glob. Change Biol., 10, 1878–1883, 10.1111/j.1365-2486.2004.00854.x, 2004.Tang, G. and Li, K.: Tree species controls on soil carbon sequestration and
carbon stability following 20 years of afforestation in a valley-type
savanna, Forest Ecol. Manag., 291, 13–19, 10.1016/j.foreco.2012.12.001, 2013.Tang, X., Zhao, X., Bai, Y., Tang, Z., Wang, W., Zhao, Y., Wan, H., Xie, Z.,
Shi, X., Wu, B., Wang, G., Yan, J., Ma, K., Du, S., Li, S., Han, S., Ma, Y.,
Hu, H., He, N., Yang, Y., Han, W., He, H., Yu, G., Fang, J., and Zhou, G.:
Carbon pools in China's terrestrial ecosystems: New estimates based on an
intensive field survey, P. Natl. Acad. Sci. USA, 115, 4021–4026, 10.1073/pnas.1700291115, 2018.Tefs, C. and Gleixner, G.: Importance of root derived carbon for soil
organic matter storage in a temperate old-growth beech forest – Evidence
from C, N and 14C content, Forest Ecol. Manag., 263, 131–137, 10.1016/j.foreco.2011.09.010, 2012.Todd-Brown, K. E., Randerson, J. T., Post, W. M., Hoffman, F. M., Tarnocai,
C., Schuur, E. A. G., and Allison, S. D.: Causes of variation in soil carbon
simulations from CMIP5 Earth system models and comparison with observations,
Biogeosciences, 10, 1717–1736, 10.5194/bg-10-1717-2013, 2013.Wang, C., Gower, S. T., Wang, Y., Zhao, H., Yan, P., and Bond-Lamberty, B.
P.: The influence of fire on carbon distribution and net primary production
of boreal Larix gmelinii forests in north-eastern China, Glob. Change Biol., 7, 719–730,
10.1046/j.1354-1013.2001.00441.x, 2001.Wang, W., Qiu, L., Zu, Y., Su, D., An, J., Wang, H., Zheng, G., Sun, W., and
Chen, X.: Changes in soil organic carbon, nitrogen, pH and bulk density with
the development of larch (Larix gmelinii) plantations in China, Glob. Change Biol., 17,
2657–2676, 10.1111/j.1365-2486.2011.02447.x,
2011.Wang, X., Ciais, P., Wang, Y., and Zhu, D.: Divergent response of seasonally
dry tropical vegetation to climatic variations in dry and wet seasons, Glob.
Change Biol., 24, 4709–4717, 10.1111/gcb.14335, 2018.Yang, Y., Li, P., Ding, J., Zhao, X., Ma, W., Ji, C., and Fang, J.:
Increased topsoil carbon stock across China's forests, Glob. Change Biol.,
20, 2687–2696, 10.1111/gcb.12536, 2014.Zhao, X., Yang, Y., Shen, H., Geng, X., and Fang, J.: Global
soil–climate–biome diagram: linking surface soil properties to climate and
biota, Biogeosciences, 16, 2857–2871, 10.5194/bg-16-2857-2019, 2019.Zheng, T. L., Zhu, J. L., Wang, S. P., and Fang, J. Y.: When will China
achieve its carbon emission peak?, Natl. Sci. Rev., 3, 8–15, 10.1093/nsr/nwv079, 2016.Zhou, G., Liu, S., Li, Z., Zhang, D., Tang, X., Zhou, C., Yan, J., and Mo, J.:
Old-growth forests can accumulate carbon in soils, Science, 314, 1417,
10.1126/science.1130168, 2006.Zhou, Z., Jiang, L., Du, E., Hu, H., Li, Y., Chen, D., and Fang, J.:
Temperature and substrate availability regulate soil respiration in the
tropical mountain rainforests, Hainan Island, China. J. Plant Ecol., 6,
325–334, 10.1093/jpe/rtt034, 2013.Zhu, J., Hu, H., Tao, S., Chi, X., Li, P., Jiang, L., Ji, C., Zhu, J., Tang,
Z., Pan, Y., Birdsey, R. A., He, X., and Fang, J.: Carbon stocks and changes
of dead organic matter in China's forests, Nat. Comm., 8, 1–10, 10.1038/s41467-017-00207-1, 2017.Zhu, J. X., Hu, X. Y., Yao, H., Liu, G. H., Ji, C. J., and Fang, J. Y.: A
significant carbon sink in temperate forests in Beijing: based on 20-year
field measurements in three stands, Sci. China Life Sci., 58, 1135–1141,
10.1007/s11427-015-4935-z, 2015.