Ecological Research

, Volume 33, Issue 2, pp 393–402 | Cite as

Imposed drought effects on carbon storage of moso bamboo ecosystem in southeast China: results from a field experiment

  • Xiaogai Ge
  • Benzhi Zhou
  • Xiaoming Wang
  • Qian Li
  • Yonghui Cao
  • Lianhong Gu
Special Feature Climate Change and Biodiversity Conservation in East Asia as a token of memory for the 7th EAFES in Daegu, Korea


Drought can severely affect carbon dynamics in forest ecosystems through impacts on carbon storage, reduced carbon fixation, abatement of the carbon sink function, and alteration of carbon sink-source relationships. Currently, little is known about the effects of drought on the productivity and spatial patterns of carbon in bamboo forests. The objective of this study was to assess the effect of imposed drought on the carbon storage and soil carbon dynamics of a bamboo forest ecosystem in subtropical area of China. Drought was imposed via throughfall exclusion in moso bamboo forest from July 2012 to April 2013. Results indicated that bamboo shoots, new culms, shoot height, and diameter at breast height were significantly lower in throughfall exclusion (TE) plots than in control check (CK) plots, with decrease of 64.6, 70.8, 10.6 and 11.3%, respectively. Annual carbon sequestration for TE plots was 58.1% lower than that for CK plots. Soil carbon storage in the 0–60-cm layer in CK and TE plots decreased by 3.7 and 12.2%, respectively, indicating that drought can decrease soil respiration by altering substrate availability. Ecosystem carbon storage increased by 4.75 t ha−1 in CK plots but decreased by 13.71 t ha−1 in TE plots. Our findings highlight that drought can reduce carbon storage and alter the spatial pattern of carbon in moso bamboo forest ecosystems, particularly when drought occurs during the development bamboo shoot. Our findings should provide a better understanding of carbon sequestration potential and aid determination of how future climate change may impact carbon budgets.


Imposed drought Throughfall exclusion Carbon storage allocation Soil respiration Moso bamboo 


Global climatic change is predicted to influence the frequency, magnitude, and duration of drought (Mann et al. 2015). Drought generally results in a fragile environment and can alter rainfall patterns during the growing season, potentially reducing total amounts of precipitation and redistributing rainfall (Sanaullah et al. 2012). Carbon cycling and storage within a broad range of ecosystems are directly affected by drought because soil moisture controls the spatio-temporal patterns of plant and soil processes (Sanaullah et al. 2012; Gorsel et al. 2013). Recently, drought has been the focus of a great deal of research worldwide (Meir et al. 2008; Zhou et al. 2013; Hinko-Najera et al. 2015). Much of this work has focused on the response to drought by carbon cycles within forest ecosystems, including changes in photosynthesis, plant physiology (Tang et al. 2004), soil respiration (Suseela et al. 2012; Hinko-Najera et al. 2015), litterfall dynamics (Brando et al. 2008), fine-root turnover (Gaul et al. 2008), soil microbial function (Pailler et al. 2014), tree mortality (Adams et al. 2009), and net ecosystem exchange using the eddy covariance method (Unger et al. 2009). However, there is minimal knowledge regarding the precise response mechanisms of forest carbon cycle processes and relative carbon allocation under drought.

Intermittent droughts can affect net primary production (NPP) and its relative allocation aboveground and in the litter and soil layers (Trenberth et al. 2014). In particular, such droughts can strongly affect soil organic carbon decomposition and carbohydrate content in plant carbon cycling and can potentially induce mortality (Molen et al. 2011). Therefore, droughts are expected to impact the carbon cycle even more severely in the future. In fact, forest carbon sinks are vulnerable to drought, particularly through natural changes in precipitation that can also lead to significant net transfers of carbon from the soil to the atmosphere. Costa et al. (2010) identified a loss of total aboveground biomass of 48.1 ± 3.4 Mg C ha−1 in long-term throughfall exclusion plots (2002–2008), while the loss in control plots was only 14.7 ± 0.8 Mg C ha−1. Brando et al. (2008) demonstrated that drought induced declines in live aboveground carbon of 32.5 Mg·ha−1 due to tree mortality as well as decreased aboveground net primary productivity (ANPP). However, if droughts occur when soil moisture is in short supply, NPP will decline and its relative allocation may change. The relationship between drought and the carbon cycle of forest ecosystems remains poorly understood.

Moso bamboo (Phyllostachys pubescens) forests are an important ecosystem in south China. These widely distributed forests exhibit rapid growth and rapid biomass accumulation (Liu et al. 2011; Zhou et al. 2011). Moso bamboo forests cover an area of 3.87 million ha, representing 70% of the country’s bamboo forest area. The carbon sequestration capacity of moso bamboo is quite high, with ecosystem annual CO2 net sequestration of up to 23.31 t ha−1, or annual carbon net sequestration of 5.09 t ha−1, which is 1.46 times the value observed in Cunninghamia lanceolata forests and 1.33 times the value in tropical rainforests (Fan 2012). Evaluating ecosystem carbon cycles in response to drought is important not only for basic knowledge of carbon cycling, but also for potential management strategies. Management strategies must be prepared to help existing bamboo forests adapt to the anticipated and uncertain future conditions by increasing and improving ecosystem carbon sinks.

Over the next 50 years, climate models project further increases in drought events in southeast China. In 2013, a severe drought occurred in south China, destroying 13,733 ha of forest through the mortality of 6.18 million bamboo culms. The production levels of rhizome shoots and winter bamboo shoots were reduced by about 40 and 20%, respectively. Therefore, studies of the effects of drought on carbon flux in moso bamboo plantations are crucial for understanding and quantifying carbon sequestration potential and for determining how future climate change may impact carbon budgets. Little is known about the effect of such drought on moso bamboo forest productivity and spatial patterns of carbon. Therefore, the objectives of the present study were: (1) to quantify changes in moso bamboo growth and biomass production caused by imposed drought; (2) to evaluate the impact of imposed drought on the vegetation carbon cycle and soil carbon content; and (3) to determine the effect of imposed drought on the carbon storage of the moso bamboo ecosystem.

Materials and methods

Study area

The study site was located in Fuyang County (119°56′–120°18′E, 30°03′–30°06′N), Zhejiang Province, China. This region has a typical subtropical humid monsoon climate, with a mean annual temperature of 16.1 °C and mean annual precipitation of 1441.9 mm, mainly occurring between April and September. The soil is classified as Haplic Luvisol soil derived from granite (Gong 2003). The area was covered by natural forest in history, which was destroyed by human for the request of timber, firewood and the agricultural development in 1950s. The moso bamboo forest was initially established around the 1960s through mother culm transplanting. The dominant herbaceous species are Digitaria sanguinalis, Paederia scandens, Ampelopsis aconitifolia, Cyperus diffomis, Dicranoqteris linearis and so on. Weeds and shrubs are cleared once every 2 years throughout the forest stands. The forest is thinned every year by harvesting mature culms older than 7 or 8 years during fall and winter. Through the removal of old culms and the recruitment of new culms each year, these bamboo forests are maintained with an uneven age structure, with culm ages ranging between 1 and 6 years old. The characteristics of the stands are summarized in Table 1.
Table 1

General information of the experimental moso bamboo stands (mean ± SD, n = 9)


Stand 1

Stand 2

Stand 3







Elevation (m)







Slope (°)














Soil depth (cm)







SOC (g kg−1)

34.37 ± 9.06

36.11 ± 7.03

36.74 ± 8.04

39.78 ± 7.39

37.82 ± 8.21

37.42 ± 9.64

Soil nitrogen (g kg−1)

2.72 ± 0.79

2.34 ± 0.27

2.72 ± 0.48

2.91 ± 0.52

2.56 ± 0.29

2.52 ± 0.41

SAP (mg kg−1)

4.25 ± 0.93

4.64 ± 0.81

5.50 ± 1.03

5.94 ± 1.23

6.48 ± 1.09

6.45 ± 2.04

DBH (cm)

13.4 ± 2.2

14.7 ± 1.6

14.4 ± 2.1

14.3 ± 1.9

13.0 ± 2.5

14.6 ± 1.3

Height (m)

17.3 ± 0.9

16.3 ± 2.3

17.0 ± 1.5

16.0 ± 2.0

15.8 ± 2.4

16.6 ± 0.9

Density (n ha−1)







CK control check, TE throughfall exclusion, DBH diameter at breast height, S sunny slope, SOC soil organic carbon, SAP soil available phosphorus

Experimental design and measurements

The experiment was conducted in three moso bamboo stands that were spatially separated (n = 3) with similar elevations (± 10 m), slope, and soil physical properties. In each stand, two 10 × 20-m plots were randomly established in July 2012, when moso bamboo stopped producing shoots and new bamboo culms stopped growing in height; one set of plots was designated as the throughfall exclusion (TE) plots and the other as the control check (CK) plots. The TE plots were sealed using transparent PVC panels supported by 1.5-m-high steel tubes. The junction gap between the bamboo and PVC sheet was filled using cement (Fig. S1). Ditches (50 cm depth, 50 cm width) were dug along the border of each plot to prevent run-off water and interflow water from draining into the plots from the outside.

At the initiation (July 2012) and termination (April 2013) of the experiment, each bamboo culm in the CK and TE plots was assessed. Bamboo culm age, height, and diameter at breast height (DBH) were recorded, and new bamboo shoot emergence, growth, and mortality were measured. Litterfall was collected using three, 0.7 × 0.7-m litter trays (1-mm mesh) that were randomly placed in each plot and suspended 100 cm above the ground (CK plots) or the PVC sheet (TE plots) using four wooden stakes fixed with iron wire. Litter was collected at monthly intervals from August 2012 through July 2013.

Soil was sampled using a hand auger (5-cm inner diameter) at the initiation of the experiment in July 2012 and at its termination in April 2013. Five soil cores were taken randomly at each sampling, and the litter layer was removed by hand. Each soil core was divided into 0–10, 10–20, 20–40, and 40–60-cm layers, and soil samples of the same layer from the five cores were pooled to obtain a mixed sample for each layer for each stand. Visible roots and organic residues were removed. Each soil sample was air-dried and finely ground to pass through a 2-mm sieve for soil organic matter analysis.

Soil bulk density was determined using the cutting-ring method, and soil organic matter content was measured using the potassium permanganate wet digestion method (Soil Science Society of China 1983; Ge et al. 2013). Monthly dynamics of soil temperature and soil moisture at the 0–10 and 10–20-cm soil layers were measured using EM50 (Decagon, USA). This work is guided on “Observation Methodology for Long-term Forest Ecosystem Research” of National Standards of the People’s Republic of China (GB/T 33027-2016).

Data analysis

Aboveground carbon storage of moso bamboo was calculated using the following formula (Zhou et al. 2010):
$$M = 0.5042 \times n \times S \times (747.787 \times D^{2.771} \times (0.148A/(0.028 + A))^{5.555} + 3.772)$$
where M is the aboveground carbon pool, n is stand density, S is area, D is diameter at breast height (cm), A is culm age, and 0.5042 is the conversion coefficient from biomass to carbon.
Litter carbon storage was calculated as follows:
$$V = B \times c$$
where V is litter layer carbon storage, B is litter biomass, and c is litter carbon content.
Soil carbon storage was estimated according to Cui et al. (2012):
$$S = A \times \sum\limits_{i = 1}^{n} {Ci} \times di \times Di$$
where S is soil carbon storage, A is soil area, i is the soil layer, C i is soil organic carbon content, d i is soil density, and D i is soil depth.

All statistical analyses were performed using SPSS (SPSS Inc., USA) and Sigmaplot version 12.5 (Systat, USA). A one-way analysis of variance (ANOVA) was conducted to test for differences between TE and CK plots in above- and belowground organic carbon storage, soil layer organic carbon content, and new culm production at the initiation and termination of the experiment. When the ANOVA indicated a significant treatment effect (P < 0.05), the least significant difference (LSD) test was used to determine differences among means.


Imposed drought effect on soil water content

Throughfall exclusion resulted in a remarkable drought effect, with soil water content in TE plots ranging from 5.5 to 18.5% and 4.9 to 17.4% (except August) lower in the 0–10 and 10–20-cm soil layers, respectively, than values in CK plots (Fig. 1). Water content in the 0–10-cm soil layer was higher than that in the 10–20-cm soil layer in the CK plots. In TE plots, however, soil water content at 0–10 cm was lower than that at 10–20 cm, except for the period between August and October 2012. In addition, the rate of decline in soil water content was greater in the 0–10-cm soil layer than in the 10–20-cm layer.
Fig. 1

Monthly changes in soil water content and rainfall in the stands under different treatments. CK control check, TE throughfall exclusion

Imposed drought effect on new culm production

In this study, the moso bamboo stands reside in an uneven-aged forest, wherein the bamboo produces new shoots at 2-year intervals; the year of new shoot production is known as an “on-year” and the following year is an “off-year”. New culm growth was strongly affected by imposed drought (Table 2). Compared with in CK plots, bamboo shoots, new culms, height and diameter at breast height in TE plots decreased by 64.6, 70.8, 10.6 and 11.3%, respectively (Table 2). Average culm height and DBH were 1416.7 and 14.4 cm in the CK plots, 1266.7 and 12.8 cm in the TE plots, respectively (Table 2).
Table 2

The moso bamboo growing status in the stands under different treatments (mean ± SD)


Shoot number (n ha−1)

Declining mortality number (n ha−1)

Declining mortality rate (%)

New culm number (n ha−1)

New culm rate (%)

New culm height (cm)

New culm DBH (cm)

CK plots

 1–5 days

1417 (75.2%)a





27.7 ± 21.1


 6–20 days

467 (24.8%)a





382.2 ± 278.5


 21–50 days






1006.8 ± 441.9



1884 (64.6%)b



684 (70.8%)


1416.7 ± 138.2 (10.6%)b

14.4 ± 2.3 (11.3%)b

TE plots

 1–5 days

450 (67.5%)a





39.2 ± 21.1


 6–20 days

150 (22.5%)a





327.5 ± 249.1


 21–50 days

67 (10.0%)a





900.0 ± 202.3








1266.7 ± 76.4

12.8 ± 3.4

CK control check, TE throughfall exclusion

aThe percentage of shoot number on the total shoot number in the same treatment

bThe percentage of (CK–TE) on CK

Most new shoots emerged during the first 5 days of the experiment, accounting for 75.2 and 67.5% of the total of new shoots in CK and TE plots, respectively (Table 2). Thereafter, the emergence of new bamboo shoots declined rapidly, with only 24.8 and 22.5% of new shoots emerging during days 6–20 in CK and TE plots, respectively (Table 2). Mortality of new shoots initially occurred on the 23rd day since new shoot emergence and increased gradually until reaching a peak at day 50. The mortality of new bamboo shoots were 63.7 and 70.0% in CK and TE plots, respectively (Table 2).

Imposed drought effect on biomass carbon storage

Annual biomass carbon sequestration significantly differed between CK and TE plots (Table 3). Annual carbon sequestration for TE plots was 4.34 t ha−1, which was 58.1% lower than that for CK plots (10.36 t ha−1). At the initiation of the experiment, biomass carbon storage did not significantly differ between CK and TE plots. At the termination of the experiment, however, the carbon storage increment significantly differed between treatments, with increase in aboveground and belowground biomass carbon storage of 2.32 t ha−1 and 0.46 t ha−1, respectively, in TE plots, which were only 31.3 and 31.1% of the increases observed in CK plots. Litter biomass carbon did not significantly differ between CK and TE plots.
Table 3

Comparison of vegetation organic carbon storage in the stands under drought treatment (mean ± SD)


Initiation (July, 2012) (t ha−1)

Termination (June, 2013) (t ha−1)

Carbon increment (t ha−1)

Litter biomass carbon (t ha−1)

Annual carbon sequestration (t ha−1 yr−1)








31.42 ± 4.00 a

6.28 ± 0.80 a

38.84 ± 4.20a

7.77 ± 0.84 a

+ 7.42 ± 0.22 a

1.48 ± 0.04 a

1.46 ± 0.18 a

10.36 ± 0.43 a


30.95 ± 3.26 a

6.19 ± 0.65 a

33.27 ± 3.91b

6.65 ± 0.78 b

+ 2.32 ± 0.88 b

0.46 ± 0.18 b

1.56 ± 0.27 a

4.34 ± 1.29 b

CK control check, TE throughfall exclusion

Values within the same column with different lowercase letters are significantly different at P < 0.05

The monthly dynamics of litterfall in TE plots was similar to that in CK plots (Fig. 2). Litterfall reached a peak in April or May, with the combined litter production in April–May accounting for about 53% of total annual litterfall. Monthly litterfall was 2.1–22.0% higher in TE plots than in CK plots for most months, with the greatest difference occurring during the months from September 2012 to January 2013 and from April to July 2013. Only 5 months experienced lower rates of litterfall (ranging between 0.2 and 6.7%) in TE plots than in CK plots (Fig. 3). Total litterfall during the experiment in TE plots was 3.09 t ha−1, which was 6.2% higher than in CK plots (2.91 t ha−1).
Fig. 2

Monthly dynamics of litterfall in moso bamboo stands under different treatments. CK control check, TE throughfall exclusion

Fig. 3

Monthly dynamics of litterfall during the moso bamboo imposed drought. CK control check, TE throughfall exclusion; Differences in monthly litterfall dynamic were calculated as CK minus TE value, while the percentage indicated the difference in litterfall dynamics divided by CK monthly litterfall

Imposed drought effect on soil carbon storage

Soil carbon content was highest in the surface layer in both CK and TE plots and decreased with increased soil depth. Soil carbon content did not differ significantly between CK and TE plots at the start of the experiment. In CK plots, soil carbon content in the upper three soil layers decreased by 1.0–12.3% but increased by 2.0% in the 40–60-cm layer (Table 4). Soil carbon content in the four soil layers decreased by 17.2–23.3% in TE plots during the experimental period. The decrease in soil carbon content was significantly higher in TE plots than in CK plots in each soil layer. Imposed drought significantly reduced the soil carbon content within each soil layer (Table 4).
Table 4

Comparison of content of soil organic carbon in the stands under different treatments (mean ± SD)

Soil layer

Initiation (g kg−1)

Termination (g kg−1)

Difference (g kg−1)









0–10 cm

37.93 ± 1.31 a

37.73 ± 2.81 a

37.57 ± 1.04 a

30.97 ± 2.07 b

− 0.37 ± 0.50 b

− 1.0

− 6.76 ± 0.99 a

− 17.9

10–20 cm

26.97 ± 0.93 a

27.13 ± 1.36 a

23.67 ± 2.50 a

20.80 ± 2.29 b

− 3.30 ± 1.82 b

− 12.2

− 6.33 ± 0.95 a

− 23.3

20–40 cm

20.53 ± 1.26 a

22.00 ± 1.31 a

20.10 ± 1.70 a

17.37 ± 2.65 b

− 0.37 ± 0.06 b

− 1.8

− 4.63 ± 1.65 a

− 21.1

40–60 cm

14.83 ± 0.60 b

16.02 ± 1.52 a

15.13 ± 1.15 a

13.27 ± 0.91 b

0.30 ± 0.67 b

+ 2.0

− 2.75 ± 0.64 a

− 17.2

CK control check, TE throughfall exclusion

aMeans the percent of soil organic carbon difference to the initiation carbon storage on throughfall exclusion plots

bMeans the percent of soil organic carbon difference to the initiation carbon storage on control plots

Soil carbon storage throughout the entire 0–60-cm soil profile decreased by 5.13 t ha−1 (3.8%) and 16.86 t ha−1 (12.2%) by the end of the experiment in CK and TE plots, respectively (Table 5), suggesting that imposed drought significantly reduced soil carbon storage. Soil carbon storage in the 0–10-, 10–20-, 20–40-, and 40–60-cm soil layers decreased by 10.7, 2.1, 10.5, and 1.9% in the CK plots and by 12.4, 1.5, 16.7, and 12.6% in the TE plots, respectively. These reductions in soil carbon storage were significantly larger in TE plots than in CK plots at each soil layer, except the 10–20-cm layer.
Table 5

Comparison of soil organic carbon storage in the stands under different treatments (mean ± SD)

Soil layer

Initiation (t ha−1)

Termination (t ha−1)

Difference (t ha−1)









0–10 cm

35.48 ± 4.48 b

36.20 ± 3.05 a

34.96 ± 4.14 a

31.69 ± 3.17 b

− 0.52 ± 0.35 b

− 1.5

− 4.50 ± 0.87 a

− 12.4

10–20 cm

24.41 ± 1.66 a

24.57 ± 2.20 a

21.33 ± 1.20 a

20.49 ± 2.13 a

− 3.08 ± 0.87 a

− 12.6

− 4.09 ± 0.29 a

− 16.7

20–40 cm

41.97 ± 2.93 a

44.35 ± 3.23 a

41.11 ± 2.85 a

39.61 ± 2.98 a

− 0.86 ± 0.13 b

− 2.1

− 4.74 ± 0.91 a

− 10.7

40–60 cm

35.46 ± 2.10 a

33.57 ± 4.09 a

34.78 ± 2.29 a

30.04 ± 4.65 b

− 0.67 ± 0.22 b

− 1.9

− 3.53 ± 1.47 a

− 10.5


136.34 ± 11.20 a

138.32 ± 12.01 a

132.18 ± 10.29 a

121.83 ± 12.33 b

− 5.13 ± 1.82 b

− 3.8

− 16.86 ± 0.6 a

− 12.2

CK control check, TE throughfall exclusion

aMeans the percent of soil organic carbon storage difference to the initiation carbon storage on throughfall exclusion plots

bMeans the percent of soil organic carbon storage difference to the initiation carbon storage on control plots

The seasonal dynamics of soil respiration exhibited a unimodal pattern in both CK and TE plots, with maximum and minimum values occurring in July and January, respectively (Fig. 4). During the experimental period, values of annual average soil respiration in CK and TE plots were 2.05 and 1.90 μmol CO2 m−2 s−1, respectively.
Fig. 4

Monthly dynamic of soil respiration during the moso bamboo imposed drought. CK control check, TE throughfall exclusion

Imposed drought effect on ecosystem carbon storage

Values of ecosystem carbon storage were 174.04 and 175.46 t ha−1 in CK and TE plots, respectively, at the initiation of the experiment, and values did not significantly differ between treatments (Fig. 5). At the end of the experiment, however, ecosystem carbon storage did significantly differ between CK and TE plots, with values of 178.79 t ha−1 and 161.75 t ha−1, respectively. Ecosystem carbon storage had increased by 2.7% (4.75 t ha−1) in CK plots but decreased by 7.8% (13.71 t ha−1) in TE plots by the conclusion of the experiment.
Fig. 5

Ecosystem total carbon storage dynamic in the moso bamboo stands under different treatments. CK control check, TE throughfall exclusion


Imposed drought effect on forest biomass carbon storage

Net primary production (NPP) and its spatial distribution during drought may strongly affect ecosystem carbon stock; however, such effect remain poorly understood in subtropical bamboo forests. Drought can affect plants by regulating endogenous factors such as root structures and plant height, as well as environmental conditions such as soil temperature and photosynthetically active radiation (Bollig et al. 2014). At the end of imposed drought experiment, annual carbon sequestration for TE plots (4.34 t ha−1) was 58.1% lower than that for CK plots (10.36 t ha−1) (Table 3), indicating that imposed drought significantly reduced aboveground tree growth and biomass accumulation. This reduced growth may have been caused by significant decrease in the net photosynthetic rate and transpiration rate during the period of drought, which were in turn driven by decreased water use efficiency (after a brief increase) during the drought (Ying et al. 2011). Our results are consistent with those of Runck (2008), who demonstrated that changes in net aboveground biomass were 17% lower in TE than in CK plots in boreal forest vegetation in interior Alaska.

The reduced growth of bamboo in TE plots could also be caused by the strong reductions in surface soil water content, which led to a sharp decline in the number of new bamboo culms (Table 2). We observed that soil water content in the 0–10- and 10–20-cm soil layers in TE plots decreased by 5.5–18.5% and 4.9–17.4% (except August), respectively, relative to that in CK plots, and the corresponding new bamboo culms in TE plots (200 culms ha−1) was more than two-thirds lower than in CK plots (684 culms ha−1) (Table 2). In addition, the effect of drought on productivity often varies with duration and season, especially for aboveground NPP (Brando et al. 2008) and especially for summer droughts (12% decrease for deciduous forest and 9% decrease for evergreen forest) (Welp et al. 2007). In contrast to our results, however, Runck (2008) showed that imposed drought (1989–2005) did not decrease aboveground tree growth but accelerated the loss of understory vegetation.

Structural changes in the vegetation under drought may also have contributed to reductions of gross primary productivity and carbon storage. For example, early leaf senescence resulted in decreased leaf area, and leaffall or the arrest of leaf expansion can alter the distribution of leaf angles in the canopy (Molen et al. 2011). In our study, annual litterfall during the experiment in TE plots (3.09 t ha−1) was 6.2% higher than in CK plots (2.91 t ha−1) (Fig. 2), which may have resulted from the decrease in net photosynthetic rate and transpiration rate due to plant self-protection mechanisms. Leaf phenology variability may alter the seasonal pattern of photosynthetic uptake, which in turn is determined by leaf gas exchange limitations (Costa-e-Silva et al. 2015). Drought affects annual tree growth in the current year as well as in following years (Granier et al. 2007); therefore, soil moisture limitation of aboveground forest carbon storage is not as likely to occur in the first year (Table 3), and effects should be monitored in subsequent years.

Imposed drought effects on forest soil carbon storage

Increased drought frequency has the potential to affect soil carbon storage and soil carbon emission, which is important for future global carbon balance (Toberman et al. 2008). Surface soil carbon storage is extremely vulnerable to changes in moisture that can alter decomposer activity (Runck 2008). Soil moisture is well known to influence decomposer activity and mineral nutrient leaching directly, indirectly affecting plant growth and thus the amount and composition of plant litter that is eventually deposited in or on the soil (Runck 2008). In our study, the organic carbon content in the 0–60-cm soil profile in TE plots was significantly lower than in CK plots, likely because the shortage of water in soil affected moisture absorption and expansion of bamboo rhizomes and roots, which further indirectly influenced nutrient leaching and the growth of new bamboo through changes in physical properties of the soil. On the other hand, reduced soil water content in TE plots may have reduced microbial activity, leading to slower litter decomposition and nutrient return (Cortez 1998; Xiao et al. 2014). Runck (2008) demonstrated that surface carbon storage (30% in the soil surface mineral layer) significantly decreased due to reduced surface decomposer activity under drought conditions. In our study, soil organic carbon was mainly distributed in the 0–20-cm soil layer, where most roots were found, in both TE and CK plots, suggesting that changes in soil moisture are likely to alter the amount and depth at which carbon is stored in the soil profile.

Pailler et al. (2014) reported that drought stress induced a significant decrease in substrate-induced respiration, which may have a toxic (e.g., ellagic acid) inhibitory effect on microbial activity. In our study, annual average soil respiration in TE plots (1.90 μmol CO2 m−2 s−1) was lower than that in CK plots (2.05 μmol CO2 m−2 s−1), although soil respiration levels did not significantly differ between CK and TE plots at the start of the drought experiment, implying that the imposed drought caused reductions in soil respiration during the growing season (April–October, with the exception of June), although the effect was not significant across the entire experimental period. The lower respiration rates observed at the end of the drought experiment could be attributed to phenolic acids, which are released through carbon and nutrients from dead biomass after drought (Pailler et al. 2014). Our results were consistent with those of Hinko-Najera et al. (2015), Unger et al. (2009), and Suseela et al. (2012), all of whom indicated that soil water content can affect soil respiration by altering substrate availability as well as the composition and activity of decomposer microbes (Williams 2007).

In addition, soil organic carbon dynamics also depends on the quantity of nutrients returned to the soil by dead roots and vegetation decomposition under drought stress (Sayer et al. 2010). Under drought conditions, fine-root biomass has been shown to decrease (Joslin and Wolfe 2003), increase (Gaul et al. 2008), or remain unchanged (Leuschner et al. 2001), while coarse root biomass (> 2 mm diameter, live and dead) has been shown to increase (Davidson et al. 2004). Compared to the growth of aboveground biomass, the response of the moso bamboo root system to drought was relatively subtle, indicating that the response and adaptability of the root system during drought differ from those of branches and leaves. During continuous drought, new moso bamboo is especially vulnerable to water loss, which softens the bamboo pole ruffle, decreases rhizome and root system biomass, and can result in the death of tall bamboo shoots due to water shortage. In addition, compared to normal weather conditions, Sanaullah et al. (2012) reported that drought caused a more than 50% decline in leaf litter decomposition due to sharply decreased decomposition rates. This study may indicate that the combination of new roots in surface soil (Runck 2008), root biomass, litterfall, and litter decomposition rate under imposed drought caused fluctuations in soil carbon storage, particularly during the growing season.

Imposed drought effects on forest ecosystem carbon cycling

Drought is a major driver of carbon fluxes in forest ecosystems, which can either suppress gross primary productivity and decrease CO2 uptake or reduce heterotrophic respiration via the transformation of CO2 sources to the atmosphere, or both (Doughty et al. 2015). As an example of the direct effects of droughts, net ecosystem exchange was estimated to increase by 270 ± 140 Tg C a−1 as a carbon source to the atmosphere during a European summer drought in 2003 (Ciais et al. 2005), as well as the decreased of total ecosystem respiration (Granier et al. 2007). In our study, ecosystem carbon storage had increased by 4.75 t ha−1 in CK plots but decreased by 13.71 t ha−1 in TE plots by the end of the imposed drought, with ecosystem carbon storage in TE plots decreasing by 10.5%. Soil carbon stock within the 0–60-cm soil profile decreased by 4.16 t ha−1 and 16.49 t·ha−1 in CK and TE plots, respectively, and the decrease in TE plots was significantly higher than in CK plots. One possible reason for this result could be that soil water content affects soil aeration and microbial processes, and reduced precipitation could alter root turnover, litterfall, litter decomposition, and mineralization, thus affecting the availability of carbon substrates (Davidson et al. 2004). The increased quantity of carbon stored in the surface soil offset the reduction of carbon captured by tree growth, indicating that the net effect of imposed drought on carbon balance was minimal (Runck 2008). Therefore, drought’s effects on gross primary production and ecosystem respiration primarily contribute to the inter-annual variability in terrestrial carbon sequestration.

The timing of the onset of drought can have a large effect on the annual carbon budget, especially during the spring to summer transition (Unger et al. 2009). Zhou et al. (2013) demonstrated that severe drought induced a lasting reduction in carbon exchange throughout the growing season, whereas severe drought at the end of the growing season did not significantly reduce carbon uptake. There are two important periods when moso bamboo requires a great deal of water for growth: during the development of bamboo shoot in fall and during bamboo shoot emergence in spring (Xiao 2009). In our study, the first 6 months of drought (August–November 2012) affected the development of shoot bamboo due to negative effects on photosynthesis, nutrient storage in bamboo rhizome, fine root biomass accumulation, and shoot differentiation, resulting in a sharp decrease of new bamboo culms.

Drought has a lasting effect on ecosystem carbon dynamics even after the initial responses of gross primary productivity and ecosystem respiration have ended (Molen et al. 2011) due to “carry-over effects”, which refer to ecological processes involved in the response of the soil and memory of plants. In our study, new bamboo shoots in TE plots was reduced by two-thirds compared to the values in the CK plots, and the DBH of new culms in TE plots was lower (10.3 cm) than in CK plots (11.6 cm) (Table 2), indicating that “carry-over effects” can actually occur for a period of time after drought has ended. The balance between carbon input (plant growth) and output (organic matter decomposition), the amount of carbon in southeast China forest, and its role in the global carbon cycle may soon rapidly change. Therefore, knowledge of the interactions between drought and the carbon cycle of forest ecosystems may be improved through better observations of specific plant-soil responses to drought and the net response of ecosystems over longer time periods (Molen et al. 2011).


Our imposed drought experiment has demonstrated that ecosystem carbon dynamics is sensitive to climate change. Ecosystem carbon storage in this subtropical bamboo forest had decreased by 10.5% in TE plots by the end of the experiment. Compared to that at the start of the experiment, ecosystem carbon storage at the end had increased by 4.75 t ha−1 in CK plots but had decreased by 13.71 t ha−1 in TE plots. The responses of these carbon distribution processes are characteristic of normal short-term responses, including direct effect on aboveground carbon, litterfall dynamics, soil respiration, and soil carbon mineralization. Future studies should examine the thorough persistent response in which drought-induced new bamboo mortality is followed by increased mineralization of carbon from dead bamboo rhizomes and fine roots. More frequent drought events in the future may cause subtropical ecosystems to become carbon sources, contributing to the positive carbon-climate feedback already anticipated in the subtropics.



This study was supported by the Fundamental Research Funds for the Central Non-profit Research Institution (CAFBB2014QA008, CAFYBB2016SY006 and RISF2013002), the National Natural Science Foundation of China (31600492 and 31670607), and the Lecture and Study Program for Outstanding Scholars from Home and Abroad (CAFYBB2011007). This paper was also supported by CFERN & BEIJING TECHNO SOLUTIONS Award Funds on excellent academic achievements.

Supplementary material

11284_2017_1529_MOESM1_ESM.pdf (97 kb)
Supplementary material 1 (PDF 97 kb)


  1. Adams HD, Guardiola-Claramonte M, Barron-Gafford GA, Villegas JC, Breshears DD, Zou CB, Troch PA, Huxman TE (2009) Temperature sensitivity of drought-induced tree mortality portends increased regional die-off under global-change-type drought. PNAS 106:7063–7066CrossRefPubMedPubMedCentralGoogle Scholar
  2. Bollig C, Feller U (2014) Impacts of drought stress on water relations and carbon assimilation in grassland species at different altitudes. Agric Ecosyst Environ 188:212–220CrossRefGoogle Scholar
  3. Brando PM, Nepstad DC, Davidson EA, Trumbore SE, Ray D, Camargo P (2008) Drought effects on litter fall, wood production and belowground carbon cycling in an Amazon forest: results of a throughfall reduction experiment. Phillos T Roy Soc A 363:1839–1848CrossRefGoogle Scholar
  4. Ciais P, Reichstein M, Viovy N, Granier A, Ogée J, Allard V, Aubinet M, Buchmann N, Bernhofer C, Carrara A, Chevallier F, Noblet ND, Friend AD, Friedlingstein P, Grünwald T, Heinesch B, Keronen P, Knohl A, Krinner G, Loustau D, Manca G, Matteucci G, Miglietta F, Ourcival JM, Papale D, Pilegaard K, Rambal S, Seufert G, Soussana GF, Sanz MJ, Schulze ED, Vesala T, Valentini R (2005) Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature 437:529–533CrossRefPubMedGoogle Scholar
  5. Cortez J (1998) Field decomposition of leaf litters: relationship between decomposition rates and soil moisture, soil temperature and earthworm activity. Soil Biol Biochem 30:783–793CrossRefGoogle Scholar
  6. Costa ACLD, Galbraith D, Almeida S, Portela BTT, Costa MD, Junior JDAS, Braga AP, Gonçalves PHLD, Oliveira ALAD, Fisher R, Phillips OL, Metcalfe DB, Levy P, Meir P (2010) Effect of 7 years of experimental drought on vegetation dynamics and biomass storage of an eastern Amazonian rainforest. New Phytol 187:579–591CrossRefPubMedGoogle Scholar
  7. Costa-e-Silva F, Correia AC, Piayda A, Dubbert M, Rebmann C, Cuntz M, Werner C, David JS, Pereira JS (2015) Effects of an extremely dry winter on net ecosystem carbon exchange and tree phenology at a cork oak woodland. Agric Forest Meteorol 204:48–57CrossRefGoogle Scholar
  8. Cui HX, Xiao WF, Pan L, Huang ZL, Wang XR, Pang HD (2012) Characteristics of soil carbon storage of Abies fargesii forest in Shennongjia. Sci Silvae Sinicae 48:107–111Google Scholar
  9. Davidson EA, Ishida FY, Nepstad DC (2004) Effects of an experimental drought on soil emissions of carbon dioxide, methane, nitrous oxide, and nitric oxide in a moist tropical forest. Global Change Biol 10:718–730CrossRefGoogle Scholar
  10. Doughty CE, Metcalfe DB, Girardin CAJ, Amézquita FF, Cabrera DG, Huasco WH, Silva-Espejo JE, Araujo-Murakami A, Costa MCD, Rocha W, Feldpausch TR, Mendoza ALM, Da Costa ACL, Meir P, Phillips OL, Malhi Y (2015) Drought impact on forest carbon dynamics and fluxes in Amazonia. Nature 519:78–82CrossRefPubMedGoogle Scholar
  11. Fan YQ (2012) An analysis on the differences of carbon storage of Phyllostachys edulis ecosystem under the terrain conditions and management intensity. Zhejiang Agric Forestry University, LinanGoogle Scholar
  12. Gaul D, Hertel D, Borken W, Matzner E, Leuschner C (2008) Effects of experimental drought on the fine foot system of mature Norway spruce. For Ecol Manag 256:1151–1159CrossRefGoogle Scholar
  13. Ge XG, Xiao WF, Zeng LX, Huang ZL, Lei JP, Li MH (2013) The link between litterfall, substrate quality, decomposition rate and soil nutrient supply in 30-year-old Pinus massoniana forests in the three gorges reservoir area, China. Soil Sci 178:442–451CrossRefGoogle Scholar
  14. Gong ZT (2003) Chinese soil taxonomy (revised proposal). China Science, BeijingGoogle Scholar
  15. Gorsel EV, Berni JAJ, Briggs P, Cabello-Leblic A, Chasmer L, Cleugh HA, Hacker J, Hantson S, Haverd V, Hughes D, Hopkinson C, Keith H, Kljun N, Leuning R, Yebra M, Zegelin S (2013) Primary and secondary effects of climate variability on net ecosystem carbon exchange in an evergreen Eucalyptus forest. Agric For Meteorol 182–183:248–256CrossRefGoogle Scholar
  16. Granier A, Reichstein M, Bréda N, Janssens IA, Falge E, Ciais P, Grünwald T, Aubinet M, Berbigier P, Bernhofer C, Buchann N, Facini O, Grassi G, Heinesch B, Ilvesniemi H, Keronen P, Knohl A, Köstner B, Lagergren F, Lindeoth A, Longdoz B, Loustau D, Mateus J, Montagnani L, Nys C, Moors E, Papale D, Peiffer M, Pilegaard K, Pita G, Pumpanen J, Rambal S, Rebmann C, Rodrigues A, Seufert G, Tenhunen J, Vesala T, Wang Q (2007) Evidence for soil water control on carbon and water dynamics in European forests during the extremely dry year: 2003. Agric For Meteorol 143:123–145CrossRefGoogle Scholar
  17. Hinko-Najera N, Fest B, Livesley SJ, Arndt SK (2015) Reduced throughfall decreases autotrophic respiration, but not heterotrophic respiration in a dry temperate broadleaved evergreen forest. Agric For Meteorol 200:66–77CrossRefGoogle Scholar
  18. Joslin JD, Wolfe MH (2003) Fine-root growth response. In: Hanson PJ, Wullschleger SD (eds) North American temperate deciduous forest response to changing precipitation regimes. Ecological Studies (analysis and synthesis), 166th edn. Springer, New York, pp 274–302CrossRefGoogle Scholar
  19. Leuschner C, Backes K, Hertel D, Schipka Schmitt U, Terborg O, Runge M (2001) Drought responses at leaf, stem and fine root levels of competitive Fagus sylvatica L. and Quercus petraea (Matt.) Liebl. trees in dry and wet years. Forest Ecol Manag 149:33–46CrossRefGoogle Scholar
  20. Liu J, Jiang PK, Wang HL, Zhou GM, Wu JS, Yang F, Qian XB (2011) Seasonal soil CO2 efflux dynamics after land use change from a natural forest to Moso bamboo plantations in subtropical China. For Ecol Manag 262:1131–1137CrossRefGoogle Scholar
  21. Mann ME, Gleick PH (2015) Climate change and California drought in the 21st century. PNAS 112:3858–3859CrossRefPubMedPubMedCentralGoogle Scholar
  22. Meir P, Metcalfe DB, Costa ACL, Fisher RA (2008) The fate of assimilated carbon during drought: impacts on respiration in Amazon rainforests. Phillos T Roy Soc A 363:1849–1855CrossRefGoogle Scholar
  23. Molen MKVD, Dolman AJ, Ciais P, Eglin T, Gobron N, Law BE, Meir P, Peters W, Phillips QL, Reichstein M, Chen T, Dekker SC, Doubková M, Friedl MA, Jung M, Hurk BJJMVD, De Jeu RAM, Kruijt B, Ohta T, Rebel KT, Plummer S, Seneviratne SI, Sitch S, Teuling AJ, Werf GRVD, Wang G (2011) Drought and ecosystem carbon cycling. Agric For Meteorol 151:765–773CrossRefGoogle Scholar
  24. Pailler A, Vennetier M, Torre F, Ripert C, Guiral D (2014) Forest soil microbial functional patterns and response to a drought and warming event: key role of climate-plant-soil interactions at a regional scale. Soil Biol Biochem 70:1–4CrossRefGoogle Scholar
  25. Runck SA (2008) Sensitivity of boreal forest carbon dynamics to long-term (1989-2005) throughfall exclusion in Interior Alaska, 78th edn. University of Alaska Fairbanks, FairbanksGoogle Scholar
  26. Sanaullah M, Rumpel C, Charrier X, Chabbi A (2012) How does drought stress influence the decomposition of plant litter with contrasting quality in a grassland ecosystem? Plant Soil 352:277–288CrossRefGoogle Scholar
  27. Sayer EJ, Tanner EVJ (2010) A new approach to trenching experiments for measuring root-rhizosphere respiration in a lowland tropical forest. Soil Biol Biochem 42:347–352CrossRefGoogle Scholar
  28. Soil Science Society of China ACC (1983) General analysis methods of soil agriculture chemistry. Science Press, BeijingGoogle Scholar
  29. Suseela V, Conant RT, Wallenstein MD, Dukes JS (2012) Effects of soil moisture on the temperature on the temperature sensitivity of heterotrophic respiration vary seasonally in an old-field climate change experiment. Global Change Biol 18:336–348CrossRefGoogle Scholar
  30. Tang ZM, Sayer MAS, Chambers JL, Barnett JP (2004) Interactive effects of fertilization and throughfall exclusion on the physiological responses and whole-tree carbon uptake of mature loblolly pine. Can J Bot 82:850–861CrossRefGoogle Scholar
  31. Toberman H, Evans CD, Freeman C, Fenner N, White M, Emmett BA, Artz RRE (2008) Summer drought effects upon soil and litter extracellular phenol oxidase activity and soluble carbon release in an upland Calluna heathland. Soil Biol Biochem 40:1519–1532CrossRefGoogle Scholar
  32. Trenberth KE, Dai A, Schrier GVD, Jones PD, Barichivich J, Briffa KR, Sheffield J (2014) Global warming and changes in drought. Nat Clim Change 4:17–22CrossRefGoogle Scholar
  33. Unger S, Máguas C, Pereira JS, Aires LM, David TS, Werner C (2009) Partitioning carbon fluxes in a Mediterrranean oak forest to disentangle changes in ecosystem sink strength during drought. Agric For Meteorol 149:949–961CrossRefGoogle Scholar
  34. Welp LR, Randerson JT, Liu HP (2007) The sensitivity of carbon fluxes to spring warming and summer drought depends on plant functional type in boreal forest ecosystems. Agric For Meteorol 147:172–185CrossRefGoogle Scholar
  35. Williams MA (2007) Response of microbial communities to water stress in irrigated and drought-prone tallgrass prairie soils. Soil Biol Biochem 39:2750–2757CrossRefGoogle Scholar
  36. Xiao JH (2009) The engineering science for bamboo forest of China. China Scinece, BeijingGoogle Scholar
  37. Xiao WF, Ge XG, Zeng LX, Huang ZL, Lei JP, Zhou BZ, Li MH (2014) Rates of litter decomposition and soil respiration in relation to soil temperature and water in different aged Pinus massoniana forests in the three gorges reservoir area, China. PLoS One 9:e101890CrossRefPubMedPubMedCentralGoogle Scholar
  38. YingYP Guo J, Wei JF, Jiang Q, Xie NN (2011) Effects of drought stress on physiological characteristics of phyllostachys edulis seedlings. Chin J Ecol 30:262–266Google Scholar
  39. Zhou GM, Jiang PK, Xu QF (2010) Carbon fixation and transformation of bamboo forest ecosystem. China Science, Beijing, pp 92–93Google Scholar
  40. Zhou BZ, Li ZC, Wang XM, Cao YH, An YF, Deng ZF, Letu G, Wang G, Gu LH (2011) Impact of the 2008 ice storm on moso bamboo plantations in southeast China. J Geophys Res 116:2005–2012Google Scholar
  41. Zhou J, Zhang ZQ, Sun G, Fang XR, Zha TG, McNulty S, Chen JQ, Jia Y, Noormets A (2013) Response of ecosystem carbon fluxes to drought events in a poplar plantation in Northern China. For Ecol Manag 300:33–42CrossRefGoogle Scholar

Copyright information

© The Ecological Society of Japan 2017

Authors and Affiliations

  • Xiaogai Ge
    • 1
    • 2
  • Benzhi Zhou
    • 1
    • 2
  • Xiaoming Wang
    • 1
    • 2
  • Qian Li
    • 1
    • 2
  • Yonghui Cao
    • 1
    • 2
  • Lianhong Gu
    • 3
  1. 1.Research Institute of Subtropical Forestry, Chinese Academy of ForestryHangzhouChina
  2. 2.Qianjiangyuan Forest Ecosystem Research Station, State Forestry Administration of ChinaHangzhouChina
  3. 3.Environmental Sciences DivisionOak Ridge National LaboratoryOak RidgeUSA

Personalised recommendations