Plant and Soil

, Volume 368, Issue 1, pp 619–627

Dynamics of soil and root C stocks following afforestation of croplands with poplars in a semi-arid region in northeast China

Authors

    • State Key Laboratory of Forest and Soil Ecology, Institute of Applied EcologyChinese Academy of Sciences
  • De-Hui Zeng
    • State Key Laboratory of Forest and Soil Ecology, Institute of Applied EcologyChinese Academy of Sciences
  • Scott X. Chang
    • Department of Renewable ResourcesUniversity of Alberta
  • Rong Mao
    • Northeast Institute of Geography and AgroecologyChinese Academy of Sciences
Regular Article

DOI: 10.1007/s11104-012-1539-2

Cite this article as:
Hu, Y., Zeng, D., Chang, S.X. et al. Plant Soil (2013) 368: 619. doi:10.1007/s11104-012-1539-2

Abstract

Background and aims

Afforestation on croplands can help sequester atmospheric CO2 through increased carbon (C) storage in the soil and vegetation. However, the dynamics of soil organic C (SOC) and root C stocks, particularly those in the deeper soil layers, following afforestation are not well documented for semi-arid regions. The aim of this study was to investigate the dynamics of soil and root C stocks to 1 m depth following afforestation with poplar (Populus × xiaozhuanica W. Y. Hsu & Liang) on croplands at the Keerqin Sandy Lands in northeast China.

Methods

Forest floor, root and mineral soil samples were collected from 23 paired plots of poplar plantations with different stand basal areas (SBA, ranging from 0.2 m2 ha−1 to 32.6 m2 ha−1) and reference croplands using a paired-site design. Changes of SOC concentration and content, and root C content were analyzed using paired t tests, and the relationships between forest floor C content, soil and root ΔC (ΔC refers to the difference in C stocks between a poplar plantation and the paired cropland) and SBA were tested with a polynomial regression analysis.

Results

Afforestation resulted in linear increases of ΔC in the forest floor and 0–10 cm mineral soil with SBA (R2 = 0.67, p < 0.001 and R2 = 0.34, p = 0.003, respectively), but there were no clear relationships between SOC stocks in the soil deeper than 10 cm and SBA. The fine root C stock increased by afforestation across all the soil layers (p < 0.05), and root ΔC had a quadratic curve (the first two mineral soil layers) or linear (the other mineral soil layers) relationship with SBA. About 73 % of the variance of ΔC in the top soil was explained by changes in the forest floor C stock, but changes in plant derived C stocks did not explain the variance of soil ΔC in the deeper layers very well.

Conclusions

Our study suggest that afforestation increased C sequestration in the forest floor and surface mineral soil, and C stocks in the forest floor and surface mineral soil and roots were strongly controlled by the SBA, which changes with stand development, in the studied semiarid region in northeast China.

Keywords

AfforestationC stockCroplandPoplar plantationSandy soilSemi-arid region

Introduction

Afforestation of degraded croplands has occurred globally and has been promoted as a means to sequester CO2 from the atmosphere to abate its rising concentration caused by large anthropogenic emissions (Post and Kwon 2000; Paul et al. 2002; Laganière et al. 2010; Li et al. 2012). For example, in China alone, 72 million ha of croplands were planted with trees between 1999 and 2003 under a large-scale afforestation program, the Grain for Green Project, in the Three-North region (Cao 2008). Afforestation of croplands can lead to rapid C accumulation in tree biomass (Nilsson and Schopfhauser 1995; Richter et al. 1999; Karhu et al. 2011), but changes in soil organic C (SOC) stocks after afforestation is not well understood (Hernandez-Ramirez et al. 2011), even though SOC stocks account for over two-thirds of C stored in forest ecosystems (Dixon et al. 1994; Arevalo et al. 2009).

Most empirical studies have suggested that the direction and magnitude of SOC stock changes had large variations after afforestation of croplands, and were affected by many factors such as previous land-use type, climate, soil texture, tree species composition, stand age, and management practices (Guo and Gifford 2002; Paul et al. 2002; Laganière et al. 2010; Arai and Tokuchi 2010; Karhu et al. 2011; Li et al. 2012). Moreover, there is little information on the change of C stocks in deeper soil layers because the majority of previous studies only reported changes of SOC stocks in the top 30 cm (Berthrong et al. 2009; Laganière et al. 2010; Mao et al. 2010; Schmidt et al. 2011). Indeed, a substantial amount of SOC (ranging from 46 % to 63 % of total SOC in 1 m depth) can be stored in deeper soil layer (>30 cm) (Jobbágy and Jackson 2000; Rumpel and Kögel-Knabner 2011), and afforestation of cropland can benefit from the deep rooting habit of the trees (Laganière et al. 2010; Olupot et al. 2010). In addition, the stability of SOC in deep soil layers can be altered by the changed distribution of roots along soil profiles following afforestation on croplands (Rasse et al. 2005; Fontaine et al. 2007; Salomé et al. 2010; Schmidt et al. 2011), and the impact of environmental conditions in deeper soil layers on root C stabilization are poorly understood (Rasse et al. 2005; Borken et al. 2007). Therefore, there is an urgent need to study the changes following afforestation of soil and root C stocks in deeper soil layers.

The SOC stocks are determined by the balance between the input of C derived from litterfall and rhizodeposition and the loss of C mainly through soil organic matter decomposition (Arai and Tokuchi 2010; Laganière et al. 2010; Persson 2012). In general, afforestation of cropland induces an initial decline in the SOC stocks as the decomposition of soil organic matter would be greater than the input of organic matter from trees, followed by a gradual increase in SOC stocks due to the increase of net primary productivity with stand age (Davis and Condron 2002; Paul et al. 2002; Hu et al. 2008; Laganière et al. 2010; Mao et al. 2010; Karhu et al. 2011). However, the mechanisms for SOC stock changes over time are not very well understood because few have studied the changes of plant derived C stocks, especially root C stocks, and their relationships with SOC stocks, following afforestation of croplands (Olupot et al. 2010).

Litterfall of above- and below-ground sources as well as the rate of soil organic C decomposition are not only affected by stand age but also tree density, which influences soil microclimatic conditions and the amount of litterfall (Helmisaari et al. 2007; Laganière et al. 2010). Therefore, in this study we use stand basal area (SBA, the cross-sectional area of all the trees at breast height per hectare) to evaluate the dynamics of SOC and root C stocks with stand development, because SBA integrates information on both stand age and tree density.

Poplar (Populus) species is one of the most widely planted trees in the Grain for Green Project, particularly in the Keerqin Sandy Lands, a semi-arid region in northern China (Hu et al. 2008; Mao et al. 2010), because this tree species has high productivity, can reduce soil erosion and is adaptable to drought environments (Pellegrino et al. 2011). While the primary objective of establishing the poplar plantations was to reduce wind erosion and the occurrence of dust storms, the potential ecological services of the poplar plantations in providing a C sink and thus mitigating climate change should be further studied beyond some of the initial research (Hu et al. 2008; Mao et al. 2010).

In this study, we selected 23 paired stands of poplar plantations with different SBA (ranging from 0.2 m2 ha−1 to 32.6 m2 ha−1) and adjacent croplands in the southeastern region of Keerqin Sandy Lands, and measured C stocks in the mineral soil and roots in different soil layers (i.e., 0–10, 10–20, 20–40, 40–60, 60–80 and 80–100 cm), in addition to forest floor C stocks. The objectives of this study were to determine (1) the dynamics of C stocks in the forest floor, mineral soil and fine roots following afforestation with poplar trees on croplands, and (2) the relationships of SOC stocks with forest floor and root C stocks.

Materials and methods

Study site

This study site is located in the southeastern region of the Keerqin Sandy Lands (42°30′–42°55′N, 122°19′–122°30′E), China. The area has a typical temperate continental monsoon climate, with a mean annual temperature of 5.7 °C, ranging from −23.2 °C in January to 32.4 °C in July (1954–2004), a mean annual precipitation of about 450 mm (ranging from 224 mm to 661 mm during 1954–2004), with more than 60 % occurring from June to August, a mean annual potential evaporation ranging from 1,300 mm to 1,800 mm, and an average length of frost-free season of about 150 days. The major soil type is classified into the Entisol order, Semiaripsamment group (according to the United States Soil Classification System) that is developed from sandy parent material through the action of wind (Zhenghu et al. 2007).

Soil sampling and chemical analysis

Forty-six 20 × 20 m plots, 23 poplar (Populus × xiaozhuanica W. Y. Hsu & Liang, a hybrid of P. nigra var. italica and P. simonii) plantations and 23 adjacent cropland plots, were selected in Kezuohouqi and Zhangwu county in June 2011 following a paired-plot experimental design. The distance between each paired poplar plantation and cropland was less than 100 m. Diameter at breast height (DBH, breast height at 1.3 m) and tree height were measured for all live trees in each poplar plot. The SBA was calculated from measurements of the DBH (cm) of all trees in a known area (A, ha) and expressed as m2 ha−1:
$$ SBA=\frac{\pi }{40000}\times \frac{{\sum {DB{H^2}} }}{A} $$
The details of the studied poplar stands were given in Table 1.
Table 1

Stand location and characteristics

Poplar plantation

Location

Elevation (m)

Tree height (m)

DBH (cm)

Density (Trees ha−1)

Stand basal area (m2 ha−1)

Cropland

Location

Elevation (m)

F1

42°53′36″N,122°24′49″E

248

10.58

9.83

1025

8.50

C1a

42°53′42″N, 122°24′53″E

248

F2

42°54′02″N, 122°25′32″E

247

6.85

7.39

1300

6.11

C2

42°54′00″N, 122°25′34″E

246

F3

42°53′59″N, 122°25′10″E

249

18.70

10.54

1575

14.62

C3

42°53′56″N, 122°25′07″E

248

F4

42°53′04″N, 122°24′08″E

250

5.26

5.73

1275

3.65

C4

42°52′58″N, 122°24′09″E

250

F5

42°53′18″N, 122°24′11″E

251

8.77

9.21

1100

8.20

C5

42°53′17″N, 122°24′07″E

252

F6

42°54′13″N, 122°23′36″E

252

15.20

15.10

700

13.15

C6

42°54′17″N, 122°23′41″E

251

F7

42°54′11″N, 122°23′30″E

252

3.35

2.92

1100

0.85

C7

42°54′04″N, 122°23′32″E

255

F8

42°54′00″N, 122°23′25″E

252

4.80

6.35

850

2.92

C8

42°53′59″N, 122°23′23″E

251

F9

42°54′25″N, 122°23′28″E

250

20.68

17.11

1150

28.51

C9

42°54′23″N, 122°23′28″E

250

F10

42°56′13″N, 122°24′28″E

245

7.34

9.16

1150

8.39

C10

42°56′10″N, 122°24′27″E

245

F11

42°55′42″N, 122°24′39″E

246

17.76

13.16

1025

15.69

C11

42°55′42″N, 122°24′37″E

246

F12

42°59′25″N, 122°20′25″E

244

8.95

11.58

825

9.24

C12

42°59′26″N, 122°20′20″E

244

F13

42°59′06″N, 122°20′52″E

246

7.63

6.39

1950

7.20

C13

42°59′04″N, 122°20′54″E

246

F14

42°37′58″N, 122°22′24″E

186

15.80

13.37

750

10.69

C14

42°38′01″N, 122°22′24″E

185

F15

42°37′26″N, 122°20′49″E

167

2.58

1.79

750

0.20

C15

42°37′25″N, 122°20′52″E

167

F16

42°36′25″N, 122°21′36″E

155

6.34

7.62

1475

7.10

C16

42°36′24″N, 122°21′33″E

155

F17

42°37′06″N, 122°20′54″E

166

4.38

4.62

1075

1.85

C17

42°37′04″N, 122°20′52″E

166

F18

42°37′06″N, 122°20′58″E

165

30.72

22.00

625

24.79

C18

42°37′04″N, 122°20′52″E

165

F19

42°37′05″N, 122°21′06″E

164

29.87

22.90

775

32.60

C19

42°37′06″N, 122°21′05″E

165

F20

42°32′52″N, 122°28′13″E

131

13.05

14.74

725

12.56

C20

42°32′51″N, 122°28′19″E

132

F21

42°32′40″N, 122°28′17″E

133

8.16

9.45

700

5.04

C21

42°32′38″N, 122°28′21″E

131

F22

42°32′34″N, 122°28′33″E

128

13.71

11.00

1500

14.98

C22

42°32′36″N, 122°28′31″E

129

F23

42°32′34″N, 122°29′21″E

126

18.68

14.14

825

13.36

C23

42°32′32″N, 122°29′22″E

125

aAll croplands were planted to corn

In each plot, soil samples were collected from six layers (i.e., 0–10, 10–20, 20–40, 40–60, 60–80 and 80–100 cm) from two pits, and thoroughly mixed to form a homogenized sample for each soil layer. At the same time, four soil cores were sampled using an auger (6 cm diameter) in each of the above mineral soil layers and fine roots (< 2 mm in diameter) were sorted out by hand and the soil was rinsed off with deionized water. The fine root was not sorted into living and dead roots or into poplar and grass roots. Because there was no obvious aboveground litter accumulation in the cropland, we only investigated forest floor organic matter accumulation in the poplar stands using four 50 × 50 cm quadrats in each plot. Soil bulk density was determined in each soil layer for calculation of SOC content.

Mineral soil samples were air dried at room temperature (20 °C). Forest floor and fine root samples were dried at 65 °C for 48 h and weighed. Soil and plant samples (i.e., forest floor and root) were ground to a fine powder with a ball mill for the analysis of C concentrations using the Walkey and Black K2Cr2O7–H2SO4 oxidation method (Nelson and Sommers 1996).

Statistical analysis

All data were analyzed using the R software package. Paired t tests were used to compare the differences in SOC concentration, SOC content, and root C content in each soil layer between croplands and poplar plantations. The relationships of SBA and C content in the forest floor, and soil and root ΔC (ΔC refers to the difference in C stocks between the poplar plantation and the paired cropland) were tested with a polynomial regression analysis to find the best fit. The relationships of soil ΔC with forest floor C content and root ΔC were analyzed with multiple linear regression analysis. The significant level was set at α = 0.05 for all the statistical analyses unless otherwise noted.

Results

Forest floor C

There was no measurable forest floor C stock in the stands with SBA < 3.65 m2 ha−1. Therefore, forest floor C ranged from 0 kg C m−2 to 0.74 kg C m−2 across all the studied poplar stands. Forest floor C increased with increasing SBA (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-012-1539-2/MediaObjects/11104_2012_1539_Fig1_HTML.gif
Fig. 1

Changes of C stocks in the forest floor with increasing stand basal area in the poplar plantations

Soil C

Soil organic C concentrations decreased from 4.38 g C kg−1 in the 0–10 cm layer to 1.20 g C kg−1 in the 80–100 cm layer in croplands, and from 6.68 g kg−1 to 1.56 g kg−1 in poplar plantations (Fig. 2a). Soil organic C concentration and content in the 0–10 cm mineral soil layer increased after afforestation (p < 0.001), but there was no afforestation effect in the other soil layers (Fig. 2a, b). Soil ΔC had an increased trend with increasing SBA in the 0–10 cm layer, but not in the other soil layers (Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-012-1539-2/MediaObjects/11104_2012_1539_Fig2_HTML.gif
Fig. 2

Changes of soil organic C concentrations (a), C content (b) and root C content (c) in different soil layers after afforestation with a hybrid poplar on croplands. The asterisk indicates a significant difference (at a level of *, P < 0.05; **, P < 0.01; and ***, P < 0.001) between cropland and poplar plantation for a given soil layer. The horizontal error bars are standard errors of the means

https://static-content.springer.com/image/art%3A10.1007%2Fs11104-012-1539-2/MediaObjects/11104_2012_1539_Fig3_HTML.gif
Fig. 3

Dynamics of soil ΔC with increasing stand basal area after afforestation with a hybrid poplar. The soil ΔC refers to the difference in soil C stocks between a poplar plantation and the paired cropland

Root C

Afforestation increased root C contents across all soil layers (Fig. 2c). Root C contents (per soil layer) ranged from 0.61 g C m−2 to 5.19 g C m−2 among soil layers in croplands and from 14.96 g C m−2 to 38.12 g C m−2 in poplar plantations. Root ΔC in the 0–10 and 10–20 cm mineral soil layers had quadratic relationships with SBA (Fig. 4). However, root ΔC in the deeper layers increased linearly with SBA, and the rates of increase of root C content decreased with increasing soil depth.
https://static-content.springer.com/image/art%3A10.1007%2Fs11104-012-1539-2/MediaObjects/11104_2012_1539_Fig4_HTML.gif
Fig. 4

Dynamics of root ΔC with increasing stand basal area after afforestation with a hybrid poplar. The root ΔC refers to the difference in root C stocks between a poplar plantation and the paired cropland

Relationships between mineral soil C and plant C stocks

Soil ΔC had a positive relationship with forest floor C content and root ΔC in the 0–10 cm layer (R2 = 0.38, p = 0.02), but not for root ΔC in the other mineral soil layers. Forest floor C contents and root ΔC explained 73 and 27 %, respectively, of the variance in soil ΔC.

Discussion

The higher SOC concentrations and contents in the forest floor and in the upper 10 cm of the mineral soil in the poplar plantations than in the croplands suggested that afforestation increased the sequestration of atmospheric CO2 into the forest floor and the surface mineral soil. Increased SOC stocks in the topsoil in the recently afforested land as compared to croplands have been reported in other regions (Jug et al. 1999; Post and Kwon 2000; Guo and Gifford 2002; Vesterdal et al. 2002; Hooker and Compton 2003; Clark and Johnson 2011). Generally, SOC stocks in the surface mineral soil initially may decrease following the conversion of croplands to forests, and would subsequently increase with stand development (Davis and Condron 2002; Paul et al. 2002; Hu et al. 2008; Laganière et al. 2010; Mao et al. 2010; Karhu et al. 2011; Li et al. 2012). However, in this study we did not find the initial decrease in SOC stocks, but SOC stocks in the topsoil linearly increased with increasing basal area (and the associated change in stand age) in the poplar stands. This result indicated that the soil C stock might begin to increase right after afforestation in the studied sandy and nutrient-poor mineral soils.

Changes of SOC in the deeper soil layers following afforestation were different from the upper layer (Laganière et al. 2010; Mao et al. 2010; Schmidt et al. 2011). In this study, SOC concentrations and contents in the deeper soil layers (i.e., > 10 cm) were not changed following afforestation (Fig. 2a, b), consistent with Arevalo et al. (2011), who also found that SOC stocks in deeper soil layers (20–50 cm) did not increase after converting an agricultural land to hybrid poplar plantations in the Parkland region in central Alberta, Canada. In addition, Clark and Johnson (2011) showed that SOC stocks in the 0–10 cm layer, but not in the 10–20 cm soil layer, increased after croplands were converted to secondary forests in western New England (WNE). In contrast, Richter et al. (1999) demonstrated that SOC stocks in the deeper soil layer (35–60 cm) decreased after 40 years of afforestation, although there was an increasing trend in the surface mineral soil layer (0–7.5 cm).

Forest floor C stocks increased with increasing SBA (Fig. 1), similar to what Vesterdal et al. (2002) reported. The large dead and decomposing leaf litter C pool in the poplar plantations of more advanced stages of development was in strong contrast to the cropland where repeated biomass removal resulted in the lack of accumulation of dead organic matter above the mineral soil surface (Paul et al. 2002; Vesterdal et al. 2002; Laganière et al. 2010).

Greater root C contents in poplar plantations than in croplands were found across all the soil layers (Fig. 2c), indicating the importance of tree root C in soil C sequestration. In addition, we found that root ΔC increased with SBA, indicating that more C could be sequestered into fine root with stand development following afforestation. It was consistent with Helmisaari et al. (2007) who found that fine root biomass of both Norway spruce (Picea abies) and Scots pine (Pinus sylvestris) increased with the increasing basal area of trees. Moreover, soil temperature and water content, and soil nutrient availability can affect the distribution of fine roots in soil layers (Rasse et al. 2005; Borken et al. 2007; Helmisaari et al. 2007; Olupot et al. 2010), which might account for the different relationships of root ΔC and SBA when comparing upper with deeper soil layers.

The changes in forest floor and tree root C stocks following afforestation can affect SOC stocks (Richter et al. 1999; Paul et al. 2002; Hu et al. 2008; Peichl et al. 2010; Clark and Johnson 2011). In this study, we demonstrated that the increased input of plant derived C, especially the forest floor C (explaining 73 % of the variance of soil ΔC), was one of the most important causes for the increased SOC stocks in the topsoil after afforestation. However, SOC stocks in the deeper soil layers could not be very well explained by the changes of forest floor and root C stocks.

The dynamics of SOC are not only controlled by the amount and the molecular structure of the organic matter input, but also by the predominate environmental and biological factors (Schmidt et al. 2011), which might lead to an inconsistent response of SOC in the deeper soil layers as compared to the surface soil layer, considering the difference in soil micro-environment and microbial communities across soil profiles (Richter et al. 1999; Agnelli et al. 2004; Laganière et al. 2010; Salomé et al. 2010). The increased input of organic matter to the forest floor and in the form of root turnover could lead to increased SOC stocks following afforestation (Richter et al. 1999; Paul et al. 2002). However, fresh C derived from tree roots and soil dissolved organic C (DOC) leaching from the forest floor could stimulate the decomposition of old soil organic C in the deeper soil layers because of the stimulation of soil microbial activities, which might lead to a loss of SOC in the deeper soil layers (Binkley and Resh 1999; Fontaine et al. 2007; Schmidt et al. 2011), particularly in the sandy soil with little protection from adsorption to organophilic clays in the deeper soil, leading to a faster turnover rate of SOC in the deeper soil as compare to the surface soil (Richter et al. 1999).

Conclusions

Afforestation with poplar trees on croplands was effective to sequester atmospheric C into soil organic matter in the organic layer and in the surface mineral soil and into root biomass in the studied semiarid region in northeast China. The accumulated forest floor C along stand development was one of the most important aspects to increasing SOC stocks in the soil after afforestation on croplands. We also concluded that C sequestration in the poplar plantations increased with stand development. In order to better understand the long-term potential of soil C sequestration following afforestation, we should investigate in the future the influence of afforestation on SOC turnover and transformation rates among the different soil C pools.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Nos. 31000297 and 41101283). We thank Gui-Gang Lin, Jing-Shi Li and Jian Ma for their considerable helps in field work and laboratory analyses, and two anonymous reviewers for their constructive suggestions on this manuscript.

Copyright information

© Springer Science+Business Media Dordrecht 2012