Wetlands (2014) 34:235–241 DOI 10.1007/s13157-013-0487-9

ASIA SPECIAL FEATURE

Soil Organic Carbon and its Fractions in Relation to Degradation and Restoration of Wetlands on the Zoigê Plateau, China Junqin Gao & Xuewen Zhang & Guangchun Lei & Guangxing Wang

Received: 5 August 2012 / Accepted: 27 September 2013 / Published online: 10 October 2013 # Society of Wetland Scientists 2013

Abstract Through restoration, degraded wetlands are thought to recover their carbon storage function over time. However, little is known about the dynamics of soil organic carbon (SOC) and its fractions after restoration of degraded wetlands. In this study, we selected four wetlands, two degraded (one grazed and one graze-released site), one restored for 6 years and one natural site from the Zoigê wetlands on the Qinghai–Tibetan Plateau, China to investigate the dynamics of SOC following restoration. The concentrations of SOC, dissolved organic carbon (DOC), light fraction organic carbon (LFOC), and heavy fraction organic carbon (HFOC) in the degraded sites were significantly lower than those in the natural wetland. In contrast, soil δ13C (i.e. 13C/12C ratio) in the degraded sites was significantly higher than that in the natural site. After 8 years of restoration, the restored and the natural wetland sites did not differ significantly in SOC, DOC, LFOC, HFOC or δ13C. Therefore, restored wetlands regained some of their role in carbon storage over time. SOC turnover was slower in the natural wetland than in the degraded and restored sites.

J. Gao (*) : X. Zhang : G. Lei (*) School of Nature Conservation, Beijing Forestry University, P.O. Box 159, Qinghuadonglu 35, Haidian District, Beijing 100083, China e-mail: [email protected] e-mail: [email protected] G. Wang Department of Geography and Environmental Resources, Southern Illinois University Carbondale, Faner Hall, 1000 Faner Dr., Carbondale 62901, USA

Keywords Labile organic carbon . Soil organic carbon . Wetland degradation . Wetland restoration . Zoigê plateau

Introduction Although wetlands occupy only 5–8 % of the Earth’s land surface, they store 20–30 % or more of the world’s terrestrial soil organic carbon (SOC) (Mitsch and Gosselink 2007). Thus, wetlands could play an important role in the reduction of carbon concentration in the atmosphere (Millennium Ecosystem Assessment 2005; Mitsch and Gosselink 2007). However, human activities such as wetland drainage and overgrazing are disturbing wetlands and degrading their biogeochemical processes and ecosystem functions (Waddington and McNeil 2002). In order to recover the function of the degraded wetlands for carbon storage, recent wetland restoration has occurred worldwide by means of restoring hydrology and vegetation and limiting overgrazing (Waddington and Price 2000; Badiou et al. 2011). Wetland restoration provides the potential to allow the degraded wetlands to function as a carbon sink (Tuittila et al. 1999; Waddington and Price 2000; Yli-petäys et al. 2007). For example, freshwater addition enhanced the accumulation of SOC in the Yellow River Delta, China (Wang et al. 2011). In a study of natural bogs in Canada, Waddington and Price (2000) also showed that the restoration of peatland recovered the potential for carbon sequestration, although net carbon storage was not achieved in a short period of time. The restoration of seasonally, semi-permanently and permanently drained wetlands could sequester a nearly equivalent amount

236

of carbon to natural wetland when restored (approximately 0.88 Mg C ha−1 year−1; Badiou et al. 2011). The conversion of wetlands to agricultural lands resulted in SOC loss in North America, but wetland restoration could provide the potential to sequester abundant SOC over a period of 10 years (Euliss et al. 2006). SOC represents the balance between the input of organic matter and output of the organic carbon losses through many processes such as decomposition, leaching and dissolved carbon export (Turetsky 2004; Wang et al. 2010). The storage patterns of labile fractions of organic carbon, such as dissolved organic carbon (DOC) and light fraction organic carbon (LFOC), also shift if the amount of organic matter changes in the soil. Therefore, changes in these components have been suggested as early indicators of the resumption of SOC dynamics in restored systems (Gregorich et al. 1994; Fang et al. 2009) and also as important indicators of soil quality. For example, LFOC was the most sensitive to the changes of organic carbon following the abandonment of cultivated lands in the northeast China (Zhang et al. 2007). Forest and grassland studies suggest that 13C abundance can be used as an indicator of SOC dynamics (Fang et al. 2009). Because respired CO2 indicates the depletion of 13 C, a shift in 13 C vs. 12C ratio indicates a change in carbon storage dynamics. The carbon component of SOC includes 12C, 13C and 14C, with 12C the most abundant carbon type. During the decomposition of soil organic matter (SOM), carbon isotope fractionation occurs. Uptake and utilization of organic substances by soil microorganisms may alter the isotopic composition of e.g. microbial biomass, respired CO2, soil organic carbon and dissolve organic carbon (Werth and Kuzyakov 2010). Little is known about the dynamics of SOC and the labile fractions of organic carbon in relation to the degradation and restoration in peatlands in China. In this study, we hypothesized that SOC content decreases during wetland degradation and increases after restoration. Moreover, we presumed that the labile fractions of SOC such as DOC and LFOC change more rapidly in degraded and restored wetlands in comparison to natural wetlands. To test these hypotheses, we investigated the dynamics of soil carbon following the degradation and restoration of Zoigê wetlands in China. Furthermore, we also analyzed the turnover times of SOC in degraded, restored and natural wetlands.

Materials and Methods Site Characteristics and Sampling The study area was located in the Zoigê wetlands (32°20′– 34°00′N, 101°36′–103°30′E) at the eastern edge of QinghaiTibet Plateau in China. The average altitude of the study area is 3,400 m (Ding et al. 2004). The region is characterized by a

Wetlands (2014) 34:235–241

mean annual temperature of 0.6 °C. Mean monthly temperatures are −10.7 °C in January and 10.9 °C in July, respectively (Ding et al. 2004). In May 2011, four sites were selected in the Zoigê wetlands. The selected sites were spatially distributed within a radius of 1 km (Fig. 1). Although the altitudes of these selected sites ranged from 3,441 m to 3,465 m, these sites were located within the same montane wetland (organic flat). Before the 1930’s, the four sites were similar in soil, vegetation and hydrology, so that these wetlands were permanently inundated with about 30–50 cm average depth of water (Chai et al. 1965; Sun et al. 1998). The dominant species in this tussock peatland is Carex spp. and the soil type is humus marsh soil (Chai et al. 1965). In the degraded wetland, a grazed site (grazed) was seriously degraded during the last several decades because of heavy grazing and invasion by Ochotona curzoniae. The surface water was drained with a ditch. The dominant species were Kobresia humilis and Potentilla anserina. Another site was fenced in 2010, and was no longer grazed (graze-released). This site was seasonally inundated and the dominant species included Potentilla anserina and Trollius chinensis. A restored site (restored) had been restored by blocking ditches for about 8 years before this study (i.e., 2005), and was also rotationally grazed by cattle; it was seasonally inundated and dominated by Blysmus sinocompressus and Kobresia tibetica. A reference site (natural wetland) was a tussock peatland in its natural condition, with rotational cattle grazing, all year inundation, and dominance by Carex muliensis. Three 2×2 m plots were arbitrarily established within each wetland type. Within each plot, nine soil samples were taken from three depths: 0–10, 10–20, and 20–30 cm. Each soil sample was split into three sub-samples in the field and placed into separate plastic bags. One sub-sample was used to measure bulk density and soil moisture. The second one was refrigerated at 4 °C for DOC analyses. The third one was air-dried for SOC and density fractionation analyses. Density Fractionation The light fraction (LF) and heavy fraction (HF) of the soil were separated by flotation in a NaI solution of 1.7 g cm−3. The concentrations of LFOC and HFOC were determined using a FLASH1112 CNS Analyzer (Zhang et al. 2007). Each sample of 100 g of soil was placed in 500 ml of NaI solution, and ultrasonicated at 400 J ml−1 with a calibrated Vibracell VCX 600 probe-type model. The supernatant was filtered through a 0.45 μm membrane filter. The fraction recovered on the filter was washed with 0.01 M CaCl2 and distilled water. The sediment from the centrifuge tubes was replaced in the beakers and re-suspended in NaI. The above procedure was repeated three times to clean the sediment. The obtained three fractions were combined, and these combined

Wetlands (2014) 34:235–241

237

China

Fig. 1 The position of the study sites on the Zoigê, Qinghai–Tibetan Plateau, China. (National Dynamic Atlas 2012)

fractions are referred to as the LF. The sediment from the centrifuge tubes and the beaker, i.e., the heavy fraction (HF), was washed once with 0.01 M CaCl2 and ten times with distilled water.

230 rpm, which was subsequently centrifuged for 20 min at 8, 000 rpm. The supernatant was filtered through a 0.45 mm filter into separate vials for carbon analysis (Ghani et al. 2003; Zhang et al. 2007).

DOC Measurement

δ13C Measurement

DOC in fresh soil was determined with a Shimadzu TOC analyzer, which had a lower detection limit of 50 ug l−1 (Shimadzu Corp., Kyoto, Japan). Moist soil samples (equivalent to 10 g oven-dried weight) were extracted with 30 ml of distilled water for 30 min using an end-over-end shaker at

For analysis of δ13C in soil organic carbon, carbonates were removed from the soil samples within a period of 3 days in a desiccator, which contained 10 M HCl. The samples were neutralized by adding deionized water and then dried. The δ13 C (i.e. 13C/12C ratio) of the samples were then determined

Table 1 Results of repeated measures ANOVA of the effects of site type, soil depth and the interaction on soil properties. P values for significant effects and interactions are in bold

Site

Bulk density pH Soil moisture Soil organic carbon (SOC) Dissolved organic carbon (DOC) Light fraction organic carbon (LFOC) Heavy fraction organic carbon (HFOC) SOC/TN δ13C

Depth

Site × depth

F 3, 8

P

F 2, 7

P

F 6, 16

P

282.57 9.78 312.20 97.29 143.28 84.91 16.45 6.21 322.99

<0.001 0.005 <0.001 <0.001 <0.001 <0.001 0.001 0.017 <0.001

42.92 0.30 2.48 5.45 23.75 0.74 9.84 5.36 19.14

<0.001 0.745 0.115 0.016 <0.001 0.490 0.002 0.017 0.001

9.10 1.55 1.77 1.18 5.61 1.11 1.18 1.07 3.03

<0.001 0.224 0.169 0.366 0.004 0.400 0.366 0.420 0.075

238

Wetlands (2014) 34:235–241

with an isotope ratio mass spectrometer (Delta Plus, Thermo Fisher Scientific, Bremen, Germany) coupled with an elemental analyzer (NC 2500, CE Instruments, Milano, Italy). Data Analysis To test the effects of sites type and soil depth on SOC, DOC, LFOC, HFOC and δ13 C (13C/12C ratio), we used ANOVA with repeated measures, and in this model soil depth was used as the repeated factor (Povin et al. 1990; Stuefer and Huber 1998; Yu and Dong 2003). When significant effects were found, Tukey’s HSD tests were used for multiple comparisons. Regression analysis was used to examine the relationship between soil δ13C and log SOC. All the significant tests were made at a significant level of p =0.05. All the statistical analyses were conducted using SPSS 18.0 (SPSS, Chicago, IL, USA) and Sigmaplot 11.0 software package (Systat Software, San Jose, CA, USA).

Results Soil Properties of the Selected Sites There was a significant interaction of site type and soil depth on bulk density (Table 1). Soil bulk density at the 0–10 cm layer differed significantly between the graze-released and restored sites, but not between the restored and reference sites. In the soil layers of both 10–20 cm and 20–30 cm, bulk density was significantly different among the restored, reference, and grazed or graze-released, but not between the grazed and graze-released sites (Fig. 2a). The pH values were significantly higher in the grazed site than in the reference site, but not significantly different among the grazed, grazed-release, and restored sites (Fig. 2b, Table 1). Soil moisture was the highest in the natural wetland (Fig. 2c). Variation of SOC and its Fractions Site type significantly affected soil organic carbon (SOC) and its fractions (Table 1, F=6.2 to 323.0, p =0.017 to <0.001). SOC, dissolve organic carbon (DOC), light fraction organic C concentrations (LFOC) or heavy fraction organic C concentrations (HFOC) did not differ significantly between the reference and the restored wetland sites, but they were significantly higher than those in the grazed and graze-released sites (Fig. 3a, c, d). In the grazed and graze-released sites, the values of SOC were 186.7 and 293.0 g C kg−1 soil, respectively, significantly (p <0.05) lower than the corresponding values in the restored (418.5 g C kg−1) and reference site (491.8 g C kg−1; Fig. 3b). SOC, LFOC and HFOC were also higher in the graze-released than in the grazed site (Fig. 3a, c, d). SOC/TN did not differ significantly between the grazed,

Fig. 2 a Bulk density, b pH and c soil moisture in the grazed, grazereleased, restored and reference wetland sites on the Zoigê plateau

graze-released and reference sites (p >0.05), but SOC/TN in the restored site was significantly lower than that in the grazed site (Fig. 3e). Soil depth significantly affected SOC (F=5.45, p =0.016), DOC (F=23.75, p <0.001), HFOC (F=9.84, p =0.002) and

Wetlands (2014) 34:235–241

239

Fig. 3 a Soil organic carbon, b dissolved organic carbon, c light fraction organic carbon, d heavy fraction organic carbon, e SOC/TN and f δ13C in the grazed, graze-released, restored and reference sites on the Zoigê plateau

SOC/TN (F=5.36, p =0.017, Table 1). Both SOC and HFOC were significantly higher in the upper soil layer (0–10 cm) than in the lower layer (20–30 cm; Fig. 3a, d). SOC/TN was significantly higher in the lower layer (20–30 cm) than in the upper soil layers (0–10 cm and 10–20 cm; Fig. 3e). The pattern of DOC differed across soil depths depending on site type (Table 1, F=5.16, p =0.004). In the upper 0–10 cm layer, dissolved organic carbon concentrations were 224.4±

27.8 mg kg−1 in the grazed site, significantly lower than that in the graze-released, restored and reference sites (Fig. 3b, p < 0.05). In the lower layers of 10–20 cm and 20–30 cm, dissolved organic carbon concentrations were 144.9±15.9 and 163.3±27.1 mg kg−1 in the grazed and graze-released sites which were significantly lower than that in the restored and reference sites, 545.7±51.2 and 474.3±57.1 mg kg−1, respectively (Fig. 3b).

240

δ13C and its Relationship with SOC Both site type (F=322.99, p <0.001) and soil depth (F=19.14, p =0.001) significantly affected δ13C (Table 1). Soil δ13C in the reference and restored site did not differ significantly, but values in these types were significantly lower than those of the grazed and graze-released site (Fig. 3f). Soil δ13C was also higher in the grazed than in the graze-released site (Fig. 3f). Soil δ13C was significantly lower in the upper soil layer (0– 10 cm) than in the lower layer (10–20 cm and 20–30 cm; Fig. 3f). Pooled soil δ13C was negatively correlated with the logarithm of SOC (Fig. 4). The reference sites had lower values of δ13C than the grazed sites (Fig. 4).

Discussion

Wetlands (2014) 34:235–241

natural wetlands. DOC in the grazed site at depths of 0–10 cm, 10–20 cm and 20–30 cm layer was 50.2 %, 38.1 % and 26.8 % of that in the natural wetland, respectively. Compared with SOC and LFOC, DOC was affected by the interaction between wetlands site and soil depth (Table 1), indicated that the environmental factors, the carbon components in the soil profile do not redevelop at exactly the same rate. These are likely affected by processes related to soil water, microbial utilization and respiration (Zhang et al. 2007; Werth and Kuzyakov 2010). The labile carbon (LFOC and DOC) has been suggested as early indicators of the effects of land use on the dynamics of SOC (Gregorich et al. 1994; Zhang et al. 2007). In our study, the change of LFOC was most highly correlated with SOC (R 2 =0.92), indicating that LFOC was the most sensitive fraction for detecting the dynamics of SOC after the restoration of the wetland (Freixo et al. 2002; Swanston et al. 2002; Roscoe and Buurman 2003; Zhang et al. 2007).

Changes of Soil Carbon due to Interference Restored wetlands may regain carbon storage functions over time, but little is known about how various carbon components change during recovery from disturbance. In our study on the Zoigê plateau in China, carbon storage indicators such as the concentrations of SOC, LFOC and HFOC were significantly lower in the grazed and graze-released wetland sites than in the natural wetland site, but did not differ in the restored and the natural sites. SOC in the grazed, grazereleased and restored sites was 40.4 %, 58.7 % and 89.2 % of that in the natural wetland, respectively. In the grazed, graze-released and restored sites, LFOC was 44.7 %, 60.2 % and 90.5 %, and HFOC was 25.9 %, 56.0 % and 86.2 % of that in the natural wetland, respectively. These comparisons implied that the trajectory of SOC, LFOC and HFOC recovery in the grazed-released and restored wetlands was toward that of

The Turnover Time of SOC Based on the regression models between log SOC and soil δ13C, we found that SOC turnover was much slower in the reference site than in the degraded and restored sites (Fig. 4). This was partially attributed to permanent inundation and higher water content in the natural site of the wetland (Fig. 2). Due to permanent inundation, the diffusion of atmospheric oxygen into the saturated wetland soil was limited, leading to a low oxygen environment that suppressed microbial activities and thus greatly affected the decomposition of soil organic matter (Daulat and Clymo 1998; Chimner and Cooper 2003). In contrast, in unflooded soils, oxygen diffuses more readily and aerobic soils accelerate the decomposition of soil organic matter (Silvola et al. 1996).

Conclusions

Fig. 4 The relationship of δ13C with log [soil organic carbon (SOC)] for the four sites

In the freshwater marsh region on the Zoigê plateau in China, the degradation of wetland resulted in a decrease in SOC content and its fractions, while the restoration of wetland increased the accumulation of SOC and its fractions. SOC turnover was slower in the natural wetland than in the degraded or restored wetland. LFOC was the most sensitive fraction for the detection of changes in SOC during wetland recovery. Results of this study suggest that restoration can influence the carbon dynamics of freshwater marsh on the Zoigê plateau, and that the initial regeneration of soil carbon pools is considerably rapid after restoration. Further work is required to address the long-term rate of recovery in the wetlands of this region in relation to land-use change.

Wetlands (2014) 34:235–241 Acknowledgments The authors gratefully thank two anonymous reviewers for their critical and constructive comments. We also thank Dr. Fei-Hai Yu and Dr. Fang-Li Luo for their help on improving the manuscript. This research is supported by the Forestry Commonweal Program (200804005) and NSFC (41071329, 30700108).

References Badiou P, McDougal R, Pennock D, Clark B (2011) Greenhouse gas emissions and carbon sequestration potential in restored wetlands of the Canadian prairie pothole region. Wetlands Ecology and Management 19:237–256 Chai X, Lang HQ, Jin SR, Zu WC, Ma XH, Zhang ZY (1965) Marshes on Zoige plateau. Science Press, Beijing Chimner RA, Cooper DJ (2003) Influence of water table levels on CO2 emissions in a Colorado subalpine fen: an in situ microcosm study. Soil Biology and Biochemistry 35:345–351 Daulat W, Clymo R (1998) Effects of temperature and watertable on the efflux of methane from peatland surface cores. Atmospheric Environment 32:3207–3218 Ding W, Cai Z, Wang D (2004) Preliminary budget of methane emissions from natural wetlands in China. Atmospheric Environment 38:751–759 Euliss NH, Gleason RA, Olness A, McDougal RL, Murkin HR, Robarts RD, Bourbonniere RA, Warner BG (2006) North American prairie wetlands are important nonforested land-based carbon storage sites. Science of the Total Environment 361:179–188 Fang HJ, Yu GR, Cheng SL, Mo JM, Yan JH, Li SG (2009) 13C abundance, water-soluble and microbial biomass carbon as potential indicators of soil organic carbon dynamics in subtropical forests at different successional stages and subject to different nitrogen loads. Plant and Soil 320:243–254 Freixo AA, Machado PLOA, Dos Santos HP, Silva CA, Fadigas FS (2002) Soil organic carbon and fractions of a Rhodic Ferralsol under the influence of tillage and crop rotation systems in southern Brazil. Soil and Tillage Research 64:221–230 Ghani A, Dexter M, Perrott K (2003) Hot-water extractable carbon in soils: a sensitive measurement for determining impacts of fertilisation, grazing and cultivation. Soil Biology and Biochemistry 35:1231– 1243 Gregorich E, Monreal C, Carter M, Angers D, Ellert BH (1994) Towards a minimum data set to assess soil organic matter quality in agricultural soils. Canadian Journal of Soil Science 74:367–385 Millennium Ecosystem Assessment (2005) Ecosystems and human wellbeing: wetlands and water synthesis. Island Press, Washington Mitsch WJ, Gosselink JG (2007) Wetlands. Wiley, Hoboken National Dynamic Atlas (2012) http://www.webmap.cn/index.php? &embeded=&map=chinaadmin

241 Povin C, Lechowicz MJ, Tardif S (1990) The statistical analysis of ecophysiological response curves obtained from experiments involving repeated measures. Ecology 71:1389–1400 Roscoe R, Buurman P (2003) Tillage effects on soil organic matter in density fractions of a Cerrado Oxisol. Soil and Tillage Research 70: 107–119 Silvola J, Alm J, Ahlholm U, Nykanen H, Martikainen PJ (1996) CO2 fluxes from peat in boreal mires under varying temperature and moisture conditions. Journal of Ecology: 219–228 Stuefer JF, Huber H (1998) Differential effects of light quantity and spectral light quality on growth, morphology and development of two stoloniferous Potentilla species. Oecologia 117:1–8 Sun G, Zhang W, Zhang J (1998) The mire and peatland of the Hengduan Mountainous region. Science Press, Beijing Swanston CW, Caldwell BA, Homann PS, Ganio L, Sollins P (2002) Carbon dynamics during a long-term incubation of separate and recombined density fractions from seven forest soils. Soil Biology and Biochemistry 34:1121–1130 Tuittila ES, Komulainen VM, Vasander H, Laine J (1999) Restored cutaway peatland as a sink for atmospheric CO2. Oecologia 120:563– 574 Turetsky MR (2004) Decomposition and organic matter quality in continental peatlands: the ghost of permafrost past. Ecosystems 7:740– 750 Waddington J, McNeil P (2002) Peat oxidation in an abandoned cutover peatland. Canadian Journal of Soil Science 82:279–286 Waddington J, Price J (2000) Effect of peatland drainage, harvesting, and restoration on atmospheric water and carbon exchange. Physical Geography 21:433–451 Wang XW, Li XZ, Hu YM, Lv JJ, Sun J, Li ZM, Wu ZF (2010) Effect of temperature and moisture on soil organic carbon mineralization of predominantly permafrost peatland in the Great Hing’an Mountains, Northeastern China. Journal of Environmental Sciences 22:1057– 1066 Wang H, Wang RQ, Yu Y, Mitchell MJ, Zhang LJ (2011) Soil organic carbon of degraded wetlands treated with freshwater in the Yellow River Delta, China. Journal of Environmental Management 92: 2628–2633 Werth M, Kuzyakov Y (2010) 13 C fractionation at the rootmicroorganisms-soil interface: a review and outlook for partitioning studies. Soil Biology and Biochemistry 42:1372–1384 Yli-petäys M, Laine J, Vasander H, Tuittila ES (2007) Carbon gas exchange of a re-vegetated cut-away peatland five decades after abandonment. Boreal Environment Research 12:177–190 Yu FH, Dong M (2003) Effect of light intensity and nutrient availability on clonal growth and clonal morphology of the stoloniferous herb Halerpestes ruthenica. Acta Botanica Sinica 45:408–416 Zhang J, Song C, Wang S (2007) Dynamics of soil organic carbon and its fractions after abandonment of cultivated wetlands in northeast China. Soil and Tillage Research 96:350–360