Although semi-arid and arid regions account for about 40% of terrestrial surface of the Earth and contain approximately 10% of the global soil organic carbon stock, our understanding of soil organic carbon dynamics in these regions is limited.
A field experiment was conducted to compare soil organic carbon dynamics between a perennial grass community dominated by Cleistogenes squarrosa and an adjacent shrub community co-dominated by Reaumuria soongorica and Haloxylon ammodendron, two typical plant life forms in arid ecosystems of saline-alkaline arid regions in northwestern China during the growing season 2010. We found that both fine root biomass and necromass in two life forms varied greatly during the growing season. Annual fine root production in the perennial grasses was 45.6% significantly higher than in the shrubs, and fine root turnover rates were 2.52 and 2.17 yr−1 for the perennial grasses and the shrubs, respectively. Floor mass was significantly higher in the perennial grasses than in the shrubs due to the decomposition rate of leaf litter in the perennial grasses was 61.8% lower than in the shrubs even though no significance was detected in litterfall production. Soil microbial biomass and activity demonstrated a strong seasonal variation with larger values in May and September and minimum values in the dry month of July. Observed higher soil organic carbon stocks in the perennial grasses (1.32 Kg C m−2) than in the shrubs (1.12 Kg C m−2) might be attributed to both greater inputs of poor quality litter that is relatively resistant to decay and the lower ability of microorganism to decompose these organic matter.
Citation: Zhou Y, Pei Z, Su J, Zhang J, Zheng Y, Ni J, et al. (2012) Comparing Soil Organic Carbon Dynamics in Perennial Grasses and Shrubs in a Saline-Alkaline Arid Region, Northwestern China. PLoS ONE 7(8): e42927. https://doi.org/10.1371/journal.pone.0042927
Editor: Ben Bond-Lamberty, DOE Pacific Northwest National Laboratory, United States of America
Received: May 16, 2012; Accepted: July 16, 2012; Published: August 10, 2012
Copyright: © Zhou et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was financially supported by the National Research Program of China (973 Program, no.2009CB825103). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Soil carbon pool, as the major part of the terrestrial carbon reservoir, plays a critical, yet poorly understood role in the global carbon cycle . Consequently, the study of soil carbon dynamics is crucial to our ability to understand the global carbon cycle and its response to future global change . Semi-arid and arid regions, which are particularly sensitive to global change, account for about 40% of the terrestrial surface of Earth ,  and contain approximately 10% of the global soil organic carbon stock . A slight change of these carbon pools could have significant impacts on regional or global carbon cycle. Although semi-arid and arid regions are important source and sinks of carbon , there is no consistent conclusion on the net carbon effect of semi-arid and arid regions on carbon sequestration partially because of the unique characteristics of arid ecosystems (e.g. exceptional vulnerability, spatial heterogeneity) and the complex responses of biogeochemical cycling to a range of environmental conditions, such as water availability, precipitation pulse, solar radiation, or even heat and wind .
The difference between inputs from primary production and the return of carbon to atmosphere through decomposition of organic matter determines the soil carbon budget in semi-arid and arid ecosystems, and other terrestrial ecosystems . Fine root production contributes to about 33% of annual net primary productivity and its turnover directly impacts the biogeochemical cycle and sequestration of carbon , . Therefore, precise estimates of fine root production and turnover rate are essential for a meaningful evaluation of soil organic carbon budgets in semi-arid and arid regions. However, the vast majority of previous studies have placed more emphasis on fine root growth and distribution which were closely related to exploit soil resources, such as soil water and nutrient in semi-arid and arid regions –. Although some studies have tried to address fine root biomass and dynamics in arid regions , , few data on the contribution of fine roots to soil carbon cycle have been reported from saline-alkali arid regions.
Litterfall and its decomposition are critical steps in forming soil organic matter, mineralizing organic nutrients, thus balancing soil carbon in terrestrial ecosystems . It is well established that litter production is largely determined by soil water availability, soil fertility, species composition, plant density in arid ecosystems ,  and litter decomposition depends on a variety of factors such as litter quality, precipitation, soil biotic activity, soil resource availability and solar radiation , , , , . While a number of recent studies have described the ecological role of litterfall in soil carbon cycling for semi-arid and arid communities in different parts of the world , , , , its role has not yet been described for saline-alkali communities in central Asia.
In arid ecosystems, perennial grass and shrub are two typical plant life forms. Selective grazing leads to the reduction of plant coverage, or even induces the replacement of plant life forms from herbaceous plants, mostly perennial grasses, to long-lived woody shrubs. Different plant life forms forming different micro-environment and providing different quantity and quality of litter may play different in nutrient cycling and ecosystem functioning within the same climatic region . Generally, perennial grasses with show-root systems would be more likely to use erratic and discontinuous water sources from the upper soil layers while shrubs with deep-root systems would use more stable water resources in deeper soil layer , , . Perennial grasses and shrubs also differ in litterfall chemistry and in the quantity and timing of litterfall , , which directly affect decomposition processes, nutrient cycling and other related ecosystem processes . Further, changes in the quality and/or quantity of litter (including dead root) inputs might shift the properties of soil microorganism like microbial biomass, activity.
Several studies have separately compared the ecological roles of fine root or litterfall between perennial grasses and shrubs , , ,  but studies addressing simultaneously the effects of fine root, litterfall and their decomposition on soil organic carbon dynamics between perennial grasses and shrubs in saline-alkali arid region are relatively scarce. In this study, our overall aim is to compare soil organic carbon dynamics between adjacent perennial grasses and shrubs in a saline-alkali arid region in northwestern China. Furthermore, our specific objectives were to examine: (1) How would the dynamics of fine root mass, litterfall and floor mass differ between the perennial grasses and the shrubs? (2) Would differences in site conditions, soil microbial properties and litter chemistry affect leaf litter decomposition rate thus result in a difference in SOC pool between the perennial grasses and the shrubs?
Soil Water Content
Gravimetric soil water content (SWC) differed significantly between life forms in the growing season 2010 (P<0.05, Table 1). Integrated over all depths and averaged over the growing season, SWC in the perennial grasses (6.0%) was 5.0% lower than the shrubs (6.3%) (Fig. 1). SWC significantly increased with soil depth (P<0.001) and varied greatly among sample dates (P<0.001, Table 1).
Fine Root Biomass, Production and Turnover Rate
Integrated over all depths and averaged over the growing season, mean fine root biomass in the perennial grasses (58.4 g m−2) was 33.6% significantly higher than the shrubs (38.8 g m−2) in 2010 (P<0.001, Table 1, Fig. 2a, c). The vertical distribution of fine root biomass significantly increased with soil depths in the perennial grasses, while in the shrubs fine root biomass was highest in 10–20 cm of the soil profile in the growing season (Fig. 2a, c). Seasonal variations of fine root biomass showed different patterns between the perennial grasses and the shrubs (Fig. 2a, c).
Vertical bars indicate standard errors of means (n = 3).
Fine root necromass did not differ significantly between the perennial grasses (17.1 g m−2) and the shrubs (20.5 g m−2) in 2010 (Table 1, Fig. 2b, d), but showed significant differences between soil layers (P<0.01, Table 1, Fig. 2b, d). Significant differences among sampling dates on fine root necromass were observed in the growing season of 2010 (P<0.001, Table 1).
Fine root production was significantly differed between life forms (P<0.001). Integrated over all depths, fine root production in the perennial grasses (142.2 g m−2 yr−1) was 45.6% higher than the shrubs (77.4 g m−2 yr−1) (Fig. 3a). Fine root turnover rate did not significantly differ between the perennial grasses (2.52 yr−1) and the shrubs (2.17 yr−1) (Fig. 3b).
Aboveground Litterfall and Floor Mass
No significant differences were found between the perennial grasses (10.6 g m−2 yr−1) and the shrubs (10.9 g m−2 yr−1) in litterfall in the growing season 2010 (Table 1, Fig. 4a), whereas the effects of sampling date were statistically significant for litterfall (P<0.05, Table 1). The seasonal pattern of litterfall in the perennial grasses exhibited bimodal peaks separately in the beginning and the drought of the growing season while the shrubs showed a clear uni-modal pattern in dry season (Fig. 4a).
Litterfall mass remained (% of original) (c) in the perennial grasses and the shrubs during more than one year period. Vertical bars indicate standard errors of means (n = 3).
Mean floor mass in the perennial grasses (19.0 g m−2) was 44.2% significantly higher than the shrubs (10.6 g m−2) in the growing season 2010 (P<0.05,Table 1, Fig. 4b). The seasonal patterns of floor mass were very similar between the perennial grasses and the shrubs, with the largest floor mass in August to October, and to next year’s April (Fig. 4b).
Leaf Litter Decomposition
Leaf litter decomposition rate was significantly different between the perennial grasses and the shrubs (Table 2, Fig. 4c). The mass loss of the perennial grasses leaves (20.9% yr−1, in comparison with its original mass) was significantly lower than the shrubs leaves (54.9% yr−1) (Table 2). The mass loss rate declined with incubation time. Significant difference in mass loss rate between bags was detected from the first three months and the first ten months of the incubation period for the perennial grasses and the shrubs, respectively. The decay constant, k (mean ± SE), of the perennial grasses was 0.19 and of the shrubs was 0.84 (Table 2).
Soil Microbial Properties
Although Soil microbial biomass carbon (SMBC) did not significantly differ between two life forms, Soil microbial activity (SMA) was significantly lower in the perennial grasses than in the shrubs (P<0.05, Table 1). More specifically, SMA in the perennial grasses was 31.7%, 14.6% and 31.8% lower than in the shrubs among the three measured months (Fig. 5b). Meanwhile, in both life forms, both SMBC and SMA were significantly higher in May and September than July 2010 (P<0.001, Table 1, Fig. 5a, b). qCO2 as a measure of microbial efficiency was significantly lower in the perennial grasses than in the shrubs but showed no variation among the three measured months (P<0.05, Table 1, Fig. 5c).
Soil Organic Stocks and Inputs
Total SOC was significantly different between two life forms (P<0.001, Table 1). Integrated over all depths, mean total SOC in the perennial grasses (1.32 Kg C m−2) was 15.2% higher than in the shrubs (1.12 Kg C m−2) in 2010. SOC significantly decreased with soil depth (P<0.001, Table 1). Soil organic inputs were calculated by multiplying production (fine root production and leaf litter production) to carbon content. Total fine root carbon input in the perennial grasses (43.2 C m−2 yr−1) was 27.8% higher than the shrubs (31.2 g C m−2 yr−1). No significant difference was detected in litterfall carbon input between the perennial grasses and the shrubs.
Fine Root Dynamics
Seasonal dynamics of fine root biomass and necromass for perennial grasses and shrubs have been reported in different parts of semi-arid and arid regions , . However, no consistent patterns have been observed. In our study, fine roots biomass in both the perennial grasses and the shrubs at the top soil (0–10 cm) fleetingly increased in spring, then gradually decreased and reached the lowest at the end of August, then slightly increased in September (Fig. 2a, c). This might be attributed to low and continuous decrease of soil water content (Fig. 1) at the top soil and increase of temperature that limited soil water availability. In contrast, the patterns of fine roots biomass in the perennial grasses and the shrubs are similar at 10–20, 20–30 cm, continuously increase from April to July, the only difference is that in the perennial grasses fine roots biomass reached the peak while the shrubs sharply decreased at the end of August (Fig. 2a, c).
This difference might largely be explained by the different responses to water stress between the perennial grasses and the shrubs. It is well established that perennial grasses with shallow-root systems would use more discontinuous and erratic water sources in the upper soil layers while shrubs with deep-root systems would use more stable water in deeper soil layers , , . However, in our study, both R. soongorica and H. ammodendron belonged to other root types instead of deep root type , . Xu and Li  reported that both R. soongorica and H. ammodendron depended on water in shallow upper soil and when long time lack of effective precipitation generated extreme drought in upper soil, both R. soongorica and H. ammodendron could shrink their roots in upper soil to effectively adjust the ratio of roots absorption area to canopy assimilation area so as to maintain regular carbon uptake , .
Therefore, the observed decrease of fine root biomass in the shrubs might be attributed to the strategies of R. soongorica and H. ammodendron to water stress after a long lack of rainfall in August. Moreover, our estimates of mean fine root biomass revealed that fine root biomass of the perennial grasses was more abundant than the shrubs near the upper soil layers, implying that the former were better competitors for water in spatial dimension, particularly for pulsed rainfall.
As for fine root necromass, we observed bimodal peaks separately in the beginning and near the end of the growing season in the perennial grasses while in shrubs the necromass slightly increased in the beginning of the growing season and did not obviously change in the remaining (Fig. 2b, d). Fine root necromass in the perennial grasses slightly increased in the beginning of the growing season might be linked to senescence of some portions of the roots after the winter while the steep increase at the end of the growing seasons was followed by the peak of fine root biomass.
Fine root production and turnover are important components of carbon budget of plants and key regulators of nutrient cycles of ecosystems. However, due largely to technical limitations, accurate measurement of fine root production is hindered by estimating methods which are based on potentially tenuous assumptions and are subject to sampling errors . In our study, we chose the Decision matrix method  to estimate fine root production. Fine root production in the perennial grasses (142.2 g m−2 yr−1) was greater than in shrubs (77.4 g m−2 yr−1) (Fig. 3a). Except for the nature of species, another possibly reason for high fine root production in the perennial grasses compared to the shrubs is the relatively favorable microclimatic conditions.
Fine root turnover rate varies within and among species and across ecosystems, also estimates vary depending on the method used. Although our fine root turnover rates (2.54 and 2.17 yr−1 for the perennial grasses and the shrubs, respectively) are larger than those reviewed by Gill and Jackson  for both grassland and shrubland, they are comparable to the values reported in a semi-arid temperate steppe (2.11∼2.23 yr-1)  and in Mojave Desert shrub communities (2.33∼2.99 yr-1) . Moreover, the result that fine root turnover rate in perennial grasses was higher than the shrubs was similar to that reported by Gill and Jackson .
Soil Microbial Properties
Soil microbial biomass (SMB) represents a small but important labile pool of nutrients in soils, and its activity exerts a key controlling influence on the rate which C, N and other nutrients cycle through ecosystems . Although not significant, slightly higher SMB-C in the perennial grasses than in the shrubs in our study might be linked to higher soil organic content in the perennial grasses. Higher SOC concomitant with higher SMB-C was also found in other studies . Significantly higher SMA in the shrubs than in the perennial grasses might be attributed to higher nutrient availability in soil . Our leaf litter bag experiment suggested that the perennial grasses produced more poor quality of organic matter than the shrubs. Similar patterns of seasonal variations (with higher values in May and September than in July) of SMB-C and SMA were also observed in other studies , . Griffiths et al.  reported that water availability affects the physiological status of bacteria and can indirectly regulate substrate availability and water stress may reduce microbial biomass and activity through induced osmotic stress, resource competing and starvation , . In our study, soil water content at 0–10 cm was extreme low in July both in the perennial grasses and in the shrubs and monthly mean air temperature reached the peak of that year. These extreme drought conditions might restrain the amounts of microbial biomass and their activity. The sharp decline observed in drought season partly suggested that the microclimate conditions, especially the soil moisture, predominate over the availability of substrates in regulating seasonal variability of microbial biomass production and activity at 0–10 cm soil layer in arid ecosystems. The lower microbial efficiency (qCO2) for the perennial grasses might be ascribed to lower quality of organic matter inputs derived from perennial grasses .
Leaf Litter Dynamics
Previous studies have indicated that litterfall could show no distinct seasonal pattern, uni-modal, bimodal, or even multi-modal peaks , . In our study, litterfall of the perennial grasses exhibited bimodal peaks while the shrubs showed a clear uni-modal pattern. Seasonal patterns of litterfall are largely attributed to inherent nature of the species and associated factors responsible for senescence and abscission. The first peak of litterfall production in the perennial grasses could be linked to the abscission of older generations of dead leaves (standing litter) when new leaves gradually appeared in the beginning of the growing season. Moreover, dry seasonal litterfall peaks observed in our study have been reported in several litterfall studies , , , . In arid ecosystems, water stress which is usually extreme in the dry season might be an important precipitating factor influencing dry season leaf fall . Moore  reported that water stress could initiate the synthesis of abscisic acid, which can stimulate the senescence of leaves and other plant parts. Zhou et al.  reported that air temperature (especially maximum and effective temperature) was an important environmental variable in affecting litterfall production. In our study, from July to September, rare rainfall concomitant with high air temperature induced a peak of litterfall production in the perennial grasses and the shrubs.
Annual litterfall production values were similar between the perennial grasses (10.6 m−2 yr−1) and the shrubs (10.9 m−2 yr−1) in our study, which were much lower than those reported in other semi-arid and arid ecosystems , , , ,  with different species, canopy development and coverage, soil and weather conditions. The low number of samples in our study might underestimate the annual litterfall production. However, we do not expect these biases will change our conclusion on the comparison of litterfall production between the perennial grasses and the shrubs. Nonetheless, our litterfall production was similar to the value of 11.6 g m−2 yr−1 reported by Weatherly et al.  in a Mojave Desert ecosystem. Meanwhile, floor mass on the soil surface serves as temporary sink for nutrients which are released gradually through decomposition, thereby providing a permanent source of nutrients to the soil . In the present study, the perennial grasses had more floor mass than the shrubs, possibly because the former had a lower decomposition rate.
Although lack of directly comparable data of leaf litter decomposition rates for C.squarrosa, R. soongorica and H. ammodendron, the percentage of 1-year weight loss from the perennial grasses (20.9%) in our study was comparable to the rate for a perennial grass community in the Patagonian steppe (31% after 20 months) . And annual decomposition rates of the shrubs (54.9%), expressed in percentage of original mass, also fell within the range rates reported for shrubs in other arid regions (35%–70% for a Northern American desert community ; 40% for a Chihuahuan Desert community ; 35%–40% for a Mojave Desert community ). Rapid initial decomposition and slow latter decomposition (Fig. 4c) observed in our study could be likely linked to rapid loss of labile compounds in leaf litter. Meanwhile, higher leaf litter decomposition rate for shrubs can be ascribed to the following reasons. First, Cornwell et al.  reviewed that plant species traits are predominant control on litter decomposition rates. Globally, herbaceous species in general did not produce litter that decomposed faster than woody species and eudicot litter decomposed on average 1.6 times faster than monocot litter . Correspondingly, in our study, Leaf litter decomposition rates were significantly higher in the shrubs (both R. soongorica and H. ammodendron are dicotyledon) than in the perennial grasses (C.squarrosa is monocotyledon). Besides, litter chemistry, particularly the concentration of nutrient and carbon compounds, is a major control on litter decomposition . C: N ratio has been identified as key features determining the rate of litter decomposition, and litter with high C: N ratio decompose at relatively low rates , . In the present study, we also detected that the C: N ratio of the perennial grasses (47.8) was significantly higher than the shrubs (17.2). In addition, higher soil microbial activity in the shrubs observed in our study indirectly suggested that higher ability of microorganism to decompose leaf litter from the shrubs.
Soil Organic Carbon Stocks and Inputs
Soil organic carbon storage, which represents the accumulated difference between inputs from primary production and outputs through decomposition, is critical to estimate effective carbon sequestration capacity of ecosystems. In our study, the observed total soil organic carbon stocks (1.32 and 1.12 Kg m−2 for the perennial grasses and the shrubs, respectively) are within the range reported by Feng et al.  for 17 locations in different types of desert in China (0.02–4.97 kg m−2 for 0–20 cm profile). Total soil carbon stocks were higher in the perennial grasses than in the shrubs. This accumulative discrepancy might ascribe to the following two main reasons. First, soil organic carbon inputs through primary production (here is the sum of fine root production and litterfall) in the perennial grasses were higher than in the shrubs. Second, our litter bag experiment and measured soil microbial properties suggested that inputs from the perennial grasses are in poor quality and relatively resistant to decay by lower ability of microorganism. Therefore, our results suggested that the perennial grasses might accumulate more soil organic carbon with time than the shrubs because larger amounts of inputs from litter and slower return of carbon through decomposition in the perennial grasses than in the shrubs.
Our results show that there are obvious differences in fine root biomass, production, turnover, floor mass, leaf litter decomposition and soil microbial properties between the perennial grasses and the shrubs. The absolute size of soil organic carbon stocks increased with time in the perennial grasses is likely due to greater inputs from litter and slower outputs through decomposition. Our findings provide sound and basic knowledge in estimating the effectiveness of soil organic carbon dynamics and sequestration in the two type of life forms in saline-alkali arid regions of China and might be useful to address the eventual effects of plant species shifting on ecosystem functioning. Nonetheless, the low number of samples taken in our study might not accurately reflect the differences between the perennial grasses and the shrubs in soil carbon dynamics due to the high heterogeneity in arid ecosystems. Also, our study only conducted one year, and annual rainfall during that year was higher than the long-term mean values. Ecological processes in water-limited ecosystems might vary sharply with water availability. These remarks should be taken into consideration in the future studies.
Materials and Methods
Fukang Desert Ecosystem Research Station is a department of Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences. This study was approved by State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, the Chinese Academy of Sciences and Fukang Desert Ecosystem Research Station.
This study was conducted during the growing season of 2010 in the vicinity of the Fukang Desert Ecosystem Research Station (44°17′N, 88°56′E) of the Chinese Academy of Sciences, which is located at a transitional area of oasis-desert and the southern edge of the Gurbantonggut Desert. This region has a typical continental arid temperate climate, with a long-term mean annual precipitation <150 mm but a potential mean annual evaporation >2000 mm, and has a long cold winter and hot summer, with a long-term mean annual temperature of 4–6°C. Soils at this region are characterized as Yermi-sodi-Haplic Luvisol . Physical properties of the soils in the perennial grasses and the shrubs sites in layers of 0–10, 10–20 and 20–30 cm are shown in Table 3.
We established three plots (each 25 m×25 m) for each plant life form for field measurements during the growing season of 2010. The perennial grasses community is dominated by Cleistogenes squarrosa which is widely distributed in the Eurasian steppe zone . C. squarrosa is a perennial bunchgrass with characteristic C4 anatomy and a fibrous root system , and is also considered as a key species for sustainable grassland development . The shrubs community is co-dominated by Reaumuria soongorica and Haloxylon ammodendron. R. soongorica, an extreme xeric semishrub of Tamaricaceae, is the constructive and dominant species of the desert ecosystems in the Central Asia, and forms the vast and distinctive landscape of the salt desert . H. ammodendron, a shrubby perennial of Chenopodiaceae with characteristic C4 anatomy, is an important component of old Mediterranean flora . The deserts communities which dominate or co-dominate by H. ammodendron are the most widely distributed vegetation types of Asian deserts . All the plots are relatively flat with a slope of <5° and an elevation of about 450 m above sea level. In the perennial grasses sites, mean density, height and ground cover are about 60000 clusters ha−1, 0.14 m and 50%, respectively. In the shrubs sites, mean density, height and ground cover of R. soongorica are about 2500 stems ha−1, 0.34 m and 23%, respectively. For H. ammodendron they are about 500 stems ha−1, 1.07 m and 10%, respectively.
Soil Sampling and Soil Properties
Soil samples (0–10, 10–20, 20–30 cm) for soil organic carbon, soil water content, and fine root mass were randomly collected from five cores (8 cm diameter) in each plot per month. Besides, soil samples (0–10 cm) for analyzing microbial properties were done during May, July and September. Fresh samples for determination of soil microbial properties were immediately transported to the laboratory with a portable ice box and stored at 4°C before analysis. In the laboratory, these soil samples for analyzing soil organic carbon, soil physical properties, and soil microbial properties were passed through a 2-mm sieve and manually cleaned off any visible plant tissues.
We calculated soil carbon stock (g m−2) for each soil sample at different depths by measuring bulk density (g m−3) and mass-based soil carbon content (%). Bulk density of soil was determined by weighing air-dried volume samples. Soil carbon content was measured by the dichromate oxidation method . Briefly, finely grinded 0.2 g soil was digested with 5 ml 2 M K2Cr2O7 and 5 ml of concentrated H2SO4 at 170°C for 10 min, then titrated the digests with 2 M standardized FeSO4.
Gravimetric soil water content (SWC) was measured by oven-drying samples at 105°C for 24 h. Soil pH was measured in the 1∶5 soil/water extract, electrical conductivity in the saturated paste extract (ECsp). The particle size distribution of the <2 mm particle fractions was determined using the laser detection technique.
Fine Root Mass, Production and Turnover Rate
Fine roots (<2 mm) were manually washed free of soil samples and separated into live and dead root fragments by their color, resilience, consistency and the degree of cohesion between the cortex and the periderm. The separated fine roots were oven-dried to constant mass at 65°C for determination of mass and chemical analysis. Fine root production was estimated using the Decision matrix method . Fine root turnover rate was calculated as the ratio of the total amount of fine root production over the growing season to the mean standing biomass of fine roots . Mean fine root biomass was estimated as the average of fine root biomass on April, May, June, July, August and September 2010.
Litterfall and Floor Mass
Litterfall samples in each plot were collected monthly from April to October 2010 using three randomly located litter traps. Each trap (50 cm×50 cm) was made up of 1 mm mesh opening nylon netting suspended from a PVC tube square and held 5 cm above the ground by four PVC tubes. Floor mass was measured on five 50 cm×50 cm subplots randomly located in each plot every month. All plant litter samples were washed and oven-dried to constant mass at 65°C for determination of mass and chemical analysis.
Decomposition of Leaf Litter
Decomposition rate of leaf litter was determined using the litterbag method . Recently fallen leaf litter was collected from each life form. Three grams of air-dried leaf litter was placed in a 1 mm mesh opening nylon bag with a dimension of 10 cm×15 cm. We placed a total of 264 bags in six plots in June 2009. Four litterbags were collected each time at each plot after1, 2, 3, 4, 10, 11, 12, 13, 14, 15, 16 months. Collected leaf litter was oven-dried to a constant mass at 65°C. We used the following exponential function: Yt = Y0×e−kt  to determine the decay constant (k) and the average rate of leaf litter loss.
At the onset of the decomposition experiments, we also determined total C and N of leaf litter. Total C was determined by the standard method of wet-combustion, and total N by semi-micro Kjeldahl method . The C and N contents, and the C:N ratios of the leaf litter in perennial grasses and shrubs are 43.0%, 0.9%, and 47.8, and 34.5%, 2.0% and 17.2, respectively.
Soil Microbial Properties
Soil microbial biomass carbon (SMBC) was estimated using the fumigation-extraction method . Briefly, the fresh soil samples after adjusting to approximate 60% of water holding capacity were incubated for one week in dark at 25°C. Then 20 g (dry weight equivalent) of fumigated with CH3Cl for 24 hours and non-fumigated soil samples were extracted with 0.5 M K2SO4. The extracts were filtered through 0.45-µm filters and analyzed for extractable organic carbon by dichromate digestion as described by Lovell et al. . SMBC was calculated as the difference in extractable organic carbon content between the fumigated and non-fumigated soil samples using a conversion factor (Kec) of 0.38 to correct the incomplete extractability .
Soil microbial activity (SMA), i.e. soil microbial respiration, was measured by determining CO2 evolution over a 2-week incubation period. Briefly, 20 g (dry weight equivalent) of soil sample after adjusting to approximate 60% water holding capacity was incubated for 2-week at 25°C. Respired CO2 was captured in 5.0 ml of 0.5 M NaOH contained in a beaker suspended inside a Mason jar . The NaOH solution was immediately titrated to determine the amount of CO2 evolved, and SMA was expressed as mg CO2 kg−1 d−1.
The metabolic quotient (qCO2: soil microbial respiration to microbial biomass carbon, mg CO2-C g−1 Cmic h−1), which representing the microbial respiration per biomass unit, was calculated according to Wardle and Ghani .
Seasonal mean values used in this study were calculated from the monthly mean values, which were first averaged from all measurements in the same month. Repeated Measures ANOVA (RMANOVA) were used to examine life form and soil layer effects on soil water content, soil organic content, fine root biomass, necromass, and also life form effects on aboveground litterfall, floor mass, soil microbial biomass carbon, soil microbial activity and qCO2 over the growing season in 2010. Two-way ANOVA was used to examine life form, soil layer and their interaction on fine root production and fine root turnover rate. Significant differences among means of measurements were determined by t test at p = 0.05. All the statistical analyses were performed with SPSS 11.5 software (SPSS, Chicago, IL, USA).
We gratefully acknowledge the State Key Laboratory of Vegetation and Environmental Change, Institute of Botany, the Chinese Academy of Sciences and Fukang Desert Ecosystem Research Station for help with logistics and access permission to the study site.
Conceived and designed the experiments: CX RW. Performed the experiments: ZP Y. Zhou JS JZ. Analyzed the data: Y. Zhou ZP. Contributed reagents/materials/analysis tools: RW Y. Zheng. Wrote the paper: Y. Zhou CX JN.
- 1. Davidson EA, Janssens IA (2006) Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature 440: 165–173.
- 2. Davidson EA, Trumbore SE, Amundson R (2000) Biogeochemistry: soil warming and organic carbon content. Nature 408: 789–790.
- 3. Austin AT, Vivanco L (2006) Plant litter decomposition in a semi-arid ecosystem controlled by photodegradation. Nature 442: 555–558.
- 4. Stone R (2008) ECOSYSTEMS: Have Desert Researchers Discovered a Hidden Loop in the Carbon Cycle? Science 320: 1409–1410.
- 5. White DA, Welty-Bernard A, Rasmussen C, Schwartz E (2009) Vegetation controls on soil organic carbon dynamics in an arid, hyperthermic ecosystem. Geoderma 150: 214–223.
- 6. Austin AT (2011) Has water limited our imagination for aridland biogeochemistry? Trends in Ecology & Evolution 26: 229–235.
- 7. Jackson RB, Mooney HA, Schulze ED (1997) A global budget for fine root biomass, surface area, and nutrient contents. Proceedings of the National Academy of Sciences 94: 7362–7366.
- 8. Gill RA, Jackson RB (2000) Global patterns of root turnover for terrestrial ecosystems. New Phytologist 147: 13–31.
- 9. Hipondoka MHT, Aranibar JN, Chirara C, Lihavhad M, Mackob SA (2003) Vertical distribution of grass and tree roots in arid ecosystems of Southern Africa: niche differentiation or competition? Journal of Arid Environments 54: 319–325.
- 10. Wilcox C, Ferguson JW, Fernandez GCJ, Nowak RS (2004) Fine root growth dynamics of four Mojave Desert shrubs as related to soil moisture and microsite. Journal of Arid Environments 56: 129–148.
- 11. Rodríguez MV, Bertiller MB, Bisigato A (2007) Are fine roots of both shrubs and perennial grasses able to occupy the upper soil layer? A case study in the arid Patagonian Monte with non-seasonal precipitation. Plant and Soil 300: 281–288.
- 12. Jha P, Mohapatra KP (2009) Leaf litterfall, fine root production and turnover in four major tree species of the semi-arid region of India. Plant and Soil 326: 481–491.
- 13. Austin AT, Ballaré CL (2010) Dual role of lignin in plant litter decomposition in terrestrial ecosystems. Proceedings of the National Academy of Sciences 107: 4618–4622.
- 14. Pavón NP, Briones O, Flores-Rivas J (2005) Litterfall production and nitrogen content in an intertropical semi-arid Mexican scrub. Journal of Arid Environments 60: 1–13.
- 15. Mlambo D, Nyathi P (2008) Litterfall and nutrient return in a semi-arid southern African savanna woodland dominated by Colophospermum mopane. Plant Ecology 196: 101–110.
- 16. Aerts R (1997) Climate, leaf litter chemistry and leaf litter decomposition in terrestrial ecosystems: a triangular relationship. Oikos 79: 439–449.
- 17. Yahdjian L, Sala OE, Austin AT (2006) Differential Controls of Water Input on Litter Decomposition and Nitrogen Dynamics in the Patagonian Steppe. Ecosystems 9: 128–141.
- 18. Maya Y, Arriaga L (1996) Litterfall and phenological patterns of the dominant overstorey species of a desert scrub community in north-western Mexico. Journal of Arid Environments 34: 23–35.
- 19. Alvarez JA, Villagra PE, Rossi BE, Cesca EM (2009) Spatial and temporal litterfall heterogeneity generated by woody species in the Central Monte desert. Plant Ecology 205: 295–303.
- 20. Vargas DN, Bertiller MB, Ares JO, Carrera AL, Sain CL (2006) Soil C and N dynamics induced by leaf-litter decomposition of shrubs and perennial grasses of the Patagonian Monte. Soil Biology and Biochemistry 38: 2401–2410.
- 21. Sala O, Golluscio R, Lauenroth W, Soriano A (1989) Resource partitioning between shrubs and grasses in the Patagonian steppe. Oecologia 81: 501–505.
- 22. Canadell J, Jackson R, Ehleringer J, Mooney HA, Sala O, et al. (1996) Maximum rooting depth of vegetation types at the global scale. Oecologia 108: 583–595.
- 23. Campanella M, Bertiller MB (2008) Plant phenology, leaf traits and leaf litterfall of contrasting life forms in the arid Patagonian Monte, Argentina. Journal of Vegetation Science 19: 75–85.
- 24. Chapin III FS, Zavaleta ES, Eviner VT, Naylor RL, Vitousek PM, et al. (2000) Consequences of changing biodiversity. Nature 405: 234–242.
- 25. Bai W, Wan S, Niu S, Liu W, Chen Q, et al. (2010) Increased temperature and precipitation interact to affect root production, mortality, and turnover in a temperate steppe: implications for ecosystem C cycling. Global change biology 16: 1306–1316.
- 26. Xu H, Li Y (2005) Water use strategies and corresponding leaf physiological performance of three desert shrubs. Acta Botanica Boreali-Occidentalia Sinica 25: 1309–1316.
- 27. Xu H, Li Y, Zou T, Xie JX, Jiang LX (2007) Ecophysiological response and morphological adjustment of Haloxylon ammodendron towards variation in summer precipitation. Acta Ecologica Sinica 27: 5019–5028.
- 28. Hendricks JJ, Hendrick RL, Wilson CA, Mitchell RJ, Pecot SD, et al. (2006) Assessing the patterns and controls of fine root dynamics: an empirical test and methodological review. Journal of Ecology 94: 40–57.
- 29. Fairley RI, Alexander IJ (1985) Methods of calculating fine root production in forests. In: Fitter AH, Atkinson D, Read DJ, Usher MB, editors. Ecological interactions in soil. Boston: Blackwell Scientific. 37–42.
- 30. Phillips DL, Johnson MG, Tingey DT, Catricala CE, Hoyman TL, et al. (2006) Effects of elevated CO2 on fine root dynamics in a Mojave Desert community: a FACE study. Global change biology 12: 61–73.
- 31. Lovell RD, Jarvis SC, Bardgett RD (1995) Soil microbial biomass and activity in long-term grassland: effects of management changes. Soil Biology and Biochemistry 27: 969–975.
- 32. Liao JD, Boutton TW (2008) Soil microbial biomass response to woody plant invasion of grassland. Soil Biology and Biochemistry 40: 1207–1216.
- 33. Tu C, Ristaino JB, Hu S (2006) Soil microbial biomass and activity in organic tomato farming systems: Effects of organic inputs and straw mulching. Soil Biology and Biochemistry 38: 247–255.
- 34. Kinsbursky RS, Steinberger Y (1989) Root and microbial biomass dynamics under the canopy of the desert shrub Zygophyllum dumosum. Oecologia 80: 498–500.
- 35. Holmes WEZ, Zak DR (1994) Soil microbial biomass dynamics and net nitrogen mineralization in northern hardwood ecosystems. Soil Science Society of America Journal 58: 238.
- 36. Griffiths RI, Whiteley AS, O’Donnell AG, Bailey MJ (2003) Physiological and community responses of established grassland bacterial populations to water stress. Applied and Environmental Microbiology 69: 6961–6968.
- 37. Chuyong GB, Newbery DM, Songwe NC (2000) Litter nutrients and retranslocation in a central African rain forest dominated by ectomycorrhizal trees. New Phytologist 148: 493–510.
- 38. Descheemaeker K, Muys B, Nyssen J, Poesen J, Raes D, et al. (2006) Litter production and organic matter accumulation in exclosures of the Tigray highlands, Ethiopia. Forest Ecology and Management 233: 21–35.
- 39. Moore TC (1979) Biochemistry and physiology of plant hormones: Springer-Verlag New York.
- 40. Zhou G, Guan L, Wei X, Zhang D, Zhang Q, et al. (2007) Litterfall production along successional and altitudinal gradients of subtropical monsoon evergreen broadleaved forests in Guangdong, China. Plant Ecology 188: 77–89.
- 41. Weatherly HE, Zitzer SF, Coleman JS, Arnone III JA (2003) In situ litter decomposition and litter quality in a Mojave Desert ecosystem: effects of elevated atmospheric CO2 and interannual climate variability. Global change biology 9: 1223–1233.
- 42. Schaefer D, Steinberger Y, Whitford WG (1985) The failure of nitrogen and lignin control of decomposition in a North American desert. Oecologia 65: 382–386.
- 43. Whitford WG, Steinberger Y, MacKay W, Parker LW, Freckman D, et al. (1986) Rainfall and decomposition in the Chihuahuan Desert. Oecologia 68: 512–515.
- 44. Cornwell WK, Cornelissen JHC, Amatangelo K, Dorrepaal E, Eviner VT, et al. (2008) Plant species traits are the predominant control on litter decomposition rates within biomes worldwide. Ecology Letters 11: 1065–1071.
- 45. Taylor BR, Parkinson D, Parsons WFJ (1989) Nitrogen and lignin content as predictors of litter decay rates: a microcosm test. Ecology 70: 97–104.
- 46. Feng Q, Endo KN, Guodong C (2002) Soil carbon in desertified land in relation to site characteristics. Geoderma 106: 21–43.
- 47. FAO-UNESCO (1988) Soil Map of the World, Revised Legend. Rome.
- 48. Wang SP, Wang YF, Chen ZZ (2003) Effect of climate change and grazing on populations of Cleistogenes squarrosa in Inner Mongolia steppe. Acta Phytoecologica Sinica 27: 337–343.
- 49. Liang C, Michalk DL, Millar GD (2002) The ecology and growth patterns of Cleistogenes species in degraded grasslands of eastern Inner Mongolia, China. Journal of applied ecology 39: 584–594.
- 50. Liu S, Liu XY (1996) The Estimating Model of Upper Plant Weight on Reaumuria soongorica Semishrub [J]. Arid Zone Research 13: 36–41.
- 51. Sheng JH, Qiao YH, liu HY, Zhai ZX, Guo YH (2004) A study on the root system of Haloxylon ammodendron (C.A. Mey) Bunge. Acta Agrestia Sinica 12: 91–94.
- 52. Wu ZY (1995) Vegetation of China. Beijing: Academic Press. 1382 p.
- 53. Aber JD, Melillo JM, Nadelhoffer KJ, McClaugherty CA, Pastor J (1985) Fine root turnover in forest ecosystems in relation to quantity and form of nitrogen availability: a comparison of two methods. Oecologia 66: 317–321.
- 54. Berg B, Berg MP, Bottner P, Box E, Breymeyer A, et al. (1993) Litter mass loss rates in pine forests of Europe and Eastern United States: some relationships with climate and litter quality. Biogeochemistry 20: 127–159.
- 55. Olson JS (1963) Energy storage and the balance of producers and decomposers in ecological systems. Ecology 44: 322–331.
- 56. Bao SD (1999) Agriculture Soil Chemical Analysis. Beijing: Science Press. 263–271 p.
- 57. Vance ED, Brookes P, Jenkinson DS (1987) An extraction method for measuring soil microbial biomass C. Soil Biology and Biochemistry. 19: 703–707.
- 58. Hu S, van Bruggen A (1997) Microbial dynamics associated with multiphasic decomposition of 14 C-labeled cellulose in soil. Microbial ecology 33: 134–143.
- 59. Wardle DA, Ghani A (1995) A critique of the microbial metabolic quotient (qCO2) as a bioindicator of disturbance and ecosystem development. Soil Biology and Biochemistry 27: 1601–1610.