Winter soil CO2efflux in two contrasting forest ecosystems on the eastern Tibetan Plateau,China

Zhenfeng Xu•Feifei Zhou•Huajun Yin•Qing Liu

Winter soil CO2efflux in two contrasting forest ecosystems on the eastern Tibetan Plateau,China

Zhenfeng Xu1,2•Feifei Zhou2•Huajun Yin2•Qing Liu2

Significant CO2fluxes from snow-covered soils occur in cold biomes.However,little is known aboutwinter soilrespiration on the eastern Tibetan Plateau of China.We therefore measured winter soil CO2fluxes and estimated annual soil respiration in two contrasting coniferous forest ecosystems(a Picea asperata plantation and a naturalforest). Mean winter soil CO2effluxes were 1.08μmolm-2s-1in the plantation and 1.16μmol m-2s-1in the natural forest. These values are higher than most reported winter soil CO2efflux values for temperate or boreal forest ecosystems. Winter soil respiration rates were similar for our two forest ecosystems butmean soil CO2efflux over the growing season washigherin the naturalforestthan in the plantation.The estimated winter and annual soil effluxes for the natural forestwere 176.3 and 1070.3 g m-2,respectively,based on the relationship between soilrespiration and soiltemperature, which were 17.2 and 9.7%greater than their counterparts in the plantation.The contributions of winter soilrespiration toannualsoilefflux were 15.4%forthe plantation and 16.5% for the natural forest and were statistically similar.Our results indicate thatwinter soil CO2efflux from frozen soils in the alpine coniferous forest ecosystems of the eastern Tibetan Plateau was considerable and was an important component of annual soil respiration.Moreover,reforestation(natural coniferous forests were deforested and reforested with P.asperata plantation)may reduce soilrespiration by reducing soil carbon substrate availability and input.

Winter soil CO2efflux·Eastern Tibetan Plateau·Natural coniferous forest·Picea asperata plantation

Introduction

Soil respiration is the second largest carbon flux in terrestrial ecosystems,and plays a critical role in global carbon cycling(Wang et al.2006).Forest ecosystems constitute a major reservoir of global soil carbon(Li et al. 2005).Soil CO2flux typically account for 30–80%of annual total ecosystem respiration in forests(Davidson etal.2006).Although considerable research has focused on quantifying these fluxes during the growing season,comparatively little efforthas focused on fluxes during the nongrowing season(Hubbard et al.2005;Wang et al.2010). This is partly because forestecosystems athigh altitudes or latitudes were once thoughtdormantduring the long winter period(Winston et al.1997).But a growing number of studies have convincingly demonstrated that winter soil respiration is a pronounced component of the annual carbon budget in these ecosystems(Mariko et al.2000; Hubbard et al.2005;Mo et al.2005;Monson et al.2006; Suzuki et al.2006;Schindlbacher et al.2007;Wang et al.2010).Moreover,Brooks et al.(2004)concluded that annual carbon sequestration is overestimated substantially when winter CO2fluxes are not included.Therefore, understanding carbon cycling in forest ecosystems during the winter period is criticalfor estimating the globalcarbon budget.

Snow covers approximately 50%of the Northern Hemisphere’s land area over the winter period(Armstrong and Brodzik 2001).Compared with boreal forest ecosystems,alpine forest ecosystems on the eastern Tibetan Plateau often have shorter winters during which snow cover generally continues for 4–5 months.Moreover,air and soil temperatures remain relatively high during the winter period.Thus,winter soil CO2fluxes in this region may exhibit patterns and magnitudes different from those of boreal forest ecosystems.However,up to now,winter soilrespiration and its contribution to annualsoil CO2flux is scarcely studied in these forest ecosystems.

Land-use changes make large contributions to CO2emissions(IPCC 2007).Land-use changes also affect soil respiration by altering environmental factors,litter inputs and soil substrates(Sheng et al.2010).Many recent studies have addressed the effects of land-use changes on soil CO2efflux.However,these studies have not shown consistent effects,with positive,negative,and neutral effects being reported(Ishizuka et al.2002;Gru¨nzweig et al.2003; Campos 2006;Liu etal.2008a;Sheng etal.2010).Due to huge demand for timber,many naturalconiferous forests of southwestern China have been deforested and subsequently reforested with Picea asperata overthe lastseveraldecades. Currently,there are vast areas of P.asperata plantations in southwestern China,accounting for a greatproportion of all forested land in this region(Liu etal.2002).To date,very few studies on soil CO2efflux response to reforestation are available on the eastern Tibetan Plateau.

The sub-alpine and alpine forest ecosystems in the Eastern Tibetan Plateau are located at the transition zone between the Qinghai-Tibet plateau and the Sichuan basin, and play important roles in regional carbon balance(Wang et al.2003).Winter CO2fluxes from alpine forest soils could be more pronounced in this region than in other ecosystems.Reforestation may affect soil CO2efflux by altering environmental factors,litter inputs and soil properties(Jiang et al.2009).Understanding responses of soil CO2efflux to land-use changes could enable more accurate estimation of the global carbon budget.In this study,we measured soil CO2effluxes in two contrasting forest ecosystems(P.asperata plantation versus spruce-fir dominated natural forest)in the eastern Tibetan Plateau from November 2007 to October 2008.Our major objectives were:(1)to determine the magnitude of winter soil respiration and its contribution to total annual soil flux for two forestecosystems;(2)to examine the seasonalvariation of soil respiration rate and its correlation with soil temperature and moisture;and(3)to explore the effect of reforestation on soil respiration.

Materials and methods

Site description

The study was conducted on two sites thatwere separated by ca.300 m(Fig.1).One site was in a P.asperata plantation(70-year-old)and the other was in a sprucedominated natural forest(200-year-old).There were some mosses and grasses(e.g.,Carex capilliformis,Deyeuxia arundinacea,Festuca ovina)growing under the plantation. Conversely,the understory of the natural forest was dominated by mosses,woody trees(e.g.,Betula albo-sinensis, Acer mono,Lonicera spp.)and grasses(e.g.,Carex capilliformis,Anemone rivularis).Both experimentalsites were located at the Miyaluo Experimental Forest of Lixian County,eastern Tibetan Plateau(31°35′N;102°35′E; 3150 m a.s.l.).This spruce plantation was planted in the 1940s.According to the local Forestry Bureau,no management(e.g.,fertilizer,irrigation)was practiced in the plantation.The localclimate is a montane monsoon thatis humid and rainy in the summer but cold and dry in the winter.The mean annualairtemperature is 6–10°C,with a relatively short,cold winter(December–March).The soilis frozen approximately from late November to early April, with a maximum snow depth of ca.30 cm(Qin and Liu 2010).Annual precipitation ranges from 600 to 1100 mm. Soils atthe two sites were classified in the mountain brown soilseries.The basic soilproperties as determined in July 2008 were as follows:organic C(44.82±3.25)g kg-1(mean value±standard deviation),total N (2.95±0.34)g kg-1,total P(0.65±0.03)g kg-1,total K(14.16±0.39)g kg-1and pH 6.19±0.47 for the plantation;organic C(145.02±14.87)g kg-1,total N (9.56±1.19)g kg-1,total P(0.67±0.07)g kg-1,total K(11.31±0.51)g kg-1and pH 5.85±0.65 for the naturalforest.

Soil CO2efflux measurements

Three independent plots(20×30 m)were selected as three replicates foreach foresttype(P.asperata plantation and natural forest).Nine subsamples(i.e.soil respiration collars)were distributed randomly in each plot.We measured soil CO2efflux with a Li-Cor 6400 portable CO2infrared gas analyzer(Li-Cor Inc.,Lincoln,NE,USA)to monitor the change in CO2concentration over time in flux chambers.A Li-Cor 6400 portable CO2infrared gas analyzer(Li-Cor Inc.,Lincoln,NE,USA)was used to measuresoil respiration between 10:00 AM and 12:00 AM(local time)from October 2007 to October 2008.Living plants inside the soil collars were clipped at the soil surface to avoid leaf respiration.Over the growing season,nine polyvinylchloride collars(10.2 cm diameter,5 cm height) were inserted 2 cm into the soilin each plot.During winter, longer soil respiration collars(determined by snow depth) were inserted into the soil surface and stabilized for 24 h before measurement of the winter soil CO2efflux(Wang et al.2010).Fluxes were measured approximately every other week from November 2007 to October 2008.Soil CO2efflux in each chamberwas measured continuously for three cycles,and the three measurements were averaged to yield a mean soil flux.

Fig.1 Location of the study site in Miyaluo area of western Sichuan,China

Soil temperature and water content measurements

Soiltemperature(5 cm below the soilsurface)was measured both discretely and continuously.Discrete soil temperature measurementswere concurrently made with measurements at 5 cm depth near each collar with a digital thermometer. Continuous soil temperature was monitored in each forest type at 60 min intervals from November 2007 to October 2008 using DS1921G Thermochron iButton dataloggers (DS1921G-F5 Maxim Integrated Products,Dallas Semiconductor inc.,Sunnyvale,California).Soil samples near the collars were sampled from 0 to 5 cm depths,and dried at 70°Cto a constantmassforsoilwatercontentdetermination.

Estimate of annual and winter soilrespiration

Based on the measured data,an exponential function was formulated to describe the relationship between soil respiration and soil temperature:R=a ebT,where R is soil respiration,T is soil temperature,coefficient a is the intercept of soil respiration when temperature is zero,and coefficient b represents the temperature sensitivity of soil respiration.

Based on the continuous soiltemperature at5 cm depth, soil CO2efflux throughout one year was estimated by integrating CO2effluxes for the period from 1 November 2007 to 31 October 2008 using the observed ecosystemspecific response equation between soilrespiration and soil temperature.Annual and winter soil respiration for each ecosystem was estimated by interpolating measured soil respiration between sampling dates for every day of the year and then computing the sum to obtain the annual or winter values(Wang et al.2010).

Statisticalanalyses

Each plot was an experimental unit,thus replicate data were averaged by plots for analysis.Before analysis,all data were tested for normal distribution and homogeneity. If data were heterogeneous,they were ln-transformed to improve homogeneity.Linearregression was used to assess the relationship between soil respiration and soil water content.Exponential regression was used to examine the correlation between soil temperature and soil respiration rate.Student’s t tests were used to compare the differences of soil temperature,soil water content and soil CO2flux between two contrasting forest types.Statistical tests were considered significant at the p＜0.05 level.All statistical analyses were performed using SPSS software(SPSS 13.0 for windows,SPSS Ins.,Chicago,IL,USA).

Results

Soil temperature and water content

Curves for soil temperature at 5 cm depth were similar in the two forest types with a minimum in winter and a maximum in summer.Soiltemperatures were similar in the two foresttypes.The daily mean soil temperatures at5 cm depth were below 0°C during winter except in early December(Fig.2).Soil temperatures also fell below 0°C from 15 to 24 November 2007 due to a protracted unseasonally cold period(Fig.2).

Soil water content at 5 cm depth exhibited a similar seasonality for two forest types,responding to changes in precipitation(Fig.3).The annual mean soil water content (gravimetrically)was 20.4±2.8%in the plantation and 24.5±3.2%in the naturalforest;the latter 20.0%greater than the former(Fig.3).Irrespective of forest type,soil moisture remained relatively stable during winter(Fig.3). Moreover,there were no clear seasonal changes in soil moisture content in either forest ecosystem(Fig.3).

Fig.2 Seasonal changes in soil temperature 5 cm below the soil surface in two forest ecosystems

Fig.3 Monthly variations in soilwater contentat5 cm depth in two forest ecosystems

Soil respiration rate related to soil temperature and water content

Soil temperature was the main determinant of seasonal dynamics in soil respiration for both forests(Fig.4; Table 1).The temporalchanges in soilCO2efflux followed the clear seasonal pattern of soil temperature for each ecosystem,being high during summer and low in winter (Figs.2,4).Soil respiration was correlated strongly with soil temperature at 5 cm depth in both forest types (Table 1).CO2effluxes during winter were relatively stable in both forest ecosystems(Fig.4).

The model based on exponential relationship between soil CO2efflux and soil temperature at 5 cm depth accurately simulated the seasonal variation in soil respiration rate,but occasionally underestimated or overestimated the values during the growing season(Fig.4).The relationships between soil water content and soil respiration rate were similar in the two forest ecosystems(Table 1).

Fig.4 Simulated(curve)and observed(open cycle)soil CO2efflux from September 2007 to October 2008 in two forest types

Table 1 Results of statistical analyses showing the R2and p values ofsoilrespiration(R)in relation to soiltemperature(T)and soilwater content(W)at 5 cm depth

Fig.5 Variation of winter CO2effluxes in forest ecosystems with latitude.Data shown in the figure were reported in our study(Suzuki et al.2006;Hubbard et al.2005;Wang et al.2010;Groffman et al. 2006;Rustad and Fernandez 1998;Schindlbacher et al.2007; McDowell et al.2000;Muhr et al.2009;Kurganova et al.2003) from low to high latitude

Soil respiration rate between ecosystems

Mean winter soil CO2effluxes were 1.08±0.15μmol m-2s-1in the plantation and 1.16±0.18μmol m-2s-1in the natural forest,both higher than the measured values reported for temperate and boreal forest ecosystems(Fig.5). WintersoilCO2efflux wassimilarin the two contrasting forest ecosystems(Table 2).Conversely,mean soil CO2efflux over thegrowing season wasmuch higherin thenaturalforestthan in the plantation(Table 2).

Contribution of winter flux to annual soil efflux

The estimated winter and annual soil effluxes for the natural forest were 176.3 and 1070.3 g m-2,respectively, 17.2 and 9.7%greater than their counterparts in the plantation,respectively(Table 2;p＜0.05).The contributions of winter soilrespiration to annualsoilefflux were similarforthe two foresttypes at15.4%forthe plantation and 16.5%for the natural forest(Table 2).

Discussion

Seasonalvariation in soilrespiration was similar in the two coniferous forests,with the highest rates observed in the warm and wetgrowing season and the lowestrates during winter.This is consistent with many results reported for coniferous forests(e.g.,Vanhala 2002;Kurganova et al. 2003).Soil temperature and moisture are two important factors that regulate temporal variations in soil respiration rates at the ecosystem level.Our results showed a significantexponential positive correlation of soilrespiration rate with soiltemperature for both forestecosystems,indicating thatsoiltemperature at5 cm depth was a good predictor of seasonal soil respiration for both ecosystems.Such strong dependence of soilrespiration on soiltemperature has been reported for subalpine coniferous forest ecosystems in our study region(Zhou et al.2009).Similar results were also observed in temperate and boreal forest ecosystems (Schindlbacher etal.2007;Wang et al.2010).

Soil respiration rate was not correlated with soil water content in either forest ecosystem.Due to the frequent rainfall and well drained topography,the mountain forest soils scarcely suffered from long-term saturation or drought.Therefore,the seasonal change in soil water content was not clear and could not restrict microbial activities in mostmonths.We interpreted the seasonality of soil respiration as an effect of temperature only,with no effect of soilmoisture.

Recently,a few studies have demonstrated that prominent soil CO2fluxes were released from snow-covered or frozen forestsoils during the winter period(Hubbard etal. 2005;Monson et al.2006;Suzuki et al.2006;Schindlbacher et al.2007;Wang et al.2010).In the present case, mean wintersoilCO2effluxes were 1.08±0.15μmol m-2s-1for the plantation and 1.16±0.15μmol m-2s-1for the natural coniferous forest.Our measured values of mean winter soil CO2efflux were higher than all of the values reported for boreal coniferous forest ecosystems or temperate deciduous broad-leaved or hardwood forest ecosystems(Rustad and Fernandez 1998;Kurganova et al. 2003;Hubbard et al.2005;Groffman et al.2006;Suzuki et al.2006;Schindlbacher et al.2007;Wang et al.2010). For example,moderate soil CO2effluxes of 0.5–0.7μmol m-2s-1were recorded during winter in coniferous forests in Austria,Germany,Russia and USA (Sommerfeld etal.1996;McDowelletal.2000;Kurganova et al.2003;Schindlbacher et al.2007;Muhr et al.2009). Comparatively low winter soil respiration rates were observed in two mixed conifer forests of different ages(50 and 300 years)in Colorado,USA(0.31μmolm-2s-1for 50 years old forest and 0.35μmol m-2s-1for 300 years old forest,Hubbard et al.2005),in a deciduous broadleaved forest in Japan(0.37μmol m-2s-1,Suzuki et al. 2006)and in three hardwood forests in New Hampshire, USA(0.21–0.3μmolm-2s-1,Groffman et al.2006).Our study showed thatsoil CO2effluxes from frozen soils were relatively stable in two forest ecosystems during winter. This resultis consistentwith many other studies conducted in forestecosystems(Suzukietal.2006;Wang etal.2010).

When reported values of winter soil CO2effluxes are arranged in order of increasing latitude,we found that winter soil CO2effluxes in forest ecosystems were similar from 35°N to 45°N and from 45°N to 55°N(Fig.5). Estimated winter soil CO2efflux seemed to be markedly lower at mid-latitude(35–45°N)than at high latitude (45–55°N)(Fig.5).However,our measured values ofmean winter soil CO2efflux at 31 N were much higher than those estimated formid and high latitudes.Wintersoil CO2efflux may be controlled by the soil substrate,root activity and subsurface temperature influenced by thickness and duration of snow cover(Brooks et al.2004; Monson et al.2006).Some findings indicate that soil microbialbiomass may actually peak in winter suggesting the presence of a soil heterotrophic community with an unknown capability of utilizing soil carbon(Lipson et al. 2000;Schadt et al.2003).The changes in air temperature and snow cover(thickness and duration)with latitude might,to some degree,determine soil temperatures during winter,which could partly control the pattern of winter CO2efflux with latitude.In our study region,the daily mean air temperature was generally above-5°C throughout winter.In addition,maximum snow cover was less than 30 cm and the duration of snow cover was generally ca.4 months.Soiltemperature at5 cm depth ranged from-2 to 0°C during winter.Under such conditions, microbial and root activity under snow may continue generating CO2fluxes(Schimel and Clein 1996;Brooks et al.1997;Schindlbacher et al.2007).Jiang et al.(2009) reported thatsoils in our two study forests contained large soil organic carbon pools and microbial biomass carbon was relatively high in winter.Also,itis importantto note thatsoilCO2efflux may,to some degree,be overestimated by using the Li-6400 method because itis difficultto keep the IRGAs warm atvery cold ambienttemperatures(Wang et al.2006).All of these factors together may cause relatively high winter soil CO2fluxes atour experimental sites that were located at relatively low latitude.

Table 2 Winter soil respiration rate and its contribution to annual soil CO2efflux for two forest ecosystems

Soil respiration is typically the most important contributor(30–80%)to annual total respiration in forest ecosystems(Davidson et al.2006).Soil CO2effluxes during the dormantseason have been found to range from 10 to 50%of total annual soil respiration in forest ecosystems(Schindlbacheretal.2007).Apparently,winter soil CO2efflux can play a significant role in the annual carbon budget of forest ecosystems.Recently,a growing number of studies have reported that carbon emissions from forest soils during winter were significant,but winter CO2fluxes from alpine coniferous forest soils in the Tibetan Plateau had notbeen quantified prior to this study. The reported values of winter CO2fluxes in forest ecosystems ranged from 40 to 232 g C m-2(Brooks etal. 2004;Schindlbacher et al.2007).Brooks et al.(2004) found that winter CO2fluxes were larger and less variable in deciduous forest than in coniferous forests.We estimated winter soil respiration at 150.4 g m-2for the plantation and 176.3 g m-2for the natural forest,which were comparable to the values reported for coniferous forests in Colorado,Wyoming,Idaho and Siberia(Sommerfeld etal. 1993;McDowell et al.2000;Monson et al.2002;Kurganova et al.2003),and higher than those found in many other forest ecosystems(Zimov et al.1996;Brooks et al. 1997;Mariko et al.2000;Hubbard et al.2005;Mo et al. 2005;Vogel et al.2005).In this study,the winter contribution to the annualamountof CO2evolved from the soil (approximately 16%in both forests)was consistent with the estimates of 10–50%for forest ecosystems(Schindlbacher et al.2007;Wang et al.2010).Moreover,our estimates for both forests were comparable to the average values(approximately 15%)reported for forest ecosystems(Schindlbacher et al.2007).The relatively large winter soil CO2effluxes and large contributions to annual soil respiration at our experimental sites were mainly attributed to relatively high soil respiration rates,because the duration of winter was relatively short.The duration of winter was shorter than for many other ecosystems athigh latitudes and/or elevations,but winter CO2effluxes nonetheless accounted for a substantial proportion of annual soil respiration.Obviously,winter CO2efflux in coniferous forestecosystems in the cold and snowy regions on the Tibetan Plateau is very important and considerable. Further studies are needed to clarify the relative role of soil microbial and root activities in soil CO2efflux during winter.

Land-use activities could strongly influence soil microclimate,plant carbon allocation patterns,substrate availability and input,and hence soil respiration(Sheng et al. 2010).Numerous studies have convincingly demonstrated that reforestation tended to decrease soil CO2efflux(Ishizuka et al.2002;Carlisle et al.2006;Werner et al.2006; Sheng et al.2010).Our results also showed a significant decline in soil respiration after the transition of natural coniferous forest to pure P.asperata plantation.Manystudies have revealed that soil CO2efflux was correlated positively with substrate availability(Wang et al.2006, 2010;Sheng et al.2010).Our prior studies reported the decreasing trends of carbon substrate(labile organic carbon,soilmicrobialbiomass),soilnutrientavailability,litter input and litter quality following the conversion from natural forest to plantation(Liu et al.2002;Jiang et al. 2009).Such land-use changes might,to some extent, induce declines in microbial respiration in the plantation. Soil respiration consists overwhelmingly of contributions from roots(autotrophic)and microbes(heterotrophic).Liu et al.(2008b)reported that live fine root biomass was significantly higher in natural forest than in a P.asperata plantation.Reduced live fine root biomass and root turnover could cause a decrease in autotrophic respiration in the plantation.As mentioned above,these substantial declines of aboveground litterfall input,litter quality, belowground soil carbon substrate and fine root turnover might directly/indirectly explain why soil respiration declined following the transition of natural coniferous forest to pure P.asperata plantation.

In conclusion,at our subalpine coniferous forest sites, very significant CO2effluxes were emitted from the frozen soils in both natural forest and plantation during winter. Winter soil respiration was about 15–17%of totalannual soil respiration.Ignoring this seasonal carbon loss would cause overestimation of the carbon sequestration potential in subalpine coniferous forests of the eastern Tibetan Plateau.Moreover,land-use changes have substantial influence on forest structures and soil properties.Functional characteristics,such as carbon and nutrient fluxes,are closely linked to associated structural characteristics such as species composition,stand age,density and biomass. Therefore,our study also demonstrated that reforestation did not affect mean winter CO2efflux but significantly reduced annualsoilrespiration.Globalairtemperatures are predicted to increase 1.8–4.0°C over the course of this century,with a greater warming occurring during winter (IPCC 2007).Temperature is a key factor that regulates almost all biogeochemical processes of terrestrial ecosystems,such as soil respiration and decomposition of soil organic matter(SOM).Thus,warmer air temperatures would likely result in warmer soil temperatures which could,in turn,increase the efflux of CO2from soils. Additionally,projected global warming would affect wintersoilCO2efflux by altering the thickness and duration of snow cover.Rates of soil CO2flux are surprisingly high during winter,accounting for significant percentages of annual soil respiration,and winter soil respiration is sensitive to global warming(Janssens and Pilegaard 2003). Alpine and sub-alpine soils are generally considered to be net sinks for atmospheric CO2(Sommerfeld et al.1993). The magnitude and direction of the net exchange of CO2between forests and the atmosphere is expected to depend greatly on changes in winter soil CO2efflux.To date,most experimentalwarming studies have focused on the growing season,with less focus on winter soil CO2efflux(Groffman et al.2006).Therefore,warming studies on winter soil respiration in differentland-use types are needed to assess interactive effects of warming and land-use changes on winter CO2efflux.

Armstrong RL,Brodzik MJ(2001)Recent Northern Hemisphere snow extent:a comparison of data derived from visible and microwave satellite sensors.Geophys Res Lett 28:3673–3676

Brooks PD,Schmidt SK,Williams MW(1997)Winter production of CO2and N2O from alpine tundra:environmental controls and relationship to inter-system C and N fluxes.Oecologia 110:403–413

Brooks PD,Mcknight D,Elder K(2004)Carbon limitation of soil respiration under winter snow packs:potential feedbacks between growing season and winter carbon fluxes.Glob Change Biol 11:231–238

Campos CA(2006)Response of soil surface CO2-C flux to land-use changes in a tropical cloud forest(Mexico).For Ecol Manage 234:305–312

Carlisle EA,Steenwerth KL,Smart DR(2006)Effects of land use on soil respiration:conversion of oak woodlands to vineyards. J Environ Qual 35:1396–1404

Davidson EA,Richardson AD,Savage KE,Hollinger DY(2006)A distinct seasonal pattern of the ratio of soil respiration to total ecosystem respiration in a spruce-dominated forest.Glob Change Biol 12:230–239

Groffman PM,Hardy JP,Driscoll CT,Fahey TJ(2006)Snow depth, soil freezing,and fluxes of carbon dioxide,nitrous oxide and methane in a northern hardwood forest.Glob Change Biol 12:1748–1760

Gru¨nzweig JM,Sparrow SD,Chapin FS(2003)Impact of forest conversion to agriculture on carbon and nitrogen mineralization in subarctic Alaska.Biogeochemistry 64:271–296

Hubbard RM,Ryan MG,Elder K,Rhoades CC(2005)Seasonal patterns in soil surface CO2flux under snow cover in 50 and 300 year old subalpine forests.Biogeochemistry 73:93–107

IPCC(2007)Climate change 2007:the physical science basis. Working Group I contribution to the IPCC Fourth Assessment Report.http://www.ipcc.ch/

Ishizuka S,Tsuruta H,Murdiyarso D(2002)An intensive field study on CO2,CH4,and N2O emissions from soils at four land-use types in Sumatra,Indonesia.Glob Biogeochem Cycles.doi:10. 1029/2001GB001614

Janssens IA,Pilegaard K(2003)Large seasonalchanges in Q10of soil respiration in a beech forest.Glob Change Biol 9:911–918

Jiang F,Sun H,Lin B,Liu Q(2009)Dynamic changes of topsoil organic carbon in subalpine spruce plantation at different succession stages in western Sichuan Province.Chin J Appl Ecol 20:2581–2587

Kurganova BI,Gerenyu VL,Rozanava L,Sapronov D,Myakshina T, Kudeyarov V(2003)Annual and seasonal CO2fluxes from Russian southern taiga soils.Tellus 55:338–344

Li Y,Xu M,Zou X,Shi P,Zhang Y(2005)Comparing soil organic carbon dynamics in plantation and secondary forest in wet tropics in Puerto Rico.Glob Change Biol 11:239–248

Lipson DA,SchmidtSK,Monson RK(2000)Carbon availability and temperature control the post-snowmelt decline in alpine soil microbial biomass.Soil Biol Biochem 32:441–448

Liu Q,Wu Y,Chen Q(2002)Ecological research on subalpine coniferous forest in China.Sichuan University Press,Sichuan, pp 28–185

Liu H,Zhao P,Lu P,Wang Y,Lin Y,Rao X(2008a)Greenhouse gas fluxes from soils of different land-use types in a hilly area of South China.Agric Ecosyst Environ 124:125–135

Liu L,Zhang J,Yang W,Wang M,Xue Q,Dong S(2008b)Biomass and carbon stock offine roots in the representative subalpine and alpine forests in Western Sichuan.J Sichuan For Sci Technol 1:7–10

Mariko S,Nishimura N,Mo W,Matsui Y,Kibe T,Koizumi H(2000) Winter CO2flux from soiland snow surfaces in a cool-temperate deciduous forest,Japan.Ecol Res 15:363–372

McDowell NG,Marshall JD,Hooker TD(2000)Estimating CO2flux from snowpacks at three sites in the Rocky Mountains.Tree Physiol 20:745–753

Mo W,Lee M,Uchida M,Inatomi M,Saigusa N,Mariko S,Koizumi H(2005)Seasonal and annual variations in soil respiration in a cool-temperate deciduous broad-leaved forest in Japan.Agric For Meteorol 134:81–94

Monson RK,Turnipseed AA,Sparks JP,Harley PC,Scott-Denton LE,Sparks K,Huxman TX(2002)Carbon sequestration in a high-elevation,subalpine forest.Glob Change Biol 8:459–478

Monson RK,Lipson DL,Burns SP,Turnipseed AA,Delany AC, Williams MW,Schmidt SK(2006)Winter forestsoilrespiration controlled by climate and microbial community composition. Nature 439:711–714

Muhr J,Borken W,Matzner E(2009)Effects of soil frost on soil respiration and its radiocarbon signature in a Norway spruce forest soil.Glob Change Biol 15:782–793

Qin J,Liu Q(2010)Oxidative metabolism-related changes during germination of mono maple(Acer mono Maxin.)seeds under seasonal frozen soil.Ecol Res 25:337–345

Rustad LE,Fernandez IJ(1998)Experimentalsoilwarming effects on CO2and CH4flux from a low elevation spruce-fir forestsoil in Maine,USA.Glob Change Biol 4:597–605

Schadt CW,Martin AP,Lipson DA,Schmidt SK(2003)Seasonal dynamics of previously unknown fungallineages in tundra soils. Science 301:1359–1361

Schimel JP,Clein JS(1996)Microbialresponses to freeze-thaw cycles in tundra and taiga soils.Soil Biol Biochem 28:1061–1066

Schindlbacher A,Zechmeister-Boltenstern S,Glatzel G,Jandl R (2007)Winter soilrespiration from an Austrian mountain forest. Agric For Meteorol 146:205–215

Sheng H,Yang Y,Yang Z,Chen G,Xie J,Guo J,Zou S(2010)The dynamic response of soil respiration to land-use changes in subtropical China.Glob Change Biol 16:1107–1121

Sommerfeld RA,Mosier AR,Musselman RC(1993)CO2,CH4,and N2O flux through a Wyoming snowpack.Nature 361:140–142

Sommerfeld RA,Massman WJ,Musselman RC(1996)Diffusionalflux ofCO2through snow;spatialand temporalvariability among alpine subalpine sites.Glob Biogeochem Cycles 10:473–482

Suzuki S,Ishizuka S,Kitamura K,Yamanoi K,Nakai Y(2006) Continuous estimation of winter carbon dioxide efflux from the snow surface in a deciduous broadleaf forest.J Geophys Res. doi:10.1029/2005JD006595

Vanhala P(2002)Seasonal variation in the soil respiration rate in coniferous forest soils.Soil Biol Biochem 34:1375–1379

Vogel JG,Valentine DW,Ruess RW(2005)Soiland rootrespiration in mature Alaskan black spruce forests that vary in soil organic matter decomposition rates.Can J For Res 35:161–174

Wang K,Kelloma¨ki S,Zha T(2003)Modifications in photosynthetic pigments and chlorophyll fluorescence in 20-year-old pine trees after a four-year exposure to carbon dioxide and temperature elevation.Photosynthetica 41:167–175

Wang C,Yang J,Zhang Q(2006)Soil respiration in six temperate forests in China.Glob Change Biol 12:2103–2114

Wang W,Peng S,Wang T,Fang J(2010)WintersoilCO2efflux and its contribution to annualsoilrespiration in differentecosystems of a forest-steppe ecotone,north China.SoilBiolBiochem 42:451–458

Werner C,Zheng XH,Tang JW,Xie BH,Liu CY,Kiese R, Butterbach-Bahl K(2006)N2O,CH4and CO2emissions from seasonaltropicalrainforests and a rubberplantation in Southwest China.Plant Soil 289:335–353

Winston GC,Sundquist ET,Stephens BB,Trumbore SE(1997) Winter CO2fluxes in a boreal forest.J Geophys Res 102:28795–28804

Zhou F,Lin B,Liu Q(2009)Characteristics of soil respiration in artificial restoration process of subalpine coniferous forest in eastern Qinghai-Tibet Plateau.Chin J Appl Ecol 20:1783–1790

Zimov SA,Davidov SP,Voropaev YV,Prosiannikov SF,Semiletov IP,Chapin MC,Chapin FS(1996)Siberian CO2efflux in winter as CO2source and cause of seasonality in atmospheric CO2. Clim Change 33:111–120

7 October 2014/Accepted:28 December 2014/Published online:18 July 2015

©Northeast Forestry University and Springer-Verlag Berlin Heidelberg 2015

Project funding:This study was supported by the National Natural Science Foundation of China(31200474,31270552),the National Key Technologies R&D in China(2011BAC09B05),Postdoctoral Science Foundation of China(2013M540714 and 2014T70880).

The online version is available at http://www.springerlink.com

Corresponding editor:Zhu Hong

✉Zhenfeng Xu sicauxzf@163.com

✉Qing Liu sichuanliuqing@163.com

1Institute of Forest&Ecology,Sichuan Agricultural University,Chengdu 611130,People’s Republic of China

2Chengdu Institute of Biology,Chinese Academy of Sciences, Chengdu 610041,People’s Republic of China