Global Change Biology Global Change Biology (2015) 21, 777–788, doi: 10.1111/gcb.12690
Considerable methane uptake by alpine grasslands despite the cold climate: in situ measurements on the central Tibetan Plateau, 2008–2013 D A W E I 1 , X U - R I 1 , T E N Z I N - T A R C H E N 1 , 2 , Y U E S I W A N G 3 and Y I N G H O N G W A N G 3 1 Key Laboratory of Tibetan Environment Changes and Land Surface Processes, Key Laboratory of Alpine Ecology and Biodiversity, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China, 2Tibet University, Tibetan Autonomous Region, China, 3Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
Abstract The uptake of CH4 by aerate soil plays a secondary role in the removal of tropospheric CH4, but it is still highly uncertain in terms of its magnitude, spatial, and temporal variation. In an attempt to quantify the sink of the vast alpine grasslands (1 400 000 km2) of the Tibetan Plateau, we conducted in situ measurements in an alpine steppe (4730 m) and alpine meadow (4900 m) using the static chamber and gas chromatograph method. For the alpine steppe, measurements (2008–2013) suggested that there is large interannual variability in CH4 uptake, ranging from 48.8 to 95.8 lg CH4 m2 h1 (averaged of 71.5 2.5 lg CH4 m2 h1), due to the variability in precipitation seasonality. The seasonal pattern of CH4 uptakes in the form of stronger uptake in the early growing season and weaker uptake in the rainy season closely matched the precipitation seasonality and subsequent soil moisture variation. The relationships between alpine steppe CH4 uptake and soil moisture/temperature are best depicted by a quadratic function and an exponential function (Q10 = 1.67) respectively. Our measurements also showed that the alpine meadow soil (average of 59.2 3.7 lg CH4 m2 h1) uptake less CH4 than the alpine steppe and produces a similar seasonal pattern, which is negatively regulated by soil moisture. Our measurements quantified – at values far higher than those estimated by process-based models – that both the alpine steppe and alpine meadow are considerable CH4 sinks, despite the cold weather of this high-altitude area. The consecutive measurements gathered in this study also highlight that precipitation seasonality tends to drive the interannual variation in CH4 uptake, indicating that future study should be done to better characterize how CH4 cycling might feedback to the more extreme climate. Keywords: alpine meadow, alpine steppe, CH4 sink, soil moisture, Tibetan Plateau Received 4 June 2014 and accepted 25 June 2014
Introduction Methane (CH4) is the second most important of the long-lived greenhouse gases behind carbon dioxide (CO2), contributing approximately 20% of the total global warming forcing (Schulze et al., 2009; Kirschke et al., 2013). Ice-core proxy and instrumental measurements of the atmospheric CH4 abundance have revealed a significant increase from 700 ppb before 1750 (Etheridge et al., 1998) to over 1819 ppb in 2012 (WMO, 2013). Recent atmospheric observations of CH4 concentrations have revealed a slowdown in the growth rate up until the 1990s, a steady state from 1999 to 2006, and renewed growth thereafter (Dlugokencky et al., 2011; WMO, 2013). To date, it is still difficult to attribute the unique growth pattern to either anthropogenic or natural sources (Bousquet et al., 2006). Although major natural and anthropogenic CH4 sources Correspondence: Dr. Xu-Ri, tel. +86-10-84097072, fax +86-10-84097079, e-mail: [email protected]
© 2014 John Wiley & Sons Ltd
and sinks have been identified, their relative strengths remain highly uncertain (Spahni et al., 2011). For example, wetlands comprise the largest single source of CH4, but the strength of their supply varied greatly from 177 to 284 Tg CH4 a1 during the past decade (IPCC, 2013). The uptake of CH4 by natural soils is the second largest sink of atmospheric CH4 (CH4 + O2 ? CO2) after OH depletion. The CH4 is primarily oxidized by high affinity methanotrophs within the aerate soil layers (King, 1997; Li, 2000; Le Mer & Roger, 2001), a process which is regulated by methanotroph activity and substrate availability (CH4 and O2) (Segers, 1998; Li et al., 2014). The methanotroph community is affected by soil aerobic conditions and temperature, while the CH4 and O2 availability are constrained by air–soil diffusion. The variation in soil moisture, therefore, strongly determines the air–soil exchange of CH4/O2 and the activity and abundance of the methanotroph community (e.g. Curry, 2007; Zhuang et al., 2013). Although a lack of correlation between soil temperature and CH4 uptake is 777
778 D . W E I et al. quite common, many measurements do reveal that the oxidation is sensitive to temperature variation (Zhuang et al., 2013). Despite current understanding of the importance of CH4 uptake, there are still great uncertainties with respect to its magnitude, spatial distribution, and temporal variation. There is no consistency in terms of its global sink strength among inventories and processbased model simulations. For the bottom-up inventory method, the aerate soil CH4 sink has been estimated at 21 Tg CH4 a1 (Potter et al., 1996), 7–120 Tg CH4 a1 (Smith et al., 2000), 30 15 Tg CH4 a1 (Ehhalt & Prather, 2001), and 20–45 Tg CH4 a1 (Laure & Verchot, 2007). Using process-based models, estimates have been 17–23 Tg CH4 a1 (Potter et al., 1996), 38 Tg CH4 a1 (20–50 Tg CH4 a1) (Ridgwell et al., 1999), 28 Tg CH4 a1 (9–47 Tg CH4 a1) (Curry, 2007), and 32–36 Tg CH4 a1 (Zhuang et al., 2013). Recent IPCC report (2013) also produces a similar value at 28 Tg CH4 a1 (9 to 47 Tg CH4 a1), based on the simulation results of Spahni et al. (2011) and Curry (2007). Uptake by grasslands constitutes a sizeable proportion of the total CH4 sink, and the Eurasian Steppe forms one of the largest examples of grassland ecosystems worldwide. The Tibetan Plateau (TP, Fig. 1), which is mostly above 4000 m in altitude, covers an area of 2 500 000 km2 (a quarter of China’s total territory). The high altitude and thus low temperature, as well as the semiarid climate, result in the dominance of alpine steppe and alpine meadow across the plateau (720 000 and 680 000 km2, respectively) (Editorial Board of Vegetation Map of China, Chinese Academy of Sciences, 2001). Currently, little is known about the role of alpine grassland on the TP in regulating CH4 cycling, especially alpine steppe (e.g. Pei et al., 2003). Thus, we conducted in situ measurements of CH4 uptake at an alpine steppe and alpine meadow site in the hinterland of the TP (Fig. 1). Existing studies already identified the sensitivities of methanotroph and CH4 oxidation to temperature variation (e.g. Segers, 1998; Zhuang et al., 2013). Thus, we predicted that the CH4 uptake would be relatively low in these cold ecosystems, due to the likely temperature constraints upon methanotrophs and subsequent CH4 oxidation. However, observations made over the course of several growing seasons in the alpine steppe (2008–2013) and alpine meadow (2012–2013) quite exceeded our initial assumption by showing considerable CH4 uptake. In the above context, the present study reports the unique CH4 flux dataset observed above 4700 m on the hinterland of the TP. Through analysis of the data, we attempt to clarify the variation in soil–atmosphere CH4 uptake and its controls in the two typical alpine ecosystems, i.e. alpine steppe and alpine meadow. To the best
Fig. 1 Location map of Nam Co Station on the Tibetan Plateau. The triangle indicates the research area of this study. The DEM map is provided by the Environmental and Ecological Science Data Center for West China, NSFC, \http://westdc.westgis.ac. cn).
of our knowledge, the dataset established in this study represents the longest in situ observation record available for the TP. The reporting of consecutive data of this kind in this region will also benefit process-based model development through parameterization and validation.
Materials and methods
Site description The measurements were taken at the Nam Co Station, run by the Institute of Tibetan Plateau Research, Chinese Academy of Sciences (Fig. 1). The local climate is controlled by the Indian monsoon in summer and Westerlies in winter, with 80–90% of precipitation falling from May to September, which is also the growing season for the alpine steppe and alpine meadow ecosystem. The local annual ambient temperature in the alpine Nam Co Station is around 0.6 °C. The cumulative precipitation is 414.6 mm, but demonstrates large interannual variation, ranging from 294.8 to 549.7 mm (Zhang et al., 2011). The alpine steppe covers an altitudinal range of approximately 4700–4900 m, while the alpine meadow is situated between approximately 4900 and 5200 m, where there is more rainfall and subsequently higher soil water content (Figure S1a, b and Table 1). The alpine steppe site was located at 4730 m. Three replicates in a conventional grazed alpine steppe (Stipa purpurea) and another three in a fenced alpine steppe (fenced since August 2005) were set up. The alpine meadow (Kobresia pygmae) site was located at approximately 4900 m at the foot of the Mt.Nyenchen Tanglha. Four replicates were set up in a grazed meadow and another four in a © 2014 John Wiley & Sons Ltd, Global Change Biology, 21, 777–788
C H 4 U P T A K E B Y A L P I N E G R A S S L A N D S D E S P I T E C O L D 779 Table 1 General characteristics of the alpine steppe and alpine meadow site at Nam Co (aboveground biomass, soil organic carbon, total nitrogen, and pH measured by Y. Liu et al., unpublished results)
Location Altitude Constructive species Aboveground biomass Soil texture Bulk density Soil organic carbon Total nitrogen pH Chamber numbers Period
Alpine steppe
Alpine meadow
N30°E900 59″ 4730 Stipa purpurea 53.0 9.8
N30°E910 01″ 4900 Kobresia pygmaea 89.1 7.2
Sand, Sandy loam 1.03 0.10 35.1 2.6
Sandy loam, Sand n.a. 29.1 2.5
3.4 0.2 7.6 0.1 6 2008–2013
2.0 0.1 7.4 0.2 8
Units
m
g m2
g m3 mg g1 mg g1
2012–2013
fenced meadow (fenced since September 2011). Stainless steel collars were installed 4 days before the first measurements were taken, and kept in place during the entire observation period.
Measurement of CH4 uptake The uptake of CH4 by the alpine steppe and meadow were measured using the static chamber and gas chromatograph (GC) method (details in Wang & Wang, 2003). Five air samples for each replicate were taken in weekly intervals at around 10:00–12:00 hours (UTC +8) during the growing seasons, and then measured using the GC within 24 h at the station (Agilent 7890A, Agilent Technologies, Santa Clara, CA, USA, Figure S2a and b). The air samples were injected into the column and flowed with a carrier gas (N2, 99.9992%; column: SS-2 m 9 2 mm, 13 XMS, 60/80), and then detection was carried out using a flame ionization detector. Following Wang et al. (2005), the relation Fi ¼ qi ðV=AÞ ðP=P0 Þ ðT0 =TÞ ðdCi =dt Þ was used to calculate the CH4 flux, where Fi is the flux rate, qi is the density under standard conditions, V is the chamber volume, A is the base area, P is air pressure, P0 is standard air pressure, T is air temperature, T0 is standard temperature, and dCi/dt is growth rate of CH4 concentration. More details are available in Wei et al. (2012). No measurements were taken in winter because of poor accessibility to the Nam Co Station. Therefore, to estimate the annual scale of CH4 uptake, a fixed factor for the nongrowing season of 31% of that in the growing season was applied based on a whole year’s measurements at an alpine steppe site at Mt. Tianshan, north to the TP (Li et al., 2012), which is similar to another annual-scale measurement taken at a typical temperate grassland site in Inner Mongolia (30%, Chen et al., 2010, 2011).
(a)
(b)
(c)
Fig. 2 Seasonal variation in (a) air/soil temperature, (b) soil moisture/precipitation, and (c) CH4 uptake in the alpine steppe. © 2014 John Wiley & Sons Ltd, Global Change Biology, 21, 777–788
780 D . W E I et al.
Meteorological measurement
(a)
The ambient temperature (1.5 m aboveground), soil temperature (5 cm), and soil moisture (5 to 0 cm) were recorded using micro weather stations (H-21, Onset Corp., Pocasset, MA, USA), while the precipitation was measured using datalogging rain gauges (RG-3, Onset Corp.). Additional soil temperature (5 cm) and soil moisture (5 to 0 cm) measurements were taken using an electric thermometer (JM-624, Jinming Corp., Tianjin, China) and soil moisture probe (Soil Moisture Analyzer, Delta-T Devices, Cambridge, UK), during the gas sampling campaign in the field. To fill the observational gap of soil moisture data (18.9%) of the auto weather station, a regression equation based on existing relationship between soil moisture and precipitation was used: soil moisture = precipitation (20-day moving average) 9 1.45 + 3.67 (R2 = 0.76, P < 0.001, n = 586).
(b)
Statistical analyses All data are displayed as the mean 1 SE (standard error) unless otherwise stated. Pearson correlation, stepwise linear regression, and linear/nonlinear regressions were applied to test the relationship between CH4 uptake and soil moisture/ temperature. The Van’t Hoff equation (y = aebt) was used to calculate the temperature sensitivity (Q10) of CH4 uptake (Q10 = e10b, soil temperature at 5 cm belowground).
(c)
Results
Alpine steppe: meteorological background
Fig. 3 Monthly averaged seasonal variability in (a) air/soil temperature, (b) soil moisture/precipitation, and (c) CH4 uptake in the alpine steppe.
The soil temperature reached its maximum in June, while the air temperature peaked in July (Figs 2a and 3a). The growing-season-averaged air and soil temperatures were 7.13 0.1 °C and 12.48 0.2 °C, respectively, during the whole observation period. The
Table 2 Summary of growing-season-averaged air temperature, soil temperature, soil moisture, and cumulative precipitation and CH4 uptake in the alpine steppe and alpine meadow at Nam Co (n.a. = not available; Soil temp. = Soil temperature; Soil mois. = Soil moisture; Precip. = Precipitation)
Year
Air temp. (°C)
Alpine steppe 2009 7.20 2010 7.71 2011 6.65 2012 7.22 2013 7.02 Mean 7.13 CV 5% Alpine meadow 2012 7.74 2013 7.27 Mean 7.50
0.3 0.2 0.2 0.3 0.3 0.1
0.2 0.2 0.2
Soil temp. (°C)
12.51 12.92 12.07 12.53 12.37 12.48 3%
0.3 0.2 0.2 0.2 0.2 0.2
10.5 0.2 10.2 0.2 10.3 0.2
Soil mois. (m3/m3)
Precip. (mm)
CH4 uptake (lg CH4 m2 h1)
6.07 7.46 7.72 7.85 7.35 7.29 10%
0.2 0.4 0.2 0.2 0.2 0.3
327 494 402 312 397 386 32.3 19%
67.7 70.2 48.8 74.0 95.8 71.5 23.6%
n.a. 11.9 0.5 11.9 0.5
311 489 400 88.9
36.8 3.3 73.7 5.2 59.2 3.7
3.0 4.5 4.8 4.5 5.3 2.5
© 2014 John Wiley & Sons Ltd, Global Change Biology, 21, 777–788
C H 4 U P T A K E B Y A L P I N E G R A S S L A N D S D E S P I T E C O L D 781 observational period. The soil moisture showed clear seasonal pattern of approximately 4.99 0.65% in May, a decrease to 4.43 0.65% in June, and a maximum of 9.59% in August (Fig. 3). The climatic indices, especially precipitation and soil moisture, showed large interannual variation (Table 2). The coefficient of variation (CV) of cumulative precipitation and average soil moisture was 18.7% and 10% respectively, while for the ambient temperature and soil temperature these values were 5% and 3%, respectively. The highest level of precipitation occurred in 2010, but there was a long drought before the highly concentrated rainy season in August, when the soil moisture reached its maximum value (~25%) of the whole measurement period. The highest average air temperature and soil temperature (7.71 0.2 °C and 12.92 0.2 °C respectively) also occurred in 2010. The highest growing-season-averaged soil moisture content occurred in 2012 (7.85 0.2%) when the precipitation was decentralized during the months of June, July and August, although the cumulative precipitation was only 312 mm. A similar situation occurred in 2011 when the average soil moisture content was 7.72% and the ambient temperature and soil temperature were 6.65 0.2 °C and 12.07 0.2 °C respectively – the warmest growing season. It is also worth noting that in 2009, when the soil moisture content was only 6.07 0.2%, it was the driest growing season of the whole observational period.
(a)
(b)
(c)
Alpine steppe: seasonal pattern of CH4 uptake Fig. 4 Dependency of seasonal variation in CH4 uptake on (a) soil moisture, (b) temperature, and (c) their interactions in the alpine steppe.
precipitation concentrated between the months of July and August, but especially August (130 27.8 mm), while the average precipitation in May was only 24.0 4.1 mm. The growing season cumulative precipitation was 386 32.3 mm, and the average soil moisture content was 7.29 0.3% during the whole
The alpine steppe soils generally uptake CH4 during the growing seasons of the measurement period (Figs 2c and 3c) at 71.5 2.5 lg CH4 m2 h1. This shows a typical seasonal pattern of stronger CH4 uptake in the dry season (the highest monthly average was 101 lg CH4 m2 h1 in June) and a decline during the rainy season in July and August (lowest monthly average was 63.8 lg CH4 m2 h1 in August). Although the exact period differed among years, CH4 uptake generally followed the variation in
Table 3 Linear and nonlinear regression of CH4 uptake with soil moisture and temperature in the alpine steppe and alpine meadow at Nam Co. SM = Soil moisture; ST = Soil temperature
Alpine steppe Stepwise linear regression Nonlinear regression Alpine meadow Stepwise linear regression
Function
R2
P
CH4 uptake = 529 9 SM 115 CH4 uptake = 464 9 SM 2.68 9 ST 77.3 CH4 uptake = 6725 9 SM2 448 9 SM 87.0
0.21 0.31 0.24
Considerable methane uptake by alpine grasslands despite the cold climate: in situ measurements on the central Tibetan Plateau, 2008–2013 D A W E I 1 , X U - R I 1 , T E N Z I N - T A R C H E N 1 , 2 , Y U E S I W A N G 3 and Y I N G H O N G W A N G 3 1 Key Laboratory of Tibetan Environment Changes and Land Surface Processes, Key Laboratory of Alpine Ecology and Biodiversity, Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing, China, 2Tibet University, Tibetan Autonomous Region, China, 3Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
Abstract The uptake of CH4 by aerate soil plays a secondary role in the removal of tropospheric CH4, but it is still highly uncertain in terms of its magnitude, spatial, and temporal variation. In an attempt to quantify the sink of the vast alpine grasslands (1 400 000 km2) of the Tibetan Plateau, we conducted in situ measurements in an alpine steppe (4730 m) and alpine meadow (4900 m) using the static chamber and gas chromatograph method. For the alpine steppe, measurements (2008–2013) suggested that there is large interannual variability in CH4 uptake, ranging from 48.8 to 95.8 lg CH4 m2 h1 (averaged of 71.5 2.5 lg CH4 m2 h1), due to the variability in precipitation seasonality. The seasonal pattern of CH4 uptakes in the form of stronger uptake in the early growing season and weaker uptake in the rainy season closely matched the precipitation seasonality and subsequent soil moisture variation. The relationships between alpine steppe CH4 uptake and soil moisture/temperature are best depicted by a quadratic function and an exponential function (Q10 = 1.67) respectively. Our measurements also showed that the alpine meadow soil (average of 59.2 3.7 lg CH4 m2 h1) uptake less CH4 than the alpine steppe and produces a similar seasonal pattern, which is negatively regulated by soil moisture. Our measurements quantified – at values far higher than those estimated by process-based models – that both the alpine steppe and alpine meadow are considerable CH4 sinks, despite the cold weather of this high-altitude area. The consecutive measurements gathered in this study also highlight that precipitation seasonality tends to drive the interannual variation in CH4 uptake, indicating that future study should be done to better characterize how CH4 cycling might feedback to the more extreme climate. Keywords: alpine meadow, alpine steppe, CH4 sink, soil moisture, Tibetan Plateau Received 4 June 2014 and accepted 25 June 2014
Introduction Methane (CH4) is the second most important of the long-lived greenhouse gases behind carbon dioxide (CO2), contributing approximately 20% of the total global warming forcing (Schulze et al., 2009; Kirschke et al., 2013). Ice-core proxy and instrumental measurements of the atmospheric CH4 abundance have revealed a significant increase from 700 ppb before 1750 (Etheridge et al., 1998) to over 1819 ppb in 2012 (WMO, 2013). Recent atmospheric observations of CH4 concentrations have revealed a slowdown in the growth rate up until the 1990s, a steady state from 1999 to 2006, and renewed growth thereafter (Dlugokencky et al., 2011; WMO, 2013). To date, it is still difficult to attribute the unique growth pattern to either anthropogenic or natural sources (Bousquet et al., 2006). Although major natural and anthropogenic CH4 sources Correspondence: Dr. Xu-Ri, tel. +86-10-84097072, fax +86-10-84097079, e-mail: [email protected]
© 2014 John Wiley & Sons Ltd
and sinks have been identified, their relative strengths remain highly uncertain (Spahni et al., 2011). For example, wetlands comprise the largest single source of CH4, but the strength of their supply varied greatly from 177 to 284 Tg CH4 a1 during the past decade (IPCC, 2013). The uptake of CH4 by natural soils is the second largest sink of atmospheric CH4 (CH4 + O2 ? CO2) after OH depletion. The CH4 is primarily oxidized by high affinity methanotrophs within the aerate soil layers (King, 1997; Li, 2000; Le Mer & Roger, 2001), a process which is regulated by methanotroph activity and substrate availability (CH4 and O2) (Segers, 1998; Li et al., 2014). The methanotroph community is affected by soil aerobic conditions and temperature, while the CH4 and O2 availability are constrained by air–soil diffusion. The variation in soil moisture, therefore, strongly determines the air–soil exchange of CH4/O2 and the activity and abundance of the methanotroph community (e.g. Curry, 2007; Zhuang et al., 2013). Although a lack of correlation between soil temperature and CH4 uptake is 777
778 D . W E I et al. quite common, many measurements do reveal that the oxidation is sensitive to temperature variation (Zhuang et al., 2013). Despite current understanding of the importance of CH4 uptake, there are still great uncertainties with respect to its magnitude, spatial distribution, and temporal variation. There is no consistency in terms of its global sink strength among inventories and processbased model simulations. For the bottom-up inventory method, the aerate soil CH4 sink has been estimated at 21 Tg CH4 a1 (Potter et al., 1996), 7–120 Tg CH4 a1 (Smith et al., 2000), 30 15 Tg CH4 a1 (Ehhalt & Prather, 2001), and 20–45 Tg CH4 a1 (Laure & Verchot, 2007). Using process-based models, estimates have been 17–23 Tg CH4 a1 (Potter et al., 1996), 38 Tg CH4 a1 (20–50 Tg CH4 a1) (Ridgwell et al., 1999), 28 Tg CH4 a1 (9–47 Tg CH4 a1) (Curry, 2007), and 32–36 Tg CH4 a1 (Zhuang et al., 2013). Recent IPCC report (2013) also produces a similar value at 28 Tg CH4 a1 (9 to 47 Tg CH4 a1), based on the simulation results of Spahni et al. (2011) and Curry (2007). Uptake by grasslands constitutes a sizeable proportion of the total CH4 sink, and the Eurasian Steppe forms one of the largest examples of grassland ecosystems worldwide. The Tibetan Plateau (TP, Fig. 1), which is mostly above 4000 m in altitude, covers an area of 2 500 000 km2 (a quarter of China’s total territory). The high altitude and thus low temperature, as well as the semiarid climate, result in the dominance of alpine steppe and alpine meadow across the plateau (720 000 and 680 000 km2, respectively) (Editorial Board of Vegetation Map of China, Chinese Academy of Sciences, 2001). Currently, little is known about the role of alpine grassland on the TP in regulating CH4 cycling, especially alpine steppe (e.g. Pei et al., 2003). Thus, we conducted in situ measurements of CH4 uptake at an alpine steppe and alpine meadow site in the hinterland of the TP (Fig. 1). Existing studies already identified the sensitivities of methanotroph and CH4 oxidation to temperature variation (e.g. Segers, 1998; Zhuang et al., 2013). Thus, we predicted that the CH4 uptake would be relatively low in these cold ecosystems, due to the likely temperature constraints upon methanotrophs and subsequent CH4 oxidation. However, observations made over the course of several growing seasons in the alpine steppe (2008–2013) and alpine meadow (2012–2013) quite exceeded our initial assumption by showing considerable CH4 uptake. In the above context, the present study reports the unique CH4 flux dataset observed above 4700 m on the hinterland of the TP. Through analysis of the data, we attempt to clarify the variation in soil–atmosphere CH4 uptake and its controls in the two typical alpine ecosystems, i.e. alpine steppe and alpine meadow. To the best
Fig. 1 Location map of Nam Co Station on the Tibetan Plateau. The triangle indicates the research area of this study. The DEM map is provided by the Environmental and Ecological Science Data Center for West China, NSFC, \http://westdc.westgis.ac. cn).
of our knowledge, the dataset established in this study represents the longest in situ observation record available for the TP. The reporting of consecutive data of this kind in this region will also benefit process-based model development through parameterization and validation.
Materials and methods
Site description The measurements were taken at the Nam Co Station, run by the Institute of Tibetan Plateau Research, Chinese Academy of Sciences (Fig. 1). The local climate is controlled by the Indian monsoon in summer and Westerlies in winter, with 80–90% of precipitation falling from May to September, which is also the growing season for the alpine steppe and alpine meadow ecosystem. The local annual ambient temperature in the alpine Nam Co Station is around 0.6 °C. The cumulative precipitation is 414.6 mm, but demonstrates large interannual variation, ranging from 294.8 to 549.7 mm (Zhang et al., 2011). The alpine steppe covers an altitudinal range of approximately 4700–4900 m, while the alpine meadow is situated between approximately 4900 and 5200 m, where there is more rainfall and subsequently higher soil water content (Figure S1a, b and Table 1). The alpine steppe site was located at 4730 m. Three replicates in a conventional grazed alpine steppe (Stipa purpurea) and another three in a fenced alpine steppe (fenced since August 2005) were set up. The alpine meadow (Kobresia pygmae) site was located at approximately 4900 m at the foot of the Mt.Nyenchen Tanglha. Four replicates were set up in a grazed meadow and another four in a © 2014 John Wiley & Sons Ltd, Global Change Biology, 21, 777–788
C H 4 U P T A K E B Y A L P I N E G R A S S L A N D S D E S P I T E C O L D 779 Table 1 General characteristics of the alpine steppe and alpine meadow site at Nam Co (aboveground biomass, soil organic carbon, total nitrogen, and pH measured by Y. Liu et al., unpublished results)
Location Altitude Constructive species Aboveground biomass Soil texture Bulk density Soil organic carbon Total nitrogen pH Chamber numbers Period
Alpine steppe
Alpine meadow
N30°E900 59″ 4730 Stipa purpurea 53.0 9.8
N30°E910 01″ 4900 Kobresia pygmaea 89.1 7.2
Sand, Sandy loam 1.03 0.10 35.1 2.6
Sandy loam, Sand n.a. 29.1 2.5
3.4 0.2 7.6 0.1 6 2008–2013
2.0 0.1 7.4 0.2 8
Units
m
g m2
g m3 mg g1 mg g1
2012–2013
fenced meadow (fenced since September 2011). Stainless steel collars were installed 4 days before the first measurements were taken, and kept in place during the entire observation period.
Measurement of CH4 uptake The uptake of CH4 by the alpine steppe and meadow were measured using the static chamber and gas chromatograph (GC) method (details in Wang & Wang, 2003). Five air samples for each replicate were taken in weekly intervals at around 10:00–12:00 hours (UTC +8) during the growing seasons, and then measured using the GC within 24 h at the station (Agilent 7890A, Agilent Technologies, Santa Clara, CA, USA, Figure S2a and b). The air samples were injected into the column and flowed with a carrier gas (N2, 99.9992%; column: SS-2 m 9 2 mm, 13 XMS, 60/80), and then detection was carried out using a flame ionization detector. Following Wang et al. (2005), the relation Fi ¼ qi ðV=AÞ ðP=P0 Þ ðT0 =TÞ ðdCi =dt Þ was used to calculate the CH4 flux, where Fi is the flux rate, qi is the density under standard conditions, V is the chamber volume, A is the base area, P is air pressure, P0 is standard air pressure, T is air temperature, T0 is standard temperature, and dCi/dt is growth rate of CH4 concentration. More details are available in Wei et al. (2012). No measurements were taken in winter because of poor accessibility to the Nam Co Station. Therefore, to estimate the annual scale of CH4 uptake, a fixed factor for the nongrowing season of 31% of that in the growing season was applied based on a whole year’s measurements at an alpine steppe site at Mt. Tianshan, north to the TP (Li et al., 2012), which is similar to another annual-scale measurement taken at a typical temperate grassland site in Inner Mongolia (30%, Chen et al., 2010, 2011).
(a)
(b)
(c)
Fig. 2 Seasonal variation in (a) air/soil temperature, (b) soil moisture/precipitation, and (c) CH4 uptake in the alpine steppe. © 2014 John Wiley & Sons Ltd, Global Change Biology, 21, 777–788
780 D . W E I et al.
Meteorological measurement
(a)
The ambient temperature (1.5 m aboveground), soil temperature (5 cm), and soil moisture (5 to 0 cm) were recorded using micro weather stations (H-21, Onset Corp., Pocasset, MA, USA), while the precipitation was measured using datalogging rain gauges (RG-3, Onset Corp.). Additional soil temperature (5 cm) and soil moisture (5 to 0 cm) measurements were taken using an electric thermometer (JM-624, Jinming Corp., Tianjin, China) and soil moisture probe (Soil Moisture Analyzer, Delta-T Devices, Cambridge, UK), during the gas sampling campaign in the field. To fill the observational gap of soil moisture data (18.9%) of the auto weather station, a regression equation based on existing relationship between soil moisture and precipitation was used: soil moisture = precipitation (20-day moving average) 9 1.45 + 3.67 (R2 = 0.76, P < 0.001, n = 586).
(b)
Statistical analyses All data are displayed as the mean 1 SE (standard error) unless otherwise stated. Pearson correlation, stepwise linear regression, and linear/nonlinear regressions were applied to test the relationship between CH4 uptake and soil moisture/ temperature. The Van’t Hoff equation (y = aebt) was used to calculate the temperature sensitivity (Q10) of CH4 uptake (Q10 = e10b, soil temperature at 5 cm belowground).
(c)
Results
Alpine steppe: meteorological background
Fig. 3 Monthly averaged seasonal variability in (a) air/soil temperature, (b) soil moisture/precipitation, and (c) CH4 uptake in the alpine steppe.
The soil temperature reached its maximum in June, while the air temperature peaked in July (Figs 2a and 3a). The growing-season-averaged air and soil temperatures were 7.13 0.1 °C and 12.48 0.2 °C, respectively, during the whole observation period. The
Table 2 Summary of growing-season-averaged air temperature, soil temperature, soil moisture, and cumulative precipitation and CH4 uptake in the alpine steppe and alpine meadow at Nam Co (n.a. = not available; Soil temp. = Soil temperature; Soil mois. = Soil moisture; Precip. = Precipitation)
Year
Air temp. (°C)
Alpine steppe 2009 7.20 2010 7.71 2011 6.65 2012 7.22 2013 7.02 Mean 7.13 CV 5% Alpine meadow 2012 7.74 2013 7.27 Mean 7.50
0.3 0.2 0.2 0.3 0.3 0.1
0.2 0.2 0.2
Soil temp. (°C)
12.51 12.92 12.07 12.53 12.37 12.48 3%
0.3 0.2 0.2 0.2 0.2 0.2
10.5 0.2 10.2 0.2 10.3 0.2
Soil mois. (m3/m3)
Precip. (mm)
CH4 uptake (lg CH4 m2 h1)
6.07 7.46 7.72 7.85 7.35 7.29 10%
0.2 0.4 0.2 0.2 0.2 0.3
327 494 402 312 397 386 32.3 19%
67.7 70.2 48.8 74.0 95.8 71.5 23.6%
n.a. 11.9 0.5 11.9 0.5
311 489 400 88.9
36.8 3.3 73.7 5.2 59.2 3.7
3.0 4.5 4.8 4.5 5.3 2.5
© 2014 John Wiley & Sons Ltd, Global Change Biology, 21, 777–788
C H 4 U P T A K E B Y A L P I N E G R A S S L A N D S D E S P I T E C O L D 781 observational period. The soil moisture showed clear seasonal pattern of approximately 4.99 0.65% in May, a decrease to 4.43 0.65% in June, and a maximum of 9.59% in August (Fig. 3). The climatic indices, especially precipitation and soil moisture, showed large interannual variation (Table 2). The coefficient of variation (CV) of cumulative precipitation and average soil moisture was 18.7% and 10% respectively, while for the ambient temperature and soil temperature these values were 5% and 3%, respectively. The highest level of precipitation occurred in 2010, but there was a long drought before the highly concentrated rainy season in August, when the soil moisture reached its maximum value (~25%) of the whole measurement period. The highest average air temperature and soil temperature (7.71 0.2 °C and 12.92 0.2 °C respectively) also occurred in 2010. The highest growing-season-averaged soil moisture content occurred in 2012 (7.85 0.2%) when the precipitation was decentralized during the months of June, July and August, although the cumulative precipitation was only 312 mm. A similar situation occurred in 2011 when the average soil moisture content was 7.72% and the ambient temperature and soil temperature were 6.65 0.2 °C and 12.07 0.2 °C respectively – the warmest growing season. It is also worth noting that in 2009, when the soil moisture content was only 6.07 0.2%, it was the driest growing season of the whole observational period.
(a)
(b)
(c)
Alpine steppe: seasonal pattern of CH4 uptake Fig. 4 Dependency of seasonal variation in CH4 uptake on (a) soil moisture, (b) temperature, and (c) their interactions in the alpine steppe.
precipitation concentrated between the months of July and August, but especially August (130 27.8 mm), while the average precipitation in May was only 24.0 4.1 mm. The growing season cumulative precipitation was 386 32.3 mm, and the average soil moisture content was 7.29 0.3% during the whole
The alpine steppe soils generally uptake CH4 during the growing seasons of the measurement period (Figs 2c and 3c) at 71.5 2.5 lg CH4 m2 h1. This shows a typical seasonal pattern of stronger CH4 uptake in the dry season (the highest monthly average was 101 lg CH4 m2 h1 in June) and a decline during the rainy season in July and August (lowest monthly average was 63.8 lg CH4 m2 h1 in August). Although the exact period differed among years, CH4 uptake generally followed the variation in
Table 3 Linear and nonlinear regression of CH4 uptake with soil moisture and temperature in the alpine steppe and alpine meadow at Nam Co. SM = Soil moisture; ST = Soil temperature
Alpine steppe Stepwise linear regression Nonlinear regression Alpine meadow Stepwise linear regression
Function
R2
P
CH4 uptake = 529 9 SM 115 CH4 uptake = 464 9 SM 2.68 9 ST 77.3 CH4 uptake = 6725 9 SM2 448 9 SM 87.0
0.21 0.31 0.24
