Quantification and biogeochemical modeling of C and N turnover processes and biosphere-atmosphere exchange of C and N compounds
Final Report Abstract
Summary of results (1st and 2nd project phase; 3-6 pages) First project phase: Microbial N-turnover processes at the different field sites During the first project phase, at four differently managed sites, i.e. ungrazed since 1979 (UG79), ungrazed since 1999 (UG99), winter grazed (WG) and overgrazed (OG), soil inorganic N content and gross rates of ammonification and nitrification were measured. Different management of the sites in form of different grazing regime did not result in significant differences in soil ammonium concentrations (0.1-5 mg N kg-1). At all sites, soil nitrate concentrations (0.4-14.5 mg N kg-1) were significantly higher than ammonium concentrations and nitrate concentrations at the UG79 site were significantly higher than at all other sites. During the vegetation period in 2005, ammonification rates were significantly higher and gross nitrification was higher at the ungrazed sites as compared to the grazed sites. Hence, microbial N turnover decreased due to the reduction of above- and below ground plant productivity by grazing. For further details see Holst et al. 2007, Ecosystems. Determination of the biosphere-atmosphere exchange of N2O, NO/NO2, NH3 Measurements of N2O fluxes from soil to atmosphere at the sites UG99 and WG did not show a significant difference. The level of N2O fluxes was very low at both sites with average values of 0.65±0.2 and 0.79±0.2 µg N m-2h-1 for UG99 and WG respectively during the growing season 2004. During the much drier growing season in 2005, average fluxes were 25-33% lower as compared to the flux level in 2004. This together with a watering experiment conducted in 2005 underlined the strong water limitation of this semi-arid ecosystem. Manual N2O flux measurements at the sites UG79 and OG were with 0.95±0.56 and 1.6±0.47 µg N m-2h-1 somewhat higher, but the differences were not significant. In 2006, measurements at the sites UG99 and WG started in March. First results revealed that N2O fluxes during spring thaw were distinctly stimulated at the ungrazed site. These results suggest that winter fluxes may dominate the annual N2O exchange in steppe ecosystems. This may be due to higher soil moisture levels at the ungrazed site as a consequence of snow capturing by the higher vegetation. If these results can be confirmed in the second phase, this would have important implications for our understanding of N2O fluxes from steppe ecosystems, since at present we assume that N2O fluxes from those ecosystems do not significantly contribute to the global atmospheric N2O budget. NO/NO2 flux measurements were conducted in August and September 2004. Ambient air concentrations of NO and NO2 were mostly below 1ppbv and NOx fluxes were always close to the detection limit of the system. P5 – Quantification and biogeochemical modeling of C- and N-turnover processes and biosphere-atmosphere exchange of C- and N-compounds 115 Gaseous losses of ammonia (NH3) were determined using concentration measurements at passive samplers located at different locations such as villages, single farmers and remote field sites. Measured air concentrations of ammonia varied considerable between time periods and sampling sites. Highest mean NH3 concentrations of up to 250 µg NH3 m-3 were measured close to a sheepfold in the village in the vicinity of the IMGERS, whereas lowest values were found for a remote site (up to 2.4 µg m-3) ~15 km from the nearest village. This indicates that the areas close to the village, where most of the sheepfolds are found, are sources of NH3, which is deposited at downwind areas. Summarizing the results of the N2O measurement campaigns, the hypothesis that grazing increases gaseous losses of nitrogen during the growing season could not be confirmed as no significant differences in N2O and NO emission between grazed and ungrazed sites were observed. For further details see Holst et al. 2007, Ecosystems, Holst et al. 2007, Plant and Soil, Holst et al., 2008, Plant and Soil, Liu et al., 2009, Agriculture, Ecosystem and Environment, and Holst et al. 2009, Biology and Fertility of Soils. Determination of the biosphere-atmosphere exchange of CH4 and CO2 The soils at all four sites (UG79, UG99, WG, OG) always functioned as a sink for atmospheric CH4. During late winter, when soil temperatures in 5 cm depth were still well below zero, CH4 uptake at UG99 and WG was in the range of 4-20 μg C m-2h-1. At both sites maximum CH4 uptake was observed in the summer months, with daily means for UG99 of up to 86 μg C m-2 h-1 and for WG of up to 51 μg C m-2h-1. Significant soil moisture effects on CH4 uptake, i.e. increasing uptake with decreasing soil moisture, were only found in the UG99 site. Our data show that in all investigated years and during all seasons except for the missing winter data, CH4 uptake was always lower at the WG site as compared to the UG99 site. In view of the comparable site history and knowing that differences in soil properties such as C content, C:N ratio etc. can be ascribed mainly to grazing, we have a strong argument that grazing is the reason for the observed reduction in CH4 uptake at the WG site. Comparing WG with UG99 we found: a) a slight, but not significant reduction in soil inorganic nitrogen, b) an increase in soil temperature by approx. 2 ºC (p<0.01) (reduced shading by vegetation), c) a significant decrease (p<0.01) in soil moisture during the period from start of thawing until the beginning of the rainy season in June by 40% (lower capturing of snow during winter times), but no differences in top soil bulk densities and organic carbon content. Therefore, P5 – Quantification and biogeochemical modeling of C- and N-turnover processes and biosphere-atmosphere exchange of C- and N-compounds 116 one could assume that CH4 uptake should be higher at the WG site as compared to the UG99 site and not vice versa. However, by characterizing soil properties in greater detail in cooperation with P8 we found that due to trampling of sheep the vertical and horizontal saturated hydraulic conductivity significantly decreased by 43.7% (p<0.01) and 44.0% (p<0.05) on the WG site as compared to the UG99 site. Furthermore, the air permeability was also significantly lower at the WG site (p<0.05) (WG: 55.5 cm day-1, UG99: 99.6 day-1). Both parameters support the hypothesis that a reduced pore continuity exists, that results in a reduction of gas diffusion. Nighttime soil CO2 efflux (respiration fluxes only) was measured at the UG99 and WG site with the automatic measuring system. The measurements in 2005 revealed a significantly higher CO2 efflux at the UG99 site throughout the entire growing season as compared to the WG site. Mean values for the whole measurement period in 2005 were 51.0 mg C m-2 h-1 and 23.6 mg C m-2 h-1 for the UG99 and the WG site, respectively. In late winter and early spring 2006 the efflux was low at both sites, but still significantly higher at the UG99 than at the WG site. Due to the very dry spring 2006 efflux rates did not rise above 20 mg C m-2 h-1 before end of May 2006, when noteworthy amounts of rain occurred for the first time in that year. Larger differences between UG99 and WG were not observed until end of June 2006. The mean values for the whole measurement period in 2006 were 17.9 mg C m-2 h-1 and 12.4 mg C m-2 h-1 for the UG99 and the WG site, respectively. Results of the first phase on exchange processes of CH4 and CO2 between steppe ecosystems and the atmosphere have been published by Liu et al., 2007, Atmospheric Environment, Liu et al., 2008, Advances in Atmospheric Science, and Liu et al., 2009, Atmospheric Environment. Landscape N- and C-trace gas fluxes and effect of landscape position In Inner Mongolia sheep are kept in sheepfolds overnight and during the dormant season, when they are fed with hay. Due to the concentration of high numbers of sheep in comparatively small areas (1-2 sheep m-2) large amounts of animal excreta accumulate on top of the soil, resulting in a thick layer of dung. This dung is “harvested” annually and piled up as dunghills. When the dung has become completely dry, it is used as fuel. As a result of high nutrient concentrations in the dung we hypothesized that the sheepfolds and dunghills could be a significant point source of the greenhouse gases CO2, CH4 and N2O. Therefore we conducted trace gas measurements on sheepfolds and dunghills in 2005 and 2006. The sheepfolds were strong sources of all three greenhouse gases, especially when it was wet P5 – Quantification and biogeochemical modeling of C- and N-turnover processes and biosphere-atmosphere exchange of C- and N-compounds 117 and warm. In late winter and early spring 2006 fluxes were close to zero due to a lack of soil moisture. The first rainfall end of May 2006 triggered the onset of trace gas emissions. Especially CH4 and N2O fluxes occasionally reached extremely high levels (e.g. for N2O up to 12 mg N m-2 h-1). Peak N2O emissions were up to three orders of magnitude higher than from the surrounding grassland. Our results show that sheepfolds cannot be neglected in a regional greenhouse gas balance of Inner Mongolian grassland. The effect of the landscape position (top, slope, bottom) on CO2, CH4 and N2O soilatmosphere exchange was determined in 2005. At nine locations along a height-gradient from the top of a hill (= position 1) down to the edge of the Xilin river (= position 9), manual chamber measurements were performed at eight different times throughout the growing season. Fluxes of all three trace gases along the hill were at a comparable level as in flat grassland. For CO2 only at the two positions closest to the river significantly higher emissions were observed. CH4 uptake rates at positions 7 and 8 were already close to zero. At position 9 CH4 emissions were two orders of magnitude higher than CH4 uptake rates in dry steppe were observed. For N2O fluxes no significant differences between landscape positions could be observed. Further details on landscape fluxes of N- and C-trace gases are summarized in Liu et al., 2009, Atmospheric Environment and Holst et al., 2007, Plant and Soil. Second project phase: Results that were achieved in the second project phase were based on three main experiments: 1. Year-round measurement of biosphere-atmosphere exchange of trace gases at differently grazed sites (2007/2008) 2. Regionalization and Parameterization experiments with intact soil cores from 6 land use classes in the Xilin-River catchment (August 2007) 3. Ultralight aircraft campaign (June to August 2009) Year-round measurements of trace gas fluxes: In the second project phase, year round measurements of trace gas fluxes were conducted at differently grazed typical steppe sites. A preliminary study during the first project phase had indicated the relevance of the spring thaw period for annual N2O emission. The yearround measurements revealed vigorous N2O emission at ungrazed sited when soils thawed after the long and cold winter period. Though these distinctly increased emission rates lasted only for about six weeks, they dominated the annual N2O emission of ungrazed steppe. The P5 – Quantification and biogeochemical modeling of C- and N-turnover processes and biosphere-atmosphere exchange of C- and N-compounds 118 relevance of the spring thaw period for the annual emission decreased with increasing stocking rate, indicating that grazing rather reduced N2O emission form steppe compared to ungrazed conditions. This was due to a lower snow accumulation on grazed plots and the reduced insulation of the soil against low air temperatures at lower snow cover. Lower soil temperatures hampered re-growth of the microbial community of which 80% had died off after the first frost in late autumn and resulted in lower microbial N turnover and associated N2O emission at grazed sites. These results were published in Nature by Wolf et al., 2010 and by Wolf et al., 2011, Journal Plant Nutrition and Soil Science. In contrast to the N2O emission, CH4 uptake by steppe soil was not altered during spring thaw. As even the sustained high water content during spring thaw did not result in net CH4 production in the soil, steppe soil acted exclusively as a sink for atmospheric methane. Even during the non-growing season, significant amounts (36% and 26% of the annual emission for the sites UG99 and WG respectively) of atmospheric methane were consumed in steppe soil. This finding indicated that the contribution of CH4 uptake during winter and autumn had been underestimated so far in general and for steppe ecosystems in particular. During the year, CH4 uptake was mainly controlled by temperature and decreased sharply after summer rainfall events. Light and moderate grazing did not change annual CH4 uptake significantly compared to ungrazed conditions, while heavy grazing reduced CH4 uptake significantly by approx. 28%. These findings imply that easing the pressure of heavily grazed steppes (e.g. moving to light or moderate stocking rates) would help restore steppe soil sinks for atmospheric CH4. More details are provided by Chen et al., 2010, Journal of Geophysical Research and Chen et al., 2011, Global Change Biology (in press) Since grazing sheep are kept in sheepfolds overnight, trace gas fluxes from these sheepfolds have to be assessed as well if the effect of grazing on trace gas emissions is to be analysed. Flux rates of N2O from sheepfolds were highest in presence of sheep, after rainfall and during spring thaw and were up to three orders of magnitude higher than flux rates measured in pastures or ungrazed steppe. For the Baiyinxile administrative region, data on animal numbers, pasture area and area of sheepfolds were available, so that the importance of sheepfolds for the regional N2O budget could be calculated. Based on these calculations, N2O emission form sheepfolds were identified as a significant regional source of N2O that can account for up to one third of the regional N2O budget. These additional N2O emissions need to be considered for a complete evaluation of livestock farming since the effect of grazing on emissions from pastures and sheepfolds have different signs. While CH4 was consumed by steppe soils and pastures, sheepfolds mainly emitted CH4, but average emission was at most factor 20 higher than the absolute of average methane P5 – Quantification and biogeochemical modeling of C- and N-turnover processes and biosphere-atmosphere exchange of C- and N-compounds 119 uptake. Therefore, the importance of CH4 emission form sheepfolds was negligible for both the total greenhouse gas balance of sheepfolds and the regional CH4 budget. These results were summarized in more detail by Chen et al., 2011, Plant and Soil. The total greenhouse gas balance of livestock farming in the Xilin-River catchment was calculated in cooperation with subprojects P1, P3 and P4. This total greenhouse gas (GHG) balance showed that grazing turned steppe from a sink to a source of GHG with GHG emission from pastures dominating the budget (Schönbach et al., 2011, Agriculture, Ecosystem and Environment (under review)). The observations made during the year-round measurement campaign were also crucial for the further development of the MOBILE framework. First simulations with the DNDC modules in the MOBILE framework had shown that soil moisture conditions during spring thaw were distinctly underestimated by the model, and the peak emissions of N2O during spring thaw were entirely missed. The underestimation of soil moisture during spring thaw was due to an overestimation of water redistribution in the soil profile during spring thaw when the soil was still frozen. This could be overcome by consideration of the reducing effect of ice in the soil matrix on hydraulic conductivity. The realistic simulation of soil moisture was followed by partly anoxic conditions in the topsoil, which had been observed in the field as well. Limited oxygen availability results in increased denitrification in the model and thus caused peak N2O emission during spring thaw. The ability of the model to reproduce these vigorous N2O emissions that may dominate the N2O budget constitutes a prerequisite for regional modelling of trace gas emissions. This study is currently under review (Wolf et al., 2011, Plant and Soil). Regionalization and Parameterization experiments with intact soil cores from six land use classes In August 2007, intact soil cores were taken from six representative land use classes within the Xilin-river catchment in order to identify effects of land use and land cover on trace gas fluxes. Per sampled site, six soil cores were taken of which 3 were used to estimate the midsummer green house gas (GHG) budget in the Xilin-River catchment. This was achieved by immediate analysis of the trace gas fluxes at IMGERS research station which revealed that N2O, CH4 and CO2 flux rates during midsummer were not significantly different between land use and land cover classes and that CO2 fluxes may dominate the global warming potential of midsummer soil-atmosphere GHG fluxes. As this midsummer-GHG budget represented only a snapshot for midsummer conditions, the remaining three intact soil cores P5 – Quantification and biogeochemical modeling of C- and N-turnover processes and biosphere-atmosphere exchange of C- and N-compounds 120 of each site were transported to the IMK-IFU in Garmisch-Partenkirchen and were incubated under different moisture and temperature treatments, thereby determining the GHG fluxes. With respect to N2O fluxes, simulated drying-rewetting cycles as well as freeze/thaw transitions were followed by pluse emissions from soil cores taken from typical steppe, mountain meadow, sand dune and marshland. This indicates that the transition period from winter to spring, i.e. the thaw period, may be of importance for the annual budget not only in typical steppe but also in other land-use or land-cover classes. However, the magnitude of N2O emission was not significantly different between land-use / land-cover classes as soil moisture, temperature and microbial biomass were the determining factors controlling N2O emission. With regard to CH4 fluxes, no significant differences were found for typical steppe soils, sand dunes and mountain meadow soils, while methane uptake was significantly lower for marshland soils and even positive for water saturated marshland soils. Therefore and in accordance with N2O fluxes, temperature and moisture conditions overwrote land-use and land-cover information. Emission of CH4 by saturated marshland soils was also observed during the measurements carried out in the context of the midsummer GHG budget. Owing to their partly strong CH4 emission by saturated marshland soils, this indicates the regional importance of saturated marshland soils. Also for CO2 emission, significant effects of soil moisture and temperature were discovered, but in contrast to N2O and CH4 fluxes, CO2 fluxes were significantly different between all analysed land-use/land-cover classes except for typical steppe soil and sand dunes which were not significantly different. Highest fluxes were observed for cores from mountain meadow soils, followed by cores from typical steppe, sand dune and marshland. Also in the case of CO2, temperature and moisture effects as well as short term events like drying/rewetting partly ruled out determining factors such as microbial biomass and soil organic carbon (SOC) content. Results of field and laboratory studies have been published by Yao et al., 2010, Journal of Geophysical Research, Yao et al., 2010, Plant and Soil, and Wu et al., 2010, Soil Biology and Biochemistry. Ultralight aircraft campaign During summer 2009, an airborne campaign was performed to investigate regional scale fluxes of momentum, heat, CO2 and water. The aim of this campaign was to provide spatial information for the extension of the local ground based measurements deployed within the MAGIM research area to a regional scale. Within the campaign between June 12 and August P5 – Quantification and biogeochemical modeling of C- and N-turnover processes and biosphere-atmosphere exchange of C- and N-compounds 121 8, 2009 46 individual flight experiments were performed and simultaneously to the airborne operations, half-hourly tower eddy covariance fluxes were measured at 3 control sites, two of them non-grazed (C4 dominated Leymus, C3 dominated Stipa) and one heavily grazed (HG). Airborne soundings followed multiple repetitions of line transects, organized in vertical stacks, typically around sun apex. The flight altitudes were 50 m, 100 m and 150 m above ground, the horizontal extend of the patterns ranged from 2 to 80 km. Operational results at lowest flight level (AC50) averaged to sensible heat flux, QH = -177±78 Wm-2, latent heat flux, QE = -78±40 Wm-2 and carbon dioxide flux, QC = 0.02±0.2 mg CO2 m- 2s-1 throughout the campaign. At an average radiation budget QS* = -412±86 Wm-2, which is 16 % lower than the control sites, the energy balance from AC50 measurements was closed to 72±18 %. AC50 measured QH was found to be comparable in magnitude to the HG control site, whereas Leymus and Stipa sites were 30 % higher on average. The magnitude of QE was comparable to the Leymus site, whereas QE from Stipa and HG were ≥35 % higher on average. This relation is remarkable, since the soil moisture on HG was actually lower (9 %vol) compared to the non-grazed sites (14 %vol). AC50, Stipa and HG measurements indicate marginal CO2 release / uptake, only Leymus displayed significant CO2 assimilation (- 0.07±0.3 mg CO2 m-2s-1). The chronology of AC50 QH, QE, QC measurements throughout the campaign correlated best to Stipa, Stipa / HG and Leymus site, respectively. During postprocessing spectral behaviour will be assessed by means of conditional wavelet analysis. In combination with footprint analysis measured fluxes will also be related to surface features so as to allow the creation and evaluation of surface flux maps. Manuscript on results of the Ultralight flight campaign, the underlying flux calculations and the regionalization will be finalized within the next two months or have already been submitted by Metzger et al., 2011 to Atmospheric Measurement Techniques (manuscript available at online discussion).
Publications
- (2007): Importance of point sources on regional nitrous oxide fluxes in semi-arid steppe of Inner Mongolia, China. Plant and Soil, 296, 209-226
Holst, J., Liu, C., Yao, Z., Bruggemann, N., Zheng, X., Han, X. Butterbach-Bahl, K.
(See online at https://doi.org/10.1007/s11104-007-9311-8) - (2007): Microbial N turnover and N-oxide (N2O/NO/NO2) fluxes in semi-arid grassland of Inner Mongolia. Ecosystems, 10, 623-634
Holst, J., Liu, C.Y., Bruggemann, N., Butterbach-Bahl, K., Zheng, X.H., Wang, Y.S., Han, S.H., Yao, Z.S., Yue, J., Han, X.G.
(See online at https://doi.org/10.1007/s10021-007-9043-x) - (2007): Winter-grazing reduces methane uptake by soils of a typical semi-arid steppe in Inner Mongolia, Chinaand Atmospheric Environment, 41, 5948-5958
Liu, C.Y., Holst, J., Bruggemann, N., Butterbach-Bahl, K., Yao, Z.S., Yue, J., Han, S.H., Han, X., Krummelbein, J., Horn, R., Zheng, X.H.
(See online at https://doi.org/10.1016/j.atmosenv.2007.03.017) - (2008): Effects of Irrigation on Nitrous Oxide, Methane and Carbon Dioxide Fluxes in an Inner Mongolian Steppe. Advances in Atmospheric Sciences, 25,748-756
Liu, C.Y., Holst, J., Bruggemann, N., Butterbach-Bahl, K., Yao, Z.S., Han, S.H., Han, X.G., Zheng, X.H.
(See online at https://doi.org/10.1007/s00376-008-0748-3) - (2008): Fluxes of nitrous oxide, methane and carbon dioxide during freezing-thawing cycles in an Inner Mongolian steppe. Plant and Soil, 308,105-117
Holst, J., Liu, C.Y., Yao, Z., Brueggemann, N., Zheng, X., Giese, M., Butterbach-Bahl, K.
(See online at https://doi.org/10.1007/s11104-008-9610-8) - (2010): Grazing-induced reduction of natural nitrous oxide release from continental steppe. Nature, 464, 881-884
Wolf, B., Zheng, X.H., Brueggemann, N., Chen, W.W., Dannenmann, M., Han, X.G., Sutton, M.A., Wu, H.H., Yao, Z.S., Butterbach-Bahl, K.
(See online at https://doi.org/10.1038/nature08931) - (2009): Dinitrogen fixation by biological soil crusts in an Inner Mongolian steppe. Biology and Fertility of Soils, 45, 679-690
Holst, J., Butterbach-Bahl, K., Liu, C.Y., Zheng, X.H., Kaiser, A.J., Schnitzler, J.P., Zechmeister-Boltenstern, S., Bruggemann, N.
(See online at https://doi.org/10.1007/s00374-009-0378-7) - (2009): Sheepfolds as "hotspots" of nitric oxide (NO) emission in an Inner Mongolian steppe. Agriculture, Ecosystems & Environment, 134, 136-142
Liu, C.Y., Holst, J., Yao, Z.S., Bruggemann, N., Butterbach-Bahl, K., Han, S.H., Han, X.G., Zheng, X.H.
(See online at https://doi.org/10.1016/j.agee.2009.06.007) - (2010): Annual emissions of greenhouse gases from sheepfolds in Inner Mongolia. Plant and Soil, 340 (1-2), 291-301
Chen, W.W., Wolf, B., Bruggemann, N., Butterbach-Bahl, K., Zheng, X.H.
(See online at https://doi.org/10.1007/s11104-010-0367-5) - (2010): Annual methane uptake by typical semiarid steppe in Inner Mongolia. Journal of Geophysical Research – Atmospheric, 115
Chen, W.W., Wolf, B., Yao, Z.S., Bruggemann, N., Butterbach-Bahl, K., Liu, C.Y., Han, S.H., Han, X.G., Zheng, X.H.
(See online at https://doi.org/10.1029/2009JD013783) - (2010): Effects of soil moisture and temperature on CO2 and CH4 soil atmosphere exchange of various land use/cover types in a semi-arid grassland in Inner Mongolia, China. Soil Biology& Biochemistry, 42, 773-787
Wu, X., Yao, Z., Bruggemann, N., Shen, Z.Y., Wolf, B., Dannenmann, M., Zheng, X., Butterbach-Bahl, K.
(See online at https://doi.org/10.1016/j.soilbio.2010.01.013) - (2010): Soil-atmosphere exchange potential of NO and N2O in different land use types of Inner Mongolia as affected by soil temperature, soil moisture, freeze-thaw, and drying-wetting events. Journal of Geophysical Research –Atmospheric,115, D17116
Yao, Z.S., Wu, X., Wolf, B., Dannenmann, M., Butterbach-Bahl, K., Brueggemann, N., Chen, W.W., Zheng, X.
(See online at https://doi.org/10.1029/2009JD013528) - (2010): Spatial variability of N2O, CH4 and CO2 fluxes within the Xilin River catchment of Inner Mongolia, China: a soil core study. Plant and Soil, 331, 341-359
Yao, Z.S., Wolf, B., Chen, W.W., Butterbach-Bahl, K., Bruggemann, N., Wiesmeier, M., Dannenmann, M., Blank, B., Zheng, X.H.
(See online at https://doi.org/10.1007/s11104-009-0257-x) - (2011): Applicability of the soil gradient method for estimating soil-atmosphere CO2, CH4 and N2O fluxes for steppe soils in Inner Mongolia. Journal of Plant Nutrition and Soil Science
Wolf, B., Chen, W.W., Brueggemann, N., Zheng, X.H., Pumpanen, J., Butterbach-Bahl, K.
(See online at https://doi.org/10.1002/jpln.201000150) - (2011): Feedback of grazing on gross rates of N mineralization and inorganic N partitioning in stepe soils of Inner Mongolia. Plant and Soil, 340, 127-139
Wu, H.H., Dannenmann, M., Fanselow, N., Yao, Z.S., Wu, X., Brueggemann, N., Zheng, X.H., Han, X., Dittert, K., Butterbach-Bahl, K.
(See online at https://doi.org/10.1007/s11104-010-0575-z) - (2011): Steppe ecosystems and climate and land-use changes - vulnerability, feedbacks and possibilities for adaptation. Plant and Soil, 340 (1-2), 1-6
Butterbach-Bahl, K., Kogel-Knabner, I., Han, X.G.
(See online at https://doi.org/10.1007/s11104-010-0651-4)
