Publications

Approximately 1.5 million km² of the Tibetan Plateau are covered with grasslands. Thereof one third is occupied by the world’s largest pastoral alpine ecosystem (Kobresia pastures). Paleo-records indicate the grazing-induced origin of this ecosystem since more than 8000 years or at least since yak domestication since 4000 years. Long-term moderate grazing by yak and sheep increased belowground C allocation of Kobresia pygmaea, caused the development of dense root-mats and finally lead to an accumulation of soil organic carbon (SOC) and nutrients such as nitrogen (N) and phosphorus (P) in the topsoil. These pastures, however, are increasingly affected by large-scale degradation caused by overgrazing of these highly sensitive ecosystems. Loss of the topsoil threatens several ecosystem functions: i.e. SOC and nutrient storage, biodiversity, provision of grazing-ground and supply of clean water for large parts of SE-Asia. Here, we present a conceptual model and results of degradation processes combining anthropogenic and natural amplifications. To evaluate losses of SOC and nutrients we synthesize field observations and surveys in the highlands and validates this with own analyses in the Kobresia core area. We show that drought- and frost-induced polygonal cracking opens the root-mats, already weakened by overgrazing. This initiates a dying of the Kobresia turf, extends the surface cracks, triggers soil erosion and promotes SOC mineralization and leaching losses. Soil erosion caused further high losses of SOC and nutrients from the topsoil (i.e. 0-10 cm: ~5.1 kg C m-2), whereas SOC loss beneath the surface cracks was primary caused by both, decreasing C-input and SOC mineralization (mineralization-derived SOC loss: ~2.5 kg C m-2). The root biomass decreased with intensity of pasture degradation and lower C input constrains the ecosystem recovery. A negative δ13C shift of SOC reflected intensive decomposition and corresponded to a relative enrichment of 13C depleted lignin components. In sum, degradation triggered high SOC loss (up to 70% of intact soil in 0-30 cm: ~7.6 kg C m-2) from this ecosystem with profound consequences for carbon sequestration, atmospheric CO2, water quality and ecosystem stability.

Plant roots interact with rhizosphere microorganisms to accelerate soil organic matter (SOM) mineralization for nutrient acquisition. Root-mediated changes in SOM mineralization largely depend on root-derived carbon (root-C) input and soil nutrient status. Hence, intraspecific competition over plant development and spatiotemporal variability in the root-C input and nutrients uptake may modify SOM mineralization. To investigate the effect of intraspecific competition on SOM mineralization at three growth stages (heading, flowering, and ripening), we grew maize (C4 plant) under three planting densities on a C3 soil and determined in situ soil C- and N-mineralization by 13 C-natural abundance and 15 N-pool dilution approaches. From heading to ripening, soil C- and N-mineralization rates exhibit similar unimodal trends and were tightly coupled. The C-to-N-mineralization ratio (0.6 to 2.6) increased with N availability, indicating that an increase in N-mineralization with N depletion was driven by microorganisms mining N-rich SOM. With the intraspecific competition, plants increased specific root lengths as an efficient strategy to compete for resources. Root morphologic traits rather than root biomass per se were positively related to C- and N-mineralization. Overall, plant phenology and intraspecific competition controlled the intensity and mechanisms of soil C- and N- mineralization by the adaptation of root traits and nutrient mining.

<p>The Paris climate agreement is pursuing efforts to limit the increase in global temperature to below 2 °C above pre-industrial level. The overall consequence of relatively slight warming (~2 °C), on soil C and N stocks will be dependent on microorganisms decomposing organic matter through release of extracellular enzymes. Therefore, the capacity of soil microbial community to buffer climate warming in long-term and the self-regulatory mechanisms mediating soil C and N cycling through enzyme activity and microbial growth require a detailed comparative study. Here, microbial growth and the dynamics of enzyme activity (involved in C and N cycling) in response to 8 years warming (ambient, +1.6 °C, +3.2 °C) were investigated to identify shifts in soil and microbial functioning. A slight temperature increase (+1.6 °C) only altered microbial properties, but had no effect on either hydrolytic enzyme activity or basic soil properties. Stronger warming (+3.2 °C) increased the specific growth rate (μ<sub>m</sub>) of the microbial community, indicating an alteration in their ecological strategy, i.e. a shift towards fast-growing microorganisms and accelerated microbial turnover. Warming strongly changed microbial physiological state, as indicated by a 1.4-fold increase in the fraction of growing microorganisms (GMB) and 2 times decrease in lag-time with warming. This reduced total microbial biomass but increased specific enzyme activity to be ready to decompose increased rhizodeposition, as supported by the higher potential activitiy (V<sub>max</sub>) and lower affinity to substrates (higher K<sub>m</sub>) of enzymes hydrolyzing cellobiose and proteins cleavage in warmed soil. In other words, stronger warming magnitude (+3.2 °C) changed microbial communities, and was sufficient to benefit fast-growing microbial populations with enzyme functions that specific to degrade labile SOM. Combining with 48 literature observations, we confirmed that the slight magnitude of temperature increase (< 2 °C) only altered microbial properties, but further temperature increases (2-4 °C) was sufficient to change almost all soil, microbial, and enzyme properties and related processes. As a consequence, the revealed microbial regulatory mechanism of stability of soil C storage is strongly depended on the magnitude of future climate warming.</p>

The effect of three land use types on decomposition of 14C-labelled maize (Zea mays L.) residues and soil organic matter were investigated under laboratory conditions. Samples of three Dystric Cambisols under plow tillage (PT), reduced tillage (RT) and grassland (GL) collected from the upper 5cm of the soil profile were incubated for 159days at 20°C with or without 14C-labelled maize residue. After 7days cumulative CO2 production was highest in GL and lowest in PT, reflecting differences in soil organic C (SOC) concentration among the three land use types and indicating that mineralized C is a sensitive indicator of the effects of land use regime on SOC. 14CO2 efflux from maize residue decomposition was higher in GL than in PT, possibly due to higher SOC and microbial biomass C (MBC) in GL than in PT. 14CO2 efflux dynamics from RT soil were different from those of PT and GL. RT had the lowest 14CO2 efflux from days 2 to 14 and the highest from days 28 to 159. The lowest MBC in RT explained the delayed decomposition of residues at the beginning. A double exponential model gave a good fit to the mineralization of SOC and residue-14C (R 2 >0.99) and allowed estimation of decomposition rates as dependent on land use. Land use affected the decomposition of labile fractions of SOC and of maize residue, but had no effect on the decomposition of recalcitrant fractions. We conclude that land use affected the decomposition dynamics within the first 1.5months mainly because of differences in soil microbial biomass but had low effect on cumulative decomposition of maize residues within 5months.

Salix cupularis is a key species for restoring desertified alpine meadows in the eastern Tibetan Plateau. However, the mechanisms driving soil restoration under S. cupularis remain unclear due to complex rhizosphere processes and harsh high-altitude conditions. This study investigated root exudation, soil carbon (C) and nitrogen (N) contents, enzyme activities, and microbial biomass across shoot biomass levels (high, medium, low), phenological stages (regreening, fruiting, withering), and plant gender (male, female) in S. cupularis. Root exudation was identified as a critical driver of increased rhizosphere soil C and N, unaffected by plant gender. Root exudation rates peaked at the fruiting stage, and the C exudation rate increased with shoot biomass. While S. cupularis with high shoot biomass showed greater rhizosphere and bulk soil organic C (SOC) content compared to low shoot biomass plants. Intensive root exudation enhanced the contents of dissolved organic C (DOC), NH₄+, microbial biomass C (MBC) and microbial biomass N (MBN), as well as extracellular enzyme activities in rhizosphere soil compared to bulk soil, with their highest levels observed at the fruiting stage. No significant effect of plant gender was discovered. Structural equation modeling revealed two main pathways for increasing rhizosphere C and N which are direct root exudates inputs and microbial biomass accumulation driven by root exudates. These findings highlight shrub shoot biomass as a key factor in soil restoration, emphasizing the roles of root exudates and microbial biomass formation in enhancing soil C and N during desertified alpine meadow recovery.

European beech (Fagus sylvatica L.) is the prevailing tree species of mesic forests in Central Europe. Increasing summer temperatures and decreasing precipitation, as climate change scenarios predict, may, however, negatively influence beech growth and induce a shift to more thermophilous forest communities. Temperatures as expected in the future for western Central Europe are currently found in parts of western Romania. In light of this climate analogy we investigated forest vegetation as an indicator for future vegetation changes in five regions of western Romania representing a climatic gradient. We related species composition to climate variables and examined if tree and understorey species composition respond similarly to the climatic gradient. We further analysed if tree species occurrences correspond with their modelled distance to the rear niche edge. We found evidence for climatic effects on vegetation composition among regions as well as within deciduous and pine forests, respectively. This underlines that vegetation composition is a useful indicator for environmental change. Tree and understorey species compositions were closely linked showing that community-based characterization of forest stands can provide additional information on tree species suitability along environmental gradients. Both, vegetation composition and a climatic marginality index demonstrate the rear niche edge occurrence of beech in the studied sites of Romania and can predict the site suitability for different tree species. While vegetation surveys indicate Quercus petraea to be associated to moderately mesic forests, the marginality index suggested an inner niche position of sessile oak along the climatic gradient. Phytosociological relevés that differentiate between subspecies (or microspecies) of sessile oak with differing habitat requirements should be considered to complement national forest inventories and species distribution maps when modelling rear distribution edges. We conclude that climate driven forest vegetation composition in western Romania is a suitable analogon and may indicate future forest development in western Central Europe.

No abstract is provided for this article.

The globally expanding colonization of grasslands by shrubs increases soil organic carbon and nitrogen, but the effects of shrubs on phosphorus (P) cycling have been rarely studied. We compared P contents in roots and soil fractions, phosphatase activity in the 1 m profile, and in situ net P mineralization between shrubby Potentilla fruticosa patches and grassy interspaces in grazed shrubby meadows at three representative sites on the eastern Tibetan Plateau. The P uptake of P. fruticosa exceeded 1 m soil depth, whereas grasses acquired P mainly within the upper 0.6 m. The P contents in shoots, aboveground litter and roots under P. fruticosa were greater than those under grasses. Litter P stock under the shrubs was 4–8 times higher than that under grasses and the root P stock doubled compared to that in grass areas. P. fruticosa generally increased the organic P (OP) content in the topsoil but decreased inorganic P (IP) in the subsoil. Phosphorus availability increased in the topsoil but decreased in the subsoil under the shrubs compared to grasses. Microbial biomass P (MBP), the activities of acid and alkaline phosphatases, and OP lability were all greater in the 1 m soil under P. fruticosa than grasses, leading to faster P mineralization and the P turnover under the shrubs. In the 1 m soil, P. fruticosa increased MBP and OP stocks but decreased IP and available P stocks. The larger and deeply distributed root system of P. fruticosa improved its P uptake ability especially from the subsoil. The subsequent greater organic matter input through litter fall and root turnover under P. fruticosa fed a larger microbial biomass that synthesized more microbial-derived OP in the topsoil. We concluded that shrubs increase the biological (plant and microbial) P transformation in the soil, the P uplift in the profile, and P cycling in shrubby grassland ecosystems. Such mechanisms structuring spatial heterogeneity of P content, transformation, turnover and fluxes are common in shrubby grasslands worldwide.

Livestock and poultry farming sectors are among the largest anthropogenic methane (CH4) emission sources, mainly from enteric fermentation and manure management. Previous inventories of CH4 emission were generally based on constant emission factor (EF) per head, which had some weaknesses mainly due to the succession of breeding and feeding systems over decades. Here, more reliable long-term changes of CH4 emissions from livestock and poultry farming in Beijing are estimated using the dynamic EFs based on the Intergovernmental Panel on Climate Change (IPCC) Tier 2 method, and high-resolution spatial patterns of CH4 emissions are also estimated with intensive field survey. The results showed that the estimated CH4 emissions derived by dynamic EFs were approximately 13–19% lower than those based on the constant EF before 2010. After 2011, however, the dynamic EFs-derived CH4 emissions were a little higher (3%) than the constant EF method. Temporal CH4 emissions in Beijing had experienced four developing stages (1978–1988: stable; 1989–1998: slow growth; 1999–2004: rapid growth and reached hot moments; 2005–2014: decline) during 1978–2014. Over the first two decades, the contributions of pigs (45%) and cattle (46%) to annual CH4 emission were similar; subsequently, the cattle emitted more CH4 compared to the pigs. At a spatial scale, Shunyi, Daxing, and Tongzhou districts with more cattle and pigs are the hotspots of CH4 emission. In conclusion, the dynamic EFs method obviously improved the spatio-temporal estimates of CH4 emissions compared to the constant EF approach, and the improvements depended on the period and aquaculture structure. Therefore, the dynamic EFs method should be recommended for estimating CH4 emissions from livestock and poultry farming in the future.