Shrubification is expanding in the alpine meadows of the Qinghai-Tibetan Plateau. Several studies have examined its effects on soil carbon (C), nitrogen (N), and phosphorus (P) content, but the stoichiometric response remains less explored, especially in deeper soils. We investigated the horizontal spatial heterogeneity of soil C:N, C:P, and N:P ratios through grid sampling down to a 1.2 m soil profile in a grazed alpine shrubby meadow dominated by grass patches and four expanding shrub species. We also analyzed the C:N:P stoichiometry in leaves and litter during the growing season. Root and microbial biomass C:N:P stoichiometry were measured throughout the soil profile. Shrub-induced stronger spatial heterogeneity of soil C:P and N:P ratios was detected to a depth of 1.2 m across the plot. Soil C:P and N:P ratios increased with plant size (height and crown diameter), but soil C:N ratios did not differ among vegetation types. Microbial C:N and C:P ratios were lower under shrubs than under grasses, corresponding to the lower C:nutrient ratios observed in leaves, litter, and roots of shrubs relative to grasses. Therefore, shrubs growing in alpine meadows amplify the C:N and C:P stoichiometric imbalance between plant-microbe and soil. Furthermore, the trend of increasing soil organic C stock weakened with higher soil C:P and N:P ratios. We conclude that shrubification increases soil C:P and N:P ratios to a depth of at least 1.2 m-a change that may constrain long-term C sequestration potential by intensifying P limitation.

We investigated whether shrubification effects on litter decomposition differ among shrub species and plant organs in alpine meadows. Two dominant shrub species of Potentilla fruticosa and Sibiraea angustata were used as study species. Their leaves and fine roots were incubated in litter bags (on the soil surface and in the soil profile) for 577 days at both home-field and common garden sites. P. fruticosa exhibited lower decomposition rates for leaves and roots than grasses, whereas S. angustata had a higher decomposition rate only for leaves. Mixtures of P. fruticosa and grass leaves/roots decomposed more slowly than grass leaf or root litter alone, while mixtures of S. angustata and grass leaves decomposed faster than grass leaf litter alone. Across all plant communities, fine roots in the soil decomposed more slowly overall than leaves on the soil surface. Decomposition at home-field versus common garden sites did not alter the patterns of decomposition rate variation among plant communities or organs. Acid detergent fibre (ADF) content is the most important predictor for leaf and root litter decomposition. Litter decomposition rate decreased with increasing ADF content. In addition to higher ADF content, lower nitrogen content in fine roots further reduced root decomposition rates relative to leaf decomposition rates. The effects of shrubification on litter decomposition rates differ by shrub species and plant organ, and cannot be generalized. When assessing the impacts of shrubification on soil carbon and nutrient contents, the species- and organ-specific effects of shrubs on litter decomposition must be considered.

Intercropping green manure optimizes the nitrogen (N) rate in high-yield maize production. However, the effect of cropping pattern on soil organic carbon (SOC) accumulation is unclear. This study investigated the interaction effects between the N rate and green manure intercropping on the SOC biochemical composition of a maize field in an oasis region of China using amino sugars, lignin phenols, and phospholipid fatty acids (PLFA) as biomarkers. Without green manure crops, decreasing N rate from 360 kg ha-1 (conventional) to 270 kg ha-1 (reduced) for 7 years decreased SOC and total N stocks in the 0.4 m upper soil by 20%-24%. In contrast, with green manure intercropping, these reductions were of negligible magnitude. Decreases in SOC and total N under reduced N fertilization in areas without green manure were linked to microbial necromass C (MNC) depletion. In plots with green manure, this loss was compensated for by MNC and lignin accumulation. The ratio between fungal- and bacterial-derived MNC was evidently lower in the plots without green manure but was maintained in plots with green manure under reduced N fertilization compared with conventional N fertilization. Reduced N did not affect microbial biomass or the fungal-to-bacterial PLFA ratio, but decreased the ratio between C- to Nacquiring enzyme activities and N availability. Intercropping green manure increased soil microbial biomass, the ratio of fungal- to bacterial-derived PLFAs, and the ratio of C- to N-acquisition enzyme activities, and buffered the decrease in N availability caused by reduced N. These results indicate that in N fertilizer-reduced plots, decreases in soil MNC without green manure were ascribed to increased mineralization of microbial (particularly fungal) debris because of intense N limitation, whereas the maintenance of MNC with green manure was associated with increased microbial (particularly fungal) debris input and reduced mineralization because of alleviated microbial N limitation. We demonstrated the link between soil MNC accumulation and N status in croplands, showing that N rate can be optimized by intercropping green manure in maize production, which increases SOC formation by reducing N loss and inputting high-quality organic components.

Abandonment is a strategy applied to increase soil organic C (SOC) in degraded cropland, but such efforts may fail because of microbial N limitation after abandonment in the absence of fertilization. In this study, we investigated the associations between SOC and microbial necromass C (MNC) dynamics in bulk soil and particlesize pools with N availability in a cropland abandonment chronosequence on the Loess Plateau. The total SOC, total MNC, and their particulate fractions (> 0.05 mm) in soil declined in the first eight years after cropland abandonment, but increased thereafter. By the 23rd year, the SOC content in abandoned soils increased towards the levels of cropland (16.5 g kg(-1)) but were still far lower than those of natural vegetation (21.5 g kg(-1)). The mineral-associated SOC (< 0.05 mm) content maintained after abandonment; but by contrast, the mineralassociated MNC profoundly decreased. This indicated that the reduction in MNC in this fraction was compensated for by plant-derived substances from the particulate fraction. Enzymatic stoichiometry analysis identified microbial N limitations in abandoned soils compared with cropland soils. As such, microbial N limitation led to increases in mineralization and/or decreases in synthesis of MNC in both particulate and mineral-associated fractions after abandonment, attributable to the decreased total SOC. Across the abandonment chronosequence, up to 20 % of particulate SOC was derived from microbes, whereas more than half of mineral-associated SOC came from plants. These findings challenge the general consensus that particulate SOC is dominated by plant residues whereas the mineral-associated fraction contains mainly microbially derived substances. The MNC contained a smaller proportion of fungal substances in mineral-associated fractions compared to particulate fractions, reflecting microbial ecological niche differentiation in the SOC formation between particle-size fractions. In conclusion, cropland abandonment decreased MNC accumulation because of microbial N limitation, and the mineral-associated SOC was stable in quantity but not in its source composition.

Insufficient precipitation and low temperatures can restrict grain yield but not necessarily vegetative growth in cold–arid regions. This indicates that forage production may be more suitable than grain cultivation in these environments while also meeting the increasing demand for livestock products. In this study, we compared the effects of cultivating forage maize (Zea mays L.) and forage oat (Avena sativa L.) with those of traditional grain crops, such as potato (Solanum tuberosum L.) and wheat (Triticum aestivum L.), in terms of aboveground biomass, crude protein yield, and water use efficiency (WUE). Across the four-year study, the results showed that aboveground biomass increased by 26–125% with oat (9.10 t ha−1) and maize (13.7 t ha−1) cultivation compared to potato (7.23 t ha−1) or wheat (6.10 t ha−1). Maize and potato exhibited greater biomass stability due to longer growing seasons and better synchronization with peak precipitation. In contrast, wheat and oat exhibited higher biomass variability, reflecting their susceptibility to early spring drought. Among the four crops analyzed, maize achieved the highest crude protein yield (1068 kg ha−1) and WUE (31.9 kg biomass ha−1 mm−1), primarily due to its superior biomass production rather than its protein concentration or elevated soil water consumption. Therefore, cultivating forage crops with longer growth periods could effectively align water demand with seasonal precipitation, thereby improving biomass accumulation and WUE in hydrothermally limited regions.

The huge soil organic C (SOC) storage (around 34 Pg in the top 0.7 m) in Qinghai-Tibet Plateau (QTP) grasslands is commonly explained by slow decomposition of litter under cold climate therein, but this view may not be reliable as humidity also affects microbial activity. We sampled the 20 cm topsoil of grasslands along an altitudinal gradient from 1286 m on the western Loess Plateau (LP) to 4200 m above sea level on the northeastern QTP. The light-fraction SOC (LFOC), composition of non-cellulosic neutral carbohydrates, and amino sugars were used as biomarkers to investigate the intensity of microbial action on SOC as a function of climate along this altitudinal gradient. From the lowest-to the highest-humidity site with rising altitude, the root biomass tripled and the SOC content increased approximately sevenfold (from 13.5 g kg(-1) to 93.3 g kg(-1)). The non-cellulosic neutral carbohydrate, microbial biomass C (MBC), and microbial necromass C (MNC) contents increased, but the LFOC content decreased. The contribution of MNC to the SOC and ratios between microbially- and plant-derived sugars in the non-cellulosic carbohydrate pool increased, but the proportion of LFOC in the SOC dropped. Consequently, besides the increased root biomass, the selective preservation of microbial compounds at colder and more humid sites contributed to SOC accruals in grasslands. The higher MBC in cold and humid grasslands perfectly explained the increased selective preservation of microbial derived C at the expense of plant C in higher-relative to lower-altitude areas. Importantly, the above humidity-controlled accumulations of microbial substances and SOC in grasslands were confirmed by results synthesized from published data across the LP and QTP. The higher SOC contents in cold and humid QTP grasslands relative to warm and dry regions were ascribed to the increased accumulation of microbial residues because of the increased humidity in QTP grasslands.

Legumes play a crucial role in nitrogen cycling; however, it is unclear how this effect is related to soil organic carbon dynamics in grassland across climatic gradients. Here, we investigate how legumes affect soil nitrogen levels and thus microbial necromass accumulation across a 2500 km climatic gradient, using nitrogen-fixing Caragana shrubs as indictors for legumes. We show that microbial necromass and its contribution to organic carbon were markedly lower in warm-arid than in cold-humid grasslands, associated with lower soil nitrogen availability in arid environments. The impacts of legumes on increasing soil nitrogen availability in warm-arid grasslands were greater than in cold-humid regions. This led to a 64–112

Nitrogen (N)-cycling microorganisms participate in nitrification and denitrification processes to regulate the emission of nitrous oxide (N2O). Use of N fertilization and film mulching (FM) are two common agricultural methods conducive to plant growth and yield increase. At present, the interaction of film mulching and N application (FM+N) on soil microbial characteristics and N2O emission has received little attention. In a meta-analysis, we synthesized 634 pairs of observations from 191 papers to investigate the effects of global N addition and FM on terrestrial N2O emissions and related microbial functional gene abundance. The results showed that N, FM, and FM+N treatments increased the N2O emission in soil by 235 %, 33 %, and 21 %, respectively. Nitrogen fertilizer treatment can reduce the abundance of ammonia-oxidizing archaea (AOA, ammonia is oxidized to nitrite) and nitrous oxide reductase gene (nosZ, N2O is reduced to N-2), increase the abundance of ammonia-oxidizing bacteria, promote ammonia oxidation, inhibit the process of converting N2O into N-2, and thus increase N2O emissions. For total N < 4 g/kg, an increase in its content promoted N2O emission. The FM+N treatment only negatively affected AOA abundance. High temperature and high SOC were more beneficial for soil N2O emission in FM+N treatment. This suggests that integrated management measures in agricultural production may have complex interactions on N2O emissions. Nitrogen fertilizer treatment increases N2O emissions by altering the abundance of related microorganisms and genes (reducing AOA and nosZ, and increasing ammonia-oxidizing bacteria), thereby promoting ammonia oxidation and inhibiting the conversion of N2O to N-2. Use of FM may offset some of the effects of N fertilizer application on N-cycle microorganisms and N2O emissions. In summary, FM+N treatment exhibited complex and interrelated effects in regulating soil N cycling and N2O emissions: fertilization had a two-way perturbation on microbial abundance, promoting N2O emission and also triggering a compensatory effect of film mulching on N2O emissions.

Shrubs are expanding in grassland ecosystems across the world, yet their influence on the spatial heterogeneity of soil phosphorus (P) transformation has seldom been investigated. We conducted grid sampling to a soil depth of 1.2 m in a grazed alpine meadow dominated by grass patches and four shrub species to analyze the spatial distributions of soil pH and total soil P and its acid-soluble (representing inorganic P) and acid-insoluble (organic) P fractions. Heterogeneities of soil pH and total, acid-soluble and acid-insoluble P were all visible to the depth of 1.2 m across the plot. The total P stock decreased in the upper 0.4 m of soil, but increased in the lower 0.8 m, under shrubs compared to grasses, resulting in only negligible differences in the total soil P stock in the entire 1.2 m profile between vegetation patch types. The acid-soluble P stock decreased under shrubs throughout the profile, responding to the lowering of soil pH by shrubs, while the acid-insoluble P stock increased in the lower 0.8 m, under shrubs compared with grasses. Soil acid-insoluble P under shrubs increased at the expense of acid-soluble P. These features led to significantly higher proportions of acid-insoluble P in the total P across the 1.2 m soil profile under shrubs (66-71 %) compared to grasses (55 %), associated with the lowered soil pH in shrubby patches. We conclude that shrubification alters the biochemistry of the soil P pool to a depth of at least 1.2 m. The lowered soil pH under shrubs drives the transformation of soil P from inorganic to organic forms in alpine meadows. The altered biochemistry of the soil P pool suggests a rapid release of available P, which supports primary productivity. We emphasize the importance of deep soil sampling in investigating the effects of shrubification on soil P cycling.

Plant diversity loss caused by climate change decreases soil organic carbon (SOC) sequestration, but the mechanism involved remains unclear. Investigating the changes in soil microbial necromass carbon (MNC) accumulation along a climate-plant species diversity gradient can help clarify this mechanism, as it is crucial for the stability of SOC. We conducted large-scale sampling across a 2500-km transect through grasslands on the Tibetan Plateau to investigate the MNC content and its contribution to SOC at depths of 0-20 and 20-40 cm in response to environmental and plant species diversity gradients. Plant species richness, plant biomass, and the proportion of above-ground biomass accounted for by Cyperaceae (sedges) increased with decreasing aridity along the gradient of rising altitude. Meanwhile, the MNC content, its contribution to SOC, and the ratio of fungal to bacterial necromass carbon in both soil layers also increased with decreasing aridity. These results indicate that, in addition to changing climate factors along the altitudinal gradient, plant species richness plays a pivotal role in facilitating soil MNC accumulation and thus the accrual of SOC. Structural equation modelling revealed that plant diversity increases the MNC content by enhancing the abundance of Cyperaceae in the grassland. A higher abundance of Cyperaceae significantly increased the root biomass and level of rhizodeposition, thereby increasing the diversity and activity of soil microbes, and ultimately the accumulation of MNC. Synthesis. We conclude that changes in plant species richness in response to aridity dominate the MNC content and its proportion in SOC on the Tibetan Plateau. These findings demonstrate the importance of incorporating plant species diversity into conservation efforts in grasslands to mitigate the negative impacts of climate change and human activities on SOC.

Saline soils are widely distributed in arid areas but there is a lack of mechanistic understanding on the effect of salinity on the formation and biochemical composition of soil organic carbon (SOC). We investigated the effects of salinity on the accumulation of microbial necromass under natural vegetation and in cropland in salt-affected arid areas stretching over a 1200-km transect in northwest China. Under both natural vegetation and cropland, microbial physiological activity (indicated by microbial biomass carbon normalized enzymatic activity) decreased sharply where the electrical conductivity approached 4 ds m(-1) (a threshold to distinguish between saline and non-saline soils), but microbial biomass was only slightly affected by salinity. These indicated that a larger proportion of microbes could be inactive or dormant in saline soils. The contribution of fungal necromass C to SOC decreased but the contribution of bacterial necromass C to the SOC increased with increasing soil salinity. Adding fungal and bacterial necromass C together, the contribution of microbial necromass C to SOC in saline soils was 32-39 % smaller compared with non-saline soils. Fungal necromass C took up 85-86 % of microbial necromass C in non-saline soils but this proportion dropped to 60-66 % in saline soils. We suggested that the activity, growth, and turnover rate of microbes slowed by salinity was responsible for the decreased accumulation of fungal necromass in saline compared with non-saline soils, while the increased accumulation of bacterial residue in saline soils could be induced mainly by its slower decomposition. Soil microbial biomass was a poor predictor for the accumulation of microbial necromass in saline soils. We demonstrated a reduced contribution of microbial necromass to SOC and a shift in its composition towards the increase in bacterial origin in saline relative to non-saline soils. We concluded that salinity profoundly changes the biochemistry of SOC in arid regions.

Distribution of shrubs expanding in grasslands – shrubification – is ongoing worldwide in grasslands and is common on the Qinghai–Tibetan Plateau (QTP). But the consequences of shrubification for plant carbon (C) input and fate in soil are unclear. We used 13C pulse labelling in a meadow on the QTP to compare photosynthetic capacity and photosynthate distribution in shoots, roots, soil and microbial functional groups between herbaceous plants (herbs) and shrubby Potentilla fruticosa. During 3 h of labeling in 13CO2 atmosphere, the 13C amount assimilated by shrubs (0.81 g C/m−2) was only 38 % of that by herbs. Over 8 days after labeling, 13C amount respired jointly by roots and soil microorganisms under shrubs (0.049 g m−2) was less than half of that under herbs. The mean residence time of 13C for respiration jointly by roots and rhizosphere microorganisms was longer under shrubs (0.61 day) than under herbs (0.44 day). Within 22 days after labelling, 13C amounts in roots, soil, and microorganisms were consistently smaller under shrubs than those under herbs. Consequently, shrub P. fruticosa had not only smaller photosynthetic potential, but also allocated less photosynthate belowground and slowed down C cycling in soil compared with herbs. The distribution of total 13C in microbial functional groups indicated by PLFA analysis was similar between herbs and shrubs. Averaged over vegetation patch types, gram negative bacteria and AMF accounted for 22 % and 4 % of the total microbial PLFAs in the 0–20 cm, respectively, but these two functional groups took up 51 % and 23 % of the total 13C absorbed by microorganisms, respectively. This indicates that gram negative bacteria and AMF are major consumers of rhizodeposits. Concluding, shrubification leads to smaller C allocation belowground and slows down C cycling in the soil.

Globally, grasslands are experiencing shrub expansion, but the effects of such plant community shifts on soil N cycling are seldom investigated. We explored the spatial heterogeneity of total soil N content by grid sampling down to a soil depth of 1.2 m, and analyzed net ammonification and nitrification rates in the top 0.2 m, during the growing season, in response to the patchiness of grasses and four shrub species in a grazed meadow on the Tibetan Plateau. The delta N-15 analyses of plant, soil, and mineral N (NH4+ and NO3-) revealed differences in N absorption strategies between shrubs and grasses. Heterogeneity in the total soil N distribution was common to at least a depth of 1.2 m. The lower delta N-15 values of the 1.2 m soil profile under shrubs were consistent with the lower delta N-15 values of leaves, aboveground litter, and roots of shrubs compared to grasses, reflecting the imprint of long-term N cycling in the soil as affected by shrubs. Increases in total soil N stock in the entire 1.2 m profile by 28%-32% under non-leguminous shrubs relative to grasses were ascribed to N accumulation in the lower 0.8 m. Consequently, deep sampling is essential for accurate estimation of soil N storage in shrubby meadows. The total N stock in soil under the leguminous shrub species was similar to that under grasses, which showed that the effect of shrubification on soil N storage depends on the shrub species. Nitrification rates in the topsoil under all shrubs were greater than under grasses, producing higher N-15-depleted NO3- contents under shrubs. Therefore, the lower delta N-15 signatures of shrub tissues compared to grasses reflected a preferential or adaptive uptake of NO3- by shrubs. The expansion of shrubs in grasslands creates high heterogeneity in N cycling and storage down to a considerable soil depth.

Shrubification is a common phenomenon in grasslands worldwide, but the effects of such a plant community shift on soil nitrification and denitrification are rarely studied. We investigated the effects of shrub species (Dasiphora fruticosa, Hippophae tibetana, Salix oritrepha, Spiraea alpina) on soil nitrification and denitrification and abundances of related microbial functional groups in an alpine meadow. All four shrub species generally increased fungal gene abundance but decreased archaeal gene abundance compared with grasses. This led to increased gene abundance ratios of bacteria to archaea, fungi to archaea, and fungi to bacteria in soils under shrubs relative to grasses. Shrub species of D. fruticosa, S. oritrepha and S. alpina generally increased soil potential nitrification rate and upregulated abundances of ammonia-oxidizing bacteria (AOB) and comammox bacteria (CAOB) compared with grasses, whereas H. tibetana did not affect nitrification or AOB and CAOB abundances. These indicated that the effects of shrubs on soil nitrification rate and related microbial function groups changed with shrub species. The increased soil nitrification potentials under shrubs relative to grasses were not related to ammonia-oxidizing archaea, because their abundances decreased under all shrubs. All four shrub species increased soil potential denitrification rates by 50 %-70 % compared with grasses, associated with increases of nirK gene abundance under D. fruticosa, H. tibetana and S. alpina and with an increase of nirS gene abundance under S. oritrepha. Changed abundances of soil microbial functional groups under shrubs were mainly ascribed to changes in soil pH and increased soil organic matter content and nutrient availability. It is concluded that shrubification increases potentials of nitrification and denitrification and related functional gene abundances in alpine meadows.

Background: In areas prone to water erosion, crop selection strategies should be based on assessment of their effects on soil structural properties.Aims: The present study compared the effects of the cultivation of forage maize (Zea mays L.) and forage oat (Avena sativa L.) and their cultivars on soil aggregation relative to potato (Solanum tuberosum L.) or wheat (Triticum aestivum L.) at a hydrothermally limited site on the Loess Plateau, China.Methods: The water-stable aggregate (WSA) distribution in soil was measured under three cultivars in each of maize, oat, wheat, and potato (a total of 12 cultivars from four crops) in their flowering stage of three cropping seasons, when root biomass was largest.Results: In each year, the water-stable macroaggregates (>0.25 mm) content and mean weight diameter (MWD) of WSAs in the top 20 cm of soil did not differ between tested cultivars of every crop but increased under maize and oat, compared with those under wheat or potato. The increased soil aggregation under maize and oat, compared with wheat or potato, was consistent with the pattern of change in root biomass but was not consistent with the changes in root length density, root surface area, or root mean diameter across the crops. The water-stable macroaggregates content and MWD of soil was positively correlated with root biomass across cultivars and crop species within each cropping season.Conclusions: We suggest that increased root biomass under maize and oat relative to potato or wheat resulted in increased soil aggregation in maize and oat cultivated soils. It is demonstrated that, in areas prone to soil water erosion, planting high-biomass-yielding crops such as maize and oat is more beneficial for increasing soil aggregation and stability, compared with low-biomass-yielding crops such as wheat or potato.

Uncertainty about winter carbon (C) fluxes and their drivers hinders the accurate estimation of C budgets in high-latitude and high-altitude ecosystems. We conducted 3-year-long field observations of soil respiration (Rs) in an alpine forest in northwestern China. The results showed that the mean winter Rs ranged from 193.79 to 233.03 g C m- 2 year- 1 and significantly increased with elevation, while the mean growing season Rs ranged from 467.29 to 711.25 g C m- 2 year-1 and significantly decreased with elevation. The ratio of winter to annual soil CO2 emissions ranged from 21.96 to 34.24 %. The Q10 value (temperature sensitivity of Rs) also varied seasonally, with significantly higher values in winter (3.00-3.85) than in the growing season (2.08-2.22). During the growing season, the Rs was directly regulated by soil temperature (ST), fine root biomass, and microbial activity and indirectly regulated by nitrate nitrogen via an increase in fine root biomass and microbial activity. In addition to the direct effects of soil moisture (SM), ST, and the ratio of fungi to bacteria (F/B) on winter Rs, we revealed another effect: the indirect effect on Rs of the interaction of ST with SM (decreasing ST and increasing SM) with elevation by reducing the soil aggregate stability, probably improving the organic substrate availability, and decreasing the F/B. The ST was the most important factor affecting the growing season Rs, and SM had the greatest effect on the winter Rs. Overall, our findings indicated that the patterns and underlying mechanisms of Rs dynamics in alpine forests in winter are different from those in the growing season and that research on winter Rs dynamics in alpine forests under future climate conditions is needed due to the greater temperature response of Rs in winter than in the growing season.

The mechanism of soil organic carbon (SOC) mineralization stimulated by anaerobic conditions is unclear. The present study was designed to test whether SOC mineralization stimulated by waterlogging is dependent on abundances of plant and microbial origins in substrate. Two soils were sampled from alpine meadows (silty texture and sandy texture), and the abundance of plant materials was mediated through maize residue addition. The soil aeration conditions were adjusted by three moisture levels: 60

The role of the increasing rodent population in grassland degradation on the Qinghai–Tibet Plateau (QTP) is the subject of continuing debate. Rodenticides are used to control the population, but this measure threatens ecosystem biodiversity. We conducted a five-year grazing experiment to investigate the response of soil permeability to livestock grazing and subterranean plateau zokor ( Myospalax baileyi ) in grasslands. Using a meta-analysis approach, we generalized the effects of livestock grazing on soil permeability in global grasslands. The objective was to test the hypothesis that underground soil disturbance by burrowing mammals helps to improve the permeability of soil that has been compacted by trampling from grazing livestock. We found that soil saturated hydraulic conductivity (Ks) in the upper 10 cm layer under livestock grazing was one half of that under no grazing in zokor-excluded plots. The Ks value on zokor mounds (> two years old) was eight times greater than that of inter-mound areas in livestock-grazed plots. These contrasting effects on soil permeability of livestock grazing and subterranean zokor disturbance were ascribed to their opposite effects on soil macroporosity. Meta-analysis confirmed a global generality of decreased soil permeability (measured as Ks and steady infiltration rate) in grazed compared with non-grazed grasslands. Our results provide a new perspective for understanding the ecological function of subterranean mammals in pastures. In the management of subterranean mammals in grazed grasslands, their beneficial effects on promoting soil permeability should be considered.

Plants modulate their phosphorus (P) acquisition strategies (i.e., change in root morphology, exudate composition, and mycorrhizal symbiosis) to adapt to varying soil P availability. However, how community- and species-level P-acquisition strategies change in response to nitrogen (N) supply under different P levels remains unclear. To address this research gap, we conducted an 8-year fully factorial field experiment in an alpine grassland on the Qinghai-Tibet Plateau (QTP) combined with a 12-week glasshouse experiment with four treatments (N addition, P addition, combined N and P addition, and control). In the field experiment (community-level), when P availability was low, N addition increased the release of carboxylate from roots and led to a higher percentage of colonisation by arbuscular mycorrhizal fungi (AMF), along with decreased root length, specific root length (SRL), and total root length colonised by AMF. When P availability was higher, N addition resulted in an increase in the plant's demand for P, accompanied by an increase in root diameter and phosphatase activity. In the glasshouse experiment (species-level), the P-acquisition strategies of grasses and sedge in response to N addition alone mirrored those observed in the field, exhibiting a reduction in root length, SRL, and total root length colonised, but an increased percentage of AMF colonisation. Forbs responded to N addition alone with increased investment in all P-acquisition strategies, especially increased root biomass and length. P-acquisition strategies showed consistent changes among all species in response to combined N and P addition. Our results suggest that increased carboxylate release and AMF colonisation rate are common P-acquisition strategies of plants in alpine grasslands under N-induced P limitation. The main difference in P-acquisition strategies between forbs and grasses/sedges in response to N addition under low-P conditions was an increase in root biomass and length.

Nutrient cycling in alpine grasslands is susceptible to climate change and anthropogenic activities, which can affect soil phosphorus (P) availability. Despite the crucial role of soil P availability in maintaining stability and productivity of grassland ecosystems, limited research has been conducted on the effects of nitrogen (N) addition and winter grazing on P transformation on the Qinghai-Tibet Plateau. In an 8-year experiment, we applied four different N addition rates (0, 25, 50, and 100 kg urea ha ? 1 year- 1) in combination with winter grazing to investigate the effects of N addition and winter grazing on the soil P fractions. The results reveal that increasing the N addition gradually reduced the resin-Pi and NaOH-Pi contents in the soil by increasing the plant P uptake and promoting the release of carboxylates in the rhizosheath, regardless of grazing. Winter grazing decreased the NaHCO3-Pi and NaOH-Pi contents compared with the no-grazing treatment by increasing the P uptake of the plants. In contrast, neither grazing nor N addition affected the HClconc.-P or residual-P content. In the no-grazing plots, the soil NaHCO3-Po content exhibited a gradual increase in response to N addition, whereas N addition had no discernible effect on the NaOH-Po content. In the grazing plots, the NaHCO3- and NaOH-Po contents gradually decreased with N addition, which was associated with the increased acid phosphatase activity in the rhizosheath and the export of forage. Thus, we conclude that N addition promotes the dissolution of NaOH-Pi to more available inorganic P forms. Under winter grazing conditions only, the transformation of P from inorganic to organic forms gradually decreased with increasing N additions.