Deep soil C and N pools in long-term fenced and overgrazed temperate grasslands in northwest China

Grazing exclusion is more effective for vegetation restoration and nutrient transfer in the heavily degraded desert steppe

BACKGROUND

Grazing exclusion is an efficient practice to restore degraded grassland ecosystems by eliminating external disturbances and improving ecosystems' self-healing capacities, which affects the ecological processes of soil-plant systems. Grassland degradation levels play a critical role in regulating these ecological processes. However, the effects of vegetation and soil states at different degradation stages on grassland ecosystem restoration are not fully understood. To better understand this, desert steppe at three levels of degradation (light, moderate, and heavy degradation) was fenced for 6 years in Inner Mongolia, China. Community characteristics were investigated, and nutrient concentrations of the soil (0-10 cm depth) and dominant plants were measured.

RESULTS

We found that grazing exclusion increased shoots' carbon (C) concentrations, C/N, and C/P, but significantly decreased shoots' nitrogen (N) and phosphorus (P) concentrations for Stipa breviflora and Cleistogenes songorica. Interestingly, there were no significant differences in nutrient concentrations of these two species among the three degraded desert steppes after grazing exclusion. After grazing exclusion, annual accumulation rates of aboveground C, N, and P pools in the heavily degraded area were the highest, but the aboveground nutrient pools were the lowest among the three degraded grasslands. Similarly, the annual recovery rates of community height, cover, and aboveground biomass in the heavily degraded desert steppe were the highest among the three degraded steppes after grazing exclusion. These results indicate that grazing exclusion is more effective for vegetation restoration in the heavily degraded desert steppe. The soil total carbon, total nitrogen, total phosphorus, available nitrogen, and available phosphorus concentrations in the moderately and heavily degraded desert steppes were significantly decreased after six years of grazing exclusion, whereas these were no changes in the lightly degraded desert steppe. Structural equation model analysis showed that the grassland degradation level mainly altered the community aboveground biomass and aboveground nutrient pool, driving the decrease in soil nutrient concentrations and accelerating nutrient transfer from soil to plant community, especially in the heavily degraded grassland.

CONCLUSIONS

Our study emphasizes the importance of grassland degradation level in ecosystem restoration and provides theoretical guidance for scientific formulation of containment policies.

Grazing practices affect phyllosphere and rhizosphere bacterial communities of Kobresia humilis by altering their network stability

The primary utilization strategy for meadow grasslands on the Qinghai-Tibet Plateau (QTP) is livestock grazing. This practice is considered as one of the major drivers of plant-associated bacterial community construction and changes in soil properties. The species of Kobresia humilis is considered as the most dominant one in grasslands. However, how different grazing practices affect the phyllosphere and rhizosphere bacterial communities of K. humilis is unknown. To address this issue, the effects of the grazing enclosure (GE), single-species grazing (YG and SG, representing yak only and sheep only, respectively), and different ratios of grazing (ratio of yak to sheep is 1:2, 1:4, and 1:6, represented by MG1:2, MG1:4, and MG1:6, respectively) on the dominant plant of K. humilis, it's phyllosphere and rhizosphere bacteria, and soil properties were investigated using artificially controlled grazing and grazing enclosure. Our data showed that grazing enclosure enhanced vegetation coverage, and rhizosphere bacterial richness and diversity, while reduced plant number and bacterial network stability of K. humilis. The NO3--N, K+, and Cl- concentrations were lower under grazing compared to GE. SG reduced the concentration of NH4+-N, TN, K+, and Na+ compared to YG. Moderate grazing intensity had a lower relative abundance of the r-strategists (Bacteroidota and Gammaproteobacteria) with higher bacterial network stability. Yak and sheep grazing showed reversed impacts on the bacterial network stability between the phyllosphere and rhizosphere of K. humilis. Proteobacteria and Actinobacteriota were identified in the molecular ecological network analysis as keystone taxa in the phyllosphere and rhizosphere networks, respectively, under all treatments. This study explained why sheep grazing has more adverse effects on grazing-tolerant grass species, K. humilis, than yak grazing, and will contribute to a better understanding of the impacts of different grazing practices and grazing enclosure on alpine grassland ecosystems on the QTP.

Long-term enclosure at heavy grazing grassland affects soil nitrification via ammonia-oxidizing bacteria in Inner Mongolia

Enclosure and grazing can significantly change the turnover of nitrogen in grassland soil. Changes of soil nitrogen mineralization and ammonium-oxidizing microorganisms caused by enclosure in different grazing intensities (about 30 years of grazing history) grassland, however, has rarely been reported. We selected the grassland sites with high and medium grazing intensity (HG and MG, 4 and 2 sheep ha^-1, respectively) and had them enclosed (45 × 55 m) in 2005 while outside the enclosure was continuously grazed year-round. A two factorial study was designed: grazing intensity (MG and HG sites) and enclosure (fence and non-fence). Nitrogen mineralization was detected through a laboratory incubation experiment. The abundance and community structure of soil ammonia-oxidizing archaea (AOA) and ammonia-oxidizing bacteria (AOB) were analyzed using quantitative PCR (q-PCR), terminal-restriction fragment length polymorphism (T-RFLP), cloning, and sequencing. Results showed that compared with MG site, at HG site the AOB abundance and community structure of AOB changed significantly while the AOA abundance and community structure did not change obviously. Enclosure significantly decreased the cumulative mineralized N, N mineralization rate, the abundance of AOB and the AOB community structure at the HG site, while at MG site, enclosure did not change these parameters. Potential nitrification rate (PNR) was positively correlated with the abundance of AOA and AOB at the MG and HG sites, respectively. The abundance of AOA was significantly correlated with soil pH; however, AOB abundance was significantly correlated with soil available N, total N, C/N ratio, pH, etc. The phylogenetic analysis showed that Nitrososphaeraceae and Nitrosomonadaceae were the dominant AOA and AOB, respectively. Totally, the responses of AOB and AOA mainly were associated to changes in soil physicochemical properties caused by different intensity grazing; AOB and AOA may be the dominant functional players in ammonia oxidation processes at HG and MG site, respectively.

Rebuilding soil organic C stocks in degraded grassland by grazing exclusion: a linked decline in soil inorganic C

Background

Our study evaluated how soil organic carbon (SOC) and soil inorganic carbon (SIC) recovered over time in deep loessial soil as overgrazed grassland was fenced and restored.

Methods

The study was conducted in the Yunwu Mountain Nature Reserve in the Ningxia Autonomous Region of China. In it we compared soil data from three grazed grassland (G) sites, three sites that were fenced for 15 years (F15), and three sites that were fenced for 30 years (F30) as a so-called ‘space for time series’.

Results and Discussion

We compared SOC accumulation in soil up to 200 cm below the surface in G, F15, and F30 plots. An increase in SOC correlated with a decrease in soil pH, and decreased soil bulk density. However, SOC sequestration in fenced plots was largely offset by a decrease in SIC, which was closely correlated (r = 0.713, p = 0.001) with SOC-driven soil pH decline. We observed no significant increase in soil total carbon in the F15 or F30 sites after comparing them to G.

Conclusions

Our data indicate that fencing causes the slow diffusion processes to intensify the soil property changes from increased litter return, and this slow diffusion process is still active 30 years after fencing at 100–200 cm soil depths in the studied deep loessial soil. These findings are likely applicable to similar sites.