Study Area

The study was carried out in a semi-arid meadow steppe at Grassland Ecological Research Station of Northeast Normal University, Jilin Province, China (44°45′N, 123°47′E). The climate is semi-arid, continental with an annual mean temperature ranging from 4.6 to 6.5ºC and annual precipitation from 280 to 400 mm. The soil is a mixed salt-alkali meadow steppe (Salid Aridisol, US Soil Taxonomy) of 29% sand, 40% silt, and 31% clay (top 10 cm) and is nutrient-poor with total N content ranging from 2.2 to 2.5 mg g^-1, and total P content ranging from 0.23 to 0.27 mg g^-1 [24]. The perennial grass L. chinensis is the dominant plant species, accounting for ≥60% of the annual total aboveground biomass in the area. Other plant species include grasses such as Calamagrostis epigejos, and Phragmites australis; forbs species such as Lespedeza davurica, Artemisia scoparia, and legume species Lathyrus quinquenervius and Melilotus suaveolens. The study area has a long history (over 30 years) of mowing, but it was fenced in 2005 to protect against uncontrolled human disturbances [25]. Mowing is the major management strategy in the region.

Experimental Design and Treatment

We established a large-scale natural gazing and artificial mowing experiment to test the different effects of grazing and mowing on soil physiochemical properties from 2016 to 2019 (Fig. 1). Within the study site, the comparatively flat land area with similar soil conditions was fenced in 2016. In June 2016, we randomly established six 400 × 150 m blocks. The distance between the two blocks was kept at 100-250 m. Each block was comprised of three sub-enclosure plots of size 100 x 100 m including control (white color), cattle grazing (red color), and mowed (green color). Each plot was 50 m apart from each other. The plots were mowed and grazed by cattle (mean weight 300±8 kg, mean±S.E.) at an equal light to moderate intensity (0.1-0.3 animal units per ha), a recommended grazing intensity by local governments. Moderate intensity of grazing is often considered to be a proper utilization method for grasslands.

Grazing treatment started each year from June to September during the first two weeks of each month from 6.00 a.m. to 8.00 a.m. and from 4.00 p.m. to 6.00 p.m.; grazing activities were similar to local grazing habits [8]. Mowing treatment started in every early-August, the peak of the growing season. The plant above-ground biomass >20 cm of height was removed (moderate intensity of mowing) with hay mowers. All litters were moved away from the study sites after the mowing treatments. Moderate intensity of mowing is also often considered to be a proper utilization method for grasslands. We used the large domestic herbivores, cattle (B. taurus), as model organisms, as well as the mowing disturbance, and tested their potential effects on soil physiochemical properties.

Soil Sampling and Measurements

Two parallel transects (100 m long and 15 m apart) were established within each of the control, mowed, and grazed plots and assessed soil, plant, and microclimate properties in five 1 × 1 m quadrats located about every 15 m along each transect. For soil properties, soil moisture was determined using a handheld soil moisture reader (OSA-1, OUSU Technology, Hebei, China), which took readings from two random locations within each of the quadrates. Soil nutrients were determined by using a 4-cm-diameter soil auger by randomly collecting two replicate 0-20 cm soil samples from each quadrat, which were pooled to homogenize the samples. For measurement of soil nutrients, a 10 g subsample was extracted with 70 mL 2 mol L^-1 KCl. Extracts were frozen at 20ºC for analysis of NH₄⁺ and NO₃^- content by continuous flow analyzer (Alliance Flow Analyzer; Futura, Frépillon, France). Total soil N was the sum of NH₄⁺ and NO₃^- concentrations. For soil total available P, another 10 g subsample soil is extracted using acidified NH₄OAc-EDTA and analyzed by ICP (Spectro Analytical Instruments, Marlborough, MA, USA) [26]. Bulk density was measured by dividing soil dry weight by its volume. This volume comprised soil particle volume and the volume of soil pores among particles and was expressed in g/cm³. The electrical conductivity of soil was measured by using a conductivity meter while soil pH was measured through a pH meter [27].

Fig. 1.

Study area Map.

Statistical Analyses

All statistical analyses were performed through the open-source software R 3.1.0 (R Development Core Team. 2014). To assess the effects of grazing and mowing on grassland soils, we used linear mixed-effect models with “treatment” included as a fixed effect (two levels: “grazed” and “control”) and “replicate block” included as a random effect. The specific response variables assessed were: soil physicochemical properties and their ratios. Models were fitted using the function lmer from the package lme 4 and the package lmer Test was used to calculate P-values [28]. All response variables were tested for homogeneity of residual variances, using Levene’s test. If needed, data were normalized by log, square root, or arcsine square root transformations.

Grazing/Mowing and Soil Physical Properties

Cattle grazing and mowing did not significantly affect soil moisture at 0-10 cm depth, as compared to control and grazed plots. It was slightly higher in mowed plots (F2,15 = 5.26, P = 0.4460; Fig. 2a), but significantly changed soil moisture at 10-20 cm soil depth (F2,15 = 18.36, P = 0.031; Fig. 2b). Soil moisture in the grazed plots was higher than mowed plots (P = 0.014), but was not different from the control plots (P = 0.224). Grazing and mowing significantly affect soil bulk density at 0-10 cm soil depth (F2,15 = 45.69, P = 0.041; Fig. 3a), but failed to affect soil bulk density at 10-20 cm depth (F2,15 = 1.088, P = 0.376; Fig. 3b). Soil bulk density at 0-10 com depth in the grazed plots was significantly higher than the control plots (P = 0.028), but was not different from the mowed plots (P = 0.118, Fig. 3a). As compared to control plots grazing and mowing slightly decreased soil bulk density in soil depth from 10-20 cm. Cattle grazing and mowing did not significantly affect soil pH at soil depth from 0-10 cm (F2,15 = 0.69, P = 0.179; Fig. 4a) nor at 10-20 cm (F2,15 = 7.55, P = 0.268; Fig. 4b) in the experimental plots. Compared to control and grazed plots soil pH was marginally higher in mowed plots at a soil depth of 0-10 cm, whereas it was higher in grazed plots at a soil depth of 10-20 cm, than control and mowed plots. Grazing and mowing did not significantly affect soil electrical conductivity neither at soil depth from 0-10 cm (F2,15 = 2.369, P = 0.224; Fig. 5a) nor at 10-20 cm (F2,15 = 1.68, P = 0.096; Fig. 5b) in the plots. Compared to control and mowed plots soil electrical conductivity was slightly higher in grazed plots at soil depth from 0-10 cm, whereas it was marginally higher in mowed plots at soil depth from 10-20 cm than control and grazed plots.

In history, numerous studies have been conducted to investigate the independent effects of grazing or mowing on soil physiochemical properties in indifferent localities, we have compared the combined effect of grazing and mowing on soil physiochemical properties in the same localities of semi-arid grasslands. Besides, we have made an attempt to choose the best-fit management system for the grassland ecosystem in northeastern China. We found that compared to mowing, grazing significantly increased soil moisture, bulk density, and N concentration. However, no significant effect of grazing was observed on soil C and P concentrations and C:N, C:P and N:P ratios. One explanation for our insignificant results was the short duration and the small number of species in the initial pool; which is relatively low compared to central Italy, Switzerland, and Poland [29].

In our study compared to mowing, grazing significantly increased soil moisture (Fig. 2(a-b)). Management practices such as grazing and mowing have profound effects from positive to negative on soil physicochemical properties due to above-ground biomass removal, pressure enforcement, and plant litter layer disruption, in grassland ecosystems for different ecosystems [30]. Grazing with moderate intensity is reported to significantly affect soil moisture [31], our results add to these findings. However, while previous studies are mostly conducted by studying the independent effect of grazing or mowing with other stresses in different localities, in contrast, we compared the effect of grazing and mowing in the same locality at the same time, we document that such patterns also exist in the same locality of natural communities. The increase in soil moisture might be due to certain mechanisms, for example: First, due to selective foraging, grazing did not remove plant cover completely in the grazed plots, on the other hand mowing totally removed plant cover by exposing the soil surface to sun rays which encouraged water loss from soil surface to atmosphere, thus decreased soil moisture in the mowed plots. Second, the study area has heavy snowfall in the winter season; therefore, due to high vegetation cover in the grazed plots as compared to mowed plots retained more snow, hence increasing soil moisture in grazed plots [32]. Third, as compared to grazing, mowing decreased the penetration of water into the soil water table due to more soil compaction and macro pores damage [33].

Grazing and mowing significantly affect soil bulk density (Fig. 3(a-b)). Grazing and mowing are generally reported to increase soil bulk density [15, 34], and our results add to these findings. This increase in soil bulk density might be due to force applied on the soil surface by livestock trampling and mowing machines by increasing soil mechanical resistance and compaction, reduction in soil infiltration, and aggregate stability [3, 4, 15, 35]. The effect of grazing and mowing on soil bulk density may also vary with respect to ecological factors, soil type, and seasons. For example, spring grazing with moderate intensity was observed with an increase in soil bulk density in the upper soil layer as compared to fall grazing, whereas under standard situations, soil with high clay particles was observed with a lower soil bulk density [6, 15, 36].

In our system, both grazing and mowing did not significantly affect soil electrical conductivity and pH (Fig. 4(a-b)). Management practices sometimes have limited effects on soil electrical conductivity and pH in grassland ecosystems [16, 37], which might be due to the short duration of our study. Besides, soil in the grassland ecosystem was tolerant to moderate and short-term managements like grazing and mowing, which did not affect soil physicochemical properties. Moreover, these management practices often have less effect on soil physicochemical properties due to soil high compaction, which decreases, increases, or has no effect on plant litter decomposition due to exposure of soil surface to disturbance and ecological factors, alongside vegetation structure can also reduce the effect of disturbance due to their short height and strong rooting system [16, 38].

Grazing/Mowing and Soil Chemical Properties

Cattle grazing and mowing did not significantly affect soil C concentration, neither at soil depth from 0-10 cm (F2,15 = 1.355, P = 0.147; Fig. 6a) nor at 10-20 cm (F2,15 = 2.018, P = 0.175; Fig. 6b). Compared to control and grazed plots C concentration was marginally higher in mowed plots at a soil depth of 0-10 cm, whereas it was higher in grazed plots than in control and mowed plots at a soil depth of 10-20 cm. Grazing and mowing significantly affect soil N concentration at soil depth from 0-10 cm (F2,15 = 78.25, P = 0.007; Fig. 7a), but failed to affect soil N concentration at 10-20 cm (F2,15 = 2.39, P = 0.869; Fig. 7b). Soil N concentration at soil depth from 0-10 cm in the grazed plots was significantly higher than that in the control plots (P = 0.017), but was not different from the mowed plots (P = 0.365, Fig. 7a), whereas soil N concentration was slightly higher in the grazed plots than control and mowed plots at soil depth from 10-20 cm depth. Cattle grazing and mowing did not significantly affect soil P concentration at soil depth from 0-10 cm (F2,15 = 2.669, P = 0.254; Fig. 8a) nor at 10-20 cm (F2,15 = 0.336, P = 0.247; Fig. 8b) in the experimental plots. Soil P concentration in soil depth from 0-10 cm was marginally found higher in the grazed plots than control and mowed plots, whereas slightly decreased in grazed plots than control and mowed plots.

Fig. 2.

Effects of grazing and mowing on soil moisture (0-20 cm depth).

Fig. 3.

Effects of grazing and mowing on soil bulk density (0-20 cm depth).

Grazing and mowing did not significantly affect soil C concentration (Fig. 6(a-b)). Our findings support those of [39], compared to N and P, grazing and mowing generally have limited impacts on soil C concentrations in a variety of ecosystems. The lack of statistical difference might be due to numerous reasons. First, grazing and mowing slow down nutrient cycling by affecting the growth rate of nutrient-rich plant species of high litter quality compared to plant species with low litter quality [40]. Second, alterations in plant species composition might influence the foraging biomass input to the soil nutrient pool and third, grazing declines the plant litter cover and plant canopy, thus affecting the decomposition of litters [41]. Fourth, the lack of statistical difference in soil carbon concentration might be due to the short duration of the study, however, if the area is subjected to long-term management, it may impact soil carbon concentration [42]. Besides; the total soil C concentrations, which is relatively stable compared to soil organic carbon. Management practices may affect the balance of soil storage by modifying the carbon inputs and outputs from the soil. Still, there is little evidence about the effect of grazing and mowing intensity on the C budget in temperate grasslandsGrazing and mowing did not significantly affect soil P concentration (Fig. 8(a-b)). Our results are consistent with the results of previous studies that management practices sometimes have less effect on soil P concentration compared to N concentration in grassland ecosystems [43]. These non-significant effects on soil P concentration might be due to certain mechanisms: first, short study duration second, the removal of above-ground biomass through grazing and mowing reduced the deposition of plant litter, causing a nutrient greater output than input in the soil ecosystem, but this deficiency was balanced through the deposition of organic matters by large herbivores and production of cavities in available vegetation and disintegration by mowing; third, grazing and mowing increased soil N concentration as compared to P concentration, it might be due to N dissolved faster than P, or mowing and grazing decreased the plant species richness of those herbaceous species which were high in P concentration and can be processed through plant-soil feedback [37, 44].

Fig. 7.

Effects of grazing and mowing on soil nitrogen concentration (0-20 cm).

Fig. 4.

Effects of grazing and mowing on soil pH (0-20 cm).

Fig. 6.

Effects of grazing and mowing on soil carbon concentration (0-20 cm).

Fig. 5.

Effects of grazing and mowing on soil electrical conductivity (0-20 cm).

Both grazing and mowing significantly affect soil N concentration. Soil N concentration in the grazed plots was significantly higher (Fig. 7(a-b)). Grazing and mowing are generally reported to increase soil N concentration in grassland ecosystems [8, 45], and our results add to these findings. These significant effects could be due to first, grazing and mowing, facilitating litter disintegration and decomposition which enriched the soil nutrient pool and hence increased soil nitrogen concentration [46]. Besides this, the soil nutrient increase or decrease is closely related to grazing and mowing intensity, duration, landscape type, soil properties, diversity, and plant community composition [47]. Several mechanisms have been proposed regarding the effect of management practices on soil nutrient concentrations. Till now, more attention has been given to grazing as compared to mowing. For example; grazing can increase the soil nutrient pool through the deposition of dung, and urines and put mechanical pressure through trampling on plant and soil structure, which enhances soil nutrient concentration [42, 48, 49-51]. On the other hand, the effect of mowing on soil nutrient concentration is poorly understood in a grassland ecosystem, as nutrient allocation and ecological stoichiometry, are always size-dependent, based on the soil and vegetation type mowing sometimes has less effect on small herbaceous plant species which cannot be affected with mowing, besides this, plants with short size are rich in nitrogen concentration as compared to large size, which can increase soil nitrogen concentration through plant-soil feedback mechanism [52].

Relationships Among Soil Nutrients and Their Ratios

Grazing and mowing did not significantly affect soil C:N neither at soil depth from 0-10 cm (F2,15 = 1.589, P = 0.364; Fig. 9a) nor at 10-20 cm (F2,15 = 3.588, P = 0.147; Fig. 9b) in the experimental plots. Compared to control and mowed plots, grazing slightly decreased the soil C:N ratio at soil depth from 0-10 cm. Cattle grazing and mowing did not significantly affect soil C:P, at soil depth from 0-10 cm (F2,15 = 5.261, P = 0.147; Fig. 10a) nor at 10-20 cm (F2,15 = 4.337, P = 0.951; Fig. 10b) in the experimental plots. Compared to control and grazed plots mowing increased coil C:P, at soil depth from 0-10 cm, whereas, grazing increased soil C:P, at soil depth from 10-20 cm. Grazing and mowing failed to significantly affect soil N:P, at soil depth from 0-10 cm (F2,15 = 1.226, P = 0.457; Fig. 11a) nor at 10-20 cm (F2,15 = 7.298, P = 0.611; Fig. 11b) in the experimental plots. Compared to control and mowed plots, grazing increased soil N:P, at soil depth from 10-20 cm.

Fig. 8.

Effects of grazing and mowing on soil phosphorus concentration (0-20 cm).

Fig. 9.

Effects of grazing and mowing on soil C:N ratio (0-20 cm).

Fig. 10.

Effects of grazing and mowing on soil C:P ratio (0-20 cm depth).

Fig. 11.

Effects of grazing and mowing on soil N:P ratio (0-20 cm depth).

In our system grazing and mowing did not significantly affect soil C:N, C:P, and N:P ratios (Fig. 9(a-b)). However, grazing decreased soil C:N and increased soil C:P whereas mowing increased soil C:P ratios (Fig. 10-11(a-b)). Generally, short-term and independent management practices such as grazing and mowing have a weak effect on soil stoichiometry and their ratios [8, 53]. This non-significant effect of grazing and mowing on soil C:N:P stoichiometry might be due to several important mechanisms for example, land management can stimulate root exudation of C-rich matters; enhanced N concentration, hence reduced C:N:P ratio in soil [38]. In contrast, grazing increased soil C:P and N: P ratios in the meadow steppe as compared to desert steppe [42], and decreased soil N:P and C:P ratios in a meadow steppe [54]. Besides, grazing and mowing may slow down nutrient cycling by affecting the growth of nutrient-rich species of high litter quality and accelerating the growth of other plant species with low litter quality which slows down nutrient cycling [40, 43]. As compared to soil C:N ratio, the effects of grazing on soil C:P and N:P ratios are less focused, whereas the non-significant effects of grazing and mowing on soil C:N:P stoichiometry were due to short duration of the study and less diversified habitat. The higher ratios recorded can probably be attributed to high soil moisture and C and N mineralization, as indicated by the positive correlations between soil C, N concentration with soil C:N, C:P and N: P ratios, C:P and N:P ratios. the effects of disturbances on stoichiometric differences on soil N:P and C:P ratios are poorly understood [55-58]. Further comprehensive experiments are required to clarify the plausibility of this mechanism.