Effects of Patchiness on Soil Properties and Degradation of Alpine Meadow on the Qinghai–Tibetan Plateau

Abstract

Patchiness is known to affect soil water and soil temperature, and may trigger or accelerate alpine meadow degradation. However, there is a lack of direct evidence concerning the role of the size and type of patches on soil conditions. Here, we estimated the effects of typical patch types on the critical factors of soil water and soil temperature through an in situ survey and then analyzed the potential impacts of patchiness on alpine meadow degradation. The results showed that (1) the soil water storage (SWS) of typical patch types was remarkably different, and vegetation patches had higher SWS than bare patches; (2) with abundant precipitation, the isolated vegetation patch (IV) had higher SWS than the original vegetation patch (OV), accompanied by a decrease in the SWS of the surrounding medium bare patch (MP); (3) patchiness significantly altered the surface soil temperature, with the IV having the highest soil temperature (ST), followed by the OV and bare patches; and (4) the maximum mean value of evapotranspiration was observed in IV (3.85 mm/day), about 12.78% and 46.66% higher than in the bare patches and OV. Our findings indicated that patchiness intensified the heterogeneity of soil water and soil temperature. The IV could absorb soil water from surrounding MP for excessive evapotranspiration through constant high ST, potentially inducing or aggravating degradation. Increasing the connectivity of IV and preventing the transition of OV to IV is important for the stability and restoration of alpine meadow.

1. Introduction

The Qinghai–Tibetan Plateau (QTP), regarded as the third pole and the highest unique territorial unit on Earth, is an important eco-region in both China and the world [1,2,3]. Alpine meadow (7 × 10⁵ km²) covers more than 50% of the available pastures on the QTP and provides important ecological functions and services [4]. Its stability plays an important role in the lives of local pastoralists, regional climate regulation, soil water retention, and biodiversity conservation [5,6].

Unfortunately, the alpine meadow ecosystem is fragile and vulnerable to external disturbances [7]. The patchiness of alpine meadow due to the coupled effects of climate change, livestock overgrazing, and rodent damage has become common on the QTP [8,9]. Patches with various sizes and types intensify the heterogeneities of vegetation coverage and soil physical and chemical properties [10,11]. These differences cause the vertical and horizontal exchange of energy and water fluxes among the atmosphere, ground surface, and underneath the soil, thereby altering pedogenic processes and affecting the eco-healthy development of alpine meadow [5,12].

In this context, the restoration of degraded alpine grassland is urgently needed. To design restoration strategies properly, a reliable diagnosis of the coupling relationship between the patchiness and degradation and the underlying soil water and soil temperature transmission mechanisms should be implemented [13]. Previous studies focus on exploring the formation mechanisms, the ecological and hydrological effects of bare patches, and their impacts on the restoration of degraded grasslands [14,15]. One typical viewpoint concluded that the “black soil land” degraded grassland would generate a black hole of soil moisture storage and soil thermal capacity, leading to the loss of soil water and soil temperature [16].

As research progresses, it becomes clear that it is not only bare patches that significantly affect soil water and soil temperature, but the ecological and hydrological effects produced by vegetation patches also vary with different scales and spatial forms [17,18]. It has been found that isolated vegetation patches differ from original vegetation patches in terms of soil moisture and physical properties, such as bulk density and hydraulic conductivity [18]. Notably, soil temperature is also an important prerequisite for the growth of alpine meadow [19,20]. Clarifying the diversity in soil temperature among various patch types and the underlying influence mechanism is of great practical significance for alpine grassland restoration and management [15,21]. Although the heterogeneity of soil physical properties and soil moisture affects soil temperature, there is still a lack of direct evidence to indicate the role of the size and type of vegetation patches on soil temperature.

Under such circumstances, we conducted comprehensive and continuous in situ observations to monitor the critical factors of soil processes in fragmented alpine meadow. The specific objectives were to (1) assess the differences in soil water storage (SWS), soil temperature (ST), soil temperature gradient (STG), and evapotranspiration among five typical patch types; (2) explore the potential water movement between isolated vegetation patches and surrounding bare soil; and (3) explore the underlying impacts on alpine meadow degradation.

2. Materials and Methods

2.1. Study Area

The study area is located in the source region of the Shule River Basin, the northeast edge of the QTP, China (see Figure 1), and the field observations were conducted at Suli Ecological Observation and Experiment Station (98°18′33.2″ E, 38°25′13.5″ N). The average elevation is approximately 3887 m above sea level, and the terrain is fairly flat, surrounded by gentle rolling hills. The area has a typical continental arid desert climate with the mean annual precipitation ranging from 200 to 400 mm (90% falls from May to September), and the mean annual air temperature is about −4 °C [22]. The soil particle size distribution is 30.96% silt and clay, 57.52% fine sand, and 10.68% coarse sand, and the soil bulk density of the top 0–40 cm soil layer is about 1.41 g cm⁻³ [17]. The land surface is typically characterized by alpine meadow, dominated by Kobresia capillifolia and Carex moorcroftii [23]. Due to permafrost degradation and external disturbance, a mosaic pattern of vegetation patches and bald patches with different sizes is common [5].

2.2. Field Work

Given that water and solar inputs are critical prerequisites for the growth of alpine meadow [19], two years with significant differences in temperature and precipitation were selected to ensure the accurate evaluation of the effects of alpine meadow patchiness (

Table 1. Comparison of precipitation and temperature in two years.

Descriptive Parameters	2019	2020
Observation period	16 July to 18 August	20 July to 15 August
Average daily precipitation/mm	1.17	2.8
Mean daily air temperature/°C	7.92	10.27
Accumulation precipitation/mm	39.7	75.5

). Specifically, our field observations were conducted during the vigorous growth stage of alpine meadow with the most concentrated precipitation (July to August) in 2019–2020. The effects of alpine meadow patchiness on SWS, ST, STG, and evapotranspiration were evaluated by classifying the land surface into five representative patch types (

Table 2. Classification of the five typical patch types.

Patch Types	Abbreviation	Area	Description
Original vegetation patch	OV	>20 m2	Large vegetation patch without obvious fragmentation
Isolated vegetation patch	IV	<20 m2	Small vegetation patch surrounded by bare soil
Large bare patch	LP	>9 m2	Large bare patch with almost no vegetation cover and accompanied by a small amount of gravel
Medium bare patch	MP	1–9 m2	Medium sized bare patch with occasional sparse vegetation
Small bare patch	SP	<1 m2	Small bare patch with sparse vegetation inside and surrounded by lush vegetation

) (Figure 2a–e) [5,19].

Within each patch type, the soil moisture and temperature in the 0–40 cm soil layer were measured with an auto-measurement system (Decagon Inc., Pullman, WA, USA). The system consisted of an EM50 logger and five 5TM sensors embedded in soil layers at 0–5 cm, 5–10 cm, 10–20 cm, 20–30 cm, and 30–40 cm, respectively. The datalogger automatically recorded the data every 30 min. Simultaneously, soil samples were collected from the corresponding soil layers using soil drills (diameter: 5 cm) and aluminum boxes, and then the samples were weighed in the laboratory to calculate the soil water content, which was used to calibrate the EM50 data. Moreover, 75 bulk soil samples (three replicates) were taken from five patch types using volumetric rings (diameter: 7 cm, height: 5 cm, and volume: 200 cm³) to determine the soil bulk density.

Evapotranspiration was measured using a micro-lysimeter (Figure 2f). The cylindrical shaped micro-lysimeter was made of metal, 2 mm thick. It was divided into an inner barrel and outer barrel. The inner barrel of each micro-lysimeter, containing isolated volumes of soil and grass mounted flush with or slightly above the soil surface, was 36 cm in diameter with a depth of 40 cm. The micro-lysimeters were weighed twice a day (8:00 a.m. and 8:00 p.m.) to determine water loss using electronic balances with 0.001 kg precision to meet the measurement requirements [18]. In addition, the automatic meteorological tower recorded precipitation and air temperature (at the height of 2.0 m above the ground) as the background data to reflect the soil water and soil temperature of sample plots. The meteorological data were also logged automatically every 30 min.

2.3. Soil Sample Analysis

The soil samples were tested and analyzed according to the national environmental protection standard HJ613-2011 of the People’s Republic of China and the agricultural industry standard NY/T 1121.4-2006 of the People’s Republic of China [24,25]. The fresh soil samples obtained from the soil drill were immediately placed in aluminum boxes for packaging, numbering, weighing, and recording of their wet weights as w₁. The weighed fresh soil samples were brought back to the laboratory for drying (constant temperature of 105 °C), and then taken out and cooled down to room temperature, weighed, and recorded as w₂. Subsequently, the aluminum boxes were washed clean and then weighed and recorded as w₃. Finally, the soil water content (SWC) was calculated using Equation (1).

$$\mathrm{SWC}={\displaystyle \frac{{\mathit{w}}_1-{\mathit{w}}_2}{{\mathit{w}}_2-{\mathit{w}}_3}}$$

(1)

The determination of soil bulk density included the following steps: (1) the volumetric rings (including 2 lids) and soil samples were weighed and recorded as w₁; (2) the non-porous lid at the end of the volumetric ring was removed and the perforated lid at the top of the volumetric ring (with filter paper inside) was placed in the container with the water slowly added until the top lid was just submerged, and then placed in a tray covered with fine sand for 48 h, weighed, and recorded as w₂; (3) the volumetric rings and soil samples were put in the oven for drying, and then taken out and cooled down to room temperature, weighed, and recorded as w₃ (including 2 lids); (4) the volumetric ring was washed clean and then weighed and recorded as w₄. The soil bulk density was calculated using Equation (2).

$$\rho ={\displaystyle \frac{w_3-w_4}{V}}$$

(2)

where ρ is soil bulk density (g/cm³) and V is the volume of the volumetric ring (200 cm³).

2.4. Data Analysis

The SWS was calculated using Equation (3):

$$\mathrm{SWS}=\rho ·h·w·10$$

(3)

where SWS is the soil water storage (mm), ρ is the soil bulk density (g/cm³), h is the thickness of the soil layer (cm), and w is the soil water content (%).

The STG was calculated using Equation (4):

$${\displaystyle \frac{\partial \mathit{T}}{\partial \mathit{z}}}={\displaystyle \frac{\frac{\mathrm{\Delta }{\mathit{T}}_1}{\mathrm{\Delta }{\mathit{z}}_1}+\frac{\mathrm{\Delta }{\mathit{T}}_2}{\mathrm{\Delta }{\mathit{z}}_2}}{2}}$$

(4)

where ${\displaystyle \frac{\partial \mathrm{T}}{\partial \mathrm{z}}}$ is the soil temperature gradient (°C/m), and ΔT (°C) and Δz (m) are the soil temperature and depth differences between the two layers, respectively.

2.5. Statistical Analysis

One-way analysis of variance (ANOVA) was applied to clarify whether there are significant differences in the effects of typical patch types on soil properties. ANOVA is a statistical method used to test for significant differences in the means of three or more groups. The main steps include (1) establishing hypotheses; (2) calculating the sum of the squared differences between all observed values and the total mean; (3) calculating the sum of the squares of the differences between the mean values of each group and the total mean; (4) calculating the sum of the squares of the differences between each observed value within the group and its mean; (5) calculating the mean squares and F-value, which are used to compare variances between and within groups; and (6) determining the significance level (usually 0.05) and drawing conclusions. The SWS, STG, and evapotranspiration of the five patch types were assessed using the ANOVA method and tested for significant differences at LSD = 5%, with the significance test parameter (p-value) being used to evaluate the performance. All statistical analyses were performed in SPSS 19.0 (SPSS Inc., Chicago, IL, USA), and plotting was performed using Origin software 2017 (Origin Lab Inc., Northampton, MA, USA).

3. Results

3.1. Variations in SWS among the Five Typical Patches

The in situ measured SWS showed obvious differences among the five typical patches in all soil layers. The vegetation patches had higher SWS values than the bare patches, with the average value for the 0–40 cm soil layer exceeding 150 mm in both observation years (Figure 3 and Figure 4). Among the vegetation patches, the IV had higher SWS than OV at 0–30 cm soil layer in years with abundant precipitation, while the OV had higher SWS in years with less precipitation (Figure 3 and Figure 4). The SWS in bare patches with different sizes also had remarkable discrepancies in 2019: the maximum mean value was found for MP, followed by SP and LP (Figure 4). However, in 2020, with abundant precipitation, except for 0–5 cm soil layer, the SWS of MP decreased significantly, with values between those of LP and SP (Figure 3). Moreover, the differences in the SWS depended on soil depth and precipitation. In the 0–40 cm soil layer, the differences in the SWS weakened with the increase in the soil depth. When the soil depth exceeded 30 cm, the SWS of the five typical patches tended to be consistent, especially in years with less precipitation (Figure 4).

3.2. Differences in ST and STG among the Five Typical Patches

The patchiness of the alpine meadow affected the ST (Figure 5 and Figure 6). The vegetation patches had a higher ST than bare patches, and the differences were mainly observed in the upper soil layer with root distribution (0–20 cm). The lack of vegetation roots resulted in homogeneous soil properties; therefore, the temperature and variation patterns of the three bare patch types were relatively similar, and the ST showed a gradually decreasing trend with the increase in the soil depth (Figure 5 and Figure 6). Unlike bare patches, although the IV and OV were both vegetation patches, they exhibited evident temperature differences and variation patterns. Generally, the IV had a higher ST; for example, in 2020, the average ST values of the IV were 13.5 °C, 12.2 °C, 11.2 °C, 10.4 °C, and 9.9 °C for the 0–5 cm, 5–10 cm, 10–20 cm, 20–30 cm, and 30–40 cm soil layers, and the differences with the average ST of OV in the corresponding soil layers were 1.3 °C, 0.4 °C, −0.07 °C, 1.2 °C, and 2.9 °C, respectively (Figure 6).

Considering the remarkable difference in ST mentioned above, the STG was used to reveal the impact of alpine meadow patchiness on ST. The results showed that the STGs of five typical patches were all less than 0, indicating that solar input was transferred from the surface to deeper soil layers (Figure 7). In 2019, three bare patch types had a greater rate of solar input than vegetation patches, and for two vegetation patches, the gradient of solar input was slightly greater for OV than for IV (Figure 7a). Notably, in 2020, with higher air temperature, the absolute value of the STG of IV increased dramatically, which was the highest among the five patch types (Figure 7b). Our study emphasized that the IV patch type was sensitive to solar input, as the STG of the IV showed obvious fluctuations with changes in the soil temperature.

3.3. In Situ-Measured Evapotranspiration

The in situ measured evapotranspiration indicated considerable variations among the five typical patches in 2020, a year with abundant precipitation and high air temperature (Figure 8b). The mean evapotranspiration ranged from 2.63 mm/day to 3.85 mm/day, and the highest value appeared in the IV, which was approximately 7.58%, 11.78%, 18.98%, and 46.66% higher than the MP, LP, SP, and OV, respectively (Figure 8b). By contrast, in 2019, a year with relatively low precipitation and low air temperatures, although the variation in evapotranspiration among the five typical patches showed the same variation trend as in 2020, the differences did not pass the significance test, which may be related to the smaller number of samples (Figure 8a).

4. Discussion

4.1. Differences in SWS between Vegetation and Bare Patches

Alpine meadow patchiness intensifies the heterogeneity of SWS in the surface soil layer (Figure 3 and Figure 4). Firstly, the differences are mainly generated by affecting the two key parameters, namely, soil bulk density and soil moisture, which are used for SWS estimation [26,27]. On the one hand, the high proportion of decayed and fresh root channels increases soil porosity [28,29], resulting in a lower soil bulk density in vegetation patches compared to the bare patches (

Table 3. Variations in soil bulk density among the five typical patches.

Soil Layer	OV/g cm−3	IV/g cm−3	LP/g cm−3	MP/g cm−3	SP/g cm−3
0–5 cm	1.13	0.99	1.27	1.43	1.19
5–10 cm	1.04	1.03	1.42	1.33	1.16
10–20 cm	1.42	1.45	1.56	1.37	1.18
20–30 cm	1.46	1.51	1.55	1.55	1.25
30–40 cm	1.49	1.59	1.64	1.62	1.31

). On the other hand, the presence of roots can also enhance soil infiltration and water-holding capacity in vegetation patches [30], ultimately leading to higher soil moisture and SWS (Figure 3 and Figure 4). Furthermore, owing to the vulnerability to erosion (wind, etc.), the gravel content is significantly higher in bare patches than in vegetation patches [31], inducing low soil water-holding capacity and SWS [32]. In addition, the higher evapotranspiration of bare patches accelerates the loss of soil water, which is also an important reason for the lower SWS (Figure 8).

4.2. Soil Moisture Migration between IV and Surrounding MP

More attention should be paid to exploring the noticeable difference in the SWS between the IV and OV, which may determine the roles of these two types of soil in maintaining the stability of alpine meadow. Our research used EM50 soil moisture data to calculate the SWS and further analyzed their synchronous changes with precipitation over the observation periods (Figure 9 and Figure 10). The results illustrate that the difference in SWS between the IV and OV is closely related to precipitation. When precipitation is abundant, IV has higher SWS in the 0–30 cm soil layer than the OV, whereas it is significantly lower than OV in the case of less precipitation (Figure 9 and Figure 10). It is worth noting that there is a corresponding change in the SWS of the MP. Under the condition that the SWS of the IV is lower than the OV, the MP has the highest SWS among the bare patch types. Conversely, the SWS of the MP is decreased significantly, with values between those of the LP and SP.

With abundant precipitation, bare patches are completely submerged and the soil water content is supersaturated as a result of the terrain being lower than that of the vegetation patches (Figure 11). For the vegetation patches, the mucus secreted by roots and attached microorganisms has a strong water-holding capacity [18], significantly affecting the hydraulic properties of the rhizosphere soil and reducing its hydraulic conductivity [33]. During the growing season, the range of rhizosphere soil can be extended to 10 cm around the roots, further expanding the hydraulic conductivity of IV and its surrounding MP. Nevertheless, the differences in hydraulic conductivity are likely to induce lateral flow from bare patches to vegetation patches [34,35]. Therefore, we assume that there is a lateral flow of soil water between IV and MP with abundant precipitation, which also provides a good insight into why MP has a lower SWS value in humid years and a higher value in dry years (Figure 3 and Figure 4).

Furthermore, the IV has relatively higher evapotranspiration compared to the OV, and had even more so than the LP and SP in 2020 (Figure 8b). In addition to higher ST (Figure 5 and Figure 6) and adequate soil moisture (Figure 3 and Figure 4), it may be attributed to surface wind, which is an essential force for the evapotranspiration process [36]. The IV is surrounded by bare soil with a lower ground level (Figure 11), lacking shelter and resulting in the acceleration of surface heat flux and evapotranspiration [18]. Similar to our research, other studies also report that differences in evapotranspiration can drive the lateral flow of subsurface soil water [5,37]. Therefore, our analysis suggests that the IV could absorb soil moisture from the surrounding bare patches through roots, and partial water is more likely to be lost via evaporation, which is excessive consumption and not conducive to maintaining alpine meadow stability.

4.3. Effects of Alpine Meadow Patchiness on Soil Temperature

It is generally believed that, under the same radiation conditions, the heat absorption capacity and heating rate of bare ground are greater than those of vegetated ground [38]. Moreover, the transpiration of vegetation leaves consumes heat, finally resulting in a lower temperature in vegetation than that in bare soil [39]. In accordance with existing studies, our results show that vegetation patches have a higher soil temperature than bare patches, which was especially the case in 2020 with sufficient precipitation (Figure 6). Through field investigation and image analysis, we believe that the feedback mechanism of the dry and wet states of soil caused by rainfall is the main reason for the higher temperatures of vegetation patches. With precipitation, the grassland emissivity changes less, while the emissivity of wet bare soil is about 0.05 higher than that of dry bare soil [40]. Thus, under continuous precipitation, the top layer of bare soil has a higher water content and is in a moist state, resulting in a lower ST.

With less precipitation in 2019, the ST of an OV is not significantly different from those of bare patches, and the changes are consistent with air temperature (Figure 5). Nevertheless, the absolute value of the STG in an OV is the smallest, which is conducive to the stability of vegetation; instead, IV still has the highest ST, with a large absolute value of STG (Figure 5 and Figure 7). Compared with the OV, the IV has lower vegetation coverage and a weaker shading effect on the underlying surface soil, which makes it difficult to block more solar radiation, leading to the accumulation of heat in the surface soil [40]. Additionally, surface wind is an essential force for surficial processes [36,41]. The QTP is windy all year round and the annual mean wind speed exceeds 2 m s⁻¹ [42]. The IV is surrounded by low bare soil, lacking shelter and resulting in the acceleration of STG and evapotranspiration [18,35].

4.4. Implications of Alpine Meadow Patchiness on Grassland Degradation and Restoration

Grassland degradation is manifested in the decline of ecological attributes, such as grassland productivity, biodiversity, and water conservation capacity [43]. Improving or maintaining the water conservation function of grassland is a critical part of the restoration of degraded grassland [44,45]. Our research indicates that OV is the key patch type for grassland stability, while IV is the more sensitive patch type in the process of alpine meadow degradation. The lateral flow of soil moisture is generated by the difference in hydraulic conductivity and soil water potential [46]. IV absorbs soil water from surrounding bare patches and accelerates evapotranspiration through sustained higher ST, causing water loss and potentially accelerating alpine meadow degradation. Traditional restoration methods include self-recovery through natural succession, artificial grassland planting, no-tillage and reseeding, fertilization, and turf transplantation [47,48]. Therefore, we prefer to emphasize that more attention should be paid to restoring the plants in bare patches around an IV, for instance, laying mulch around the edge of an IV to reduce soil evaporation and improve soil fertility, followed by artificial planting or turf transplantation. A valuable understanding could be that increasing the connectivity between IV patches may be more beneficial for maintaining the eco-healthy development of alpine meadow than directly restoring bare soil vegetation.

5. Conclusions

Alpine meadow patchiness caused by permafrost degradation, rodent disturbance, and overgrazing is common on the QTP. Patches with various sizes and types intensify the spatial heterogeneous distribution of soil gravel content, root biomass, and hydraulic conductivity. These differences will cause the vertical exchange of energy and water fluxes among the atmosphere, ground surface, and underneath the soil, and may trigger or accelerate grassland degradation. However, there is a lack of direct observational evidence concerning the role of the size and type of patches on soil moisture and soil temperature, which are prerequisites for the growth of alpine meadow. Here, we evaluated the effects of patchiness on the critical factors of soil processes and discussed the implications of isolated vegetation patches on grassland degradation and restoration based on in situ sampling and auto-measurement data.

Based on the analysis of field observation data, our conclusions can be summarized in the following points: (1) original vegetation patches had high soil water storage and low soil temperatures and evapotranspiration, playing a critical role in maintaining the stability of the alpine meadow; (2) medium bare patches had the highest soil water storage in years with less precipitation; otherwise, the soil water storage decreased significantly, with values between those of large bare patches and small bare patches; (3) isolated vegetation patches absorbed soil moisture from surrounding bare patches for excessive evapotranspiration through constant higher soil temperatures, which might be an inducement for grassland degradation.

To conclude, although other soil parameters such as soil water potential data need to be supplemented to verify the lateral flow of soil moisture, our findings have guiding significance for grassland management. We suggest that increasing the connectivity of isolated vegetation patches through artificial seeding and preventing the transition of original vegetation to isolated vegetation patches is important for the stability and restoration of alpine meadow.