Distribution Characteristics of Carbon Density in Plant–Soil System of Temperate Steppe and Temperate Desert in the Longzhong Loess Plateau

Abstract

Grassland, as a key component of the carbon cycle in terrestrial ecosystems, is vital in confronting global climate change. Characterising the carbon density of grassland ecosystems in the Longzhong Loess Plateau is important for accurately assessing the contribution of grasslands to global climate change and achieving the goal of “peak carbon” and “carbon neutral”. In this study, the Longzhong Loess Plateau was used as the research object to explore changes in the plant–soil system carbon density in two grassland types by analysing the aboveground vegetation biomass carbon density, belowground vegetation biomass carbon density, 0–100 cm soil carbon density, and ecosystem carbon density of temperate steppe and temperate desert. The results showed that the vegetation biomass (standing and living, litter, and belowground biomass), soil, and ecosystem carbon densities of the temperate steppe were significantly higher than those of the temperate desert (p < 0.05). Their carbon densities were 700.51, 7612.95, and 8313.45 g·m⁻², respectively. The vertical distribution of belowground biomass and soil carbon density in the temperate steppe was significantly higher than that in the temperate desert. The overall trend of belowground biomass carbon density in the temperate steppe and temperate desert showed a gradual decrease, whereas soil carbon density showed a steady increase. More than 91% and 96% of the carbon was stored in soil in the temperate steppe and temperate desert, respectively, and the belowground biomass carbon stock accounted for more than 84% of the total biomass carbon pools in both temperate steppe and temperate desert. Temperate steppe has a significant effect in improving the carbon stock of grassland ecosystems, so ecological protection and restoration of grassland should be strengthened in the future to enhance the capacity of grassland to sequester carbon and increase sinks.

1. Introduction

Terrestrial ecosystems, as key atmospheric carbon sinks, assume an important carbon sink function [1], which has attracted extensive attention at the academic and policy-making levels in the context of setting carbon peaking and carbon neutrality targets [2]. Grassland, as the largest terrestrial ecosystem type in China with an area of about 394.93 × 104 km² and accounting for 41.1% of the national land area [3], plays an important role in the country’s carbon cycle [4]. Previous studies have found that the carbon stocks of grassland aboveground, root, and 0–100 cm soil layer were 230, 1380, and 29,370 Tg, respectively [5], with the total carbon stock accounting for about 10% of the global carbon stock of grassland [6,7]. In addition, about 90.0% of the carbon stock of grassland is distributed in the soil layer [8], which is recognised as an important carbon sink [9] and has a role in climate change that cannot be ignored. Therefore, an in-depth study of the carbon stock in grassland ecosystems, especially the distribution and changes in different vegetation types, has a crucial role to play in advancing the realisation of the “double carbon” goal in China and actively responding to global climate change through the rational use of grassland ecosystems.

Changes in grassland type alters not only the species composition and structure of grassland ecosystems but also converts carbon stocks [10], which in turn affects the carbon source and sink functions of the entire terrestrial ecosystem [11]. Some studies have showed that the carbon density of grassland ecosystems varies greatly depending on the type of grassland [12]. Studies in the Loess Plateau have found that grasslands have a higher carbon sequestration capacity compared to woodlands [13]. This phenomenon is mainly attributed to the high biodiversity of grasslands and good ground cover, which helps maintain soil moisture and reduce the risk of soil erosion, thereby providing a favourable environment for microbial activities, which further enhances their carbon sequestration efficiency. In addition, the carbon density of natural grasslands is higher than that of shrublands [14]. Xu et al. [15] found that carbon stocks in typical grasslands in Inner Mongolia were higher than those in desert grasslands. Meanwhile, some scholars have carried out relevant studies on ecosystem allocation patterns, and the results showed that soil carbon density of different grassland types accounted for a large proportion of the total carbon density, reaching more than 85% [16].

The Longzhong Loess Plateau is located in the arid and semiarid region of northern China. It has a complex topography and long gullies and is an important ecological security barrier in the northwestern region of China [17]. The main grassland types are temperate steppe and temperate desert, and their plant–soil system carbon stocks contribute significantly to the carbon balance of grasslands in China, which is important for maintaining the ecological environment of the grasslands in the Loess Plateau in Longzhong and coping with global warming [18]. However, the impacts of global climate change and human activities, coupled with the fragile ecological environment of grasslands in the Longzhong Loess Plateau, has resulted in intense grassland soil erosion, serious loss of organic carbon from grasslands, and serious threats to the ecological security of grasslands [19]. Therefore, there is an urgent need for in-depth research on the characteristics of the changes in carbon density of grassland ecosystems in the Longzhong Loess Plateau. In view of this, this study selected the temperate steppe and temperate desert grassland ecosystems in the Loess Plateau in Longzhong to investigate the changes in vegetation and soil carbon density as well as the distribution pattern of carbon density and carbon sequestration potential. The aim was to assess the contribution of these two grassland types to the global carbon balance and propose targeted management strategies to enhance the carbon sink capacity of grasslands so as to provide a scientific basis for coping with climate change and achieving China’s “dual-carbon” goal.

2. Materials and Methods

2.1. Study Site Description

The study area is located in the Longzhong Loess Plateau region (32°11′–42°57′ N, 92°13′–108°46′ E), and the administrative boundaries include Lanzhou City, Dixi City, Baiyin City, and Linxia Hui Autonomous Prefecture in Gansu Province (Figure 1) with an elevation of 1300–4200 m above sea level and a total area of about 4.26 × 105 km². It has a temperate continental (semiarid) climate with an average annual temperature of 8.7 °C, an annual precipitation of 240 mm, an average annual evaporation of 1900 mm, 2600 h of sunshine, and a frost-free period of 152 d. The soil types are complex and varied, with Sierozem and Heilu soil predominating. The main grassland types are temperate steppe and temperate desert with grassland vegetation such as Stipa. bungeana, S. capillata, Agropyron cristatum, Kalidium foliatum, Salsola passerinum, Reaumuria songarica, and Sympegma regelii (

Table 1. Study locations and ecosystem characteristics.

Grassland Type	Utilization Pattern	Longitude (° E)	Latitude (° N)	Altitude (m)	Administrative Division	Soil Type	Plant Community
Temperate steppe	Seasonal grazing	103.413	36.6239	2033	Lanzhou	Sierozem	Stipa capillata + Poa annua
		103.098	36.802	2589	Lanzhou	Heilu soil	Agropyron cristatum + Stipa grandis
		103.06	36.505	2245	Lanzhou	Sierozem	Stipa bungeana + Artemisia frigida
		104.4633	37.07766	1830	Baiyin	Sierozem	Stipa capillata + Poa annua
		105.192	36.77812	2030	Baiyin	Sierozem	Agropyron cristatum + Leymus chinensis
		104.574	36.268	2191	Baiyin	Sierozem	Stipa bungeana + Leymus chinensis
		103.9776	35.1708	2290	Dingxi	Sierozem	Stipa capillata + Aster altaicus
		104.191	35.041	2288	Dingxi	Sierozem	Agropyron cristatum + Stipa grandis
		104.4519	35.33396	2402	Dingxi	Heilu soil	Stipa bungeana + Gueldenstaedtia verna
Temperate desert	Seasonal grazing	104.237	36.914	1744	Baiyin	Sierozem	Sympegma regelii + Krascheninnikovia ceratoides
		104.202	36.578	1754	Baiyin	Sierozem	Sympegma regelii + Stipa sareptana
		104.984	36.82	2265	Baiyin	Sierozem	Sympegma regelii + Reaumuria songarica
		104.385	36.785	1622	Baiyin	Sierozem	Reaumuria songarica + Ephedra przewalskii
		104.699	37.045	1958	Baiyin	Sierozem	Reaumuria songarica + Ephedra przewalskii
		103.895	36.871	2196	Baiyin	Sierozem	Reaumuria songarica + Salsola passerinum
		104.036	36.669	1982	Baiyin	Sierozem	Kalidium foliatum + Salsola passerinum
		104.424	37.311	1620	Baiyin	Sierozem	Kalidium foliatum + Kalidium gracile
		104.606	36.726	1511	Baiyin	Sierozem	Salsola passerinum + Reaumuria songarica

) [20].

2.2. Experimental Design

In this study, the grassland type was used as the basic unit in July–September 2021, and the distribution area of each grassland type was used as the basis for selecting grassland types that were broadly representative and zonal. The sample plots were set up according to the grassland utilisation mode and utilisation intensity. Nine sample plots, i.e., nine replicates, were selected for each grassland type. In each survey sample plot, we selected a 100 m × 100 m area for sampling and investigation. For this survey, we set up a 100 m long sample line on the diagonal of the area. In the temperate grassland part, we randomly set up five 1 m × 1 m herbaceous sample squares along this sample line. In the temperate desert section, on the other hand, we set up five 20 m × 5 m shrub sample squares at randomly selected locations on the diagonal [21]. In the selected samples, we collected standing and living and litter parts of all plants. Herbaceous were harvested directly for aboveground standing and living matter; for shrubs, we used a standard plant-based method to collect standing and living matter biomass, which included all fractions (leaves, new shoots, and old shoots). Both herbaceous and shrub litter portions were mowed for their dead residual parts. The collected samples were processed in the laboratory. Firstly, they were killed at 105 °C for 30 min to stop the biochemical process. Subsequently, the samples were dried at 80 °C until the weight was constant, a process aimed at removing all the water from the samples. Finally, the dried samples were weighed and the value obtained was the dry weight biomass (g·m⁻²) of the standing dead and living littering material [22].

After harvesting the aboveground parts of plants in five sample plots and removing the debris on the ground surface, we took seven layers of soil samples with a 5 cm diameter soil auger perpendicular to the ground (0–5, 5–10, 10–20, 20–30, 30–50, 50–70, and 70–100 cm; five augers in one). Then, we rinsed the samples taken with the root system into a 40-mesh mesh net and packed them into a mesh bag according to the layer and recorded the number of the sample plots. The samples were brought back to the laboratory, and the field-treated roots were rinsed again with a 100-mesh sieve so that the fine soil particles attached to the roots were reduced to the minimum and the purpose of complete separation of the roots and the soil samples was achieved. Roots were picked out of the sieve with forceps, individually bagged and labelled, dried at 65 °C to a constant mass, and weighed, and the belowground biomass per unit area (g·m⁻²) was calculated. The soil sample sampling point was set near the underground biomass sampling point, and the number of sampling points and levels were the same as that of underground biomass. The samples were stratified into self-sealing bags, brought back to the laboratory for natural air-drying and removal of roots and gravel, and then preserved after sieving (0.25 mm). The organic carbon content of the soil was determined using the external heating method of potassium dichromate (g·kg⁻¹) [23]. At the same time, a 1.5 m × 0.5 m × 1 m (length × width × depth) capacity pit was dug in each sample plot. After smoothing the soil profile with a soil trimmer, the soil was then sampled from top to bottom with a ring knife of known weight in seven soil layers (0–5, 5–10, 10–20, 20–30, 30–50, 50–70, and 70–100 cm) with five replicates in each layer and then brought back to the laboratory to be labelled. The ring knife containing the soil samples was dried in an oven at 105 °C to a constant mass and weighed, and the soil bulk density (g·cm⁻³) was calculated. The size of the ring cutter was 5 cm in height and 100 cm³ in volume. The formula for calculating soil bulk density is as follows:

Soil bulk density = weight of dried soil sample (g)/volume of ring knife

2.3. Estimation of Carbon Density

We calculated the aboveground, belowground, and total vegetation carbon density based on Equations (1)–(3) [24]:

$$D_{\mathit{HA}}={\sum }_{i=1}^2{B}_{\mathit{HAi}}\times C_{\mathit{HAi}}$$

(1)

$$D_{\mathit{HB}}={\sum }_{j=1}^n{B}_{\mathit{HBj}}\times C_{\mathit{HBj}}$$

(2)

$$D_H=D_{\mathit{HA}}+D_{\mathit{HB}}$$

(3)

In the formula,

• D_HA is the aboveground carbon density of vegetation (g·m⁻²);
• B_HAi is the biomass (kg·m⁻²) of the ith portion of the aboveground vegetation, where i = 1 and 2 denote the aboveground vegetation standing and living material and litter material, respectively;
• C_HAi is the carbon content of aboveground part i of the vegetation (g·kg⁻¹);
• D_HB is the belowground carbon density of vegetation (g·m⁻²);
• B_HBj is the belowground biomass of the jth layer of vegetation (kg·m⁻²);
• C_HBj is the root carbon content in the jth layer of vegetation (g·kg⁻¹);
• D_H is the total carbon density of vegetation (g·m⁻²).

The formula for soil carbon density (SCDi) (g·m⁻²) for a soil layer i is given below [25]:

$${\mathrm{SCD}}_i=\left({\mathrm{BD}}_i-{\mathrm{ST}}_i\right)\times {\mathrm{h}}_i\times {\mathrm{SC}}_i\times 10$$

In the formula, SCDi is the soil carbon density (g·m⁻²), BDi is the soil bulk density in layers i (g·cm⁻³), STi is the soil gravel content in layers i (g·cm⁻³), hi is the soil depth in layers i (cm), and SCi is the soil organic carbon content in layers i (g·kg⁻¹).

2.4. Statistical Analysis

One-way analysis of variance (one-way ANOVA) was used to compare the carbon densities of the vegetation–soil systems of two grassland types, namely, temperate steppe and temperate desert, in the Loess Plateau in Longzhong. The raw data were organised using Excel 2010 software. Data were analysed using the SPSS 26.0 statistical software(SPSS Inc., Chicago, IL, USA), and the significance of differences between the data was tested by the least significant difference comparison (LSD) method (p < 0.05). Graphs were plotted with the help of Origin 2018 (OriginLab, Northampton, MA, USA) and SigmaPlot 12.5 (Systat Software Inc., San Jose, CA, USA) software.

3. Results

3.1. Analysis of Biomass Carbon Density of Vegetation in Temperate Steppe and Temperate Desert

The results of the analysis of variance (ANOVA) of biomass carbon density of vegetation in the temperate steppe and temperate desert of the Loess Plateau in Longzhong showed that the carbon sequestration capacity of vegetation in different grassland types on the Loess Plateau in Longzhong was different, and the carbon content and carbon density of standing and living grassland in the temperate steppe were significantly higher than those in the temperate desert (p < 0.05, Figure 2a,b), with carbon densities of 41.54 ± 5.33 and 17.81 ± 1.87 g·m⁻², respectively. The results of the ANOVA for litter carbon content (Figure 2c) showed that there was no significant difference between the litter carbon content of the temperate steppe and temperate desert (p > 0.05), but the litter carbon density of the temperate steppe (4.40 ± 0.17 g·m⁻²) was significantly higher than that of the temperate desert (3.43 ± 0.13 g·m⁻²) (p < 0.05, n = 9, Figure 2d).

Influenced by climatic conditions and plant community composition, the aboveground biomass carbon density of the temperate steppe was significantly higher (by 116.4%) than that of the temperate desert (Figure 3a). The ranges of belowground biomass carbon density of the temperate steppe and temperate desert were 527.0–737.3 and 137.3–187.64 g·m⁻², respectively (Figure 3b), and the belowground biomass carbon density of the temperate steppe was 4.03 times higher than that of the temperate desert. In summary, the biomass carbon density of vegetation of the two grassland types in the Longzhong Loess Plateau showed that the biomass carbon density of vegetation of the temperate steppe was better than that of the temperate desert grassland.

3.2. Distribution of Soil Bulk Density and Carbon Content in Temperate Steppe and Temperate Desert

The soil bulk density of Longzhong temperate steppe and temperate desert showed a gradual increase with the depth of the soil layer. Among them, the minimum and maximum values of soil bulk density appeared at 0–5 and 70–100 cm, respectively (Figure 4a). The fluctuation range of soil bulk density in each layer was 1.10–1.24 and 1.14–1.41 g·cm⁻³ for the temperate steppe and temperate desert, respectively. One-way ANOVA analyses showed that soil bulk density in the 5–10, 30–50, 50–70, and 70–100 cm layers were significantly lower (p < 0.05) in the temperate steppe than in the temperate desert, but the differences in bulk density in the 0–5, 10–20, and 20–30 cm layers were not significant (p > 0.05).

The results of one-way ANOVA for soil organic carbon in different grassland types on the Loess Plateau in Longzhong showed that soil organic carbon content was significantly higher in the 0–5, 5–10, 20–30, and 50–70 cm soil layers of the temperate steppe than that of the temperate desert (p < 0.05, Figure 4b), whereas no significant differences existed in the 10–20, 30–50, and 70–100 cm layers. In addition, there was a significant vertical trend between the different soil layers. The soil organic carbon content of the temperate steppe decreased with the deepening of the soil layer, and the temperate desert showed a tendency of first increasing and then decreasing, reaching the highest value at 10–20 cm. Soil organic carbon content fluctuated from 3.81 to 12.69 g·kg⁻¹ and 2.51 to 6.15 g·kg⁻¹ at different layers in the temperate steppe and temperate desert, respectively. Regression analyses pooling the organic carbon content of all soil horizons with the bulk density revealed a significant negative correlation (R² = 0.48, n = 126), i.e., the soil organic carbon content was lower as the bulk density increased (Figure 5).

3.3. Distribution of Belowground Biomass and Soil Carbon Density in Temperate Steppe and Temperate Desert

The carbon density of belowground biomass was higher in all layers within 1 m in the temperate steppe (3.57 to 201.57 g·m⁻²) than in the temperate desert (3.28 to 32.29 g·m⁻², Figure 6a). The temperate steppe showed a decreasing trend layer by layer except for the 10–20 cm soil layer, whereas the carbon density of belowground biomass in the temperate desert showed a first increasing and then decreasing trend with the deepening of the soil layer, reaching the highest value at 10–20 cm. Belowground vegetation biomass carbon stocks in the 0–50 cm soil layer accounted for 97% of the total belowground biomass carbon stocks in the temperate steppe and 92% in the temperate desert, and both the temperate steppe and temperate desert belowground biomass carbon density showed significant differences (p < 0.05) in all soil layers above 50 cm.

Soil carbon density was higher in all layers within 1 m in the temperate steppe than in the temperate desert (Figure 6b). There were significant differences in the 0–5, 5–10, and 50–70 cm soil layers, with soil carbon density in the temperate steppe decreasing in the deeper layers of the 50–100 cm soil and reaching the highest content in the 30–50 cm soil layer (1557.62 g·m⁻²). In the surface layer of the temperate desert, the 0–5 cm soil carbon density was the lowest (261.97 g·m⁻²), and the overall trend of carbon density from 5–100 cm showed a steady increase.

3.4. Carbon Density and Its Allocation Pattern in Temperate Steppe and Temperate Desert Ecosystems

The fluctuation ranges of vegetation carbon density, soil organic carbon density, and ecosystem carbon density in the temperate steppe and temperate desert were 563.4–807.4, 4452.4–13,449.8, and 5110.0–14,239.0 and 160.6–217.6, 2726.0–6308.4, and 2886.6–6502.1 g·m⁻², respectively (Figure 7a–c). The vegetation carbon density, soil (0–100 cm) organic carbon density, and ecosystem carbon density of the temperate steppe were significantly higher than those of the temperate desert (p < 0.05). In the study, the 0–100 cm soil organic carbon density in the temperate grassland and temperate desert accounted for 91% and 96% of the carbon density in their ecosystems, respectively. Meanwhile, the proportion of vegetation carbon density in these two ecosystems was 9% and 4%, respectively. In addition, the vegetation carbon densities of the temperate grassland and temperate desert were 700.51 and 182.42 g·m⁻², respectively (Figure 7d). Belowground biomass carbon density in the temperate steppe and temperate desert accounted for 90.22%–95.33% and 84.88%–92.27% of total vegetation biomass carbon density, respectively (Figure 3b and Figure 7a).

4. Discussion

4.1. Biomass Carbon Density of Vegetation in Temperate Steppe and Temperate Desert in the Longzhong Loess Plateau

Grassland vegetation growth in the semiarid Loess Plateau region has a significant carbon sink effect [26] and plays a crucial role in targeting regional climate change and maintaining the carbon balance of grassland ecosystems. In this study, we found that the vegetation biomass carbon pools of temperate steppe and temperate desert differed, and aboveground biomass, belowground biomass, and litter biomass carbon density (g·m⁻²) showed significantly higher variations in temperate steppe than in the temperate desert (p < 0.05). This was mainly in the temperate steppe with higher soil moisture content and better vegetation growth. The dominant species of this grassland are Stipa bungeana, S. capillata, etc., whose biomass is higher among the forage grasses due to the strong tillering property [20]. This improves soil (e.g., soil texture and nitrogen content) and environmental factors (e.g., increased moisture content) by increasing the aboveground vegetative biomass and favours the accumulation of soil organic carbon content, thus increasing the aboveground biomass carbon stock [27,28]. At the same time, grasses dominate the grassland vegetation of temperate steppe, whose grasses have tall plants and well-developed root systems that can be as deep as 85 cm [29,30], resulting in higher biomass content of the underground root systems of the temperate steppe. Also, the increase in litter biomass continuously improves the soil environment and is facilitated by soil microorganisms, leading to the decomposition of litter matter and thus increasing the nutrient content of the soil. Meanwhile, it has been found that sustained improvement in the soil environment can promote the growth of herbaceous vegetation, thereby increasing vegetation biomass [31]. This in turn leads to increasing vegetation (aboveground, belowground, and litter) carbon stocks, which were found to be higher than those in the temperate desert in this study. Aboveground biomass carbon density of the temperate steppe in Longzhong (45.94 g·m⁻²) was higher than aboveground biomass carbon density of typical grassland sample site in Inner Mongolia (32.00 g·m⁻²) [32], but aboveground biomass carbon density of the temperate desert (21.23 g·m⁻²) was lower than the aboveground biomass carbon density of shrublands in Northwest China (98.16 g·m⁻²) [33], and both the temperate steppe and temperate desert in the Loess Plateau in Longzhong were lower than the average aboveground biomass carbon density of grasslands in Northern China (36.9 g·m⁻²) [34]. There may be two reasons for the differences in the findings: (1) Precipitation is the most important climatic factor affecting the productivity of grassland vegetation [35]. The study area is located in the hinterland of Eurasia with a fragile ecological environment, which is one of the sensitive areas of climatic impacts. The region is poorly adapted to climate change [36], and the amount of precipitation is a key climatic factor in the distribution of temperate deserts. The Longzhong Loess Plateau region has a limited growth rate of vegetation due to low precipitation, which in turn causes a lower level of aboveground biomass compared to other study areas. (2) Differences in the extent of aboveground plant foraging by domestic animals in different regions alters aboveground biomass content [37], i.e., high and low levels of livestock carrying rate in different study areas may have led to differences in aboveground biomass. In the Longzhong Loess Plateau region, overgrazing by domestic animals leads to increased vegetation depletion, which in turn causes biomass reduction [38]. Thus, the combination of precipitation deficit and livestock foraging activities resulted in the significant differences observed in the carbon density of aboveground biomass in grasslands of the Longzhong Loess Plateau.

In this study, the vertical distribution of carbon density of belowground biomass in the temperate steppe was greater than that of the temperate desert in different soil depths, indicating that the proportion of plant photosynthetic organs in the temperate desert has declined and the growth of plants has slowed down in comparison to the temperate steppe [39], which in turn has changed the strategy of resource allocation in the plant community [40], leading to a higher allocation of belowground biomass in the temperate steppe and a high content of carbon density. In this study, both the temperate steppe and the temperate desert showed significantly higher belowground biomass than aboveground biomass, which is due to the fact that this type of grassland is mainly distributed in the Loess Plateau region of Longzhong, where there is less precipitation but sufficient light, resulting in its belowground biomass showing well-developed root systems and a higher distribution of belowground biomass when it is subjected to a combination of factors, such as the characteristics of the community and the composition of the species. Therefore, total vegetation carbon density is mainly derived from total root carbon density, which also indicates that the root system is an important component of vegetation carbon density. The main focus of the current research on carbon density of vegetation biomass was on the use of the average carbon density approach [34], that is, the aboveground biomass of the grassland obtained through the model estimation and the belowground biomass of the grassland were estimated based on the root–crown ratio calculated for different types of grassland. The carbon density (g·m⁻²) was then expressed by multiplying the biomass data by 0.45 (an internationally accepted conversion factor). However, a drawback of this approach was that it ignored differences in carbon content between species and sites, which could lead to an overestimation of the ecosystem carbon density. In fact, vegetation biomass carbon density is influenced by vegetation biomass [41]. In order to calculate the aboveground biomass carbon density of grassland vegetation more accurately, the carbon content and biomass of different species and different parts were used in this study. The method effectively determined the biomass and carbon content of the vegetation, thus increasing the credibility of the results. Therefore, the vegetation carbon density in the Longzhong temperate grassland is higher compared to that in the temperate desert area, and the means of calculating carbon density applied in this study is trustworthy.

4.2. Soil Carbon Density in Temperate Steppe and Temperate Desert in the Longzhong Loess Plateau

Soil organic carbon content influences the soil organic carbon density, and the soil organic carbon content of different grassland types is influenced by vegetation characteristics (vegetation cover and height of grass layer) and the root and litter biomass of different grassland types, resulting in differences in the soil organic carbon density of different grassland types. The organic carbon density of the temperate steppe soil in this study was significantly higher than that of the temperate desert soil (p < 0.05), which is similar to a previous study [12]. This may be due to the fact that in the semiarid region of Longzhong, the vegetation of temperate steppe grows better, with higher vegetation cover and grass layer height [20], and the root system of this type of grassland is more developed, resulting in higher root biomass content. This, coupled with the litter being higher than that of the temperate desert, leads to the temperate steppe soil being more prone to the accumulation of soil organic carbon. Therefore, the soil organic carbon content and soil organic carbon density of temperate steppe is higher than that of the temperate desert. The vegetation characteristics of sparsely distributed and species-poor plants in temperate desert grasslands [42] leads to a relatively slow rate of litter accumulation in temperate desert grasslands, which in turn maintains the soil organic carbon density at a low level. The soil organic carbon density contents at 1 m depth in the temperate steppe and temperate desert of the Loess Plateau in Longzhong were 7612.95 and 4721.51 g·m⁻², respectively. Both values were lower than the soil organic carbon density content (between 8500 and 15,100 g·m⁻²) in China [43], presenting a lower level in the whole country. This was mainly due to the fact that the Longzhong Loess Plateau area is constrained by geographic conditions such as aridity, windy conditions, and soil erosion [19], which results in low and sparse vegetation growth and low productivity, thus making soil carbon sources limited. In addition, the extreme climatic conditions (cold winters and hot summers) constrains the rate and amount of biological conversion of organic carbon by soil microorganisms, resulting in a slow rate of accumulation of soil organic carbon and consequently a low soil organic carbon density content. The present study further revealed that belowground biomass carbon density showed more significant variability between temperate desert and temperate steppe, while the trend of soil carbon density was relatively similar. This phenomenon is mainly due to the fact that belowground biomass carbon density is directly affected by the cycle of plant growth and decline [44], and its trend is thus relatively drastic. In contrast, soil carbon density reflects more complex carbon cycling processes, such as decomposition, mineralisation, and humification of organic carbon, which are usually more peaceful, thus keeping soil carbon density relatively stable. In addition, the biomass of belowground vegetation in temperate steppe changes drastically, which leads to large differences in the carbon density of belowground biomass at different levels.

The study showed that the soil organic carbon content and soil organic carbon density in the Longzhong Loess Plateau area were lower. In the vertical profile, the soil organic carbon content gradually decreased from the surface layer downwards with an obvious phenomenon of surface agglomeration, while the soil bulk density and soil organic carbon density showed the trend of gradually increasing with the deepening of the soil layer. This finding is consistent with Lan et al. [45], who found that the organic carbon content of grassland soils was highest in the surface layer and gradually decreased with the depth of the profile in the Loess Plateau, and Wu W J et al. [46], who found that the soil bulk density and organic carbon density of grassland soils gradually increased with deepening of the soil layer in the Loess Plateau. The reason for this lies, on the one hand, in the fact that plant roots, litter material, and plant residues are enriched in the top layer of the soil [47], which means the top layer of the soil has high microbial activity and is rich in nutrients. However, as the soil layer deepens, nutrient supply gradually decreases, which in turn leads to a decrease in soil organic carbon content. At the same time, the reduction in deep soil moisture contributes to an increase in soil bulk density by reducing vegetation cover and plant root vigour. On the other hand, soil organic carbon density is affected by soil bulk density, soil layer thickness, and soil organic carbon content. As the thickness of the soil layer gradually increased during the sampling process, it resulted in an overall increasing trend of soil organic carbon density.

4.3. Carbon Density of Temperate Steppe and Temperate Desert Ecosystems in the Longzhong Loess Plateau

As an important link in the terrestrial carbon cycle, the carbon sequestration potential of grassland ecosystems has always received widespread attention from the scientific research community. This is of great significance for the in-depth understanding of the carbon cycle process and the formulation of effective emission reduction policies. In this study, the carbon density distribution characteristics of temperate steppe and temperate desert grassland ecosystems in the Loess Plateau region of Longzhong were analysed based on field survey data from field sample plots. The results showed that the average carbon densities of the temperate steppe and temperate desert ecosystems in the Longzhong Loess Plateau were 8313.45 and 4905.36 g·m⁻², respectively. The carbon density of the temperate steppe ecosystem was similar to the national average carbon density of grassland ecosystems (8820.5 g·m⁻²) [43], but the carbon density of the temperate desert ecosystem was significantly lower than the national average of grassland ecosystems. Meanwhile, both grassland ecosystems were lower than the average carbon density of grassland ecosystems in Qinghai Province (18,080 g·m⁻²) [48] and higher than the carbon density of temperate steppe (5940 g·m⁻²) and temperate desert (2370 g·m⁻²) ecosystem in Ningxia [9]. This may be due to geographic differences in the hydrothermal conditions, ecology, grassland composition, and soil types [49].

The total ecosystem carbon density included soil organic carbon density and vegetation carbon density, which was dominated by soil organic carbon density, which was similar to the results of some studies at the southern edge of the Tengger Desert [50]. Grasslands in the Longzhong Loess Plateau area are affected by degradation, leading to grassland ecosystem carbon stocks [51], and the ecosystem carbon density and its dynamics are mainly determined by the balance between the input and output of organic matter [50]. However, this balance is disrupted during grassland ecosystem degradation, leading to a decrease in organic matter input and an increase in output. Grassland degradation leads to the destruction of plant growth, which in turn progressively reduces productivity and decreases the cumulative input of carbon from vegetation biomass, thereby reducing ecosystem carbon stocks [52]. At the same time, grassland degradation reduces the protective effect of vegetation on the soil, leading to the deterioration of the soil aggregate structure [53], which reduces the soil organic carbon content. Therefore, grassland degradation in the Loess Plateau in Longzhong not only directly affects the content of vegetation biomass carbon (VBC) and soil organic carbon (SOC) but also indirectly affects the distribution of VBC and SOC by altering the structure of the vegetation and soil, the depth of the soil layer, and the soil moisture content. Ultimately, there is an impact on the carbon density of grassland ecosystems, and this impact is mainly manifested in changes in soil organic carbon density. Temperate steppe had a greater carbon sequestration potential (3408.09 g·m⁻²) compared to temperate desert. Therefore, in response to future global climate change, we should implement reasonable grassland conservation measures in the Loess Plateau region in Longzhong as a key strategy to enhance the carbon stock of grassland ecosystems, help our country achieve carbon peak and carbon neutrality, and respond to global climate change.

5. Conclusions

The carbon sequestration potential of the Longzhong temperate steppe was higher than that of the temperate desert. The carbon density of the temperate steppe ecosystem was higher, with 41.54, 654.57, 4.40, and 7612.96 g·m⁻² for aboveground standing and living biomass, belowground biomass, litter biomass, and soil carbon density, respectively. In the carbon allocation pattern, 0–100 cm soil organic carbon density accounted for the highest proportion, reaching 91% of the ecosystem carbon density. This shows that soil carbon density dominates among the carbon densities of grassland ecosystems. From the vertical distribution of soil carbon pools, the soil carbon density of all layers within 1 m of the temperate steppe was higher than that of the temperate desert, and the soil carbon pools of the temperate steppe and the temperate desert showed an increasing trend with the deepening of the soil layer. In summary, temperate steppe plays an important role in the increase in carbon pools in grassland ecosystems and has a higher potential for carbon accumulation compared to temperate desert. In addition, by studying the interrelationships between grassland vegetation, soil, and ecosystem carbon density, a scientific basis can be provided for the development of effective strategies to enhance carbon storage and reduce carbon emissions.