Vertical distribution characteristics of soil organic carbon and vegetation types under different elevation gradients in Cangshan, Dali

Study area

The Cangshan study area (25°34′–26°00′N, 99°55′–100°12′E) is located in Dali Bai Autonomous Prefecture, western Yunnan Province, across Dali City. It has a subtropical southwest monsoon climate. The average annual temperature is 15.5 °C, and the annual precipitation is more than 1,000 mm. Cangshan has a steep terrain that consists of 19 peaks. The highest point, Malone Peak, is 4,122 m above sea level. In this study, a spline with a length of about 1,200 m near Cangshan National Geological Park in Dali was selected. The elevation range was 2,400–3,600 m. Sampling sites were set every 200 m along the transect, and three parallel sampling sites within a range of 10 km² were selected for soil sample collection (Fig. 1).

Collection and pretreatment of soil samples

On the basis of numerous field investigations, soil samples were collected from 2,400, 2,600, 2,800, 3,000, 3,200, 3,400, and 3,600 m elevations by using the hand-held stainless steel drills bit from December 2022 to February 2023. The sampling point information is shown in Table 1. Three parallel sample points were selected from each sampling elevation region. The soil samples were collected from a depth of 0–30 cm. The litter on the soil surface was removed before sampling, and soil samples were collected successively at sampling depths of 0–5, 5–10, 10–20, and 20–30 cm. The vegetation types were recorded during soil sample collection. The soil profile was divided into layers 0–5, 5–10, 10–20, and 20–30 cm on the basis of sampling depth. The soil samples were packed into sample bags, numbered, and brought back to the laboratory for subsequent analysis. Moreover, GPS (G120BD, Guangdong) was used to determine the elevation (Alt), latitude (Lat), and longitude (Lon), and soil water content (SWC) was measured with a soil moisture detector (SOONDA TR-6, Guangdong). The soil samples’ physicochemical properties are shown in Table 2.

Measurement and analysis of soil samples

The soil samples were air dried, sieved through a two mm sieve (10 mesh), and weighed for the determination of indicators.

SOC content

SOC content was determined through potassium dichromate titration with reference to the “Potassium Dichromate Oxidation Spectrophotometric Method” (HJ 615-2011) (Bierer et al., 2021; Bahadori & Tofighi, 2016; Weaver, Birdsey & Lugo, 1987; Beltrame et al., 2016; Lã et al., 2023; Tatzber et al., 2015; Mustapha et al., 2023), and light fraction organic carbon (LFOC) and heavy fraction organic carbon (HFOC) were extracted with the relative density method (Azam et al., 2020). Water-soluble organic carbon (WSOC) was extracted at a soil–water ratio of 1:6, and a TOC analyzer (TOC-L; Shimadzu, Tokyo, Japan) was used.

Pretreatment of soil samples: Twenty grams of the air-dried soil samples that had passed through a two mm sieve were placed in a 250 mL plastic bottle with a fine mouth, and 50 mL of sodium iodide solution with a density of 1.8 g cm⁻³ was added. The samples were shaken at 200 rpm for 2 min on a vibrating machine. Substances that adhered to the wall of the tube were washed with an additional 10 mL sodium iodide solution. The tube was left overnight then centrifuged at 5,000 rpm for 30 min. Immediately after centrifugation, the supernatant was poured into the filter device of a 0.45 um filter membrane for vacuum filtration, washed with 50 mL of 0.5M CaCl₂ solution, and rinsed with 50 mL of distilled water. The 0.45 um component was dried and weighed for 24 h at 50°C. The organic carbon of the dried sample was analyzed with the Black–Wely method after grinding.

Soil recombination and light group separation: Exactly 100 g of the air-dried soil samples were weighed and divided into three equal parts, which were placed in heavy liquid with a density of 1.70 g cm⁻³, shaken by hand for 5 min, shaken ultrasonically at 400 Jml⁻¹ for 3 min, centrifuged, siphoned to obtain the supernatant, and filtered; this process was repeated three times. The obtained samples were washed with 100 mL 0.01 mol L⁻¹ of CaCl₂ solution and then repeatedly with 200 mL of distilled water to obtain the light group. The rest was reconstituted, washed with 100 mL 0.01 mol L⁻¹ of CaCl₂ solution, and rinsed repeatedly with 200 mL of distilled water. The sample recovery rate was above 95%.

Potassium dichromate external heating of organic carbon: Exactly 2 g of the air-dried soil sample was weighed, subjected to a 0.25 mm sieve, placed in a 250 mL triangular bottle, added with 10 mL of 1N potassium dichromate, mixed evenly, added with 20 mL of concentrated sulfuric acid, boiled for 1 min by “blowing bubbles” steam, cooled for half an hour, added with 50 mL of distilled water, added with three drops of linfenolin indicator, and titrated with 0.5 N ferrous sulfate solution. FeSO₄ was added drop by drop until the color became gray–blue–green. Two blank measurements were made at the same time (no added soil samples).

(1)${\displaystyle \mathrm{SOC}\left(\text{\%}\right)=\frac{\left({\mathrm{V}}_0-\mathrm{V}\right)\times \mathrm{N}\times 0.003\times 1.3}{\mathrm{m}}\times 100\text{\%}.}$ (2)${\displaystyle \mathrm{SOC}\left(\mathrm{g}{\mathrm{kg}}^{-1}\right)=\frac{\left({\mathrm{V}}_0-\mathrm{V}\right)\times \mathrm{N}\times 0.003\times 1.3}{\mathrm{m}}\times 1000.}$

where V₀ is the volume of ferrous sulfate consumed by two blanks, V is the volume of ferrous sulfate consumed by the sample, m is the drying soil weight, N is the equivalent concentration of FeSO₄, and 0.003 is the number of grams of 1 mg equivalent C.

Soil pH, water, and organic matter

Soil pH: Determination of soil pH was performed with reference to the “Determination of Soil pH Potentiometric Method” (HJ 962-2018). Ten grams of the air-dried soil samples were subjected to a 20-mesh sieve and placed in a 50 mL beaker, which was then added with 25 mL of water (soil–water ratio of 1:2.5). The container was sealed with a sealing film, vibrated violently in a horizontal oscillator for 5 min, and allowed to stand for 30 min. The pH of the supernatant was determined with a pH meter within 1 h.

Soil water content (SWC): Soil samples were collected successively at sampling depths of 0–5, 5–10, 10–20, and 20–30 cm by using a soil drill, and the wet soil weight onsite was measured. The samples were brought back to the laboratory for drying at 105 °C, the dry weight was measured, and the weight of the box was determined to calculate SWC. The formula is as follows: (3)${\displaystyle \mathrm{SWC}=\frac{\mathrm{W}-\mathrm{D}}{\mathrm{W}-\mathrm{B}}\times 100\text{\%}}$

where W is the wet soil weight, D is the dry weight, B is the weight of the box, the units are all g.

Soil organic matter (SOM): SOM content was determined with the “Solid Waste–Determination of Organic Matter and Ignition Loss Method” (HJ 761-2015). The crucible was burned to a constant weight in a muffle furnace at 600°C in advance, and 2 g of air-dried, ground, and sifted soil samples were weighed in the crucible and burned in the muffle furnace at 600 °C for 3 h. After cooling, the samples were removed from the furnace and weighed. SOM content was calculated by measuring the change in the weight of the soil before and after burning. The SOM calculation formula is as follows: (4)${\displaystyle \mathrm{SOM}=\frac{{\mathrm{m}}_2-{\mathrm{m}}_3}{{\mathrm{m}}_1}\times 100\text{\%}}$

where the weight of the crucible was recorded as m₀, the weight of the soil was recorded as m₁, the weight of the soil and crucible was recorded as m₂, and the weight of the soil and crucible after burning was recorded as m₃, the units are all g.

Vegetation types

Above-ground vegetation types were recorded when the soil samples were collected. These vegetation types were used to identify the vertical distribution types and spatial changes in vegetation at different elevations.

Statistical analysis

One-way analysis of variance (ANOVA) followed by least significant difference (LSD) tests were used to analyze the significant difference, taking p-values lower than 0.05 as the significance level, and all representations of variance are standard errors (±SE). The figures were generated using Origin85 (Origin Lab), the topographic map of sampling points was drawn by Arcgis. All analyses were conducted using SPSS 26 (IBM, Armonk, NY).

Changes in SOC content at different elevations

The SOC content in Cangshan showed regional differences in distribution characteristics. As indicated in Fig. 2, in the study area, the organic carbon content at each sampling point generally decreased with the increase in soil depth. At the same sampling point, the SOC content of surface soil (0–20 cm) was the highest, and the SOC content in layer A+B+C+D varied between 80.7625 and 96.51525 g kg⁻¹. The average SOC at the different sampling points was in the order of 3,600 (80.7625) <3,400 (86.71) <3,200 (88.127) <2,400 (90.662) <2,600 (90.85583333) <2,800 (96.24875) <3,000 (96.51525). The SOC distribution at the different elevations had similar and different points. The SOC content was the highest in layer A+B+C (0–20 cm). The SOC content increased then decreased from layers A to D. At different elevations (2,400–3,600 m), the SOC content varied with the increase in soil depth. A gradually increasing trend was observed from layers A to B, which increased by 19.20%, 13.05%, 21.32%, 21.46%, −1.73%, 4.09%, and 19.95%. Layers B and C decreased by 29.37%, −1.44%, 8.83%, 18.45%, −14.62%, 4.71%, and 0.97%, indicating a gradually decreasing trend. Layers C and D decreased by 13.54%, 29.07%, 24.49%, −2.96%, 1.70%, 27.66%, and 18.62%, showing a sharp decline. The difference between different soil layers below 2,600 m is significant, indicating that soil profile depth has a significant impact on SOC, while the difference between different soil layers above 2,800 m is basically not significant, indicating that soil profile depth has no significant impact on SOC, and SOC content above 2,800 m is mainly affected by other environmental factors.

Vertical profile changes in SOC components at different elevations

Figure 3 shows that the distribution of SOC components in the soil profiles varied at different elevations. Figure 3A indicates that the LFOC content ranged from 1.287 g kg⁻¹ to 7.3515 g kg⁻¹. The average LFOC content at 2,400, 2,600, 2,800, 3,000, 3,200, 3,400, and 3,600 m above sea level was 4.2283, 4.4476, 4.5435, 4.6134, 5.1788, 4.6393, and 3.6156 g kg⁻¹, respectively. The LFOC content generally showed a decreasing trend, and the LFOC profile distribution did not change much. Figure 3B shows that the HFOC content ranged from 12.9727 g kg⁻¹ to 23.3708 g kg⁻¹. The average HFOC content at 2,400, 2,600, 2,800, 3,000, 3,200, 3,400, and 3,600 m were 18.4373, 18.2663, 19.5853, 19.4488, 16.8529, 17.0382, and 16.5750 g kg⁻¹, respectively. The HFOC profile distribution at different elevations was similar to the SOC distribution; it initially increased and then decreased with the increase in profile depth, indicating that the different elevations played an important role in the turnover process of SOC. Figure 3C shows that the WSOC content ranged from 235.5783 mg kg⁻¹ to 392.3925 mg kg⁻¹ at elevations of 2,400, 2,600, 2,800, 3,000, 3,200, 3,400, and 3,600, and the average WSOC content was 295.0617, 338.3092, 392.3925, 309.2242, 270.0225, 345.96, and 235.5783 mg kg⁻¹, respectively. Different soil layers of LFOC showed different significant differences under different elevation gradients. There were significant differences between different elevations under 0–5 cm soil profile, no significant differences between 5–10 and 10–20 cm soil profiles, and significant differences between 20–30 cm soil profiles. The results showed that LFOC accumulation was positive in 0–5 and 20–30 cm soil profiles, HFOC accumulation was positive in 5–10 and 20–30 cm soil profiles, and WSOC accumulation was positive in 0–5, 10–20 and 20–30 cm soil profiles, and there was no significant difference in other layers, which may be related to the amount of plant residues and surface vegetation litter.

Distribution ratio of organic carbon components in soil profiles at different elevations

Distribution ratio of SOC components at different elevations are shown in Table 3. The LFOC/SOC values at the different elevations ranged from 13.53% to 24.09%, the HFOC/SOC values ranged from 74.41% to 86.47%, and the WSOC/SOC values ranged from 0.97% to 6.02%. HFOC was the main component of SOC and accounted for the largest proportion. No significant difference in LFOC/SOC, HFOC/SOC, and WSOC/SOC was observed under different elevation gradients of 5–10, 10–20, and 20–30 cm, but the LFOC/SOC and HFOC/SOC in the 0–5 cm soil layer differed. With the increase in elevation, WSOC/SOC and LFOC/SOC showed a similar change trend in general, but the change trend of HFOC was the opposite. These results indicate that the active organic carbon content at 3,600 m exhibited a lower trend than that at 2,400 m, and the storage of active organic carbon content at middle elevation was favorable.

Factors that affect SOC content and its components

Interaction of SOC content and soil physicochemical properties at different elevations

The content distribution of SOC and its components under different soil physical and chemical properties at different elevations in the study area is shown in Fig. 4. The content distribution of SOC and its components under different soil physical and chemical properties (pH, SWC, and SOM) differed. The effects between pH and LFOC, SWC and HFOC, SOM and WSOC were obvious.

Relationship between SOC and soil properties

Figure 5 presents the relationship between SOC content and soil basic physical and chemical properties. In terms of soil basic physical and chemical properties, pH, EC, and SWC had no significant correlation with SOC content, but SOM and WSOC had a significant positive correlation (p < 0.001). The content of each organic carbon component was correlated with SOC content. Specifically, SOC and LFOC had a significant positive correlation (p < 0.001), HFOC also had a significant positive correlation (p=0.01), and SOC was correlated with WSOC (p < 0.05). The organic carbon components also exhibited different degrees of correlation. LFOC and HFOC were significantly correlated (p = 0.01), and HFOC was correlated with WSOC (p < 0.05).

Distribution types of above-ground vegetation at different elevations

The distribution of above-ground vegetation at different elevations in the study area is shown in Fig. 6. The ground vegetation at 2,400–2,800 m was mainly pine forest, in which Pinus yunnanensis and Pinus armandii were the dominant species. Many evergreen and mixed coniferous broadleaved forests were distributed at 3,000–3,200 m. The common plant species were Lithocarpus leucostachyus A. Camus + Schima argentea Pritz. ex Diels and Tsuga dumosa (D. Don) Eichler + Betula delavayi Franch. With the increase in elevation, Abies delavayi became the main forest at 3,400–3,600 m, and Abies delavayi-Rhododendron simsii Planch was the main common plant species. The vegetation distribution along the sampling sites mainly showed that Pinus yunnanensis was widely distributed below 2,500 m. Most of the species were positive and barren tolerant, and there were patches of shrub grass slopes in some areas. The vegetation at 2,500–2,900 m is dominated by Pinus yunnanensis and Pinus armandii. The mountains in this area are steep and there is a small amount of evergreen broad-leaved forest, indicating that this is a transitional zone from Pinus yunnanensis and Pinus armandii forest to evergreen broad-leaved forest. From 2,900 to 3,400 m, it is mainly mixed forest, and many areas are replaced by Abies delavayi., but still maintain the mixed forest phase, and then transition to Cangshan Abies delavayi Diels- Rhododendron simsii Planch. forest at higher elevations. 3,400–3,700 m are mainly Abies delavayi Diels-Rhododendron simsii Planch forest, which has less environmental damage and similar community composition to other locations in Cangshan.

The main distribution patterns of vegetation are as follows: (1) The native vegetation below 2,900 m is more affected by human disturbance, and the environmental damage degree is greater than that in the high elevation area, which is as follows: the evergreen broad-leaved forest is replaced by Pinus yunnanensis. Most of the wet evergreen broad-leaved forest was replaced by the warm coniferous Pinus armandii, and there were a few evergreen broad-leaved forests in the transition zone, (2) The dominant vegetation of 2,900–3,400 m is Lithocarpus leucostachyus A. Camus, Schima argentea Pritz. ex Diels, Tsuga dumosa (D. Don) Eichler, Betula delavayi Franch and Rhododendron simsii Planch above Abies delavayi, and the native vegetation in this area is Picea asperata Mast forest. Due to the distribution elevation of Picea asperata Mast. is slightly lower than that of Abies delavayi. and its wood is better, the Picea asperata Mast. forest has largely disappeared under large-scale human disturbance, and has been replaced by the less disturbed Abies delavayi- Rhododendron simsii Planch. The transition area is usually Cangshan Abies delavayi, which basically preserves the main species of native vegetation - Tsuga dumosa (D. Don) Eichler, Lithocarpus leucostachyus A. Camus and Abies delavayi and (3) In the zone above 3,400 m, man-made damage is less, the characteristics of native vegetation are more obvious, and the vegetation characteristics formed by long-term natural evolution are more retained, and the Abies delavayi. is limited by 3,400 m. The SOC content in the study area was affected by the type of above-ground vegetation. In general, the level of human activity in the 2,400–3,600 m range was low, the above-ground vegetation productivity was high, and the decomposition of soil litter was slow, which was conducive to the accumulation of organic matter in the soil.

Effects of different elevations on SOC profile distribution

In the forest ecosystem, the change of altitude is one of the important factors affecting SOC (Chen et al., 2016). The change of altitude causes the change of vegetation type. With the increase of elevation, the air temperature shows a downward trend, and the lower soil temperature at high elevation results in the lower decomposition rate of SOC (Weaver, Birdsey & Lugo, 1987). Since elevation difference reflects different air temperature and affects vegetation types, which in turn affects SOC content, the impact of elevation difference on SOC content is essentially the influence of climate factors. However, SOC in this study area did not show a trend of gradual increase with the increase of elevation. It showed that SOC content at middle elevation was significantly higher than that at high and low elevations. The main reason for this phenomenon is that the middle region is rich in vegetation (Lasanta et al., 2020; Sokołowska et al., 2020) and rich in soil water resources. In the elevation area of the study area, the vegetation type is mainly mixed forest, and the surface litter produced is the main source of soil surface organic carbon, and the vegetation roots are the important source of soil deep organic carbon. The influence of vegetation type on SOC content is mainly reflected in two aspects: input mode and decomposition path. Meanwhile, the soil in the low-elevation region is affected by water and soil erosion, which take the organic carbon components in the soil and deposit them. The low temperature and humid environment at high elevations limit the decomposition of soil microorganisms, and organic carbon is preserved in the cryogenic frozen soil.

Effects of different elevations on SOC component profile distribution

The influence of elevation difference on SOC components was found in the whole section, and the overall content was WSOC > HFOC > LFOC. However, the profile distribution of LFOC from a to d does not change much, and the profile distribution of HFOC content is similar to that of organic carbon, showing a general trend of first rising and then decreasing with the increase of profile depth (Xi et al., 2020). This may be because different elevations have an important effect on the turnover of SOC, which is related to the amount of plant residues and vegetation litter on the surface. Due to the abundant vegetation and litter on the ground in the middle elevation area, more plant residues are imported into the soil in the form of organic matter. Meanwhile, WSOC content does not change significantly with elevation. There is a significant difference between WSOC at low elevation and WSOC at middle elevation, but no significant difference between WSOC content and WSOC at high elevation. WSOC is a carbon source that can be directly utilized by soil microorganisms and participates in biogeochemical cycles. This result may be due to the fact that microorganisms can quickly convert WSOC into their own carbon (Xiang et al., 2015). The middle-elevation area with abundant vegetation is conducive to the accumulation of organic carbon, and the soil in the low-elevation area is likely to lose organic carbon components due to abundant human activities, which not only reduces the content of LFOC but also impedes the accumulation of HFOC with high stability. The low temperature and humid environment at high elevations limit the decomposition of soil microorganisms, and organic carbon is preserved in the low-temperature frozen soil. Moreover, the level of human activities and the amounts of animal and plant remains at high elevations are low, so the organic carbon content is lower than that in the central region of the study area.

Influencing factors of SOC and its components

No significant correlation was observed among soil pH, EC, SWC, and SOC content possibly because no direct complex exists among soil pH, EC, SWC, and SOC content (Xi et al., 2020). Moreover, this study only conducted sampling in the wet season and lack sample data from other seasons. Therefore, the effect of soil physical and chemical properties on organic carbon content cannot be fully reflected. The relationship between SOC and its component content at different elevations and soil physical and chemical properties needs to be further studied. Soil physical and chemical properties can affect SOC content and the amount of soil dead leaves returned. In this study, soil pH value did not change much at different elevations, and the overall significant difference was not obvious. Changes in soil pH value would change the microbial activity in soil, and then change the accumulation of SOC by affecting the turnover capacity of SOC. The water content is the highest at 3,600 m, and the SOC content is slightly lower than 3,000 m. This may be because there is less human interference at high elevations and a lot of water is frozen in the soil. However, the distribution of above-ground vegetation and litter determine that the SOC content at 3,600 m is slightly lower than 3,000 m. However, SOM content mainly shows that the surface content is higher than other soil layers, which is also directly related to surface contact.

Types of above-ground vegetation at different elevations

Soil at different elevations has different types of vegetation cover, and environmental conditions at different elevations are not exactly the same. The vegetation distribution characteristics of the Cangshan study area generally show that the low-elevation area is more affected by human disturbance, and the environmental damage degree is greater than that of the high-elevation area. Most of the original vegetation is replaced by existing vegetation (Pinus yunnanensis. and Pinus armandii.), and the middle elevation area shows a mixed forest, while the high-elevation area has less human disturbance, and more original vegetation types are preserved. Plant types at middle elevations produce more litter of different types, which makes the SOC content at middle elevations slightly higher than that at other elevations. Vegetation varies at different elevations, that is, the vegetation cover in the above-ground part of the soil differs, which changes the biomass and organic carbon of above-ground plants in the soil (Bojko & Kabala, 2017). The influence of above-ground vegetation on SOC depends on plant productivity and litter amount. Generally, the decomposition products of litter from above-ground vegetation are transferred to soil, which increases the organic carbon content in soil (Xiong, Zhou & Zhang, 2020). If a large forest area is lost due to human deforestation, the biomass of above-ground vegetation will decrease, and the carbon cycle of soil will change. Studies have shown that when forests are destroyed and transformed into farmland, SOC is lost at different degrees (Tolimir et al., 2020; Fusaro et al., 2019; Vanacker et al., 2022). Therefore, we need to protect forests and promote the accumulation of organic carbon by afforestation and returning farmland to the forest. Doing so could mitigate global warming (Bastin et al., 2019).