Distribution of Phosphorus Forms Along the Altitude Gradient in the Soil of the Qinghai–Tibetan Plateau and the Influencing Factors

Abstract

Phosphorus (P) is a key limiting nutrient in alpine meadows. Analyzing the spatial distribution of soil P and its forms along altitudinal gradients is crucial to understand soil nutrient cycling and sustain productivity under climate change. In this study, changes in the total P, available P, inorganic P (Pi), and organic P (Po) contents in soil along an altitudinal gradient of 4400–5200 m on the Qinghai–Tibetan Plateau were investigated using sequential chemical fractionation and solution ³¹P nuclear magnetic resonance (NMR). The results showed that the contents of total soil P, available P, Pi, and Po forms showed vertical distribution patterns. At an altitude of 4400–4950 m, the dominance of NaOH-Po was observed, whereas HCl-Pi was predominant at 5200 m. With increasing elevation, total soil P, orthophosphate, NaHCO₃-Pi, NaOH-Pi, HCl-Pi, and HCl-Po contents increased gradually. In contrast, the concentrations of available P, H₂O-Pi, H₂O-Po, NaHCO₃-Po, NaOH-Po, pyrophosphate, orthophosphate monoester, and diester initially increased, peaked at approximately 4950 m, and subsequently decreased. Both climatic factors (i.e., mean annual temperature and precipitation) and biological factors (aboveground biomass and enzyme activity) jointly regulated the vertical distribution of soil P forms in the alpine ecosystems.

1. Introduction

Global warming has emerged as one of the most pressing environmental challenges in the 21st century. According to projections by the Intergovernmental Panel on Climate Change [1], global temperatures may rise by 1.4–5.8 °C by the end of this century. This change can profoundly alter the structure and functioning of terrestrial ecosystems, particularly affecting nutrient cycling and vegetation productivity [2,3,4,5]. Phosphorus (P) is an indispensable element for plant growth, as it plays critical roles in cellular division, energy metabolism, and genetic material synthesis [6,7,8]. P bioavailability directly limits the productivity of ecosystems. However, due to P loss through plant uptake and erosional runoff, approximately 75% of the global grassland ecosystems currently face P limitation, which is a critical bottleneck for productivity enhancement in terrestrial ecosystems [9,10].

The soil P pool primarily comprises inorganic phosphorus (Pi) and organic phosphorus (Po), with Po requiring biochemical conversion to Pi before becoming bioavailable [11]. The chemical speciation of these forms (e.g., labile P, moderately labile P, and stable P) governs P bioavailability in soil [2]. P forms are influenced by various factors, such as source material, climatic conditions, biological activity, and soil physicochemical properties [12,13]. Notably, climatic factors, such as temperature and precipitation, can modulate P mineralization and immobilization processes by altering microbial activity and enzymatic reactions in the soil. For instance, in alpine meadows, available P exhibits a significant negative correlation with mean annual temperature (MAT), while total P shows strong associations with thermal gradients along the rising altitudes. In addition to abiotic factors, soil biological processes are an indispensable factor regulating P cycling in the alpine ecosystems. Particularly in low-temperature environments, microbial communities and related enzymes (e.g., phosphatases) often assume dominance in regulating P cycling and its availability [11]. Despite these findings, critical knowledge gaps persist regarding the altitudinal variations in soil P forms and their driving mechanisms in high-elevation ecosystems.

The Qinghai–Tibet Plateau is a climatically sensitive region, which is critical to global climate change [14]. With a fragile ecosystem and warming response rate approximately threefold faster than the global average, this region functions as a natural laboratory for investigating the coupling mechanisms of P cycling and climate change [15,16,17]. In the alpine meadows, which constitute over 50% of the plateau’s terrestrial area, productivity is tightly coupled with soil P availability [18]. The dominant forage grasses on the Qinghai–Tibet Plateau (such as Kobresia humilis) exhibit typical symptoms like purplish-red leaves and reduced tillering under P deficient conditions. These symptoms directly limit the production of forage grass biomass [19,20,21]. Clarifying the driving mechanisms of soil P forms and their availability is crucial to assess the productivity of this vulnerable ecosystem and its resilience to climate change. Although prior research has offered preliminary understanding of soil phosphorus stocks on the Plateau, understanding of the altitudinal patterns of P fractions and their biogeochemical drivers remains limited [22,23,24]. Recent investigations have demonstrated pronounced vertical differences in the distribution of P forms along the elevational gradients, with Pi exhibiting dominance in soil P composition and Po displaying nonlinear altitudinal variations. These characteristics suggest that synergistic climate–biological interactions likely govern these elevational differentiation [11,23].

In this study, sequential chemical fractionation and solution ³¹P-NMR were combined to systematically investigate the changes in the distribution of total soil P, available P, Pi, and Po forms along an altitudinal gradient of 4400–5200 m in the alpine meadows of the Tibetan Plateau. This research aims to address two critical scientific questions: (1) whether distinct altitudinal zonation governs the spatial heterogeneity of P fractions in soil, and (2) how climatic and biological factors synergistically regulate the dynamics of P speciation in the soil. By revealing the distribution characteristics of P forms and the main driving factors in different altitude regions, this study not only offers a theoretical framework of P cycling in high-altitude ecosystems but also provides critical datasets for assessing the ecological risks associated with intensified P limitation under ongoing climate change scenarios.

2. Materials and Methods

2.1. Study Sites

The study area was located on the south-facing slope of the Qinghai–Tibetan Plateau (30°30′–30°32′ N, 91°03′ E) in the north of the Northern grassland station, Damxung County, Lhasa (Figure 1). The MAT in the region is 1.3 °C, with peak temperature in July. The mean annual precipitation (MAP) is approximately 476.8 mm, of which 90% rainfall is mainly received during May to September. The region has a semi-arid continental monsoon climate. The area with an altitude of 4400 to 5200 m above sea level is covered by pasture, while bare land predominates at an altitude above 5200 m. The type of grassland is an alpine meadow. The soil of the region has been classified as alpine meadow soil (Chinese Genetic Soil Classification), Cambosol (Chinese Soil Taxonomy), Inceptisol (USDA Soil Taxonomy), or Cambisol (WRB Soil Classification). At the start of experiment, six sampling sites were set up at altitudes of 4400, 4500, 4650, 4800, 4950, and 5200 m (Figure 1). The main vegetation types in the grassland and relevant environmental data, including MAT, MAP, soil pH, aboveground biomass (AGB), soil organic carbon (SOC), soil particle size fractions (sand: 2–0.02 mm; silt: 0.02–0.002 mm; clay: <0.002 mm), and various P-related fractions (Fed, Ald, Feo, Alo), have been described in a previous study [17].

2.2. Sample Collection

At each altitudinal gradient, five sampling points were randomly selected using an “S”-shaped layout. Surface soil samples were collected from a depth of 0–20 cm at these sampling points. Prior to sampling, we meticulously removed the surface litter layer to minimize potential interference from organic residues on the sampling results. Equal volumes of soil were collected from each point and thoroughly mixed to account for small-scale spatial heterogeneity within the sampling site. To ensure the reliability and representativeness of data, the entire sampling process was replicated three times at each altitudinal gradient. The collected samples were gently broken apart by hand and then passed through a 2 mm sieve to carefully eliminate the impurities, such as plant roots and gravel. The sieved samples were homogenized and air-dried for subsequent analysis. Further grinding treatments were carried out according to the specific requirements for measurement of different P fractions (such as Pi and Po), strictly ensuring that the sample particle size met the needs of subsequent analyses.

2.3. Determination of Total P, Available P, Inorganic and Organic P Concentrations in Soil Samples

To measure the total P content, soil samples were digested with a mixture of concentrated H₂SO₄ and HClO₄ [25]. The available P content was extracted from soil using 0.5 M NaHCO₃ (pH 8.5) [26]. Soil organic P was measured by the ignition method and calculated as the difference between the P concentrations in acid extracts (50 mL of 1.0 M H₂SO₄) of ignited (550 °C, 1 h) and unignited soil samples [27,28]. Soil inorganic P was determined by subtracting the organic P content from the total P concentration. The P concentrations in extracts were measured using molybdate colorimetry [29]. The phosphorus activation coefficient (PAC), expressed as the percentage of available P relative to total P, was used to indicate soil P effectiveness. Higher PAC values correspond to greater availability of P [8].

2.4. Sequential Fractionation

Soil chemical P fractionation was conducted according to the procedure described by Hedley [30] and modified by Tiessen and Moir [31]. First, 1.0 g of air-dried soil (through a 0.25 mm sieve) was sequentially extracted using 30 mL of H₂O, 0.5 M NaHCO₃ (pH 8.5), 0.1 M NaOH, and 1.0 M HCl for 16 h at 25 °C. The sequentially extracted fractions were digested with H₂SO₄ (0.9 M) and (NH₄)₂S₂O₈ in an autoclave (120 kPa, 120 °C), and then the total P concentration was measured by molybdate colorimetry [32]. The concentration of inorganic P in the extracts was also measured by molybdate colorimetry [29], while organic P concentration was determined by deducting inorganic P concentration from the total P content.

2.5. Solution ³¹P NMR Spectroscopy

Solution ³¹P NMR analysis was performed according to the procedure reported previously [33,34,35]. First, 4.0 g of air-dried (through a 0.25 mm sieve) soil sample was extracted with 40 mL of 0.25 M NaOH and 0.05 M EDTA solution through gentle shaking for 16 h. The concentration of total P in the extracted solution was determined by the molybdate blue colorimetric method after digestion with H₂SO₄ and potassium persulfate in an autoclave (120 kPa, 120 °C, 1 h). The residual extracted solution was frozen at −80 °C and lyophilized. The lyophilized samples (0.3 g) were solubilized in 1.0 mL of 0.25 M NaOH and 50 µL of deuterium oxide, and then 0.8 mL of this solution was added to the NMR tubes (diameter 5-mm) for measurement. The NMR spectra were acquired using an AVANCE III HD 500 NMR spectrometer (Bruker Biospin, Fällanden, Switzerland) at a ³¹P frequency of 202.5 MHz. The operating parameters were: 13.0 μs pulse width, 0.4 s acquisition and 3.0 s delay time, and 15,000 scans.

Phosphorus compounds were identified based on the chemical shifts (ppm) after standardizing the orthophosphate peak of each sample to 6 ppm [6]. The spectral areas were determined by using the integrated MestReNova software (version 5.3.1) and manual calculation. The concentration of each identified P compound was determined by multiplying the concentration of total P (extracted using NaOH-EDTA) by the percentage of the integrated area of the corresponding compound. To further facilitate peak identifications, the samples were spiked with adenosine 5′monophosphate, myo-inositol hexakisphosphate (myo-IHP), phosphocholine, α- and β-glycerophosphate, α-D-glucose 1-phosphate, and D-glucose 6-phosphate, as suggested by the previous reports [27,33,36,37]. All these chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA).

2.6. Determination of Phosphatase Activity

The activities of soil acid phosphatase (AcP), neutral phosphatase (NeP), alkaline phosphatase (AlP), and phosphodiesterase (PD) were measured using a colorimetric method. A commercial enzyme activity determination kit (Suzhou Grace Biotechnology Co., Ltd., Suzhou, China) was employed. The assay was based on the p-nitrophenyl phosphate method described by Tabatabai [38], with modifications adapted for this study.

In this method, phosphatases catalyze the hydrolysis of p-nitrophenyl phosphate, releasing p-nitrophenol, a yellow product with maximum absorbance at 405 nm. The absorbance intensity is proportional to enzyme activity in the soil sample. To ensure accuracy and reproducibility, substrate concentration, pH conditions (optimized for each enzyme), and incubation time were standardized according to the manufacturer’s protocol.

Enzyme activity was calculated from the quantity of p-nitrophenol released during the reaction and expressed in terms of nanomoles (or equivalently, micrograms, considering the molecular weight of p-nitrophenol) of substrate hydrolyzed per gram of dry soil per hour (µg g⁻¹ h⁻¹). This standardized unit of measurement provides consistency across soil samples and enables direct comparison among different studies.

2.7. Statistical Method and Data Analysis

To detect whether the impacts of altitudinal gradient on soil P and its forms were statistically significant, one-way analysis of variance (ANOVA) followed by Fisher’s least significant difference (LSD) test (p < 0.05) was conducted for data that showed a normal distribution (Shapiro–Wilk test, p > 0.05) and homogenous variances (Levene test, p > 0.05). When necessary, the measured data were transformed (square root or logarithm) and reanalyzed. Relationships between P components and environmental factors (i.e., climate, vegetation, and soil properties) were explored by Pearson correlation analysis, and the significance level was set at p < 0.05 and p < 0.01. All statistical analyses were completed using SPSS 16.0 for Windows (Chicago, IL, USA). Canoco 5.0 software was used for redundancy analysis (RDA) to identify the important environmental factors affecting the spatial variations in Pi and Po forms. The level of significance was set at p = 0.05.

3. Results

3.1. Total Soil P, Available P, and Po Concentrations in Soil Samples

The concentration of total P gradually increased with elevation. The total P content in the soil at 5200 m (0.65 g kg⁻¹) was significantly higher compared to other altitudes (p < 0.05) (Figure 2a). Available P concentration increased initially with elevation, reaching a peak at 4800 (4.28 mg kg⁻¹), and subsequently decreased (Figure 2b). The trend of Po content was also similar to available P along the altitudinal gradient (Figure 2c). Furthermore, PAC was significantly larger at 4800 m than at other altitudes (p < 0.05) (Figure 2d).

3.2. Concentrations of Soil P Fractions Obtained Through Sequential Chemical Fractionation

The concentrations of Pi forms obtained by chemical fractionation were in the following order: HCl-Pi > NaOH-Pi > NaHCO₃-Pi > H₂O-Pi (Figure 3a,b). With increasing elevation, H₂O-Pi and NaHCO₃-Pi concentrations initially increased, reaching the highest levels at 4800 m, and then declined (p < 0.05). The NaOH-Pi concentration gradually increased with elevation, while the HCl-Pi concentration first decreased and then increased, reaching a significantly higher level at 5200 m (p < 0.05).

At the elevation of 4400–4950 m, concentrations of Po forms obtained by chemical fractionation showed the following order: NaOH-Po > NaHCO₃-Po > HCl-Po > H₂O-Po. However, at 5200 m, the order was NaOH-Po > HCl-Po > NaHCO₃-Po > H₂O-Po (Figure 3c,d). The concentrations of H₂O-Po, NaHCO₃-Po, and NaOH-Po showed a unimodal pattern along the altitudinal gradient, with significantly higher values at 4800 and 4950 m (p < 0.05). The HCl-Po content gradually increased with elevation, reaching the highest level at 5200 m (p < 0.01).

3.3. Soil P Fractions Measured by ³¹P NMR Spectroscopy

The NMR spectra revealed an array of inorganic and organic P compounds in the soil (Figure 4). Among them, inorganic P compounds were orthophosphate (6.25–5.80 ppm) and pyrophosphate (−4.20 to −4.50 ppm), while organic P compounds included orthophosphate monoesters (7.00–6.25 ppm, 5.80–2.50 ppm) and diesters (2.50 to −1.50 ppm). Moreover, other peaks in the orthophosphate monoester region were attributed to myo- and scyllo-IHP, α- and β-glycerophosphate, phosphocholine, and mononucleotide, while unidentified peaks were partitioned into three groups (monoester 1, 2, and 3). The peaks identified in the diester region indicated DNA, while the unidentified were partitioned into two groups (diester 1 and 2).

The proportions of NaOH-EDTA-extracted Pi compounds and orthophosphates were significantly larger than those of pyrophosphates at all elevations (

Table 1. Proportions (%) and concentrations (mg kg⁻¹) of different phosphorus (P) compounds in NaOH-EDTA-extracted soils along the elevational gradient in alpine meadow.

Treatment	EPt	Pi		Po
		Ortho	Pyro	Monoesters							Diesters
				myo-IHP	scyllo-IHP	α-Glyc	β-Glyc	Nucl	Pchol	Others	DNA	Others
Proportion (%)
4400	68.1 ± 2.11 b	17.0 ± 2.20 b	2.34 ± 0.81 b	7.04 ± 1.60 c	1.58 ± 0.36 c	6.59 ± 0.22 c	4.16 ± 1.34 ab	1.83 ± 0.37 cd	1.10 ± 0.25 b	51.1 ± 4.61 a	1.86 ± 0.30 bc	5.44 ± 1.00 ab
4500	78.2 ± 0.89 a	15.8 ± 1.80 b	1.89 ± 0.49 b	10.9 ± 2.34 b	3.00 ± 0.31 b	8.16 ± 1.48 bc	3.59 ± 0.52 bc	2.17 ± 0.22 bc	1.21 ± 0.16 ab	48.0 ± 2.60 a	1.53 ± 0.46 c	3.71 ± 1.14 b
4650	79.0 ± 3.32 a	15.7 ± 0.57 b	2.16 ± 0.24 b	15.4 ± 0.65 a	4.18 ± 0.91 a	9.51 ± 0.97 b	4.62 ± 0.50 ab	2.66 ± 0.70 ab	1.46 ± 0.18 a	37.4 ± 1.73 bc	2.04 ± 0.31 bc	4.89 ± 1.13 ab
4800	81.4 ± 0.94 a	15.2 ± 0.99 b	2.49 ± 0.34 b	16.7 ± 0.57 a	4.27 ± 0.19 a	12.9 ± 0.61 a	5.28 ± 0.33 a	3.24 ± 0.58 a	1.21 ± 0.14 ab	30.2 ± 3.16 c	2.94 ± 0.39 a	5.55 ± 0.84 a
4950	69.4 ± 2.75 b	17.2 ± 0.66 b	1.88 ± 0.58 b	14.7 ± 2.17 a	4.58 ± 0.08 a	7.15 ± 0.95 c	4.30 ± 0.62 ab	3.24 ± 0.10 a	0.69 ± 0.10 c	39.2 ± 2.88 b	3.13 ± 0.73 a	3.89 ± 0.34 ab
5200	31.0 ± 0.78 c	42.1 ± 7.75 a	3.52 ± 0.35 a	5.81 ± 0.22 c	1.54 ± 0.33 c	3.74 ± 0.94 d	2.72 ± 0.41 c	1.35 ± 0.05 a	0.53 ± 0.15 c	32.2 ± 6.83 bc	2.40 ± 0.41 ab	4.06 ± 0.83 ab
Concentrations (mg kg−1)
4400	267.0 ± 5.93 d	45.3 ± 5.06 c	6.21 ± 2.04 b	18.8 ± 4.14 d	4.22 ± 0.96 c	17.6 ± 0.52 d	11.0 ± 3.39 d	4.89 ± 0.92 d	2.95 ± 0.71 c	136.5 ± 15.1 b	4.95 ± 0.70 b	14.6 ± 3.01 b
4500	356.1 ± 10.3 c	56.2 ± 4.97 bc	6.70 ± 1.65 b	38.9 ± 9.32 c	10.7 ± 1.42 b	29.2 ± 6.13 bc	12.8 ± 2.16 cd	7.76 ± 1.01 c	4.32 ± 0.69 b	170.9 ± 5.00 a	5.43 ± 1.54 b	13.2 ± 3.99 bc
4650	369.8 ± 5.93 bc	57.9 ± 1.52 b	8.01 ± 1.01 ab	57.0 ± 1.62 b	15.4 ± 3.18 a	35.2 ± 3.81 b	17.1 ± 1.65 b	9.81 ± 2.45 bc	5.40 ± 0.65 a	138.4 ± 7.70 b	7.55 ± 1.27 b	18.1 ± 4.49 ab
4800	417.8 ± 10.3 a	63.7 ± 5.41 b	10.4 ± 1.64 a	69.6 ± 2.92 a	17.9 ± 1.15 a	54.0 ± 3.12 a	22.1 ± 1.90 a	13.6 ± 2.74 a	5.07 ± 0.44 ab	125.9 ± 10.5 b	12.3 ± 1.89 a	23.2 ± 3.98 a
4950	376.7 ± 10.3 b	64.8 ± 2.51 b	7.08 ± 2.29 b	55.4 ± 8.27 b	17.2 ± 0.22 a	26.9 ± 3.45 c	16.2 ± 1.88 bc	12.2 ± 0.05 ab	2.61 ± 0.32 c	147.8 ± 13.0 b	11.8 ± 2.96 a	14.7 ± 1.12 b
5200	201.9 ± 10.3 e	84.6 ± 12.2 a	7.13 ± 1.06 b	11.7 ± 1.05 d	3.10 ± 0.66 c	7.55 ± 1.86 e	5.53 ± 1.12 e	2.72 ± 0.04 d	1.07 ± 0.24 d	65.4 ± 15.7 c	4.88 ± 1.07 b	8.32 ± 1.92 c

Different lowercase letters within the same column indicate significant differences at p < 0.05 (mean ± standard deviation, n = 3). EPt: NaOH-EDTA extractable total P; Po: organic P; Pi: inorganic P; Ortho: orthophosphate; Pyro: pyrophosphate; myo-IHP: myo-inositol hexakisphosphate; scyllo-IHP: scyllo-inositol hexakisphosphate; α-Glyc: α-glycerophosphate; β-glyc: β-glycerophosphate; Nucl: mononucleotides; Pchol: choline phosphate; DNA: deoxyribonucleic acid.

). With increasing elevation, orthophosphate concentration gradually increased, reaching significantly greater levels at 5200 m (p < 0.05). In contrast, pyrophosphate concentration increased to a maximum level and then decreased, showing the highest content at 4800 m (p < 0.05).

Among the Po compounds, the proportion of orthophosphate monoesters was significantly larger than that of diesters at all elevations (Table 1). Furthermore, non-identified monoester showed the highest proportion, followed by IHP (myo- and scyllo-IHP), glycerophosphates (α- and β-glycerophosphates), non-identified diester, mononucleotides, and DNA. Choline phosphate accounted for the smallest proportion of Po compounds. Moreover, the proportion of myo-IHP was greater than that of scyllo-IHP in the phosphomonoester region, while α-glycerophosphate showed a higher proportion than β-glycerophosphates. Trend of each Po compound (except the non-identified orthophosphate monoesters) was similar to the trend of available P concentration, with a unimodal pattern along the altitudinal gradient and the highest concentrations around 4800 and 4950 (p < 0.05).

3.4. Phosphatase Activity

The activity patterns of soil phosphomonoesterases across the elevation gradient of the Qinghai–Tibet Plateau are shown in Figure 5a. Among the measured enzymes, AcP displayed the highest activity levels, ranging from 97.0 to 145.0 µg g⁻¹ h⁻¹, followed by NeP with activities of 44.0–120.8 µg g⁻¹ h⁻¹, and AlP, which had the lowest activity (20.8–36.0 µg g⁻¹ h⁻¹).

AcP and NeP exhibited clear elevation-dependent trends: their activities increased with elevation, peaked at 4800 m, where both were significantly higher than at other sites, and then declined. On the other hand, AlP demonstrated an opposite pattern, initially decreasing and then increasing with the elevation gradient, with the lowest activities recorded at 4500 m and 4650 m.

The spatial distribution of soil PD activity along the elevation gradient resembled that of AcP (Figure 5b). PD activity increased initially, reached a pronounced maximum at 4800 m, and then declined, indicating a similar elevation response to that of AcP.

3.5. Relationships Between P Forms and Environmental Factors

Total soil P had significant positive correlations with MAP, phosphodiesterase (PD), and sand, and significant negative correlations with MAT, pH, clay, and neutral phosphatase (NeP). Concentrations of available P, H₂O-Pi, and NaHCO₃-Pi were significantly positively correlated with MAP, AGB, SOC, PD, moisture, and Fe/Al cations and oxides. Moreover, NaOH-Pi exhibited significant positive correlations with MAP, AGB, and PD, and significant negative correlations with MAT and pH. Most organic P components were significantly positively related to AGB, SOC, moisture, silt, acid phosphatase (AcP). NeP, and Fe/Al cations and oxides (Figure 6).

The RDA results showed that the environmental factors could explain 87.2% of the total variations in Pi forms (Figure 7a). RDA1 and RDA2 explained 85.6% and 1.58% of the variations in Pi form, respectively. Notably, MAT was identified as an important factor controlling the distribution of Pi form. Furthermore, environmental factors explained 82.6% of the total variations in the Po forms (Figure 7b). RDA1 and RDA2 explained 58.3% and 24.3% of the variations in Po form, respectively. Moisture was identified as the most important environmental factor affecting the distribution of Po form, followed by AGB and PD.

4. Discussion

4.1. Spatial Variability of Soil Total P and Influencing Factors

The total soil P in alpine meadows showed an evident vertical distribution, with significantly higher values at higher altitude (Figure 2a). Total P content followed a linear increase, largely controlled by slow geological weathering and long-term accumulation of parent material, processes that are relatively insensitive to short-term climatic or biological fluctuations. However, total P may also be influenced by climate, plant biomass, soil properties, and soil erosion [7,10]. First, P mainly exists in soil minerals and is released by weathering [39,40]. As the temperature gradually declines, weathering weakens, leading to reduced P loss from primary minerals, which ultimately results in larger reserves of mineral P [41,42,43]. In this study, total P concentration was negatively related to MAT, indicating that temperature was a crucial climatic factor influencing the vertical distribution of total P in the soil. Second, the loss of P caused by runoff and erosion was also an important factor affecting the distribution of total P. With the gradual increase in AGB, vegetation coverage and root/litter biomass were enhanced significantly, thereby increasing the number of macroaggregates and microaggregates [17]. This, in turn, resulted in higher rainfall interception and soil water-holding capacity, as well as lower P loss due to runoff [10,44]. When the stable P (only soluble in HCl) concentration in soil was higher, P loss was less, which explains the high total P concentration observed at 5200 m.

4.2. Spatial Variability of Soil Pi Forms and Influencing Factors

Among the sequentially extracted Pi fractions, NaHCO₃-Pi is considered a labile Pi, which can be easily adsorbed onto crystalline surfaces. In contrast, NaOH-Pi is a moderately labile Pi, which can bind with amorphous iron and aluminum [45,46]. HCl-Pi is considered a stable Pi form, which mainly combines with calcium [2]. In this study, stable Pi content was the highest among the Pi components, followed by moderately labile Pi and labile Pi (Figure 3a,b). This result was consistent with a previous study, which reported HCl-Pi as the main Pi form in alpine meadows [23].

Soil available P is commonly regarded as the labile P fraction, which can be easily absorbed and utilized by plants [23,47]. The PAC is used to evaluate P transformation and availability in the soil. A high PAC value indicates a larger proportion of available P and a stronger capacity of soil to supply P for plant growth [8,48]. The spatial distribution of available P and PAC along the altitudinal gradient showed a unimodal pattern, which was consistent with the pattern of labile Pi. However, the PAC values of only 0.46–0.86% indicated that the potential supply of available P was low and that P was a critical factor limiting the productivity of alpine meadows. There are three plausible explanations for the vertical distribution of labile Pi forms in soil along the altitudinal gradient (Figure 8). First, with increasing AGB, the litter and root biomass in the soil increased significantly. Since the soil was largely covered by vegetation, the loss of soluble P through surface runoff was less [44]. In addition, available P in plants was directly returned to the soil through the litter, thus increasing soil available P [49]. The significant positive correlations of AGB with available P and labile Pi further confirmed our hypothesis. Second, increasing MAP enhanced soil phosphatase activities and accelerated the release of orthophosphate from phosphate esters [50,51]. The positive correlations of soil labile Pi with MAP and PD (p < 0.05) indicated that the MAP and PD were important factors contributing to greater P availability in the soil. In addition, MAT was negatively correlated with labile Pi. This finding contradicted the results of a previous simulated-warming experiment in the field, which showed a positive impact of rising temperature on available P [23]. The reason for this apparent discrepancy is that short-term warming increases available P concentration in soil, while long-term warming reduces the available P concentration by increasing the water limitation. Meanwhile, MAT may also influence the soil microenvironment and microbial activities by altering AGB, thereby regulating Po mineralization and ultimately enhancing P availability in the soil. Third, the transformation of various P forms critically affects the spatial distribution of labile P [52,53]. As the AGB increases, vegetation roots and microbial biomass also increase significantly, promoting the secretion of organic acids into the soil [54]. The organic acids promote the desorption of NaOH-Pi adsorbed by amorphous iron and aluminum and then release available P [45,51]. On the other hand, they can dissolve refractory IHP and promote its mineralization through phytase [55].

In addition, slope gradient regulates soil phosphorus availability by regulating soil moisture, erosion–deposition processes, and organic matter accumulation [56,57]. However, this study did not account for slope differences across altitudes but instead focused on identifying large-scale altitudinal patterns in soil P speciation. Future studies of alpine grasslands at the watershed scale should integrate slope gradient, which represents a critically important factor in shaping the vertical differentiation of soil P forms.

4.3. Spatial Variability of Soil Po Forms and Influencing Factors

Soil Po is considered a bio-available P pool. Under P-deficient conditions, Po can be supplemented through organic P mineralization [44]. In this study, the concentrations of H₂O-Po, NaHCO₃-Po, and NaOH-Po in soil showed an evident vertical distribution along the altitudinal gradient (Figure 3c,d), which was consistent with the patterns of available P and labile Pi contents in alpine meadows. We suggest that the unimodal distributions of available P and labile Po arise from nonlinear synergistic effects of temperature, moisture, and biological processes. An optimal hydrothermal window occurs at mid-elevations (4800–4950 m), where aboveground biomass peaks providing abundant carbon inputs that stimulate soil phosphatase activity. This synergy promotes Po mineralization, thereby expanding the available P pool. Furthermore, NaOH-Po was the dominant fraction among all Po fractions, indicating the important role of moderately labile Po in the regulation of P availability.

There are four possible explanations for the variations in Po along the altitudinal gradient. First, plant litter is the primary source of Po in natural ecosystems [58], and higher AGB increases the organic matter input, such as litter and root biomass, into the soil, thereby increasing Po accumulation in the soil surface [51,54]. In this study, H₂O-Po, NaHCO₃-Po, and NaOH-Po showed significant positive correlations with AGB and SOC. Second, temperature is the key climatic factor limiting Po mineralization and decomposition in the soil [23]. Climate warming may increase Po mineralization [52,59], while hypoxia and cold climate conditions in the alpine meadow may limit Po mineralization by reducing the activities of phosphatase and microorganisms in the soil [7]. Therefore, higher root/microbial input and lower decomposition rate may be the main reasons for Po accumulation in the alpine meadows. Temperature showed a significant negative correlation with stable Po (Figure 5), indicating that climate was an important factor for stable Po cycling and transformation. Long-term climate warming may lead to greater P availability in soil, which could worsen the effects of phosphorus limitation on ecosystem productivity. Third, phosphate monoester can easily combine with iron and aluminum hydroxide. Thus, organic P (NaOH-Po) is readily adsorbed on mineral surfaces [60,61], which protects it from mineralization to a certain extent and increases its retention in soil [40,62]. Fourth, higher content of plant root residues and microbial biomass can provide more substrates for phosphatase activity and promote Po turnover [63], thereby increasing the concentrations of labile Po forms (such as H₂O-Po and NaHCO₃-Po) in soil. The activities of acid phosphatase and phosphodiesterase showed significant positive correlations with H₂O-Po, NaHCO₃-Po, and NaOH-Po, which further indicated that Po was more sensitive to phosphatase activity. The spatial distribution of NaOH-EDTA-extractable Po compounds along the altitudinal gradient showed a unimodal pattern, which was consistent with that observed for the sequentially extracted Po fractions.

In addition, with increasing altitude, the dominant P form shifted from NaOH-Po to HCl-Pi in this study (Figure 3). This transformation is likely associated with environmental factors at high altitudes. Firstly, carbonates and calcium ions are often enriched in high-altitude regions, where stable calcium phosphate salts are readily formed, leading to the accumulation of HCl-Pi [42]. Moreover, the low temperature at high altitudes inhibits microbial activity, slows down the rate of organic matter mineralization, and affects specific mineral weathering processes [7], thereby reducing NaOH-Po sources and indirectly increasing the proportion of HCl-Pi. Therefore, this altitudinal differentiation in the distribution of P forms represents a comprehensive response of biogeochemical processes to environmental gradients on the plateau.

The myo-IHP (derived from plant seeds and P-containing organic compounds) and sycllo-IHP (derived from microbial synthesis) are the main forms of orthophosphate monoester in soil, which are strongly stabilized by their interactions with minerals (Fe, Al, and Ca) and organic matter [64,65]. With the increase in AGB, IHP accumulation in the soil increased, which led to a higher proportion of orthophosphate monoester. Pearson correlation analysis showed significant positive correlations of IHP with AGB, SOC, and Fe/Al cations and oxides, further confirming the above theory. In addition, the IHP has shown a significant linear positive correlation with available P. According to previous studies, when the soil available P is low and cannot meet the growth needs of crops, it can accelerate the solubilization of recalcitrant IHP into the available P pool through phytase activity [55] and promote the accumulation of phytate in the soil [66]. Glycerophosphates are alkaline hydrolysates of phospholipids, mainly originating from phospholipids in plant residues and microbial cells [36,67], and the spatial distribution of these compounds has been reported to be closely related to plant litterfall and microbial activity. The observed peaks of orthophosphate monoesters and diesters at mid-elevations (Table 1) likely indicate accelerated microbial turnover of Po. This pattern suggests that environmental conditions, such as favorable temperatures and humidity at these altitudes, enhanced microbial activity, promoting the mineralization of Po into intermediate products. Moreover, the dynamics of these phosphorus compounds are closely associated with the broader cycling of soil organic matter [68,69]. Their accumulation reflects not only microbial phosphorus mineralization but also a key stage in soil organic matter decomposition, during which microbial processing of Po is coupled with the breakdown of carbon skeletons [70,71]. This association highlights an important direction for future research.

5. Conclusions

Total P, available P, Pi and Po forms in alpine meadows showed obvious vertical distribution characteristics along the altitudinal gradient. With increasing elevation, concentrations of total P, NaHCO₃-Pi, NaOH-Pi, and HCl-Pi, HCl-Po, and orthophosphate increased gradually, while the concentrations of H₂O-Pi, H₂O-Po, NaHCO₃-Po, and NaOH-Po, pyrophosphate, orthophosphoric monoester and diester showed a unimodal distribution pattern. Meanwhile, MAT, MAP, and PD were identified as the key factors influencing the vertical distribution and transformation of P forms in the soil. The labile Po was the most significant fraction of bioavailable Po pools, with acid phosphatase and phosphodiesterase playing pivotal roles in regulating P availability in the soil. Under persistent climate warming scenarios, alterations in P speciation characteristics may lead to diminished phosphorus supply capacity of soil in alpine grassland ecosystems, thereby exacerbating the effects of P limitation on ecosystem productivity.