Alpine shrublands in the Three Rivers Source Region (TRSR) store substantial soil total nitrogen (N); however, limited information is available regarding its storage and controlling factors. To quantify the storage and controlling factors of soil total N stock, we analysed 66 soil profiles from samples obtained from 22 shrubland sites located across the TRSR on the Tibetan Plateau. Analytical methods, such as ordinary least squares regression, one-way analysis of variance, curve estimation, and variation partitioning were used to evaluate the effects of soil characteristics (soil organic carbon), vegetation characteristics (community types and ground cover of shrublands), climatic factors (mean annual temperature - MAT), and topographical features (slope) on soil N stock. Our results showed that soil N storage at a soil depth interval of 0-100 cm was 63.10 ± 27.41 Tg (Tg = 1012 g), with an average soil N stock of 2.44 ± 1.06 kg m-2 in the TRSR shrublands. Although the type of vegetation community had a small effect on soil N stock, the latter increased with increasing shrubland ground cover and soil organic carbon. However, soil N stock decreased with increasing topographical slope and MAT. Furthermore, changes in MAT primarily affected the N stock of topsoil. Among all the controlling factors, soil organic carbon explained most of the variation in the soil N stock. Considering the effects of global warming, an increase in MAT has decreased the soil N stock. Long-term monitoring of changes in soil N stock should be conducted to improve the precise estimation of soil N storage across the shrublands in the TRSR of the Tibetan Plateau.
Nitrogen (N) is an important limiting nutrient in northern ecosystems and is widely considered to be one of the most important elements in nutrient cycles (
Human activities, such as fertiliser use, industrial development, burning of biomass, expansion of agriculture, and deforestation have affected the N cycle (
Soil characteristics (
Geomorphic disturbances and terrain characteristics (
In Mediterranean regions, vegetation types, including native and reforested plants, have also been found to affect soil N storage (
On the Tibetan Plateau, the following factors affect the soil N storage: yak grazing in the alpine meadows (
Although uncertainties in the estimation of global N stocks usually exist, regional evaluations increase the precision of estimates (
The study area was situated between latitudes 31.65 °N to 37.02 °N and longitudes 89.40 °E to 102.45 °E. The TRSR, also known as the “water tower of China” due to it consisting mainly of the Lancang (Mekong), Yellow, and Yangtze rivers, is situated in the central Tibetan Plateau which has the highest altitude, largest area, and most crowded distribution of rivers, lakes, and glaciers worldwide (
The climate of the TRSR represents a typical continental plateau climate (
To estimate the soil N stock in the alpine shrublands of the TRSR, a total of 66 soil profiles were systematically sampled from 22 typical sites in the shrublands from July to August from 2011 to 2013 (
To analyse the effects of climatic factors on soil N stock, the MAT data for each site were obtained from the WorldClim database (http://www.worldclim.org/), with a spatial resolution of 1 × 1 km (
To determine the soil N stock and storage (their results and corresponding standard deviations using the 22 sites) at each soil depth, the following formulas were used (
where
The effects of MAT, SOC, and topographical slope of the shrublands on soil N stock were determined using ordinary least squares regression. Curve estimation was used to estimate the relationship between the ground cover of shrublands and soil N stock. Normality and homogeneity of variance were satisfied, and one-way analysis of variance (ANOVA) was used to compare the differences between various dominant community types. These analyses were performed using SPSS® v. 22.0 (IBM Corporation, Armonk, NY, USA) and the graphs were drawn using Origin® 2017 (OriginLab, Northampton, MA, USA). Variables that did not significantly contribute to the soil N stock were excluded from the variation partitioning analysis. The variation in soil N stock at depths of 0-30, 30-50, and 50-100 cm layers was partitioned using four explanatory factors that included SOC, slope, ground cover, MAT, and their combined effects. This analysis was performed using the “vegan” R software package (
The soil N storage in the TRSR shrublands was 63.10 ± 27.41 Tg at 0-100 cm soil depth interval, with an average soil N stock of 2.44 ± 1.06 kg m-2 (
SOC significantly affected soil N stock (
Slope was also found to significantly affect soil N stock (
The ground cover of shrublands can significantly stimulate the accumulation of soil N stock; indeed, it was observed to increase with ground cover at all soil depth intervals of 0-30 (
Soil N stock differences were observed for different shrubland community types, especially for
Soil N stock was found to significantly decrease with increasing MAT (
Variables, such as slope, ground cover, SOC, and MAT, can significantly affect soil N stock at soil depth intervals of 0-30 and 0-50 cm; thus, these variables were selected to conduct variation partitioning analyses. Considering that the effect of MAT on soil N stock was insignificant at the soil depth interval of 0-100 cm, it was not included in the variation partitioning analyses for this depth interval. The results showed that the soil N stock was effectively explained by the selected variable, supporting the hypothesis. The amount of variation captured by all selected factors was 94.52%, 94.71%, and 94.31% for soil N stock at soil depth intervals of 0-30, 0-50, and 0-100 cm, respectively (
Soil N stock was observed to increase with increasing SOC at all soil depth intervals of 0-30, 0-50, and 0-100 cm across the shrublands of the TRSR (
In the TRSR, SOC has been demonstrated to contribute to a potential sink (
Variations in topography can result in different microclimates, inducing different plant growth and soil properties (
The negative relationship between soil N stock and slope indicated that conservation of soil by revegetation or reduced grazing to restrain soil erosion is necessary to maintain soil fertility (
The amount of soil N storage is related to biotic processes, including decomposition of organic matter and productivity of plants (
Higher ground cover of shrublands can stimulate the accumulation of soil N storage in the shrublands of the TRSR. Increased ground cover is derived from a growing shrubland canopy, which can stimulate more litter input to soil (
Increasing temperatures can accelerate microbial metabolism, which can contribute to gaseous losses by denitrification and volatilisation, decreasing the soil N stock (
It has been found that soil N stock at soil depth intervals of 0-30 cm decreases with increasing MAT in the alpine shrublands of the Tibetan Plateau (
Amid the growing effects of global warming, MAT has been increasing by 1.5 °C in the TRSR (
The controlling factors of soil N stock in the shrublands can explain most soil N stock variables, with the unexplained variations being only 5.48%, 5.29%, and 5.69% at soil depth intervals of 0-30, 0-50, and 0-100 cm, respectively (
Across the shrublands in the TRSR, total soil N storage at the soil depth interval of 0-100 cm was 63.10 ± 2 7.41 Tg and the average soil N stock was 2.44 ± 1.06 kg m-2. Increasing the ground cover of shrublands stimulated soil N accumulation. However, MAT and topographical slope were negatively related to soil N stock. Furthermore, the effect of MAT on soil N stock was concentrated on the topsoil. Most soil N stock variations were explained by climatic factors, soil characteristics, topography, and vegetation type in the TRSR, with SOC explaining the largest variation. In the global climate change scenarios, long-term monitoring of changes in soil N stock is necessary to improve the estimation of soil N stock across the shrublands of the TRSR in the Tibetan Plateau.
We thank Zebing Zhong, Wenzhu Song, Hechun Liu, Changbin Li, Yi Ning and Feng Xiong for facilitating our field surveys on the Tibetan Plateau and for their laboratory assistance. This study was funded by the National Key Research and Development Program of China (2019YFC0507404), the National Program on Basic Work Project of China (2015FY11030001-5), Natural Science Foundation of Qinghai (2019-ZJ-910), Qinghai Province International Exchange and Cooperation Project (2019-HZ-807), and the Strategic Priority Research Program of CAS (XDA0505030304).
The distribution of 22 sampling sites in the shrublands of the Three Rivers Source Region shown on a 1:1.000.000 scale background vegetation map (
Soil nitrogen (N) content at different soil depths across the shrublands of the Three Rivers Source Region.
Relationship between soil organic carbon (SOC) and soil nitrogen (N) stock in the shrublands of the Three Rivers Source Region at the soil depth intervals of (A) 0-30 cm, (B) 0-50 cm, and (C) 0-100 cm.
Relationship between slope and soil nitrogen (N) stock in the shrublands of the Three Rivers Source Region at the soil depth intervals of (A) 0-30 cm, (B) 0-50 cm, and (C) 0-100 cm.
Relationship between ground cover and soil N stock in the shrublands of the Three Rivers Source Region at the soil depth intervals of (A) 0-30 cm, (B) 0-50 cm, and (C) 0-100 cm.
Relationship between vegetation types and soil N stock in the shrublands of the Three Rivers Source Region at the soil depth intervals of (A) 0-30 cm, (B) 0-50 cm, and (C) 0-100 cm. The solid black squares and error bars represent the mean of each measurement and corresponding standard deviations, respectively.
Relationships between mean annual temperature (MAT) and soil nitrogen (N) stock in the shrublands of the Three Rivers Source Region at the soil depth intervals of (A) 0-30 cm, (B) 0-50 cm, and (C) 0-100 cm.
Results of variation partitioning analysis for soil nitrogen (N) stock at the soil depth intervals of in (A) 0-30 cm, (B) 0-50 cm, and (C) 0-100 cm across the shrublands of the Three Rivers Source Region. Variation partitioning analysis consists of explained variation, including effects of slope, ground cover, soil organic carbon (SOC), mean annual temperature (MAT), and their joint effects, and unexplained variations.
Stock and storage of soil nitrogen (N) from soil surface to a depth of 100 cm in the shrublands of the Three Rivers Source Region. The results of soil N storage are shown as mean values ± standard deviation (SD).
Soil depth(cm) | Soil N stock (kg m-2) | Soil N storage(Tg) | |||
---|---|---|---|---|---|
Min | Max | Mean | SD | ||
0-10 | 0.14 | 0.79 | 0.40 | 0.17 | 10.43 ± 4.44 |
0-20 | 0.35 | 1.45 | 0.80 | 0.28 | 20.78 ± 7.35 |
0-30 | 0.45 | 1.87 | 1.14 | 0.38 | 29.44 ± 9.72 |
0-50 | 0.61 | 2.69 | 1.64 | 0.54 | 42.40 ± 14.02 |
0-70 | 0.61 | 3.58 | 2.02 | 0.75 | 52.22 ± 19.39 |
0-100 | 0.61 | 4.72 | 2.44 | 1.06 | 63.10 ± 27.41 |