ISWC OpenIR  > 水保所知识产出(1956---)
黄土高原地区土壤养分的空间分布及其影响因素
刘志鹏
Subtype博士
Thesis Advisor邵明安
2013-05
Degree Grantor中国科学院研究生院
Place of Conferral北京
Keyword黄土高原 土壤养分 空间变异性 尺度依赖性 土地利用 地质统计学 状态空间模拟
Abstract

严重的水土流失、土壤肥力和生产力下降、土地退化和荒漠化等一系列生态环境
问题威胁着黄土高原地区自然、社会和经济的可持续发展。土地利用调整、植被恢复
以及水土保持工程的大力开展是实现该区生态恢复和重建的根本途径。在土壤贫瘠的
黄土高原,土壤养分是制约植被生长和土地生产力的重要因素,同土壤水分一起影响
着该区土地利用和植被类型的空间分布。从区域尺度上,系统而科学地认识整个黄土
高原地区土壤养分资源空间分布特征及其影响因素,是各项生态重建工程有效实施的
有力保障,对区域宏观决策具有重要的指导意义。
本论文以整个黄土高原地区为研究区域,旨在摸清现阶段主要土壤养分的含量及
储量水平,明晰其空间变异特征及空间分布格局,并揭示不同尺度下土壤养分与相关
环境因素之间的关系。研究团队于 2008 年在黄土高原地区开展了大规模、高密度的
野外试验,在 382 个代表性采样点上共采集表层 0-20 cm 和 20-40 cm 以及深层 0-100
cm 和 100-200 cm 土壤混合样品共 1528 个;表层 0-20 cm 和 20-40 cm 原装土壤样品
764 个;同时获取了各采样点经纬度、海拔高度、坡向、坡位、坡度、土地利用、植
被类型等相关信息。室内统一测定的指标有土壤有机碳、土壤全氮、土壤全磷、土壤
全钾、土壤 pH 值、土壤机械组成(美国制分级)、土壤容重等。使用经典统计学(相关
分析、线性回归、方差分析、多重比较等)、地质统计学(半方差函数、克里格插值、
因子克里格分析)、状态空间模拟等方法,对各土壤养分指标的空间变异特征、空间
插值及预测、不同空间尺度下的主要影响因素进行了分析,主要研究结果如下:
(1) 黄土高原地区表层土壤有机碳浓度变化在 0.38-54.03 g kg -1 之间,0-20 cm 和 20-40
cm 土层中的均值分别为 10.34 g kg -1 和 6.78 g kg -1 ;表层 0-20 cm 和 0-40 cm 土壤
中有机碳密度分别为 2.64 kg m -2 和 4.57 kg m -2 ,相应深度的有机碳总储量分别为
1.64 Pg 和 2.86 Pg (1Pg = 10 15 g)。深层 0-100 cm 和 0-200 cm 土壤有机碳密度分别
为7.70 kg m -2 和12.45 kg m -2 ,相应深度的有机碳总储量分别为4.78 Pg和 5.85 Pg。
黄土高原地区 0-20 cm 和 0-100 cm 深度土壤有机碳总储量分别占世界总储量的0.36%和 0.31%,占我国土壤有机碳总储量的 8.21%和 5.32%。表层土壤有机碳浓
度和密度的变异系数表明,它们在区域范围内均表现为中等程度变异。皮尔逊相
关分析结果表明表层土壤有机碳浓度与相关环境因素之间存在显著(p<0.05)相关
关系,其中与土壤全氮、土壤 pH 值以及粘粒含量关系最为密切。多元逐步线性
回归(p<0.05)结果表明,土壤全氮、粘粒含量、土壤 pH 值、海拔高度以及气温对
土壤有机碳浓度具有显著影响。方差分析(p<0.05)结果表明,气温、降雨、海拔高
度、土壤细颗粒(<20μm)以及土地利用类型对土壤有机碳储量影响显著。区域尺度
上,农地中土壤有机碳含量显著高于林地和草地。地质统计学结果表明,表层土
壤有机碳浓度(0-20 cm 和 20-40 cm)和密度(0-40 cm)的最大空间相关距离分别为
384 km、393 km 和 339 km。块金基台比分别为 0.52, 0.50 和 0.45,表明土壤有机
碳在区域尺度上表现为中等程度空间依赖性。使用克里格插值方法绘制了整个黄
土高原地区表层土壤有机碳浓度和密度的空间分布图。总体而言,整个区域中心
有一个土壤有机碳含量相当低的区域,它被由中心向区域边界发散的含量递增的
同心圆包围着。这种分布特点与区域尺度上大地形区的分布相对应,同时也影响
着土地利用类型的分布:黄土高原的四周边界上分布有山地;河谷平原区主要分
布在中部地区;鄂尔多斯风沙台地位于整个区域的中部。
(2) 不同土地利用类型下,表层 0-20 cm 土层中土壤全氮和全磷浓度均值变化在
0.58-0.81 g kg -1 和 0.50-0.73 g kg -1 之间;20-40 cm 深度土层中,土壤全氮和全磷
浓度均值的变化范围分别为 0.46-0.60 g kg -1 和 0.48-0.61 g kg -1 ;0-40 cm 深度土壤
全氮和全磷密度均值变化范围分别为 0.27-0.39 kg m -2 和 0.27-0.38 kg m -2 。黄土高
原地区 0-40 cm 深度土壤全氮和全磷的总储量为 0.217 Pg 和 0.205 Pg,占到我国
总储量的 5.4%和 7.3%。变异系数表明,不同土地利用类型下土壤全氮和全磷含
量均表现为中等程度变异。方差分析(p<0.05)结果表明,气温、降雨以及土地利用
类型对表层土壤全氮和全磷含量具有显著影响。降雨和气温的影响在不同土地利
用类型下不同,土地利用类型的影响在不同降雨和气温区内也表现出不同的特点。
总体而言,农地中土壤全氮和全磷含量高于林地和草地;具有较多降雨和较高气
温的地区土壤全氮和全磷含量更高。相关和线性回归结果表明,土壤全氮和全磷
与相关环境因素,如土壤有机碳、降雨、气温、海拔高度、经纬度、坡度、粘粒
和粉粒含量以及土壤 pH 值之间存在显著的相关关系,而它们之间的关系随土地
利用类型的变化而不同。针对不同土地利用类型,建立了土壤全氮和全磷的线性
预测方程。区域尺度上,土壤全氮和全磷浓度和密度的块金基台比表明,它们具有中等程度空间依赖性,最大空间相关距离分别在 374-461 km 以及 546-664 km
之间。通过克里格插值绘制了黄土高原地区表层 0-40 cm 深度土壤全氮和全磷密
度的空间分布图。
(3) 土壤全钾浓度在表层 0-20 cm 和 20-40 cm 土层中变化范围分别为 10.07-30.97 g
kg -1  和 12.82-32.39 g kg -1 ,均值分别为 19.25 g kg -1 和 19.10 g kg -1 ;变异系数分别
为 31.7%和 26.9%,表现为中等程度变异;块金基台比分别为 31.7%和 26.9%,在
最大相关距离 546 km 和 564 km 范围内表现出中等程度空间依赖性。使用经典线
性回归和状态空间模拟方法分析了表层 0-20 cm 土壤全钾含量与土壤容重、粘粒
和粉粒含量、土壤 pH 值、降雨、气温以及海拔高度之间的相互关系。最优状态
空间方程能够解释 97%的土壤全钾总变异,而最优线性回归模型仅能解释 26%的
总变异。使用相同变量,所有的状态空间方程在预测土壤全钾含量时均优于对应
的线性回归方程。气温、容重以及粘粒含量被认为是影响土壤全钾局地变异的最
重要因素,因为它们均出现在最优状态空间方程中。结果表明,空间状态方程可
以作为有效的工具很好地模拟区域尺度上土壤养分的局地空间变异。
(4) 土壤 pH 值在表层 0-20 cm 土层中变化范围为 6.06-10.76,均值为 8.49,中位数为
8.48。土地利用类型对土壤 pH 值具有显著(p<0.05)影响,草地土壤 pH 值显著高
于农地和林地。区域尺度上,土壤 pH 值表现出较弱程度的变异和较强的空间依
赖性,其变异系数为 5%,块金基台比为 0.243。以土壤 pH 值为研究对象,分析
了四种空间插值方法,即反距离法、样条函数法、普通克里格法和协克里格法及
其相关参数对空间插值精度的影响。交叉检验结果表明,克里格方法相对于反距
离法和样条函数法具有更高的精度,而使用土壤有机碳作为协变量的协克里格法
能够进一步提高插值精度。四种空间插值方法绘制的土壤 pH 值空间分布图,在
整体上具有相似性,在细节上存在不同。整体而言,土壤 pH 值的相对低值区分
布在黄土高原的东南部,可能与该区较多降雨、土壤淋溶以及较高的土壤有机质
含量有关。土壤 pH 值的相对高值区分布在黄土高原的中北部,与该区干旱的环
境和不合理灌满以及严重的土壤盐渍化有关。
(5) 经典统计学中的相关分析并没有考虑各变量自身以及相互间的空间位置关系以
及它们的区域化特征。使用结合了多元统计和地质统计学的因子克里格方法研究
了各土壤性质(土壤有机碳、全氮、全磷、全钾、土壤 pH 值、容重、粘粒和粉粒
含量)与相关环境因素(气温、降雨、海拔高度、土壤类型和土地利用类型)之间的
尺度依赖性相关关系。使用含有块金效应和两个球状结构的协同区域化线性模型拟合单变量以及双变量交互的半方差函数,并据此分别研究了块金尺度(<30-50
km),短变程尺度(最大相关距离 200 km)以及长变程尺度(最大相关距离 400 km)
上各变量之间存在的尺度依赖性相关关系。使用了主成分分析以及单位圆投影的
方法表达了多变量之间复杂的相关关系。结果表明,各变量之间的相关关系随尺
度的变化表现出不同特征。总体而言,土壤有机碳和全氮在块金尺度和长变程尺
度上紧密相关,而在短变程尺度相关性不明显。降雨和土壤粘粒含量在块金尺度
和长变程尺度上与土壤全磷含量关系密切。土壤全钾在各尺度上与其它变量之间
的相关关系均不明显,但其与土壤类型在长变程尺度上关系密切。土壤 pH 值在
块金尺度、短变程尺度以及长变程尺度上分别与土壤容重、土壤类型以及海拔高
度密切相关。土壤容重与土地利用类型在各尺度上都具有紧密关系。土壤利用类
型和土壤类型被认为是控制短变程尺度上土壤性质空间变异的主要影响因素,而
影响长变程尺度土壤变异的主要因素为降雨、气温和海拔高度。
本论文以大量的野外试验数据为基础,从整体上认识了黄土高原地区土壤有机碳、
全氮、全磷、全钾以及土壤 pH 值的空间变异特征及其与相关环境因素之间的关系。
可靠的土壤养分空间数据丰富了黄土高原土壤数据库,为该区土壤养分空间变异及相
关研究的深入开展提供了整体框架指导,也为今后该区大尺度上数字土壤制图、碳氮
循环模拟、面源污染评估等提供了可靠的数据支持,相关研究结果也将为黄土高原地
区各项生态重建工程的宏观决策提供理论和实践指导。
关键词:黄土高原;土壤养分;空间变异性;尺度依赖性;土地利用;地质统计学;
状态空间模拟

Other Abstract

A series of eco-environmental problems, such as severe soil erosion, decreases in soil
fertility and productivity, land degradation and desertification, have threatened the
sustainable development of the ecosystems and social economy on the Loess Plateau of
China. Regional and local projects aiming on ecological restoration and reconstruction
have been launched to combat these problems on this area, practically through land use
optimization, vegetation recovery and soil and water conservation. Soils on the Loess
Plateau are poverty due to the scarcity of both soil nutrients and water resources. These
two factors together greatly limit plant growth and agricultural production, controlling the
spatial variations of land use and vegetation. From regional perspective, systematic and
accurate information on the spatial variations of soil nutrients and the impact factors at
different scales is basic and essential for effective applications of these ecological projects,
and would be helpful in related macro policy makings.
The main purposes of this dissertation were to (1) explore the current level and stocks
of several main soil nutrients across the entire Loess Plateau region; (2) to reveal the
spatial variability of these soil nutrients and illustrate their distribution patterns; (3) to
investigate the relationships between these soil nutrients and pertinent environmental
factors at different scales; (4) and to generate accurate prediction models using
easy-to-measure variables. An intensive soil survey with a sampling interval of about
30-50 km was accomplished within one year during 2008, by investigating 382
representative sampling sites across the entire Loess Plateau region (62.4 × 10 4 km 2 ). A
total of 1528 composite soil samples were corrected using a handy soil auger (5 cm in
diameter) from 0-20 cm and 20-40 cm topsoil layers, and 0-100 cm and 100-200 cm deep
soil layers. Additionally, 764 undisturbed soil cores were collected with cutting rings (100  cm 3 in volume). The environmental conditions of each sampling site were recorded, such
as latitude, longitude, elevation, aspect, slope gradient, slope position, land use type and
vegetation type. All the soil samples were taken to the laboratory for measurements of soil
organic carbon (SOC), soil total nitrogen (STN), soil total phosphorus (STP), soil total
potassium (STK), soil pH, soil mechanical composition and bulk density (BD). Traditional
statistics methods (correlation analysis, linear regression, ANOVA and post-hoc),
geostatistics methods (semivariogram, kriging interpolation, factorial kriging analysis) and
state-space modeling approach were used for spatial analysis and generation of prediction
models. The main resultes are listed as follows:
(1) Soil organic carbon concentrations (SOCC) varied within a wide range throughout the
region from 0.38 g kg -1  to 54.03 g kg -1 , with mean values of 10.34 g kg -1 and 6.78 g
kg -1 for the topsoil (0-20 cm) and subsoil (20-40 cm), respectively. The mean SOC
density (SOCD) was 2.64 kg m -2 in the 0-20 cm soil layer and 4.57 kg m -2 in the 0-40
cm soil layer, and it was estimated that 1.64 and 2.86 Pg (1Pg = 10 15 g) of organic
carbon were stored in these soil layers, respectively. Estimates for deeper soil layers
indicated that mean SOCD in the 0-100 cm and 0-200 cm layers was 7.70 and 12.45 kg
m -2 , respectively, while the total organic carbon stocks amount to 4.78 Pg (0-100 cm)
and 5.85 Pg (0-200 cm), respectively. The SOC stocks in the 0-20 cm and 0-100 cm of
soils in the Loess Plateau region contribute 0.36% and 0.31% to the global SOC stored
in these respective layers. In addition, our results indicated that the 0-20 and 0-100 cm
soil layers of the Loess Plateau, which covers nearly 6.5% of the area of China,
currently holds about 8.21% and 5.32% of the total SOC stocks in these layers in China,
respectively. Coefficient of variation values showed moderate variation for both SOC
concentration and density values in both 0-20 cm and 20-40 cm soil layers. Significant
correlations were detected between SOCC and these environmental variables, notably
with soil total nitrogen (STN), soil pH and clay content. Multiple linear regression
analysis indicated that STN, clay content, soil pH, elevation and temperature had
greater effects on regional SOCC variability, among all the selected soil and site
variables. The results of ANOVA showed that precipitation, temperature, elevation,
clay plus silt contents and land use showed significant regional impacts on SOCD. The
results also show that human activities have heavily affected SOC accumulation.  Measured SOCD under cropland was relatively higher than under grassland and
forestland.
Geostatistics analysis showed that the maximum autocorrelation ranges were 384
km, 393 km and 339 km for SOCC (0-20 cm and 20-40 cm) and SOCD (0-40 km),
respectively. Nugget-to-sill ratios were 0.52, 0.50 and 0.45, which also indicated
moderate spatial dependence. The distribution maps of SOCC in both topsoil layers and
SOCD in 0-40 cm soil layers were produced by geostatistical method showed that the
overall spatial pattern was characterized by an area of low SOC content surrounded by
bands with higher values that generally increased towards the region’s boundaries. The
distribution pattern corresponded to that of the major regional landforms, which also
influenced land use, whereby the sandy Ordos Plateau is surrounded by relatively
fertile plains and valleys, where the human population density is highest, and the
regional boundary is mountainous.
(2) In 0-20 cm soil layers, mean STN concentrations (STNC) and STP concentrations
(STPC) ranged from 0.58 g kg -1 to 0.81 g kg -1 and from 0.50 g kg -1 to 0.73 g kg -1 ,
respectively, under different land types. In 20-40 cm soil layers, mean STNC and STPC
ranged from 0.46 g kg -1 to 0.60 g kg -1 and from 0.48 g kg -1 to 0.61 g kg -1 , respectively.
Mean STN and STP densities in 0-40 cm soil layers ranged from 0.27 kg m -2 to 0.39 kg
m -2 and from 0.27 kg m -2 to 0.38 kg m -2 , respectively, under different land use types.
All the concentrations and densities of STN and STP under different land use types
showed moderate variations, which was indicated by the values of coefficient of
variation. We detected significant (p<0.05) effects of land use, precipitation and
temperature on both STN and STP. But the results varied among different precipitation
and temperature regions and different land use types. Generally, croplands had higher
concentrations and densities of STN and STP than forestlands and grasslands, and
regions with more precipitation and higher temperature had higher STN and STP
densities. Significant correlations were found between STN, or STP, with selected
factors, i.e. soil organic carbon, precipitation, temperature, elevation, latitude, longitude,
slope gradient, clay content, silt content and soil pH. The results were not consistent
within either the variable or the land use types. We generated land-use specific linear
models to predict STN and STP using these related variables. Geostatistical analysis showed moderate spatial dependence of both STN and STP, indicated by the values of
nugget to sill ratio. The spatial range of STN and STP ranged from 374 km to 461 km
and 546 km to 664 km, respectively. This range was much larger than our sampling
intervals (30-50 km). The distribution maps of STN and STP densities were made with
kriging interpolation. Finally, the stock of STN and STP was estimated to be 0.217 Pg
and 0.205 Pg in the upper 0-40 cm soil layers, which was about 5.4% and 7.3% of the
total nitrogen and phosphorus stocks in China. Our study suggests that it is important to
take land use into account when considering variation of STN and STP at regional
scale.
(3) In 0-20 cm and 20-40 cm soil layers, soil total potassium (STK) concentration varied
from 10.07-30.97 g kg -1 and 12.82-32.39 g kg -1 , with mean values of 19.25 g kg -1 and
19.10 g kg -1 , respectively. The coefficients of variation for STK were 13.4% and 13.3%,
defined as moderate variation. The spatial ranges of STK were 546 km and 564 km.
The nugget-to-sill ratios were 31.7% and 26.9%, showing moderate spatial dependence.
Two methods, state-space modeling and classical linear regression, were used to
quantify the relationships between STK (0-20 cm) and bulk density, clay and silt
content, soil pH, precipitation, temperature, and elevation. The best state-space models
explained more than 97% of the STK variation, while the best linear regression model
explained less than 26% of the STK variation. The results showed that all the
state-space models described the spatial variation of STK much better than the
corresponding linear regression models. Temperature, bulk density and clay content
were identified as important factors that affected localized variation of STK, since they
were connected to the better performance of the state-space models. State-space
modeling is recommended as a useful tool for quantifying spatial relationships between
soil properties and other environmental factors in large-scale regions.
(4) In 0-20 cm soil layers, soil pH values ranged from 6.06 to 10.76, with a mean of 8.49
and a median of 8.48. Land use type had a significant effect (p < 0.01) on soil pH;
grassland soils had higher pHs than cropland and forestland soils. From a regional
perspective, soil pH showed weak variation and strong spatial dependence, indicated
by the low values of the coefficient of variation (5%) and the nugget-to-sill ratios
(<0.25). Indices of cross-validation, i.e. average error (AE), mean absolute error  (MAE), root mean square error (RMSE) and model efficiency coefficient (MEC) were
used to compare the performance of the four different interpolation methods, i.e.
inverse distance weighting (IDW), splines, ordinary kriging and cokriging. The results
showed that kriging methods interpolated more accurately than IDW and splines.
Cokriging performed better than ordinary kriging and the accuracy was improved by
using soil organic carbon as an auxiliary variable. Regional distribution maps of soil
pH were produced. The southeastern part of the region had relatively low soil pH
values, probably due to higher precipitation, leaching, and higher soil organic matter
contents. Areas of high soil pH were located in the north of the central part of the
region, possibly associated with the salinization of sandy soils under inappropriate
irrigation practices in an arid climate.
(5) Traditional statistical analysis of the correlations between spatially distributed variables
takes no account of their regionalized nature. Factorial kriging analysis (FKA) was
applied to investigate scale-dependent correlations between selected soil properties (i.e.
soil organic carbon (SOC), soil total nitrogen (STN), soil total phosphorus (STP), soil
total potassium (STK), soil pH, bulk density (BD), and clay and silt contents) and
environmental factors (i.e. elevation, precipitation, temperature, land use type and soil
type). A linear model of coregionalization, including a nugget effect and two spherical
structures (effective ranges of 200 and 400 km), was fitted to the experimental auto and
cross-variograms of the variables. Scale-dependent correlations were calculated for
nugget effect scale (<30-50 km), short-range scale with a range of 200 km and
long-range scale with a range of 400 km. Principal component analysis was conducted
to clearly illustrate the correlations at each spatial scale. The scale-dependent
correlations were different from the general correlations and varied at different scales.
Generally, SOC and STN were strongly correlated at the nugget effect scale and the
long-range scale, but not at the short-range scale. Precipitation and clay content showed
close correlations with STP at the nugget effect scale and long-range scale. The STK
was weakly correlated with the other variables at each spatial scale, but closely
correlated with soil type at the long-range scale. Soil pH was closely correlated with
BD, soil type and elevation at the nugget effect, short and long spatial scales,
respectively. Close correlations were found between BD and land use type at each spatial scale. Land use and soil type were considered to be the important factors
controlling spatial variation of soil properties at the short-range scale while at the
long-range scale the likely factors were identified as precipitation, temperature and
elevation.
Based on intensive filed sampling and uniform laboratory measurements, our study
provided an overview on the spatial variation and impact factors of soil organic carbon,
soil total nitrogen, soil total phosphorus, soil total potassium and soil pH across the entire
Loess Plateau region of China. The reliable spatial data updated the soil database for the
study region, and can be used as important input layers in regional digital soil mapping,
carbon and nitrogen cycle modeling and evaluation of the potential non-point source
pollution associated with soil erosion. Moreover, the results presented in this dissertation
can serve as an important background for the future studies in related fields, and can be
useful in macro decision making for regional eco-environment restoration on the Loess
Plateau.
Keywords: The Loess Plateau, Soil nutrients; Spatial variability; Scale dependency, Land
use; Geostatistics; State-space modeling 

Language中文
Document Type学位论文
Identifierhttp://ir.iswc.ac.cn/handle/361005/8981
Collection水保所知识产出(1956---)
Recommended Citation
GB/T 7714
刘志鹏. 黄土高原地区土壤养分的空间分布及其影响因素[D]. 北京. 中国科学院研究生院,2013.
Files in This Item:
File Name/Size DocType Version Access License
黄土高原地区土壤养分的空间分布及其影响因(4678KB)学位论文 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[刘志鹏]'s Articles
Baidu academic
Similar articles in Baidu academic
[刘志鹏]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[刘志鹏]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.