Content uploaded by Kaihua Liao
Author content
All content in this area was uploaded by Kaihua Liao on Jun 05, 2015
Content may be subject to copyright.
Monitoring and prediction of soil moisture spatial–temporal
variations from a hydropedological perspective: a review
Qing Zhu
A,D
, Kaihua Liao
A
, Yan Xu
B
, Guishan Yang
A
, Shaohua Wu
C
, and Shenglu Zhou
C
A
State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography
and Limnology, Chinese Academy of Sciences, Nanjing 210008, China.
B
Department of Municipal Engineering, School of Civil Engineering, Southeast University,
Nanjing 210096, China.
C
School of Geographic and Oceanographic Sciences, Nanjing University, Nanjing 210093, China.
D
Corresponding author. Email: qzhu@niglas.ac.cn
Abstract. Accurate prediction of soil moisture spatial–temporal variations remains critical in agronomic, hydrological,
pedological, and environmental studies. Traditional approaches of soil moisture monitoring and prediction have limitations
of being time-consuming, labour-intensive, and costly for direct field observation; and having low spatial resolution for
remote sensing, and inconsistent accuracy and reliability for landscape feature (e.g. topography, land use, vegetation)
modelling. Innovative and effective approaches for accurate soil moisture simulation are needed. Pedological properties,
including soil structure, particle size distribution, porosity, horizon, redox feature, and organic matter content, have been
accepted as important factors controlling soil moisture and can be potentially used in soil moisture prediction. However,
pedological properties mostly lack quantification (e.g. redox feature, horizon, soil structure), and soil sampling and analysis
are time-consuming and costly, especially at large spatial scale. These limitations have restricted the utilisation of
pedological information to predict soil moisture spatial–temporal variations at different spatial scales. To overcome
these difficulties, new tools including geophysical tools and computed tomography, and new methods including mining
soil survey information and integrating pedological information with landscape features and modelling, are proposed in
this paper.
Additional keywords: geophysics, hydropedology, soil hydrology, soil water content.
Received 16 August 2012, accepted 30 December 2012, published
online
5 February 2013
Introduction
Why should
we investigate soil moisture variation?
Soil moisture is a fundamental hydrological state variable and
its spatial pattern is important for understanding hydrological
processes. However, spatial–temporal variability of soil
moisture remains a challenge to be predicted accurately (Owe
et al. 1982; Grayson et al. 1997). Determination and prediction
of soil moisture at different scales have therefore received
much attention in a wide range of agronomic, hydrological,
pedological, and environmental studies (e.g. Davidson et al.
2000; Grayson et al. 2002; Lin et al. 2006a, 2006b; Gish et al.
2011). The importance of studying soil moisture variation can
be simply shown from two perspectives:
First, from the biogeochemistry perspective, soil moisture
variation controls microbial activity, which determines
processes such as respiration, CO
2
efflux, nitrification,
denitrification, mineralisation, and other chemical
transformations (Rodriguez-Iturbe et al. 1999; Schjønning
et al. 2003; Lin et al. 2005a). These processes consequently
affect environmental quality and agricultural economics,
including greenhouse gas emission, nutrient transformation and
storage, fertilisation rate, and crop yield. For example, in the
‘Hole-in-the-Pipe’ conceptual model developed by Davidson
et al.(2000), soil moisture variation determines anaerobic and
aerobic microbial activities, denitrification and nitrification
processes, and hence soil nitrogen (N) availability, N
x
O gas
emission, and nitrate leaching (Fig. 1). Other more practical
research has also reported the importance of soil moisture
variation in controlling nutrient losses and availability, and CO
2
emission at different spatial scales. For example, Schmidt et al.
(2007, 2011) observed that corn yield and optimal N fertilisation
rate were highly correlated with soil moisture variation in June
and July at a typical hillslope of the northern Appalachian Ridge
and Valley physiographic province in the USA. Castellano
et al.(2011), using soil column studies in Maryland, USA,
reported that maximum CO
2
efflux occurred when soil moisture
content was close to field capacity.
Second, from the physics perspective, soil moisture variation
controls soil physical properties, which determine processes
such as infiltration and runoff, transportation and storage of
chemicals and energy, mechanical strength, and surface albedo
(Loague 1992; Lin et al. 2005a). Antecedent soil moisture has
been recognised as one of the controlling factors of infiltration,
Journal compilation CSIRO 2012 www.publish.csiro.au/journals/sr
CSIRO PUBLISHING
Soil Research, 2012, 50, 625–637
Review
http://dx.doi.org/10.1071/SR12228
surface runoff, and soil erosion processes (e.g. Li et al. 2011 ;Wu
et al. 2011; Defersha and Melesse 2012). These processes, as a
consequence, are critical in the transportation of sediments,
nutrients, heavy metals, and organic chemicals from soil to
surface and ground water. For example, van Verseveld et al.
(2009) observed that drier antecedent soil moisture with less
prior flushing resulted in the highest dissolved organic carbon
(C) and N concentrations in storm runoff. Blake et al.(2003)
found that dissolved heavy metal losses were dominant in storms
that occurred with high antecedent soil moisture, whereas
particulate heavy metal losses were dominant in storms that
occurred in dry periods. Soil moisture variation also affects
heat exchange and storage by influencing surface albedo,
evapotranspiration, and soil heat storage capacity (Graser and
Vanbavel 1982; Mintz and Walker 1993). Water has the greatest
heat storage capacity of ~4.2 10
3
JL
–1
K
–1
, whereas that of
soil minerals is 2.1–2.6 10
3
JL
–1
K
–1
. Therefore, soil moisture
determines the soil heat storage capacity and heat exchange
between different spheres in the critical zone. For example, Roxy
et al.(2010) found that surface albedo decreases exponentially
with an increase in soil moisture, while soil thermal diffusivity
increases first and then decreases with the increase of soil
moisture. Guan et al.(2009) observed a power relationship
between soil thermal conductivity and soil moisture in the
Loess Plateau in China.
As well as the two aspects discussed above, soil moisture
plays an important role in other scientific investigations and
practices. Hedley and Yule (2009) and Cardenas-Lailhacar and
Dukes (2010) used soil moisture monitoring to precisely control
irrigation under field condition. Soil moisture has also been
used to identify and infer subsurface flow paths in different
landscapes. Gish et al.(2005) and Zhu and Lin (2009) used soil
moisture monitoring to validate the lateral subsurface flow path
above the clay layer and the soil–bedrock interface. In addition,
soil moisture is one of the key parameters in a wide range of
modelling. For example, Choi et al.(2012) used microwave-
derived soil moisture to adjust evapotranspiration models for
water-limited cases, and Bisselink et al.(2011) initialised
regional climate simulations over Europe with satellite-
derived soil moisture. Therefore, soil moisture is a
fundamental hydrological state variable for understanding
different critical zone processes.
Advantages and limits of traditional approaches
Direct field soil moisture measurement
Among the approaches for investigating soil moisture variation,
in situ observation is the most straightforward and accurate.
Traditional field soil moisture measurement can be classified
into two types: destructive and non-destructive. Destructive
measurement uses a core sampler to collect soil samples with
a certain volume and then weighs soil samples before and after
drying (at 1108C) to calculate the gravitational and volumetric
soil water contents. With advances in geophysical technique
and application, non-destructive methods have been developed
and widely used. They mostly rely on a sensor placed in the soil
with an evaluation unit connected to the sensor cables at the
time of measurement (Kutilek and Nielsen 1994). Electrical
resistance, capacitance, gamma radiation, and neutron methods
are typical, non-destructive, soil moisture monitoring
techniques. These methods use different physical principles to
develop the relationship between soil water content and
geophysical parameters. For some of these methods, site-
specific calibration for different soil textures and salt contents
is needed to achieve better accuracy. Details of these methods
can be found in various soil physics text books (e.g. Kutilek and
Nielsen 1994).
In situ soil moisture measurement, both manual and
automatic, has been widely used in several studies (see examples
in Figs 2 and 3). Manual soil moisture measurement can
Atmosphere
Gaseous Phase of Soil
Aqueous Phase
Denitrification
Nitrification
Biological Assimilation
Abiological Reactions
Plants and
Soil Microorganisms
N
2
N
2
O
N
2
O
NO
3
–
NH
4
+
NO
NO
N
2
Fig. 1. Diagram of the ‘hole-in-the-pipe’ conceptual model proposed by Davidson et al.(2000). In this model, soil water
content was conceptualised by the relative size of the hole in the pipe through which nitric oxide and nitrous oxide ‘leak’.
With greater soil water content and less oxygen in the soil, denitrification dominated, and thus more nitric oxide and
nitrous oxide.
626 Soil Research Q. Zhu et al.
generally cover a relative large area (e.g. several hectares) with
sparse monitoring sites (usually 20–500 sites) and low
temporal resolution (usually weekly or monthly). Examples of
using manual soil moisture measurement include: use of time
domain reflectometry (TDR) sensors mounted on an all-terrain
vehicle with an integrated differential global positioning
system to measure the soil moisture content over the top 30 cm
of the soil profile at ~500 sites in Tarrawarra catchment in
Australia (Western et al. 1999, 2004); and measurement of
soil water storage (up to 140 cm) with a neutron probe at
128 sites along a 576-m-long transect in St. Denis National
Wildlife Area, Saskatchewan, Canada (Biswas and Si
2011). However, manual in situ soil moisture measurement
can be time-consuming and labour-intensive and, thus,
generally has low temporal resolution. For example, soil
moisture observations were made only 13 times from
September 1995 to November 1996 (roughly monthly) in the
study of Western et al.(1999), and only 16 times from July 2007 to
August 2009 (roughly bimonthly) in the study of Biswas and
Si (2011).
In contrast to manual soil moisture monitoring, automatic
soil moisture measurement has much higher temporal resolution
(e.g. every 1–10 min), and can well capture the temporal
dynamics of soil moisture. However, automatic soil moisture
probes and data loggers are usually expensive, and these
measurement systems can be installed only in a very limited
number and with poorer spatial coverage than manual soil
moisture monitoring. For example, Lin and Zhou (2008) used
an automatic soil moisture monitoring system to identify
subsurface flow paths in Shale Hills catchment in central
Pennsylvania, USA. However, due to the high cost,
installation of many automatic soil moisture monitoring sites
(e.g. 30–100 sites) was not feasible, and they established only
seven such sites in a small swale (~0.2 ha) instead of covering
the entire catchment (~7.9 ha). Similarly, in Maryland, De
Lannoy et al.(2006) installed only 12 soil moisture probes in
each sub-watershed (~4 ha) of the OPE3 field to automatically
monitor soil moisture content at 10-min intervals. Therefore,
trade-offs between spatial and temporal resolutions determine
the selection between manual and automatic monitoring of soil
moisture.
Remote sensing
Remote sensing is a fast and effective way to obtain soil moisture
variation at large spatial scales (e.g. watershed, regional, and
continental scales). The spectral band of electromagnetic
radiation most sensitive to soil water content is utilised, i.e.
wavelengths equal, to or longer than, those of visible radiation.
Remote sensing detection of soil moisture has the advantages of
geometric accuracy, easy interpretation, relative simple and fast
procedures, good temporal resolution (can be as high as daily),
and acceptable resolution for large-scale investigation (Kutilek
and Nielsen 1994).
The remote sensing technique has been widely used in
detecting soil moisture using various satellite and airborne
images with different wavelengths. One typical example of
using remote sensing to derive soil moisture variation is the
Fig. 2. An example of a manual soil moisture monitoring system in the Shale Hills Critical Zone observatory (Lin
et al. 2006a). Soil moisture was monitored at 77 sites and to a depth of 1.0 m or bedrock. All monitoring sites were
classified into wet, moderate wet, moderate dry, and dry. Wetness index was calculated using the algorithm
proposed by Tarboton (1997).
Spatial-temporal variations of soil moisture from hydropedological perspective Soil Research 627
study by Hain et al.(2009), in which thermal infrared remote
sensing was used to retrieve the available soil water fraction in
Oklahoma, USA, with a coarser spatial resolution of 10 km but
large spatial extent (state scale). In the study by Champagne
et al.(2011), surface agricultural soil moisture was derived
through AMSR-E passive microwave satellite image in Alberta
province, Canada. The temporal resolution of that study, which
was as high as daily (but with substantial noise) and weekly (with
low noise), can exceed most manual soil moisture monitoring,
and its spatial extent, which is the province, can also exceed
regular automatic soil moisture monitoring. Among these
different remote-sensing wavelengths, passive microwave
(1.0–11.0 GHz) is one of the most effective and most used
approaches (Cashion et al. 2005; Das and Mohanty 2006).
Any moisture that is present in the top 5 cm of soil can affect the
amount of microwave radiation that is emitted at low frequencies
(Schmugge et al. 2002).
The use of remote sensing to detect soil moisture variation
also has limitations and constraints. Generally, spatial resolution
of remote sensing is relatively coarse. In published studies, the
spatial resolution of remote sensing used to detect soil moisture
is mostly from 0.1 to 10 km depending on the system used
and spatial scales investigated. Therefore, soil moisture derived
from remote sensing is more suitable, and has always been used,
for large-scale (basin, continental, and even global) climate,
ecological, and hydrological investigations and modelling. For
example, Koyama et al.(2010) derived surface soil moisture
from Multitemporal Envisat satellite Advanced Synthetic
Aperture Radar (ASAR) with a spatial resolution of 150 m
for the River Rur basin, Germany (2364 km
2
). Mladenova
et al.(2011) evaluated C-band (6.6 GHz) based AMSR-E soil
moisture estimates using 1-km resolution in Yanco, New South
Wales (3600 km
2
). In comparison, in-situ soil moisture
monitoring has always been used at the scale of small
catchment, watershed, and farm with an area from several
hectares to square kilometers (e.g. Western et al. 1999; Gish
et al. 2005; Lin and Zhou 2008).
Compared with automatic and even manual soil moisture
monitoring, soil moisture derived from remote sensing has a
much lower temporal resolution. When using the visible
spectrum to derive soil moisture, observations can be made
only during daytime without cloudiness, and detected soil
moisture is limited to a surface layer of ~1 cm or less
(Kutilek and Nielsen 1994). Usually, the temporal resolutions
of common satellites (e.g. MODIS, SPOT5, and Landsat) are
from 1 day to several weeks. Thus, while in-situ soil moisture
monitoring has been aimed at revealing theoretical principles
and controls of storage, transportation, and transformation of
water, chemicals, and heat among different spheres (e.g.
Schmidt et al. 2007; Gish et al. 2011; Zhu and Lin 2011),
soil moisture derived from remote sensing has always been used
for more practical management and modelling of agricultural,
climate, ecological and hydrological issues (e.g. Das et al. 2008;
Pollacco and Mohanty 2012).
In addition, remote sensing can detect only the very surface
soil moisture, and its usage is constrained by other environment
factors including vegetation, surface roughness, soil salinity,
land use, etc. (McColl et al. 2012). Whereas visible spectrum
sensors are restricted by weather and light, microwave sensors
have obvious advantages, such as the ability to retrieve
through non-precipitating cloud cover, which provides shorter
repeat cycles. However, it captures surface soil moisture
(~5 cm) conditions at frequent temporal time scales but at
coarse spatial resolution. Lack of ground truth for validation
is another problem of using remote sensing to detect soil
moisture. As pointed out by Jin and Yan ( 2007 ), Heathman
et al.(2009), and others, soil moisture measurement, which is
critical in developing and validating satellite-based soil moisture
algorithms, is always lacking at different depths.
Using landscape features to predict soil moisture
(indirect approach)
Landscape factors, including topography, vegetation, and
land use are important controlling factors of soil moisture
spatial and temporal variations. Soil moisture spatial variation
was found to be significantly correlated with terrain attributes
(e.g. slope, elevation, and topographic wetness index).
Therefore, terrain attributes have been used to predict soil
moisture variation via regression, geospatial, and hydrological
modelling in several studies (Western et al. 1999, 2004; Lin
et al. 2006a; Takagi and Lin 2012). Influence of vegetation (e.g.
type, cover, distribution, and growth period) on soil moisture
variation has also been reported in several studies, and spatial
information on vegetation (usually interpreted from remote
sensing image) has been used to simulate soil moisture
Fig. 3. An example of automatic soil moisture monitoring system installed
at Kepler Farm in central Pennsylvania, USA (Q. Zhu, H. Lin, unpubl. data).
In this system, EM50 data logger, 5TE soil moisture probes, and MPS-2 soil
matric probes (Decagon Devices, Inc., Pullman, WA) were installed at
critical soil horizons.
628 Soil Research Q. Zhu et al.
variation (Mohanty et al. 2000; Hupet and Vanclooster 2002).
Soil moisture variation was also observed to be affected by and
could be interpreted from spatial information on land use,
including types of land use (e.g. fallow, crop, forest, and
grass) and agricultural practices (e.g. tilling and terracing)
(Famiglietti et al. 1999;Fuet al. 2003). Topography,
vegetation, and land use can now be mapped with high
spatial resolution (e.g. using laser-induced differential
absorption radar for elevation and satellite images for land
use and land cover). Therefore, speed and convenience are
important advantages of using landscape factors to predict
soil moisture variation.
However, since landscape is not the only factor controlling
soil moisture variation, the reliability and accuracy of using
landscape information to predict soil moisture variation are not
consistent in all cases. The study by Qiu et al.(2010)isan
example that has successfully used landscape factors (terrain
and land use) to predict soil moisture variation. However,
even in this study, R
2
values of the prediction models were
generally <0.7, which means that >30% of the spatial variation
of soil moisture has not been captured or explained by terrain
and land use. In other related studies, static landscape factors
alone rarely explained >60% of soil moisture variability.
Western et al.(1999) found that combinations of terrain
indices explain up to 61% of the spatial variation of soil
moisture during wet periods, but only up to 22% during dry
periods in an agricultural watershed in Australia. In a study
conducted at the Kepler Farm in central Pennsylvania, USA,
R
2
values of the multiple linear regressions (stepwise)
between soil moisture and terrain attributes ranged from 0.1
to 0.4 at different depths and seasons (Fig. 4a; Q. Zhu, H. Lin,
unpubl. data). All of these examples suggest that landscape
features are not the only parameters needed for simulating and
predicting soil moisture variation in places with various
topographic characteristics, land use/cover, and climate
conditions. Even in the one location but under different
seasons, antecedent weather conditions, and management
practices, the landscape features cannot be the only
parameters needed in soil moisture simulation and prediction
of variation.
0.1-m depth 0.4-m depth 0.8-m depth
(a) Terrain attributes
(b) Soil properties
(c) Terrain attributes + soil properties
0.00
0.10
0.20
0.30
0.40
0.50
R
2
value
Non-growing season
Growing season
0.00
0.10
0.20
0.30
0.40
0.50
0.00
0.10
0.20
0.30
0.40
0.50
0.15 0.20 0.25 0.30 0.35 0.15 0.20 0.25 0.30 0.35 0.15 0.20 0.25 0.30 0.35
Mean soil moisture (m
3
m
–3
)
Fig. 4. The R
2
values of multiple linear regressions for predicting soil moisture using (a) terrain attributes (topographic wetness
index, slope, elevation, and profile curvature), (b) soil properties (contents of organic matter, sand, silt, clay, and rock fragment, and
depth to bedrock), and (c) terrain attributes plus soil properties, at Kepler Farm in central Pennsylvania, USA (Q. Zhu, H. Lin,
unpubl. data). A stepwise process was adopted and the significance entry level for an independent variable in the stepwise
regression was P < 0.15. The x-axis is the mean soil moisture of the entire farm calculated from all observed soil moisture and used
as an indicator of wetness condition of the entire farm.
Spatial-temporal variations of soil moisture from hydropedological perspective Soil Research 629
Using pedological properties to predict soil moisture
variation
Potentials
Pedological properties, such as texture, depth to bedrock, and
rock fragment content, exert a first-order control on the ability of
a soil to store and transmit water (Famiglietti et al. 1998; Buttle
et al. 2004; Maeda et al. 2006). A common belief is that
topography becomes increasingly important in wet periods,
but that during dry periods, soil moisture distribution depends
primarily on soil or pedological properties (Fig. 5) (Western
et al. 1999; Grayson et al. 2002). In addition, some soil
morphological properties have been widely recognised as
influencing factors or indicators of soil moisture variation.
For example, Broderson (2003) constructed a conceptual
model of water movement in soils with different structures
(granular, prismatic, blocky, and platy); Gish et al .(2005)
found that depth to clay horizon had a significant influence
on surface soil moisture; Walker and Lin (2008) demonstrated
that redox feature can be a good indicator of a temporary
saturation zone of soil; and Zhu and Lin (2009 ) used
manganese mottle content to identify subsurface soil water
movement. These studies suggested that pedological
properties can be promising information for predicting soil
moisture variation. Some studies have attempted to adopt
pedology in soil moisture modelling, although mixed results
have been found and reported. For example, in the study by
Takagi and Lin (2012), soil moisture prediction was significantly
improved when both pedological and terrain parameters were
considered in regression kriging in the Shale Hills catchment,
USA (with root mean squared errors mostly <0.07). Zhao et al.
(2007) used multiple linear regression to predict soil moisture
with only soil properties (soil organic carbon, bulk density, and
hydraulic conductivity), and the R
2
value reached as high as
0.82.
Difficulties
Several drawbacks constrain the use of soil or pedological
properties to predict soil moisture spatial and temporal
variations. First, some pedological properties can only be
qualitatively described, and thus their accuracy and reliability
are doubtable in soil moisture prediction. Traditional soil
morphological description is a qualitative process following
standard procedure and chart. The ‘Soil Survey Manual ’ and
‘Field Book for Describing and Sampling Soils (Version 2.0)’
developed by the United States Department of Agriculture (Soil
Survey Division Staff 1993; Schoeneberger et al. 2002) are two
field soil description books widely used by soil scientists. In
these books, some morphological features including quantities
and sizes of redox feature, mottles, rock fragments, pores, and
roots are described following specific criteria and charts,
whereas other morphological features including texture,
structure, horizon, and consistence are described following
standard procedures and guidance. However, even following
the same chart and procedure, the description results are still
much related to the training, experience, judgment and
preference of the describers. Therefore, soil morphological
description is always considered qualitative, not quantitative.
Second, accurate and reliable spatial pedological information
relies heavily on soil sampling, analysis, and description, which
are time-consuming and labour-intensive. Kravchenko (2003)
50%
(a)
(b)
Wet condition
Dry condition
45%
40%
35%
30%
25%
20%
15%
26%
24%
22%
20%
18%
16%
14%
0 100 m
Fig. 5. Soil moisture patterns at the top 0.3 m in a 10-ha catchment in S.E. Australia, collected
during (a) wet conditions in winter and (b) dry conditions in summer (from Grayson et al. 1997,
2002). Under wet conditions, the soil moisture spatial pattern was correlated with topography, while
under dry conditions, it was correlated with soil properties.
630 Soil Research Q. Zhu et al.
and Zhu and Lin (2010) found that spatial variation of a soil
property can be reliably simulated and interpolated when
sampling space is less than half of its spatial correlation
range, which requires a sufficient number of samples in a
certain area. Webster and Oliver (1992) reported that the
spatial structure of a soil property can be well captured
through regular variogram when sample size is >100, while
Kerry and Oliver (2007) showed that at least 50 sample points at
an appropriate distance appear adequate when using residual
maximum likelihood variogram. All of these studies suggest that
to acquire the spatial variation of a soil property, intensive
sampling, analysis, and description are unavoidable. Such
work is even harder to accomplish at larger spatial scales.
Therefore, the spatial coverage of soil sampling, analysis, and
description is either sparse at large spatial scale or intensive but
at small spatial scale. In other words, intensive pedological data
are not readily available. High-intensity pedological data are not
even as readily available as in-situ soil moisture observations.
Therefore, how to conveniently acquire spatial variation of
pedological features and effectively utilise the available soil
database are critical in predicting soil moisture with pedological
information.
Third, the trade-off between soil sampling density and spatial
scale constrains the use of pedological data in predicting spatial
and temporal variation of soil moisture. High-intensity soil
sampling is generally only available at small spatial scales (e.
g. plot, small catchment, and farm scales). However, at large
spatial scales (e.g. watershed, regional, and continental scales),
pedological information such as a soil map is not widely
available, or its resolution, accuracy, and reliability cannot
meet the requirements of natural resources management and
landscape process modelling, including prediction of soil
moisture spatial and temporal variations (Sawyer 1994;
Rossel and McBratney 1998; Lin et al. 2005b; Zhu et al.
2010a). For example, second-order soil maps (called
SSURGO) in the United States have cartographical scales of
1 : 12 000 to 1 : 63 360, with minimum delineation sizes of
0.6–16.2 ha (Soil Survey Division Staff 1993). While highly
valuable for general land use planning and many other
applications, these soil maps have encountered challenges for
more precise applications in landscape studies, precision
agriculture, catchment hydrology, and ecosystem services (e.g.
Robert 1993;Franzenet al. 2002).
Alternative approaches
To solve the problem of the qualitative nature of some
information, new tools have been developed or introduced
into soil morphological description. Computed tomography
(CT) using X-ray or gamma-ray is one of the techniques to
quantify soil morphological features (Fig. 6a). This technique
provides very accurate and high resolution investigation of soil
columns non-destructively. Usually, resolution of medical CT is
1 mm, while that of industrial CT can be as great as 5 mm.
Numerous studies have used CT to quantify soil structure, pores,
and roots at column scale (e.g. Phogat and Aylmore 1989; Duliu
1999; Luo et al. 2010). In these studies, 3D images of soil
columns were analysed, modelled, and reconstructed to reveal
the distributions of soil pores and roots. This information from
CT scanning is then used for the modelling of soil water and
solute transport (e.g. Luo et al. 2010). In addition, other tools, for
example desktop 3D scanner, have also been developed to scan
soil samples three-dimensionally and analyse sizes and
quantities of mottles. However, CT and these scanning tools
can only be used for small spatial scale (e.g. column and soil
aggregate scales) investigation; up-scaling of CT scanning
results is critical but has not yet been well documented and
understood.
At intermediate scales (e.g. small catchment and farm scale),
geophysical tools such as ground-penetrating radar (GPR) can
be used to quantitatively detect the spatial information of soil
morphological features (e.g. horizon and depth to bedrock)
(Fig. 6
b).
Geophysical surveys are generally done on foot or
on
a mobile platform (e.g. towed behind an all-terrain vehicle).
Therefore, they have been used to survey an area from several
square meters to square kilometers, bridging a scale gap between
small-scale non-invasive sensors and large-scale remote sensing.
For example, Gish et al.(2005) used GPR to map the depth to
clay horizon at farm scale in Maryland, USA, and Lin et al.
(2006a) used GPR to map depth to bedrock and soil types in the
Shale Hills catchment in Pennsylvania, USA. Pedological
properties derived from GPR in those two studies were used
to derive and model soil moisture variation and preferential
flow paths. While CT and GPR have been used to quantify some
soil morphological features (e.g. soil structure, mottles, pores,
roots, and horizons) at small and intermediate spatial scales,
more advanced techniques are still needed to quantify other
soil morphological features, and efforts should also be invested
into translating the quantified soil morphological features
from small scales to large scales. In addition, since most
geophysical tools are indirect approaches to measure or
derive soil properties, their accuracies are highly site-specific.
For example, since the amount and type of salts and the clay
content affect the suitability of GPR survey, Doolittle et al.
(2007) developed a soil suitability map of the US for GPR
survey.
The first approach to solve the problem of time-consuming
and costly soil sampling, analysis, and description is the use of
non-invasive geophysical tools including electromagnetic
induction (EMI) and GPR (Fig. 6b, c). The principles of EMI
and GPR can be found in the literature (e.g. Davis and Annan
1989; Sudduth et al. 2001; Corwin and Lesch 2003). These tools
provide copious, direct or indirect data on soil physiochemical
properties as direct input parameters in soil mapping. The EMI
measures soil apparent electrical conductivity (ECa), which can
be correlated and used to map different properties including
texture (James et al. 2003; Saey et al. 2009), soil type
(Anderson-Cook et al. 2002), and soil moisture (Sherlock and
McDonnell 2003; Zhu et al. 2010b). The GPR can image
subsurface conditions, which can be used to identify
pedological properties including critical soil horizons (Gish
et al. 2005), rock surface (Lin et al. 2006a), and soil
moisture (Minet et al. 2012). Although in some cases soil
moisture can be predicted using EMI and GPR, in most
others it cannot be reliably and accurately modelled using
EMI and GPR. For example, in the study by Zhu et al.
(2010b), EMI could only be used to predict soil moisture
under wet condition or in wet areas. In the review by
Spatial-temporal variations of soil moisture from hydropedological perspective Soil Research 631
Huisman et al.(2003), uncertainties and limitations were also
pointed out for measuring soil moisture with GPR approaches.
Both EMI and GPR have some limitations that require more
investigation for proper survey design and data mining. The EMI
measures soil ECa, which can be affected by different
hydropedological properties. In areas with different geology,
soil genesis, management, land use, climate, and topography, or
in the same area but in different seasons and wetness conditions,
or using different EMI meters or dipole orientations, the
measured ECa may reflect different properties (Mueller et al.
2003; Zhu et al. 2010b). Therefore, repeated EMI surveys and
proper design of survey timing, equipment, and setting is needed
to acquire useful pedological information. Relationships
between soil ECa and soil properties are always constructed
using various models (e.g. regression). In previous studies, R
2
values of these models are usually 0.3–0.8 (e.g. Mueller et al.
2003; Saey et al. 2009 ), indicating that only part of the spatial
variation of soil properties can be explained by EMI survey.
Therefore, advanced modelling and data mining are needed to
improve the accuracy. Similarly, the use of GPR is restricted by
the geology and soil background of the study area. For example,
Doolittle et al.(2007) pointed out that a thin, conductive soil
horizon with high clay content and salinity causes high rates of
signal attenuation, severely restricting penetration depths and
limiting the suitability of GPR. Those authors even generated a
GPR soil suitability map of the conterminous United States
from the soil survey database. In addition, subsurface conditions
derived from GPR image are subjective and usually based on
the researcher’s experience. Therefore, ground truth is always
needed for validation.
The second approach is optimal sampling design and
interpolation to capture the spatial variation and reduce
sample size. Various studies have investigated the optimal
sample sizes and sampling methods for different pedological
m
m
5.00
–2.50
0.0
2.50
5.00
7.50
10.0 15.0 20.0 25.0
(a)
(b)
(c)
Fig. 6. Example of using (a) computed tomography to image soil macropores at column scale (Luo et al. 2008), (b) ground
penetration radar to map the soil-bedrock interface of different soil types (Weikert, Berks, and Rushtown) at swale scale (Lin et al.
2006b), and (c) a combination of different electromagnetic inductions to refine the second-order soil map at farm scale (Zhu et al.
2010a).
632 Soil Research Q. Zhu et al.
properties and different application purposes. One group of
optimisation methods is known as the space-filling sample
(Royle and Nychka 1998), which provides an optimal
dispersion of sampling points in geographical space and
optimises the distance between individual sampling points.
Vašát et al.(2010) pointed out that an efficient
implementation of this procedure is to partition the area of
interest into approximately equal units and apply the k-means
clustering algorithm. The centroids of the clusters can then be
used as sampling points (Brus et al. 2006). Another group of
methods takes a geostatistical model of reality as the starting
point and optimises the sampling scheme based on minimisation
of the average kriging variance or maximum kriging variance
(Vašát et al. 2010). This yields an optimisation in geographic
space under the ordinary kriging models (McBratney and
Webster 1981), and it yields an optimisation in both attribute
and geographic space under universal or regression kriging
(Brus and Heuvelink 2007). Reliable and accurate spatial
interpolation is also an alternative approach for investigating
the spatial variation of different pedological properties with
limited sample size. With different spatial structures,
sampling sizes, and available auxiliary variables, optimal
interpolation methods can be varied for different pedological
properties. For example, Kravchenko (2003) found that soil
properties with nugget over sill ratio (N/S) <0.1 can be mapped
more accurately by ordinary kriging than those with N/S >0.1.
Zhu and Lin (2010) suggested that N/S ratio, sampling distance
over spatial correlation range, sample size, and relationship
between target soil property and available auxiliary variables
should be considered when selecting an optimal interpolation
method.
Different approaches should be combined to acquire spatial
pedological information at various spatial scales. At large scale
(e.g. watershed, regional, and continental scales), the spatial
variability can be acquired from existing soil surveys and maps.
For example, in the United States, two orders of soil maps, State
Soil Geographic (STATSGO) and Soil Survey Geographic
(SSURGO) databases, have been developed and they provide
spatial distribution of pedological properties for different
purposes (Soil Survey Division Staff 1993). Similar soil
surveys and soil map databases are also available in other
countries. Therefore, data mining for the current soil
database, as well as remote sensing and digital elevation
model, is an important way to acquire spatial information of
pedology at large scales. At small scales (e.g. column, pedon,
and plot scales), advanced tools have been developed and
introduced to investigate pedological properties. For example,
using CT to quantify soil morphological features (Phogat and
Aylmore 1989; Duliu 1999; Luo et al. 2010), and using GPR to
image subsurface soil variations at pedon and plot scales (Lin
et al. 2006a). At intermediate scale (e.g. catchment and farm
scales), traditional soil sampling is not efficient, while remote
sensing and soil-landscape modelling are too coarse to depict the
spatial distribution of soil. Therefore, various studies have
demonstrated that geophysical tools including EMI and GPR
are preferable at intermediate scales (Gish et al. 2005; Zhu et al.
2010a).
Approaches proposed here are used to acquire pedological
properties at different spatial scales. Appropriate models are still
needed to transfer the pedological property information into soil
moisture characteristics. Pedotransfer functions (PTFs) have
been applied to interpret soil-water release and retention
curves from texture, organic matter, bulk density, etc.
(Pachepsky et al.
2006).
More advanced models and
approaches
are still needed to translate pedological data into
hydrological properties including soil moisture variation.
Integrating pedology with other information to predict
soil moisture variation
Pedology itself is not adequate in predicting soil moisture
variation. First, pedological information acquired at different
spatial scales has uncertainties and errors to some extent,
especially soil survey and soil map at large scales. Second,
pedological properties are static parameters, while soil moisture
is temporally varied. Therefore, pedological information should
be combined with other factors and incorporated in various
models to predict the spatial–temporal variation of soil moisture.
With the assistance of pedological data, the modelling of soil
moisture and its related characteristics becomes feasible and
reliable. Various PTFs have been developed and applied to
predict spatial soil hydraulic properties, for example soil
water retention characteristics and hydraulic conductivity (e.g.
Hassler et al. 2011; Tóth et al. 2012). In PTFs, pedological
properties acquired from soil maps and soil surveys are used as
the basic information and different models (e.g. wavelet, neural
network, and fuzzy logic) are adopted to ‘transfer’ pedological
information to soil hydraulic properties. Spatial interpolation
and modelling are then used to investigate the spatial patterns of
soil hydraulic properties. These soil hydraulic properties
calculated from PTFs can be further used to predict the
spatial and temporal variations of soil moisture. For example,
Loosvelt et al.(2011) used continuous PTFs to construct discrete
probability of soil hydraulic parameters with soil texture, and
then modelled and simulated soil moisture variations. Brimelow
et al.(2010) found that soil hydraulic properties generated from
PTFs can improve the soil moisture simulation in the prairie
agrometeorological model (PAMII). All of these previous
studies suggested that soil moisture simulation can be benefit
from integrating pedological information in the modelling.
By combining the information on pedology, landscape
features, and even remote sensing, soil moisture prediction
can be largely improved at intermediate and large scales. As
we discussed earlier, landscape features can explain only part of
the soil moisture variation, while pedological properties are
important controlling factors of soil moisture, especially
during the dry season (Western et al. 1999; Grayson et al.
2002; Zhu and Lin 2011). In our previously mentioned
unpublished study, the topography itself explained 10–40%
of the soil moisture variation (Fig. 4a), while the pedology
itself explained 0–30% of the soil moisture variation in a farm in
central Pennsylvania (Fig. 4b). However, using the combined
information of topography and pedology, ~20–50% of the soil
moisture variation could be explained by stepwise multiple
linear regressions (Fig. 4c). In addition, Zhu and Lin ( 2009 ,
2011) pointed out that when available auxiliary variables were
sufficient, prediction of soil moisture was signifi cantly improved
in an agricultural site in central Pennsylvania, USA. Other
Spatial-temporal variations of soil moisture from hydropedological perspective Soil Research 633
similar reports include those of Jawson and Niemann (2007),
who found that spatial patterns of soil moisture at a large scale
can be well explained by integrated information of pedology,
land use, and topographic properties; and Korres et al.(2010),
who reported that pedological properties, topography, land
management, and vegetation together controlled the soil
moisture variation in a grassland and an arable test site
within the Rur catchment in western Germany. All of these
studies agreed that prediction of soil moisture spatial and
temporal variations can be largely improved when combined
information of pedology and landscape was considered.
Conclusions
Prediction of soil moisture spatial and temporal variations can
benefit from pedological information. Traditional approaches of
soil moisture monitoring and prediction, which include in-situ
observation, remote sensing, and using landscape features (e.g.
topography, land use, and vegetation) were compared and
investigated in this paper. Based on this, potentials and merits
of using pedological properties to predict soil moisture spatial
and temporal variations, as well as its limits and possible
solutions, were analysed and discussed. Through proper use
of non-invasive geophysical tools, soil survey information,
and computed tomography tools, pedological data can be
conveniently and quantitatively acquired. Such data, when
integrated with landscape features and spatial modelling, can
provide a promising approach for predicting spatial and temporal
variations of soil moisture at different spatial scales. This
paper proposed an alternative way of simulating soil moisture
variations, which will be the important for agricultural
management, and hydrological and ecological modelling.
Acknowledgments
This work was supported by the National Natural Science Foundation of
China (41030745 and 41271109) and Key ‘135’ Project of Nanjing Institute
of Geography and Limnology, Chinese Academy of Sciences
(NIGLAS2012135005).
References
Anderson-Cook CM, Alley MM, Roygard JKF, Khosla R, Noble RB,
Doolittle JA (2002) Differentiating soil types using electromagnetic
conductivity and crop yield maps. Soil Science Society of America
Journal 66, 1562–1570. doi:10.2136/sssaj2002.1562
Bisselink B, van Meijgaard E, Dolman AJ, de Jeu RAM (2011) Initializing a
regional climate model with satellite-derived soil moisture. Journal of
Geophysical Research, D, Atmospheres 116, D02121. doi:10.1029/
2010JD014534
Biswas A, Si BC (2011) Revealing the controls of soil water storage at
different scales in a hummocky landscape. Soil Science Society of
America Journal 75 , 1295–1306. doi:10.2136/sssaj2010.0131
Blake WH, Walsh RPD, Barnsley MJ, Palmer G, Dyrynda P, James JG
(2003) Heavy metal concentrations during storm events in a rehabilitated
industrialized catchment. Hydrological Processes 17, 1923–1939.
doi:10.1002/hyp.1218
Brimelow JC, Hanesiak JM, Raddatz R (2010) Validation of soil
moisture simulations from the PAMII model, and an assessment of
their sensitivity to uncertainties in soil hydraulic parameters. Agricultural
and Forest Meteorology 150, 100–114. doi:10.1016/j.agrformet.2009.
09.006
Broderson WD (2003) ‘From the surface down: An introduction to soil
surveys for agronomic use.’ (USDA-NRCS: Washington, DC)
Brus DJ, Heuvelink GMB (2007) Optimization of sample patterns for
universal kriging of environmental variables. Geoderma 138,86–95.
doi:10.1016/j.geoderma.2006.10.016
Brus DJ, de Gruijter JJ, van Groenigen JW (2006) Designing spatial
coverage samples using the k-means clustering algorithm. In ‘Digital
soil mapping: An introductory perspective’. Ch. 14. (Ed. P Lagacherie)
pp. 183–192. (Elsevier: Amsterdam)
Buttle JM, Dillon PJ, Eerkes GR (2004) Hydrologic coupling of slopes,
riparian zones and streams: an example from the Canadian Shield.
Journal of Hydrology 287, 161–177. doi:10.1016/j.jhydrol.2003.09.022
Cardenas-Lailhacar B, Dukes MD (2010) Precision of soil moisture sensor
irrigation controller under field conditions. Agricultural Water
Management 97, 666–672. doi:10.1016/j.agwat.2009.12.009
Cashion J, Lakshmi V, Bosch D, Jackson TJ (2005) TRMM microwave
imager (TMI) satellite for Little River Watershed Tifton, Georgia.
Journal of Hydrology 307, 242–253. doi:10.1016/j.jhydrol.2004.10.019
Castellano MJ, Schmidt JP, Kaye JP, Walker C, Graham CB, Lin H, Dell C
(2011) Hydrological controls on heterotrophic soil respiration across
an agricultural landscape. Geoderma 162, 273–280.
doi:10.1016/j.geo
derma.201
1.01.020
Champagne C, McNairn H, Berg AA (2011) Monitoring agricultural soil
moisture extremes in Canada using passive microwave remote sensing.
Remote Sensing of Environment 115, 2434–2444. doi:10.1016/j.rse.20
11.04.030
Choi M, Kustas WP, Ray RL (2012) Evapotranspiration models of different
complexity for multiple land cover types. Hydrological Processes 26,
2962–2972. doi:10.1002/hyp.8346
Corwin DL, Lesch SM (2003) Application of soil electrical conductivity to
precision agriculture: Theory, principle, and guidelines. Agronomy
Journal 95, 455 –471. doi:10.2134/agronj2003.0455
Das NN, Mohanty BP (2006) Root zone soil moisture assessment using
remote sensing and vadose zone modeling. Vadose Zone Journal 5,
296–307. doi:10.2136/vzj2005.0033
Das NN, Mohanty BP, Cosh MH, Jackson TJ (2008) Modeling and
assimilation of root zone soil moisture using remote sensing
observations in Walnut Gulch watershed during SMEX04. Remote
Sensing of Environment 112, 415–429. doi:10.1016/j.rse.2006.10.027
Davidson EA, Keller M, Erickson HE, Verchot LV, Veldkamp AE (2000)
Testing a conceptual model of soil emissions of nitrous and nitric oxides.
Bioscience 50, 667–680. doi:10.1641/0006-3568(2000)050[0667:TAC
MOS]2.0.CO;2
Davis JL, Annan AP (1989) Ground-penetrating radar for high-resolution
mapping of soil and rock stratigraphy. Geophysical Prospecting 37,
531–551. doi:10.1111/j.1365-2478.1989.tb02221.x
De Lannoy GJM, Verhoest NEC, Houser PR, Gish TJ, van Meirvenne M
(2006) Spatial and temporal characteristics of soil moisture in an
intensively monitored agricultural field (OPE
3
). Journal of Hydrology
331, 719–730. doi:10.1016/j.jhydrol.2006.06.016
Defersha MB, Melesse AM (2012) Effect of rainfall intensity, slope and
antecedent moisture content on sediment concentration and sediment
enrichment ratio. Catena 90,47–52. doi:10.1016/j.catena.2011.11.002
Doolittle JA, Minzenmayer FE, Waltman SW, Benham EC, Tuttle JW,
Peaslee SD (2007) Ground-penetrating radar soil suitability map of the
conterminous United States. Geoderma 141, 416–421. doi:10.1016/j.
geoderma.2007.05.015
Duliu OG (1999) Computer axial tomography in geosciences: an overview.
Earth-Science Reviews 48, 265–281. doi:10.1016/S0012-8252(99)
00056-2
Famiglietti JS, Rudnicki JW, Rodell M (1998) Variability in surface
moisture content along a hillslope transect: Rattlesnake Hill, Texas.
Journal of Hydrology 210, 259–281. doi:10.1016/S0022-1694(98)00
187-5
634 Soil Research Q. Zhu et al.
Famiglietti JS, Devereaux JA, Laymon CA, Tsegaye T, Houser PR, Jackson
TJ, Graham ST, Rodell M, van Oevelen PJ (1999) Ground-based
investigation of soil moisture variability within remote sensing
footprints during the Southern Great Plains 1997 (SGP97) Hydrology
Experiment. Water Resources Research 35, 1839–1851. doi:10.1029/
1999WR900047
Franzen DW, Hopkins DH, Sweeney MD, Ulmer MK, Harvorson AD (2002)
Evaluation of soil survey scale for zone development of site-specific
nitrogen management. Agronomy Journal 94, 381–389. doi:10.2134/
agronj2002.0381
Fu BJ, Wang J, Chen LD, Qiu Y (2003) The effects of land use on soil
moisture variation in the Danangou catchment of the Loess Plateau,
China. Catena 54, 197–213. doi:10.1016/S0341-8162(03)00065-1
Gish TJ, Walthall CL, Daughtry CST, Kung KJS (2005) Using soil moisture
and spatial yield patterns to identify subsurface flow pathways. Journal
of Environmental Quality 34, 274–286. doi:10.2134/jeq2005.0274
Gish TJ, Prueger JH, Daughtry CST, Kustas WP, McKee LG, Russ AL,
Hatfield JL (2011) Comparison of field-scale herbicide runoff and
volatilization losses: an eight-year field investigation. Journal of
Environmental Quality 40, 1432–1442. doi:10.2134/jeq2010.0092
Graser EA, Vanbavel CHM (1982) The effect of soil-moisture upon soil
albedo. Agricultural Meteorology 27,17–26. doi:10.1016/0002-1571
(82)90015-2
Grayson RB, Western AW, Chiew FHS, Blöschl G (1997) Preferred states in
spatial soil moisture patterns: local and non-local controls. Water
Resources Research 33, 2897–2908. doi:10.1029/97WR02174
Grayson RB, Blöschl G, Western AW, McMahon TA (2002) Advances in
the use of observed spatial patterns of catchment hydrological response.
Advances in Water Resources 25, 1313–1334. doi:10.1016/S0309-1708
(02)00060-X
Guan XD, Huang JP, Guo N, Bi JR, Wang GY (2009) Variability of soil
moisture and its relationship with surface albedo and soil thermal
parameters over the loess plateau. Advances in Atmospheric Sciences
26, 692–700. doi: 10.1007/s00376-009-8198-0
Hain CR, Mecikalski SR, Anderson MC (2009) Retrieval of an available
water-based soil moisture proxy from thermal infrared remote sensing.
Part I: methodology and validation. Journal of Hydrometeorology 10,
665–683. doi:10.1175/2008JHM1024.1
Hassler
SK,
Lark RM, Milne AE, Elsenbeer H (2011) Exploring the variation
in soil saturated hydraulic conductivity under a tropical rainforest using
wavelet transform. European Journal of Soil Science 62, 891–901.
doi:10.1111/j.1365-2389.2011.01400.x
Heathman GC, Larose M, Cosh MH, Bindlish R (2009) Surface and profile
soil moisture spatio-temporal analysis during an excessive rainfall period
in the Southern Great Plains, USA. Catena 78, 159–169. doi:10.1016/
j.catena.2009.04.002
Hedley CB, Yule IJ (2009) A method for spatial prediction of daily soil water
status for precise irrigation scheduling. Agricultural Water Management
96, 1737–1745. doi:10.1016/j.agwat.2009.07.009
Huisman JA, Hubbard SS, Redman JD, Annan AP (2003) Measuring soil
water content with ground penetrating radar: a review. Vadose Zone
Journal 2, 476–491.
Hupet F, Vanclooster M (2002) Intraseasonal dynamics of soil moisture
variability within a small agricultural maize cropped field. Journal of
Hydrology 261,86–101. doi:10.1016/S0022-1694(02)00016-1
James IT, Waine TW, Bradley RI, Taylor JC, Godwin RJ (2003)
Determination of soil type boundaries using electromagnetic induction
scanning techniques. Biosystems Engineering 86, 421–430. doi:10.1016/
j.biosystemseng.2003.09.001
Jawson SD, Niemann JD (2007) Spatial patterns from EOF analysis of soil
moisture at a large scale and their dependence on soil, landuse, and
topographic properties. Advances in Water Resources 30, 366–381.
doi:10.1016/j.advwatres.2006.05.006
Jin YQ, Yan FH (2007) A change detection algorithm for terrain surface
moisture mapping based on multi-year passive microwave remote
sensing (Examples of SSM/I and TMI Channels). Hydrological
Processes 21, 1918–1924. doi:10.1002/hyp.6401
Kerry R, Oliver MA (2007) Comparing sampling needs for variograms of
soil properties computed by the method of moments and residual
maximum likelihood. Geoderma 140, 383–396. doi:10.1016/j.geo
derma.2007.04.019
Korres WC, Koyama N, Fiener P, Schneider K (2010) Analysis of surface
soil moisture patterns in agricultural landscapes using Empirical
Orthogonal Functions. Hydrology and Earth System Sciences 14,
751–764. doi:10.5194/hess-14-751-2010
Koyama CN, Korres W, Fiener P, Schneider K (2010) Variability of surface
soil moisture observed from multitemporal C-band Synthetic Aperture
Radar and field data. Vadose Zone Journal 9,
1014–1024.
doi:10.2136/
vzj2009.0165
Kravchenko AN (2003) Influence of spatial structure on accuracy of
interpolation methods. Soil Science Society of America Journal 67,
1564–1571. doi:10.2136/sssaj2003.1564
Kutilek M, Nielsen DR (1994) ‘Soil hydrology.’ (Catena Verlag:
Cremlingen-Destedt, Germany)
Li XY, Contreras S, Sole-Benet A, Canton Y, Domingo F, Lazaro R, Lin H,
Van Wesemael B, Puigdefabregas J (2011) Controls of infiltration-runoff
processes in Mediterranean karst rangelands in SE Spain. Catena 86,
98–109. doi:10.1016/j.catena.2011.03.003
Lin HS, Zhou XB (2008) Evidence of subsurface preferential flow using
soil hydrologic monitoring in the Shale Hills catchment. European
Journal of Soil Science 59,34–49. doi:10.1111/j.1365-2389.2007.
00988.x
Lin HS, Bouma J, Wilding L, Richardson J, Kutílek M, Nielsen D (2005a)
Advances in hydropedology. Advances in Agronomy 85,1–89.
doi:10.1016/S0065-2113(04)85001-6
Lin HS, Wheeler D, Bell J, Wilding L (2005b) Assessment of soil spatial
variability at multiple scales. Ecological Modelling 182, 271–290.
doi:10.1016/j.ecolmodel.2004.04.006
Lin HS, Kogelmann W, Walker C, Bruns MA (2006a) Soil moisture patterns
in a forested catchment: a hydropedological perspective. Geoderma 131,
345–368. doi:10.1016/j.geoderma.2005.03.013
Lin HS, Bouma J, Pachepsky Y, Western AW, Thompson JA, van
Genuchten MT, Vogel H, Lilly A (2006b) Hydropedology:
synergistic integration of pedology and hydrology. Water Resources
Research 42, W05301. doi:10.1029/2005WR004085
Loague K (1992) Soil water content at R-5, 1, spatial and temporal
variability. Journal of Hydrology 139, 233–251. doi:10.1016/0022-
1694(92)90204-9
Loosvelt L, Pauwels VRN, Cornelis WM, De Lannoy GJM, Verhoest NEC
(2011) Impact of soil hydraulic parameter uncertainty on soil moisture
modeling. Water
Resources
Research 47, W03505. doi:10.1029/
2010WR009204
Luo LF, Lin HS, Halleck P (2008) Quantifying soil structure and preferential
flow in intact soil using X-ray computed tomography. Soil Science
Society of America Journal 72, 1058–1069. doi:10.2136/sssaj2007.0179
Luo LF, Lin HS, Li SC (2010) Quantification of 3-D soil macropore
networks in different soil types and land uses. Journal of Hydrology
393,53–64. doi:10.1016/j.jhydrol.2010.03.031
Maeda K, Tanaka T, Park H, Hattori S (2006) Spatial distribution of soil
structure in a suburban forest catchment and its effect on spatio-temporal
soil moisture and runoff fluctuations. Journal of Hydrology 321,
232–256. doi:10.1016/j.jhydrol.2005.08.003
McBratney AB, Webster R (1981) The design of optimal sampling schemes
for local estimation and mapping of regionalized variables. II. Program
and examples. Computers & Geosciences 7, 335–365. doi:10.1016/
0098-3004(81)90078-9
Spatial-temporal variations of soil moisture from hydropedological perspective Soil Research 635
McColl KA, Ryu D, Matic V, Walker JP, Costelloe J, Rudiger C (2012) Soil
salinity impacts on l-band remote sensing of soil moisture. Geoscience
and Remote Sensing Letters 9, 262–266. doi:10.1109/LGRS.2011.
2165932
Minet J, Bogaert P, Vanclooster M, Lambot S (2012) Validation of ground
penetrating radar full-waveform inversion for field scale soil moisture
mapping. Journal of Hydrology 424–425, 112–123. doi:10.1016/j.jhy
drol.2011.12.034
Mintz Y, Walker GK (1993) Global fields of soil-moisture and land-surface
evapotranspiration derived from observed precipitation and surface
air-temperature. Journal of Applied Meteorology 32, 1305–1334.
doi:10.1175/1520-0450(1993)032<1305:GFOSMA>2.0.CO;2
Mladenova I, Lakshmi V, Jackson T, Walker JP, Merlin O, de Jeu RAM
(2011) Validation of AMSR-E soil moisture using L-band airborne
radiometer data from National Airborne Field Experiment 2006.
Remote Sensing of Environment 115, 2096–2103. doi:10.1016/j.rse.
2011.04.011
Mohanty BP, Famiglietti JS, Skaggs TH (2000) Evolution of soil moisture
spatial structure in a mixed vegetation pixel during the Southern Great
Plains 1997 (SGP97) Hydrology Experiment. Water Resources Research
36, 3675–3687. doi:10.1029/2000WR900258
Mueller TG, Hartsock NJ, Stombaugh TS, Shearer SA, Cornelius PL,
Barnhisel RI (2003) Soil electrical conductivity map variability in
limestone soils overlain by loess. Agronomy Journal 95, 496–507.
doi:10.2134/agronj2003.0496
Owe M, Jones EB, Schmugge TJ (1982) Soil moisture variation patterns
observed in Hand Country, South Dakota. Water Resources Bulletin 18,
949–954. doi:10.1111/j.1752-1688.1982.tb00100.x
Pachepsky YA, Rawls WJ, Lin HS (2006) Hydropedology and pedotransfer
functions. Geoderma 131, 308–316. doi:10.1016/j.geoderma.2005.
03.012
Phogat VK, Aylmore LAG (1989) Evaluation of soil structure by using
computer-assisted tomography. Australian Journal of Soil Research 27,
313–323. doi:10.1071/SR9890313
Pollacco JAP, Mohanty BP (2012) Uncertainties of water flux in soil-
vegetation-atmosphere transfer models: inverting surface soil moisture
and evapotranspiration retrieved from remote sensing. Vadose Zone
Journal 11,
doi:10.2136/v
zj2011.0167
Qiu Y, Fu B, Wang J, Chen L, Meng Q, Zhang Y (2010) Spatial prediction of
soil moisture content using multiple-linear regressions in a gully
catchment of the Loess Plateau, China. Journal of Arid Environments
74, 208–220. doi: 10.1016/j.jaridenv.2009.08.003
Robert P (1993) Characterization of soil conditions at the field level for soil
specific management. Geoderma 60,57–72. doi:10.1016/0016-7061(93)
90018-G
Rodriguez-Iturbe I, D’Odorico P, Porporato A, Ridolfi L (1999) On the
spatial and temporal links between vegetation, climate, and soil moisture.
Water Resources Research 35, 3709–3722. doi:10.1029/1999WR
900255
Rossel RAV, McBratney AB (1998) Soil chemical analytical accuracy and
costs: Implications from precision agriculture. Australian Journal of
Experimental Agriculture 38, 765–775. doi:10.1071/EA97158
Roxy MS, Sumithranand VB, Renuka G (2010) Variability of soil moisture
and its relationship with surface albedo and soil thermal diffusivity at
Astronomical Observatory, Thiruvananthapuram, south Kerala. Journal
of Earth System 119, 507–517. doi:10.1007/s12040-010-0038-1
Royle JA, Nychka D (1998) An algorithm for the construction of spatial
coverage designs with implementation in SPLUS. Computers &
Geosciences 24, 479–488. doi:10.1016/S0098-3004(98)00020-X
Saey T, Simpson D, Vermeersch H, Cockx L, Van Meirvenne M (2009)
Comparing the EM38DD and DUALEM-21S sensors for depth-to-clay
mapping. Soil Science Society of America Journal 73,7–12. doi:10.2136/
sssaj2008.0079
Sawyer JE (1994) Concepts of variable rate technology with considerations
for fertilizer application. Journal of Production Agriculture 7, 195–201.
Schjønning P, Thomsen IK, Moldrup P, Christensen BT (2003) Linking soil
microbial activity to water- and air-phase contents and diffusivities.
Soil Science Society of America Journal 67, 156–165. doi:10.2136/
sssaj2003.0156
Schmidt JP, Hong N, Dellinger A, Beegle DB, Lin HS (2007) Hillslope
variability in corn response to nitrogen linked to in-season soil
moisture redistribution. Agronomy Journal 99, 229–237. doi:10.2134/
agronj2006.0187
Schmidt JP, Sripada RP, Beegle DB, Rotz CA, Hong N (2011) Within-field
variability
in
optimum nitrogen rate for corn linked to soil moisture
availability. Soil Science Society of America Journal 75, 306–316.
doi:10.2136/sssaj2010.0184
Schmugge TJ, Kustas WP, Ritchie JC, Jackson TJ, Rango A (2002) Remote
sensing in hydrology. Advances in Water Recourses 25, 1367–1385.
doi:10.1016/S0309-1708(02)00065-9
Schoeneberger PJ, Wysocki DA, Benham EC, Broderson WD (2002) ‘Field
book for describing and sampling soils. Version 2.0.’ (Natural Resources
Conservation Service, National Soil Survey Center: Lincoln, NE)
Sherlock MD, McDonnell JJ (2003) A new tool for hillslope hydrologists:
Spatially distributed groundwater level and soil water content measured
using electromagnetic induction. Hydrological Processes 17,
1965–1977. doi:10.1002/hyp.1221
Soil Survey Division Staff (1993) ‘Soil survey manual.’ (US Government
Printing Office: Washington, DC)
Sudduth KA, Drummond ST, Kitchen NR (2001) Accuracy issues in
electromagnetic induction sensing of soil electrical conductivity for
precision agriculture. Computers and Electronics in Agriculture 31,
239–264. doi:10.1016/S0168-1699(00)00185-X
Takagi K, Lin HS (2012) Changing controls of soil moisture spatial
organization in the Shale Hills Catchment. Geoderma 173–174,
289–302. doi:10.1016/j.geoderma.2011.11.003
Tarboton DG (1997) A new method for the determination of flow directions
and contributing areas in grid digital elevation models. Water Resources
Research 33, 309–319. doi:10.1029/96WR03137
Tóth B, Mako A, Guadagnini A, Tóth G (2012) Water retention of salt-
affected soils: quantitative estimation using soil survey information. Arid
Land Research and Management 26, 103–121. doi:10.1080/15324982.
2012.657025
van Verseveld WJ, McDonnell JJ, Lajtha K (2009) The role of hillslope
hydrology in controlling nutrient loss. Journal of Hydrology 367,
177–187. doi:10.1016/j.jhydrol.2008.11.002
Vašát R, Heuvelink GBM, Boruvka L (2010) Sampling design optimization
for multivariate soil mapping. Geoderma 155, 147–153.
doi:10.1016/
j.geoderma
.2009.07.005
Walker C, Lin HS (2008) Soil property changes after four decades of
wastewater irrigation: a landscape perspective. Catena 73,63–74.
doi:10.1016/j.catena.2007.09.002
Webster R, Oliver MA (1992) Sample adequately to estimate variograms of
soil properties. Journal of Soil Science 43, 177–192. doi:10.1111/j.1365-
2389.1992.tb00128.x
Western AW, Grayson RB, Blöschl G, Willgoose GR, McMahon TA (1999)
Observed spatial organization of soil moisture and its relation to terrain
indices. Water Resources Research 35, 797–810. doi:10.1029/1998
WR900065
Western AW, Zhou SL, Grayson RB, McMahon TA, Blöschl G, Wilson DJ
(2004) Spatial correlation of soil moisture in small catchments and its
relationship to dominant spatial hydrological processes. Journal of
Hydrology 286, 113–134. doi:10.1016/j.jhydrol.2003.09.014
Wu XY, Zhang LP, Zhang HS, Wang XY, Fu XT, Zhang FF (2011) Factors
on particulate phosphorus loss in surface runoff under simulated
rainfalls. Asian Journal of Chemistry 23, 2922–2926.
636 Soil Research Q. Zhu et al.
Zhao Y, Peth S, Krümmelbein J, Horn R, Wang Z, Steffens M, Hoffmann C,
Peng X (2007) Spatial variability of soil properties affected by grazing
intensity in Inner Mongolia grassland. Ecological Modelling 205,
241–254. doi:10.1016/j.ecolmodel.2007.02.019
Zhu Q, Lin HS (2009) Simulation and validation of concentrated subsurface
lateral flow paths in an agricultural landscape. Hydrology and Earth
System Sciences 13, 1503–1518. doi:10.5194/hess-13-1503-2009
Zhu Q, Lin HS (2010) Comparing ordinary kriging and regression kriging for
soil properties in contrasting landscapes. Pedosphere 20, 594–606.
doi:10.1016/S1002-0160(10)60049-5
Zhu Q, Lin HS (2011) Influences of soil, terrain, and crop growth on soil
moisture variation from transect to farm scales. Geoderma 163,45–54.
doi:10.1016/j.geoderma.2011.03.015
Zhu Q, Lin HS, Doolittle JA (2010a) Repeated electromagnetic induction
surveys for improved soil mapping. Soil Science Society of America
Journal 74, 1763–1774. doi:10.2136/sssaj2010.0056
Zhu Q, Lin HS, Doolittle JA (2010b) Repeated electromagnetic induction
surveys for determining subsurface hydrologic dynamics in an
agricultural landscape. Soil Science Society of America Journal 74,
1763–1774. doi:10.2136/sssaj2010.0056
Spatial-temporal variations of soil moisture from hydropedological perspective Soil Research 637
www.publish.csiro.au/journals/sr
A preview of this full-text is provided by CSIRO Publishing.
Content available from Soil Research
This content is subject to copyright.
