Vous êtes sur la page 1sur 14

The Egyptian Journal of Remote Sensing and Space Sciences 21 (2018) 1–14

Contents lists available at ScienceDirect

The Egyptian Journal of Remote Sensing and Space Sciences

journal homepage: www.sciencedirect.com

Review Article

Application of near-infrared reflectance for quantitative assessment of


soil properties
E.S. Mohamed ⇑, A.M. Saleh, A.B. Belal, Abd_Allah Gad
National Authority for Remote Sensing and Space Sciences (NARSS), Cairo, Egypt

a r t i c l e i n f o a b s t r a c t

Article history: Beginning with a discussion of reflectance spectroscopy, this article attempts to provide a review on fun-
Received 24 July 2016 damental concepts of reflectance spectroscopic techniques. Their applications as well as exploring the
Revised 31 January 2017 role of Near-infrared reflectance spectroscopy that would be used for monitoring and mapping soil char-
Accepted 1 February 2017
acteristics. This technique began to be used in the second half of the 20th century for industrial purposes.
Available online 20 February 2017
Moreover, this article explores the potentiality of predicting soil properties based on spectroscopic mea-
surements .Quantitative prediction of soil properties such as; salinity, organic carbon, soil moisture and
Keywords:
heavy metals can be conducted using various calibration models – such models were developed depend-
Near infrared spectroscopy
Soil salinity
ing on the measured soil laboratory analyses data and soil reflectance spectra thereby resampled to satel-
Soil moisture lite images - to predict soil properties. The most common used models are stepwise multiple linear
Soil organic carbon regression (SMLR), partial least squares regression (PLSR), multivariate adaptive regression splines
Soil surface features and soil contamination (MARS), principal component regression (PCR) and artificial neural networks (ANN). Those methods
are required to quickly and accurately measure soil characteristics at field to improve soil management
and conservation at local and regional scales. Visable-Near Infra Red (VIS-NIR) has been recommended as
a quick tool for mapping soil properties. Furthermore, VIS-NIR reflection spectroscopy reduces the cost
and time, therefore has a wonderful ability and potential use as a rapid soil analysis for both precision
soil management and assessing soil quality.
Ó 2017 National Authority for Remote Sensing and Space Sciences. Production and hosting by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-
nd/4.0/).

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Spectroscopy definitions and history of spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3. Near-infrared reflectance spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
4. Predictive models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
5. Applications of NIRS in soil sciences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.1. Soil salinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.2. Soil moisture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
5.3. Soil organic carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
5.4. Clay minerals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
5.5. Soil surface features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.6. Soil contamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
6. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Peer review under responsibility of National Authority for Remote Sensing and Space Sciences.
⇑ Corresponding author.
E-mail addresses: Salama55_55@yahoo.com, Salama55@mail.ru (E.S. Mohamed).

http://dx.doi.org/10.1016/j.ejrs.2017.02.001
1110-9823/Ó 2017 National Authority for Remote Sensing and Space Sciences. Production and hosting by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2 E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14

1. Introduction response of interest as a function of wavelength or frequency


(Herrmann and Onkelinx, 1986; Clark, 1999).
Soil is described as a heterogeneous system, their mechanisms The history of spectroscopy began in the 17th century with
and processes are complex and difficult to be fully understood. Isaac Newton’s discovery of the with Isaac Newton’s discovery of
Numerous traditional methods are used in an endeavour to the light nature and color basics. He introduced the word ‘‘spec-
describe the relationship between different soil properties such trum” at first application to describe the rainbow of colors combi-
physical, chemical and principal soil components. Consequently, nation to form white light. During the early 1800s, Joseph von
simple and accurate soil testing procedures are required in field Fraunhofer made experimental advances with dispersive spec-
and laboratory. Near-infrared reflectance spectroscopy (NIRS) is a trometers that enabled spectroscopy to become a more precise
nondestructive systematic strategy for characterizing and identify- and quantitative scientific technique. Since then, spectroscopy
ing soil properties. Those techniques have been used since the has played and continues to play a significant role in chemistry,
1960 to estimate moisture, protein, and oil in agricultural products physics and astronomy (Brand, 1995). As far as the development
(Ben-Gera and Norris, 1968). During last three decades, numerous of instrumentation and its breakthrough for industrial applications
studies illustrated that the spectral reflectance property of soil in the second half of the 20th century were concerned, NIR pro-
samples in laboratory conditions, as well as field investigation of ceeded in technology jumps (Fig. 1). In this respect, credit has lar-
soils’ characteristics, can be assessed where remote photography gely to be given to researchers in the field of agricultural science.
materials occupy an increasingly significant place in the organiza- At the same time, with few exceptions, comparatively low priority
tion of soil cover monitoring (Mohamed, 2013; Mohamed et al., has been given to NIR spectroscopy in the chemical industry
2015; Saleh et al., 2015; Savin, 1993). Recently, this technique is (Siesler et al., 2002). This technique recently has been developed
widely used in several fields as an amazing tool for evaluating such into essential methods for scientific research and industrial quality
agriculture, food, polymer .pharmaceuticals and petrochemicals. control in a different applications such chemistry, environmental
Moreover, the technique, NIR method can be applied to predict soil analysis, agriculture and as well as life sciences.
properties as additional (to laboratory analysis) or initial assess-
ment of soil quality (Demattê and da Silva Terra, 2014; Mateusz
3. Near-infrared reflectance spectroscopy
Kania and Piotr gruba, 2016). Near-infrared reflectance spec-
troscopy (NIRS) has been used to predict several soil properties
The fundamental principle of VisNIR is based on the differences
such soil organic carbon, soil moisture content, soil contemenation,
in molecular characteristics, where spectral signatures of different
soil salinity, etc. Soil electrical conductivity can be detected using
materials are categorized based on their reflectance and absor-
visible, near infrared, or short-wave infrared spectral bands from
bance spectra. The change in signatures is referred to vibrational
optical sensors to be promising for the detection of surface soil
extending and bending of atoms that arrange molecules and crys-
salinity. The intensive of reflectance is related to concentration of
tals. Most soil components are usually observed in the mid-
soluble salts in salt-affected soils (Dwivedi and Rao, 1992; Khan
infrared region vibrations (2500–25,000 nm), with overtones and
et al., 2005; Nield et al., 2007; Abdi et al., 2016). Many authors sug-
combinations found in the near-infrared region (400–2500 nm)
gested that, infrared and red channels are applicable methods to
(Clark, 1999; Shepherd and Walsh, 2002). The electromagnetic
monitor soil characteristics such iron oxides and soil moisture
(EM) spectrum ranges from gamma (c) rays, at the shortest wave-
are considered (Samsonova and Meshalkina, 2011; Sonia et al.,
lengths, to radio-waves, at the longest wavelengths (Fig. 2). Most
2012 and Niederberger et al., 2015). Near infrared (NIR) and mid-
common sensing systems operate in one or several of the visible,
infrared (MIR) ranges are promising technologies considered as a
infrared (IR) and microwave portions of the spectrum. Sensor data
quantitative ones that gives good results for heavy metals concen-
covering those wavelengths are readily available from both
tration as there is a high correlation between pollutants and their
satellite and airborne platforms (NASA, 2014). The energy of infra-
spectral indicators. Reflectance spectroscopy techniques have been
red light corresponds to the energy required to cause molecular
used for retrieving and mapping the distribution of heavy metals
vibrations. Moreover, the far-IR region (A = 4  l04  106 nm)
such as Pb at high accuracy (Samsonova and Meshalkina, 2011).
harmonize to molecular variations and the mid-IR (A = 2500 
Many of regression models are used to estimate quantitative and
4  l04 nm) corresponds to fundamental molecular vibrations,
qualitative analyses of the various soil elements, based on investi-
such as stretching, bending, wagging, and scissoring. The energy
gating the correlation between each element properties and the
of near-IR light corresponds to overtones and combination bands
observance for each selected wavelength. However, the most wide-
of fundamental molecular vibrations from the mid-IR. (Drago,
spread regression models are partial least square regression (PLSR),
1992; Workman, 1996). Vibrational spectroscopy is depending on
multivariate adaptive regression splines (MARS), ordinal logistic
interactions between the molecules and electronic field compo-
regression, stepwise multiple linear regression (SIMR), artificial
nents of incident light in the mid- and near-IR region. Such inter-
neural networks (ANN), locally weighted regression (LWR) and
actions result in absorption of light by molecules when the
principal components regression (PCR) (Chang et al., 2001;
energy of incident light (Ep) is equal to the energy difference
Ciurczack, 2001; Nawar et al., 2014; Fikrat et al., 2016; Zheng
(AE) between the quantized energy levels of different vibrational
et al., 2016).
states of the molecule (Fig. 3). Their relationship can be expressed
as:

Ep ¼ hv ¼ hc=A: ¼ AE; ð1Þ


2. Spectroscopy definitions and history of spectroscopy
where:
Spectroscopy is the science that studies the interaction between v is the frequency of incident light,
matter and its electromagnetic radiation (Crouch and Skoog, 2007). c is velocity of light,
Reflectance spectroscopy is the study of light as a function of wave- A is the wavelength, and
length that has been reflected or scattered from a solid, liquid, or h is Plank’s constant
gas. This concept was expanded greatly to include any interaction
with radiative energy as a function of its wavelength or frequency. The energy difference, AE, is specified by chemical bonds of
Spectroscopic data is often represented by aspectrum, a plot of the functional groups in the molecules. A molecule must undergo a
E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14 3

Fig. 1. Development of near-infrared spectroscopy (Source: Siesler et al., 2002).

Fig. 2. Electromagnetic spectrum (Source; NASA 2014).

change in dipole moment in order to absorb IR light. Based on a A ¼ logðl=TÞ ¼ logðIo =IÞ ¼ klc; ð3Þ
harmonic oscillator, the permitted energy states of a molecule
where:
are given by:
Io is the intensity of the incident light,
Ev ¼ hvðv þ 1=2Þ ð2Þ I is the intensity of the transmitted light,
k is the molecular absorption coefficient, and
where:v is the vibrational quantum number (v = 0, l, 2,. . .) l is the path length of light through the sample.
The fundamental vibration means that the transition from
v = 0 to v = l, according to the selection rule for a harmonic oscil- The molecular absorption coefficient, k, is the characteristic of
lator. Furthermore, if the chemical bond is too weak or the each molecule and is dependent on the wavelength of the incident
atoms are too heavy, the fundamental vibration will occur at light.
very low frequency. As a result, the higher overtones, in the However, the reflectance of radiation from one type of surface
near-IR region, may not be detectable. Therefore, the near-IR is material, such as soil, varies over the range of wavelengths in the
dominated by the overtones and the combinations of fundamen- electromagnetic spectrum and known as the spectral signature of
tal vibrations for O–H, C–H, and N–H found in mid-IR (Wetzel, the material (Fig. 4).
1983).
The amount of light absorbed is a function of the absorber con- 4. Predictive models
centration. Based on the Beer-Lambert law, the relationship
between absorbance (A), transmittance (T), and concentration (c) Prediction of different soil characteristics using spectral reflec-
for monochromatic light can be expressed as follows; tions depends on statistical models that explain the relationship
4 E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14

Fig. 3. Stretching and bending vibrations.

Fig. 4. Spectral resolution of some materials (Source: Short, 2011).

between them, most common used models are stepwise multiple elimination, and bidirectional elimination. The forward selection
linear regression (SMLR), partial least squares regression (PLSR), involves starting with no variables in the model, testing the addi-
multivariate adaptive regression splines (MARS), principal compo- tion of each variable using a chosen model fit criterion, adding
nent regression (PCR) and artificial neural networks (ANN). SMLR is the variable (if any) whose inclusion gives the most statistically
a statistical method of regressing multiple variables while simulta- significant improvement of the fit, and repeating this process until
neously removing those that aren’t important. The choice of pre- none improves the model to a statistically significant extent. Back-
dictive variables is carried out by an automatic procedure ward elimination involves starting with all candidate variables,
(Efroymson, 1960; Hocking, 1976; Draper and Smith, 1981; and testing the deletion of each variable using a chosen model fit crite-
SAS, 1989). The variable that considered for addition to or subtrac- rion, deleting the variable (if any) whose loss gives the most statis-
tion from the set of explanatory variables in each step is based on a tically insignificant deterioration of the model fit, and repeating
form of a sequence of F-tests or t-tests. The widely used algorithm this process until no further variables can be deleted without a sta-
was first proposed by Efroymson (1960). The main types of Step- tistically significant loss of fit. The bidirectional elimination is a
wise multiple linear regression are forward selection, backward combination of the forward selection and backward elimination
E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14 5

Fig. 5. (a) PLSR in 2006; (b) MARS in 2006; (c) PLSR in 2012; and (d) MARS in 2012 (Source: Nawar et al., 2014).

types. The accuracy of SMLR model is measured as the actual stan- ables. MARS is also called EARTH in many implementations.
dard error (SE) or the mean error between the predicted value and MARS consists of two phases: the forward and the backward pass.
the actual value in the hold-out sample (Mayers and Forgy, 1963) The forward pass starts with a model consists of the mean of the
Fig. 5. response values and then repeatedly adds basis function in pairs
PLS is a statistical method that finds a linear regression by pro- to the model. At each step it finds the pair of basis functions that
jecting the predicted variables and the observable variables to a gives the maximum reduction in sum-of-squares residual error.
new space (Tenenhaus et al., 2005; Vinzi et al., 2010). PLS regres- The two basis functions in the pair are identical except that a dif-
sion is today most widely used in chemometrics, sensometrics, ferent side of a mirrored hinge function is used for each function.
and other related areas (Rönkkö et al., 2015). Each new basis function consists of a term already in the model
MARS is a form of non-parametric regression analysis multiplied by a new hinge function. This process of adding terms
(Friedman, 1991). MARS is an extension of linear models that auto- continues until the change in residual error is too small to continue
matically models nonlinearities and interactions between vari- or until the maximum number of terms is reached (Friedman,
6 E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14

1993). The backward pass removes terms one by one, deleting the diagnostic power of the sensor and improve the classification. For
least effective term at each step until it finds the best sub-model. example, a study utilized Landsat Thematic Mapper (TM) data and
MARS models are more flexible than linear regression models. Digital elevation model (DEM) obtained topographical indices for
PCR is a regression method that considers regressing the depen- mapping soil salinity in Western Australian (Caccetta et al.,
dent variable on a set of independent variables based on a standard 2000). Furthermore, hyperspectral data increase the capability of
linear regression model, but uses PCA for estimating the unknown remotely sensed information, thereby, can be applied more inde-
regression coefficients in the model (Jolliffe, 1982). Instead of pendently of other data sets. The absence of spectral features of
regressing the dependent variable on the explanatory variables salt still makes classification difficult. However, several researchers
directly, the principal components of the explanatory variables have concluded that soil salinity can be mapped based on other
are used as regressors. PCR is some kind of a regularized procedure. properties of soil as alternatives. Ben-Dor et al. (2002) reported
The principal components with the higher variances are selected as that, hyperspectral scanner data was used for mapping soil salinity,
the regressors. The major use of PCR lies in overcoming the multi- also there was a correlation between soil moisture and salinity
collinearity problem which arises when two or more of the reached (r = 0.58) in cultivated crops and was able to develop reli-
explanatory variables are close to being collinear (Dodge, 2003). able prediction equations. Moreover, hyperspectral remote sensing
PCR can result in dimension reduction through substantially low- data have been utilized to monitor soil salinity under different
ering the effective number of parameters characterizing the under- environmental conditions, as well as other halophyte species such
lying model. PCR can lead to efficient prediction with the as Sea Blite and Sea Barley Grass (Dehaan and Taylor, 2003). Nawar
appropriate selection of the principal components to be used for et al. (2014) coupled MARS, PLSR and NIR soil spectra and geo-
regression. statistics to map spatial variation of soil salinity in El-Tina Plain,
ANN is a computing system made up of a number of simple, north Sinai, Egypt. They measured electrical conductivity (ECe)
highly interconnected processing elements, which process infor- data and eflectance spectra of soil samples resampled to satellite
mation by their dynamic state response to external inputs sensor’s resolution (Fig. 5). The study reported good results for
(Caudill, 1987). ANNs are processing algorithms that are loosely the prediction of soil salinity; MARS (R2 = 0.73), RMSE = 6.53, and
modeled after the neuronal structure of the mammalian cerebral ratio of performance to deviation (RPD) = 1.96), while PLSR model
cortex but on much smaller scales. Neural networks are typically (R2 = 0.70, RMSE = 6.95, and RPD = 1.82).Moreover, the authors
organized in layers. Layers are made up of a number of intercon- emphasized that MARS gives very good results for prediction of soil
nected ‘nodes’ which contain an ‘activation function’. Patterns are salinity, especially under high salinity levels. Thus, it is important
presented to the network via the ‘input layer’, which communi- to monitor and map soil salinity at an early stage to enact effective
cates to one or more ‘hidden layers’ where the actual processing soil reclamation program that helps to lessen or prevent future
is done via a system of weighted ‘connections’. The hidden layers increase in soil salinity. Remote sensing has more informative
then link to an ‘output layer’ where the answer is the output. Most and professional rapid assessment of soil salinity, compared with
ANNs contain some form of ’learning rule’ which modifies the traditional methods offering more informative and professional
weights of the connections according to the input patterns. ANNs rapid assessment techniques for monitoring and mapping soil
provide an analytical alternative to conventional techniques which salinity. Soil salinity can be identified from remote sensing data
are often limited by strict assumptions of normality, linearity, and obtained by different sensors based on visible direct indicators that
variable independence. refer to salt features at soil surface indicators, such as the presence
of halophytic plant.

5. Applications of NIRS in soil sciences


5.2. Soil moisture

As mentioned above, Near-infrared reflectance spectroscopy


Previous numerous studies have shown the role of reflectance
(NIRS) has been used to predict several soil properties. There are
spectroscopy for monitoring soil moisture. Many studies illus-
many authors focuses their works on predicting soil characteristics
trated the inverse relationship between soil moisture and spectral
based on Near-infrared reflectance spectroscopy, some of them
reflectance (Post et al., 2000; Galvao et al., 2001). Furthermore, the
selected one parameter of soil with reflectance spectroscopy. Fur-
inverse relationship means the decrease of reflectance with the
thermore other authors predicted soil parameters as relation with
increase of soil moisture content. This relationship is due to two
reflectance spectroscopy (Zbíra et al., 2016). Some applications will
reasons; soil particles covered with thin films of water and water
be discussed as follows:
on the lattice sites of some minerals present in the soil (Stoner
and Baumgardner, 1981). With the improvement of measurement
5.1. Soil salinity tools, the change in spectral reflectance with change in soil mois-
ture levels became more pronounced at longer wavelengths
Salinization is an overall issue that influences the physical and (>1450 nm) (Weidong et al., 2002). The same study also showed
chemical soil properties that leads to loss in yield efficiency. that, at higher moisture contents the trend is changed and the
Throughout the previous two decades, remotely detected symbol- reflectance increased with the increasing of moisture content. They
ism has exhibited its capacity to evaluate saltiness changes at dif- determined this type of reversal to be somewhere around field
ferent scales (Metternicht and Zinck, 2008; Elnaggar and Noller, capacity, while it changed for different soils, and happens before
2009). Numerous studies have illustrated the ability of Vis-NIR the point where water retention is saturating the reflectance
reflection spectroscopy bands – from the optical sensors – for signal.
detecting surface soil salinity. Furthermore, hyperspectral data Bogrekci and Lee (2006) investigated the possibility of estimat-
have been used in several approaches for quantitative assessment ing phosphorus by spectral reflectance under the influence of dif-
of soil salinity and different soil properties (Dehaan and Taylor, ferent levels of soil moisture with different phosphorus (P)
2003; Farifteh et al., 2008; Feyziyev et al., 2016). It has been sho- concentrations (0, 12.5, 62.5, 175, 375, 750, and 1000 mg kg1)
wen that, effective prediction of saltiness is administered by the using ultraviolet (UV), visible (VIS), and near-infrared (NIR) absor-
relationship between other soil properties, such as soil moisture bance spectroscopy (Fig. 6). The authors illustrated that the mois-
(Ben-Dor et al., 2002). For multispectral image studies, the inclu- ture content affected the absorbance spectra, where correlation
sion of topographic data is sometimes used to mitigate the poor coefficient between spectra absorbance and P concentrations
E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14 7

Fig. 6. Average Spectral reflectance at different level of soil moisture (Source: Bogrekci and Lee, 2006).

Fig. 7. Correlation coefficient spectra between absorbance and P concentration at different moisture contents within the 225–2550 nm range. (Source: Bogrekci and Lee,
2006).

showed high values within the 1982–2550 nm range. In addition, sharp peaks (Fig. 8). The broad unordered bands are more common
spectral signal processing by removing the moisture content effect in naturally occurring soils in addition, the highest significant vari-
enhanced P prediction in soils considerably (Fig. 7). able in determining the reflectance located within a range 2080–
The study of any soil property is related to the understanding of 2320 mm (Baumgardner et al., 1985 and Galvao et al., 2001). The
sensitive areas at the spectrum due to presence of water. The broad unordered bands are more common in naturally occurring
vibrational frequencies of water molecules after 2500 nm affect soils. Furthermore, the highest significant variable in determining
the water absorption wavelengths (Baumgardner et al., 1985). the reflectance changes in the 2080–2320 mm. However, other
The 1450 and 1950 nm wavelengths are the absorption bands with studies emphasized on the importance role of reflectance
8 E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14

Fig. 8. Atmospheric water absorption bands.

spectroscopy and remote sensing to develop spectral models for 5.4. Clay minerals
detecting soil moisture content (Ben-Dor et al., 2002; Whiting
et al., 2004). Soil chemistry affects clay minerals, thereby the soil develop-
ment and their fertility. Many of clay minerals have unique spectral
5.3. Soil organic carbon reflectance at visible wavelengths and NIR-SWIR (Hunt, 1980). Silva
et al. (2016) illustrated that near-infrared region can be used to
Soil organic carbon (SOC) is a key characteristic of soil quality predicate soil attribute with PLSR using a limited spectral region
which impacts the assortment of organic compounds and physical (325–1075 nm) performed poorly for sand while more promising
properties of soils (Carter, 2002). The evaluation of greenhouse gas when considering the capabilities to predict silt and clay. The appli-
emissions from soils requires a precise information on the fate of cation of visible and part of the (400–980 nm) for clay prediction in
carbon and nitrogen in soils. Near-infrared reflectance spec- Oxisols achieved relative good results where regression coefficients
troscopy (NIRS) is a quick and non-damaging explanatory proce- showed good relation to the spectral behavior of weathered soils in
dure that includes diffuse-reflectance estimation in the near visible and near-infrared region .The components of soil minerals
infrared region (1000–2500 nm). Visible–NIR spectroscopy with affect the spectral reflectance of the soil through the absorption
decision-tree modeling can fairly and accurately, with small to bands and overall spectral brightness. Quartz is the biggest and
moderate uncertainty, predict soil organic carbon (Viscarra- most regular part of soils; it shows no unique absorption feature
Rossel and Hicks, 2015 and Hu et al., 2015). Soil organic matter over Vis-NIR-SWIR range although it does increase the overall
(SOM) decrease the vis-NIR spectral reflectance range (520– brightness. Clay minerals have unique absorption bands that are
800 nm), especially if the SOM content is bigger than 2% (Stoner effected by distinctive vibrational overtones, electronic and charge
and Baumgardner, 1981; Henderson et al., 1992). Humic acid con- transfers, and conduction processes (Clark, 1999). The wavelengths
sidered the most dark pigment of SOM and reduces the spectral around 2200 nm for the spectral characteristics of clay minerals
reflectance over the visible to short-wave spectral range. Other- were extracted from AVIRIS data for the identification of smectite,
wise, fulvic acid has no influence on soil reflectance (Henderson kaolinite and illite clay minerals (Chabrillat et al., 2002). The alter-
et al., 1992). A study of soils in Thailand using artificial neural net- ation phases were mapped based on absorption band position,
works found that vis-NIR VNIR spectrum (400–1100 nm) as a pre- depth and asymmetry from AVIRIS data (van-der-Meer, 2004). As
cise detector of SOM (R2 = 0.86) (Daniel et al., 2003). Furthermore, vegetation obscure the target material partially by the large distinc-
other study (Ben-Dor et al., 2002) has used hyperspectral images tive absorption features, the Mineralogical identification achieved
for mapping SOM based on the reflectance spectra of heavy clay features (Chabrillat et al., 2002). Similarly, absorption band posi-
soils in Israel where the root mean square of the prediction equa- tion, depth and asymmetry have been used to map alteration
tions was (R2 m > 0.82). A support vector machine regression phases with AVIRIS imagery (van-der-Meer, 2004). Mineralogical
(SVMR) and a successive projections algorithm (SPA) model (SPA- identification has been achieved when the target material is par-
SVMR model) have been used for improving the accuracy of soil tially obscured by vegetation due the largely distinctive absorption
organic carbon (SOC) which has resulted from integrating the features (Chabrillat et al., 2002). In the spectrum of hematite (an
laboratory-based visible and near-infrared (VIS/NIR, 350– iron-oxide mineral), the strong absorption in the visible light range
2500 nm) spectroscopy of soils (Xiaoting et al., 2014). Another is caused by ferric iron (Fe+3). In calcite, the major component of
image study used digitized color aerial photography to successfully limestone, the carbonate ion (CO=3) is responsible for a series of
map SOM based on two approaches. The first attempt was to study absorption bands between 1.8 and 2.4 mm (mm). The most common
the individual pixels thereby describe the spatial distribution; the clay minerals in soil are kaolinite and montmorillonite, these min-
second attempt was applying the relationship on image classifica- erals are distinct from others depending on the absorption spec-
tion to determine the classes units (Fig. 9). troscopy bands where the highest absorption band around 1.4 mm
E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14 9

Fig. 9. Spatial distribution of soil organic carbon (Source: Chen et al., 2000).

in wavelength, along with the weak 1.9 mm band in kaolinite, refer- 5.5. Soil surface features
ring to hydroxide ions (OH-), while the stronger 1.9 mm band in
montmorillonite is affected by bound water molecules in this The identification of surface soil features and land resources are
hydrous clay (Fig. 10). On the other hand, feldspar, the dominant very important for precise management in different scales. The
mineral in granite – shows no significant absorption features in spectral signature of each soil property influenced by spatial and
the vis-NIR-SWIR (Hauff et al., 1991; Masinter and Lyon, 1991). temporal variability of surface processes however, it is difficult to
The combination of spectroscopy reflectance data and hyperspec- measure directly from their reflectance spectra even under con-
tral satellite images give remarkable results for deriving dominant trolled laboratory conditions (Silva and ten Caten, 2016). Soil
clay mineral. The results from modeling dominant clay minerals by Vis-NIR (350–2500 nm) reflectance spectra contain valuable infor-
random forests and mapping of hyperion data using Spectral Angu- mation for predicting soil textural fractions (sand, silt, and clay
lar Mapper (SAM) illustrated the dominance of kaolinite clay min- content). Chemometrics techniques and multivariate calibration
eral followed by montmorillonite in Madhya Pradesh India (PLSR) allowed researchers to extract the relevant information
(Fig. 11) (Janaki et al., 2014). from the reflectance spectra and to correlate this with the soil
10 E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14

Fig. 10. Spectral reflectance of different clay minerals. (Source: Clark, 1999).

Fig. 11. Dominant clay mineral in Madhya Pradesh India. (Source: Janaki et al., 2014).

texture fractions. The same author illustrated that soil texture can pixel, the pixel spectra is the signatures of the ground material
be predicted where sand content (R2 = 0.81) and clay content (Roberts et al., 1993). Saleh et al. (2013) used a linear spectral
(R2 = 0.80) and less satisfactory for silt content (R2 = 0.70).The unmixing analyses to discriminate different surface soil types in
spectral signature from an image pixel is a mixture ofsurface mate- north sinia – Egypt by Near-infrared reflectance spectroscopy tech-
rials affected by their chemical components. The spectral proper- niques (Fig. 12). The spectra of the soil types were significantly
ties of a single image pixel is the representation of the surface influenced by the different surface features presented in the area.
components. Each pixel retains the characteristic features of the The same author concluded that linear spectral unmixing is very
individual spectra from each of the component reflective materials. helpful tool for identifying and mapping the different surface soil
When the ground material- such as soil types – occupies the whole types from ETM + by discriminating the different mixture spectra
E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14 11

Fig. 12. Map of surface soil types using unmixing spectra model with field spectral data. (Source: Saleh et al., 2013).

of the image into distinct soil types spectra. Furthermore, linear ing model allowed all the bands in the ETM+ image to be used, it
spectral unmixing model allowed all the bands in the ETM+ image has been proven to be reasonable in mapping soil surface features.
to be used, it has been proven to be reasonable in mapping soil sur- The authors recommended that further research is needed to eval-
face features. The spectral signature from an image pixel is a mix- uate the spectral unmixing technique for different soil types and
ture of surface materials and it affected by their chemical different image types.
components. the spectral properties of a single image pixel is the
representation of the mixed spectra of the surface components. 5.6. Soil contamination
Each pixel retains the characteristic features of the individual spec-
tra from each of the component reflective materials. When the Soil contamination by heavy metals is considered the main
ground material- such as soil types – occupies the whole pixel, environmental problem, most of them have toxic effects on plant
the pixel spectra is the signatures of the ground material and microorganisms in soil when allowing increase of their con-
(Roberts et al., 1993). Saleh et al. (2013) used a linear spectral centration levels (Mohamed et al., 2016). Conventional methods
unmixing analyses to discriminate the different surface soil types for measuring heavy metal such as inductively coupled plasma
in north sinia – Egypt by Near-infrared reflectance spectroscopy (ICP), atomic absorption spectrometry (AAS) or operationally
techniques. The spectra of the soil types were significantly influ- defined sequential extraction can estimate physico-chemical data
enced by the different surface features presented in the area. The directly, but take a long time and are very expensive. Comparing
author concluded that linear spectral unmixing is very helpful tool furthermore, the spectral analyses techniques can be used in a
for identifying and mapping the different surface soil types from wide range. Diffuse reflectance spectroscopy (DRS) in visible-near
ETM+ by discriminating the different mixture spectra of the image infrared (VNIR) region (400–2500 nm) has been used to quickly
into distinct soil types spectra. Furthermore, linear spectral unmix- analyse soil characteristics both appropriately and precisely

Fig. 13. The areas affected by concentration of heavy elements. (Source; Mohamed et al., 2016).
12 E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14

Fig. 14. Spatial distribution of selected heavy metals (Source; Mohamed et al., 2016).

through VNIR DRS, soil constituents such as iron oxides (Madeira the studied area while, an increase in concentration of Cr in the
et al., 1997; Ji et al., 2002). Heavy metals, often absorbed or areas located near Fakous city as showen in (Fig. 14).
bounded, are characterized by spectrally active constituents based
on the environmental condition which make it possible to investi-
gate their characteristics using VNIR- DRS. Previous works showed 6. Conclusion
the availability for predicting of heavy metals in soil by spectro-
scopic reflectance (Wu et al., 2005; Xia et al., 2007; Moros et al., Under the current situations, the conventional methods of soil
2009). On the other hand, heavy metals such Cr, Cu, Zn and As have analyses takes a long time, in addition to their expensive costs.
negative correlation coefficients with the spectral bands which Near-infrared reflectance spectroscopy, as advanced tools may
attributable to the absorption features of iron oxides, clay, and manage to overcome previous obstacles .This article addresses
organic matter, suggesting they are strongly bound to these soil the fundamental concepts of spectroscopic reflectance techniques
constitutes. Significant relationship between Pb, Cd, and Hg and and some applications on soil sciences. It also explores the capabil-
TOC spectral bands was observed. That indicates the important ity of Near-infrared reflectance spectroscopy for detecting and
organic matter binding for these elements. (Junfeng et al., 2010). mapping soils characteristics. The main limiting factor in assess-
Moreover, the best relationship between Pb-Zn-Mn and ratio of ment of the soil properties is finding certain data pre-treatment
610/500 nm range while Ni-Cr have the highest correlation that and calibration procedures, where, correlation between soil reflec-
associated with slope in range of 980 nm, however the authors tance data and values of each soil properties could be achieved.
indicated that spectral parameters and reflectance values for Mn, Quantitative prediction of soil properties (e.r. salinity, organic car-
Pb and Zn within 400–2500 nm range have a better prediction abil- bon, soil moisture and heavy metals) can be conducted using var-
ity for contaminated soil other than for Cr and Ni but not suitable ious calibration models which were developed depending on the
for Fe, Cu, Cd, EC and pH (McCarty et al., 2002; Vodyanitskii, 2013). measured soil analysis and soil reflectance spectra then resampled
Mohamed et al. (2016) investigated the capability of vis-NIR (350– to satellite images to predict soil properties. moreover good predic-
2500 nm) for calibration procedures and predicted contaminated tions for some chemical, physical and biological properties can be
soil in the area closed to Bahr El-Baqar, east of Nile Delta. The achieved based on laboratory analysis and NIRS data. The integra-
authors used stepwise multiple linear regression (SMLR) to tion of different remote sensing data and NIRS can be maximized
develop calibration models for Cr, Mn and Cu. The most affected the ability to cover and investigate large surfaces in a single flight
regions on spectral reflectance were 2010, and 2149 nm for copper campaign and thus produce a complete picture of surface soil
also, 2139 and 2072 nm for manganese as shown in (Fig. 13). The properties. Several predictive models such partial least squares
concentrations of heavy metals were estimated with high accuracy regression (PLSR), Stepwise multiple linear regression (SMLR),
where, R2 was recorded 0.82, 0.75 and 0.65 for Cr, Mn and Cu, multivariate adaptive regression splines (MARS) principal compo-
respectively. In addition, the authors highlighted the environmen- nent regression (PCR) and artificial neural networks (ANN) were
tal hazards occurred along with Bahr El-Baqar drain where the spa- recommended to model the relation between soil variability and
tial distribution maps of heavy metals were produced based on reflectance. However Reflectance spectroscopy is a good and sim-
reflectance spectroscopy data. Moreover, the results showed an ple analytical method that can be utilized to measure several soil
increase in both Mn and Cu concentration towards the north of properties at the same time. Vis-NIR reflection spectroscopy
E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14 13

reduces the cost and time cost effective that could be alternatives spectroscopy to define ore host environments. In: Proc. Eighth Thematic Conf.
on Geologic Remote Sensing (Denver), 447–458.
to the traditional methods for mapping soil characteristics.
Henderson, T.L., Baumgardner, M.F., Franzmeier, D.P., Stott, D.E., Coster, D.C., 1992.
Transparency Note: Prof. Gad is one of the authors of this article. High dimensional reflectance analysis of soil organic matter. Soil Sci. Soc. Am. J.
56, 865–872.
Herrmann, R., Onkelinx, C., 1986. Quantities and units in clinical chemistry,
References Nebulizer and flame properties in flame emission and absorption spectrometry
(Recommendations 1986). Pure Appl. Chem. 58, 1737–1742.
Hocking, R.R., 1976. The Analysis and Selection of Variables in Linear Regression.
Abdi, D., Cade-Menun, B.J., Ziadi, N., Tremblay, G.F., Parent, L.É., 2016. Visible near Biometrics 32.
infrared reflectance spectroscopy to predict soil phosphorus pools in Hu, W., Chau, H.W., Si, B.C., 2015. Vis-Near IR reflectance spectroscopy for soil
chernozems of Saskatchewan, Canada. Geoderma Reg. 7 (2), 93–101. organic carbon content measurement in the Canadian Prairies. Clean: Soil, Air,
Baumgardner, M.F., Silva, L.R., Biehl, L.L., Stoner, E.R., 1985. Reflectance properties of Water 43, 1215–1223.
soils. Adv. Agron. 38, 1–44. Hunt, G.R., 1980. Electromagnetic radiation – the communication link in remote
Ben-Dor, E., Patkin, K., Banin, A., Karnieli, A., 2002. Mapping of several soil sensing. In: Siegal, B.S., Gillespie, A.R. (Eds.), Remote Sensing in Geology. Wiley,
properties using DAIS- 7915 hyperspectral scanner data – a case study over New York.
clayey soils in Israel. Int. J. Remote Sensing 23, 1043–1062. Janaki, R.S., Sreenivas, K., Sivasamy, R., 2014. Hyperspectral analysis of clay
Ben-Gera, I., Norris, K.L.H., 1968. Determination of moisture content in soybeans by minerals. The International Archives of the Photogrammetry, Remote Sensing
direct spectrophotometry. Isr. J. Agric. Res 18, 124–132. and Spatial Information Sciences, Vol. XL-8, ISPRS Technical Commission VIII
Bogrekci, I.W., Lee, S., 2006. Effects of soil moisture content on absorbance spectra Symposium, 09–12 Dec. Hyderabad, India.
of sandy soils in sensing phosphorus concentrations using Uv-Vis-Nir Ji, J., Balsam, W., Chen, J., Liu, L., 2002. Rapid and quantitative measurement of
Spectroscopy. Am. Soc. Agric. Biol. Eng. (ASABE) 49, 1175–1180. hematite and goethite in the Chinese loess-paleosol sequence by diffuse
Brand, J.C.D. (Ed.), 1995. Lines of Light, the Sources of Dispersive Spectroscopy. reflectance spectroscopy. Clays Clay Miner. 50, 208–216.
Gordon and Breach Publishers, ISBN 2884491627. 1800-1930. Jolliffe, Ian T., 1982. A note on the Use of Principal Components in Regression. J. R.
Caccetta, P.A., Allen, A., Watson, I., 2000. The land monitor project. In: 10th Stat. Soci. Ser. C 31 (3), 300–303.
Australasian Remote Sensing and Photogrammetry Conference, Adelaide, Junfeng, J., Yinxian Song, A., Xuyin-Yuan, B., Yang Z., 2010. Diffuse reflectance
Australia, 21–25 August. spectroscopy study of heavy metals in agricultural soils of the Changjiang River
Carter, M.R., 2002. Soil quality for sustainable land management, organic matter Delta, China Junfeng. In: 19th World Congress of Soil Science, Soil Solutions for a
and aggregation interactions that maintain soil functions. Agron. J. 94, 38–47. Changing World, 1–6 August 2010, Brisbane, Australia.
Caudill, M., 1987. Neural networks primer, part I. AI Expert 2 (12), 46–52. Khan, N.M., Rastoskuev, V.V., Sato, Y., Shiozawa, S., 2005. Assessment of hydrosaline
Chabrillat, S., Goetz, A.F.H., Krosley, L., Olsen, H.W., 2002. Use of hyperspectral land degradation by using a simple approach of remote sensing indicators.
images in the identification and mapping of expansive clay soils and the role of Agric. Water Manag. 77, 96–109.
spatial resolution. Remote Sens. Environ. 82, 431–445. Madeira, J., Bedidi, A., Cervelle, B., Pouget, M., Flay, N., 1997. Visible spectrometric
Chang, C., Laird, D.A., Mausbach, M.J., Hurburgh, C.R.J., 2001. Near-infrared indices of hematite (Hm) and goethite (Gt) content in lateritic soils, the
reflectance spectroscopy-principal components regression analyses of soil application of a Thematic Mapper (TM) image for soil mapping in Brasilia,
properties. Soil Sci. Soc. Am. J. 65, 480–490. Brazil. Int. J. Remote Sensing 18, 2835–2852.
Chen, F., Kissel, D.E., West, L.T., Adkins, W., 2000. Field-scale mapping of surface soil Masinter, R.A., Lyon, R.J.P., 1991. Spectroscopic confirmation of increasing illite
organic carbon using remotely sensed imagery. Soil Sci. Soc. Am. J. 64, 746–753. ordering with hydrothermal alteration of argillaceous ore in the Gold Bar Mine,
Ciurczack, E.W., 2001. Principles of Near Infrared Spectroscopy. In: Burns, D.A., Eureka CO, Nevada. In: Proc. Eighth Thematic Conf. on Geologic Remote Sensing
Ciurczack, E.W. (Eds.), Handbook of Near-Infrared Analysis. Marcel Dekker, New (Denver), 563–571.
York, pp. 7–18. Mateusz Kania and Piotr gruba, 2016. Estimation of selected properties of forest
Clark, R.N., 1999. Spectroscopy of Rocks and Minerals and Principles of soils using near-infrared spectroscopy (NIR). Soil Sci. Ann. 67 (1), 32–36.
Spectroscopy. In: Rencz, A.N. (Ed.), Remote Sensing for the Earth Sciences, McCarty, G.W., Reeves, J.B., Reeves, V.B., Follet, R.F., Kimble, J.M., 2002. Mid-Infrared
Manual of Remote Rensing, vol. 3. John Wiley and Sons, New York, pp. 3–58. and near infrared diffuse reflectance spectroscopy for soil carbon measurement.
Crouch, S., Skoog, D. (Eds.), 2007. Principles of Instrumental Analysis. Thomson Soil Sci. Soc. Am. J. 66, 640–646.
Brooks/Cole, Australia, ISBN 0-495-01201-7. Metternicht, G.I., Zinck, J.A. (Eds.), 2008. Remote Sensing of Soil Salinization. Impact
Daniel, K.W., Tripathi, N.K., Honda, K., 2003. Artificial neural network analysis of on Land Management. CRC Press, Boca Raton, FL, USA, p. 337.
laboratory and in situ spectra for the estimation of macronutrients in soils of Mohamed, E.S., 2013. Spatial assessment of desertification in north Sinai using
Lop Buri (Thailand). Aust. J. Soil Res. 41, 47–59. modified MEDLAUS model. Arabian J. Geosci. 6 (12), 4647–4659.
Dehaan, R., Taylor, G.R., 2003. Image-derived spectral endmembers as indicators of Mohamed, Elsayed Said, Belal, Abdelaziz, Shalaby, Adel, 2015. Impacts of soil
salinization. Int. J. Remote Sensing 24, 775–794. sealing on potential agriculture in Egypt using remote sensing and GIS
Demattê, J.A.M., da Silva, Terra F., 2014. Spectral pedology: a new perspective on techniques. Eurasian Soil Sci. 48 (10), 1159–1169.
evaluation of soils along pedogenetic alterations. Geoderma 217–218, 190– Mohamed, E.S., Ali, A.M., El-Shirbeny, M.A., Abd El-Razek, A.A., Savin, I., 2016. Near
200. infrared spectroscopy techniques for soil contamination assessment in the Nile
Drago, R.S., 1992. Physical Methods for Chemists. second ed., Saimders College delta. Eurasian Soil Sci. 49, 632–639.
Publishing. ISBN: 10: 0030751764. Moros, J., Vallejuelo, D.S.F.O., Gredilla, A.D.A., Madariaga, J.M., Garrigues, S., de-la-
Dodge, Y., 2003. The Oxford Dictionary of Statistical Terms, OUP. ISBN 0-19- Guardia, M., 2009. Use of reflectance infrared spectroscopy for monitoring the
920613-9. metal content of the estuarine sediments of the nerbioi-ibaizabal river
Draper, N., Smith, H., 1981. Applied Regression Analysis, second ed. New York: John (metropolitan bilbao, bay of biscay, basque country). Environ. Sci. Technol. 43,
Wiley & Sons Inc. 9314–9320.
Dwivedi, R.S., Rao, B.R.M., 1992. The selection of the best possible Landsat TM band NASA’s, 2014. Comparison of Wavelength, Frequency and Energy for the
combination for delineating salt-affected soils. Int. J. Remote Sens. 13, 2051– Electromagnetic Spectrum. <https://imagine.gsfc.nasa.gov/science/toolbox/
2058. emspectrum1.html>.
Elnaggar, A.A., Noller, J.S., 2009. Application of remote-sensing data and decision- Nawar, S., Henning, B., Joachim, H., Jacek, K., 2014. Modelling and mapping of soil
tree analysis to mapping salt-affected soils over large areas. Remote Sens 2, salinity with reflectance spectroscopy and Landsat data using two quantitative
151–165. methods (PLSR and MARS). Remote Sens. 6, 10813–10834.
Efroymson, M.A., 1960. Multiple regression analysis, Mathematical Methods for Niederberger, J., Todt, B., Boca, A., Nitschke, R., Kohler, M., Kühn, P., Bauhus, J., 2015.
Digital Computers. In: Ralston, A., Wilf, H.S. (Eds.), Wiley. New York. Use of near-infrared spectroscopy to assess phosphorus fractions of different
Farifteh, J., van-der-Meer, F., van-der-Meijde, M., Atzberger, C., 2008. Spectral plant availability in forest soils. Biogeosciences 12, 3415–3428.
characteristics of salt-affected soils. A laboratory experiment. Geoderma 145, Nield, S.J., Boettinger, J.L., Ramsey, R.D., 2007. Digitally mapping gypsic and natric
196–206. soil areas using Landsat ETM data. Soil Sci. Soc. Am. J. 71, 245–252.
Feyziyev, F., Babayev, M., Priori, S., L’Abate, G., 2016. Using visible-near infrared Post, D.F., Fimbres, A., Matthias, A.D., Sano, E.E., Accioly, L., Batchily, A.K., Ferreira, L.
spectroscopy to predict soil properties of mugan plain, Azerbaijan. Open J. Soil G., 2000. Predicting soil albedo from soil color and spectral reflectance data. Soil
Sci. 6, 52–58. http://dx.doi.org/10.4236/ojss.2016.63006. Sci. Soc. Am. J. 64, 1027–1034.
Friedman, J.H., 1991. Multivariate Adaptive Regression Splines. Ann. Stat. 19, 1. Roberts, D.A., Smith, M.O., Adams, J.B., 1993. Green vegetation, non-photosynthetic
Friedman, J.H., 1993. Fast MARS. Stanford University Department of Statistics. vegetation, and soils in AVIRIS data. Remote Sens. Environ. 44, 255–269.
Technical Report 110. Saleh, A.M., Belal, A.B., Arafat, S.M., 2013. Identification and mapping of some soil
Fikrat, F., Maharram, B., Simone, P., Giovanni, L., 2016. Using visible-near infrared types using field spectrometry and spectral mixture analyses: a case study of
spectroscopy to predict soil properties of mugan plain, Azerbaijan. Open J. Soil North Sinai, Egypt. Arabian J. Geosci. 6 (6), 1799–1806.
Sci. 6, 52–58. Saleh, A.M., Belal, A.B., Mohamed, E.S., 2015. Land resources assessment of El-
Galvao, L.S., Pizarro, M.A., Epiphanio, J.C.N., 2001. Variations in reflectance of Galaba basin, South Egypt for the potentiality of agriculture expansion using
tropical soils. Spectral-chemical composition relationships from AVIRIS data. remote sensing and GIS techniques. Egypt. J. Remote Sensing Space Sci. 18 (1),
Remote Sensing Environ. 75, 245–255. S19–S30.
Hauff, P., Kruse, F.A., Madrid, R., Fraser, S., Huntington, J., Jones, M., Watters, S., 1991. Samsonova, V.P., Meshalkina, Y.L., 2011. Quantitative method of comparing soil
Illite crystallinity – case histories using X-ray diffraction and reflectance maps and schematic maps. Eurasian Soil Sci. 3, 3–5.
14 E.S. Mohamed et al. / Egypt. J. Remote Sensing Space Sci. 21 (2018) 1–14

Savin, I., 1993. Formation of reflectance properties of tilled chernozem soil’s surface Vodyanitskii, Y.N., 2013. Contamination of soils with heavy metals and metalloids
// Optical Monitoring of the Environment, CIS (selected papers), SPIE, US, No. and its ecological hazard (Analytic Review). Eurasian Soil Sci. 46, 793–801.
2107, pp. 304-313. Weidong, L., Baret, F., Xingfa, G., Qingxi, T., Lanfen, Z., Bing, Z., 2002. Relating soil
Shepherd, K.D., Walsh, M.G., 2002. Development of reflectance spectral libraries for surface moisture to reflectance. Remote Sens. Environ. 81, 238–246.
characterization of soil properties. Soil Sci. Soc. Am. J. 66, 988–998. Wetzel, D., 1983. Near-infrared reflectance analysis, Sleeper among spectroscopic
Short, N.M., 2011. The LANDSAT Tutorial Workbook, Basics of Satellite Remote techniques. Anal. Chem. 55, 1165A–1176A.
Sensing, NASA Reference Publication. 1078. NASA, USA, p. 553. Whiting, M.L., Li, L., Ustin, S., 2004. Predicting water content using Gaussian model
Siesler, H.W., Ozaki, Y., Kawata, S., Heis, H.M., 2002. Near-In Frared Spectroscopy. on soil spectra. Remote Sens. Environ. 89, 535–552.
Principles, Instruments, Applications. WILEY-VCH Verlag GMbH, p. 361. Workman, J.J., 1996. Interpretive spectroscopy for near infrared. Appl. Spec. Rev. 31,
Silva, Elisângela Benedet, ten Caten, Alexandre, 2016. Estimating Soil Texture from a 251–320.
Limited Region of the Visible/Near-Infrared Spectrum. Digital Soil Morphometr., Wu, Y., Chen, J., Ji, J.F., Tian, Q.J., Wu, X.M., 2005. Feasibility of reflectance
73–87 spectroscopy for the assessment of soil mercury contamination. Environ. Sci.
Sonia, G., Newman, R., Saadi, A., 2012. A spectroscopic approach to assess heavy Technol. 39, 873–878.
metals contents of the mine waste of Jalta and Bougrine in the north of Tunisia. Xia, X.Q., Mao, Y.Q., Ji, J., Ma, H.R., Chen, J., Liao, Q.L., 2007. Reflectance spectroscopy
J. Geograph. Inform. Syst. 4, 242–253. study of Cd contamination in the sediments of the Changjiang River, China.
Stoner, E.R., Baumgardner, M.F., 1981. Characteristic variation in reflectance of Environ. Sci. Technol. 41, 3449–3454.
surface soils. Soil Sci. Soc. Am. J. 45, 1161–1165. Xiaoting, P., Tiezhu, S., Aihong, S., Yiyun, C., Wenxiu, G., 2014. Estimating soil
Tenenhaus, M., Esposito Vinzi, V., Chatelinc, Y.-M., Lauro, C., 2005. PLS path organic carbon using VIS/NIR spectroscopy with SVMR and SPA methods.
modeling (PDF). Comput. Stat. Data An. 48 (1), 159–205. Remote Sens. 6, 2699–2717.
van-der-Meer, F., 2004. Analysis of spectral absorption features in hyperspectral Zbíra, Jiří, David, Čižmár, Stanislav, Malý, Elena, Obdržálková, 2016.
imagery. Int. J. Appl. Earth Obs. Geoinf. 5, 55–68. Characterization of soil organic matter by near infrared spectroscopy -
Vinzi, V., Chin, W.W., Henseler, J., et al., (Eds.), 2010. Handbook of Partial Least determination of glomalin in different soils. Soil Discuss. http://dx.doi.org/
Squares. ISBN 978-3-540-32825-4. 10.5194/soil-2016-9.
Viscarra-Rossel, R.A., Hicks, W.S., 2015. Soil organic carbon and its fractions estimated Zheng, G., Jiao, C., Zhou, S., Shang, G., 2016. Analysis of soil chronosequence studies
by visible–near infrared transfer functions. Euro. J. Soil Sci. 66, 438–450. using reflectance spectroscopy. Int. J. Remote Sens. 37 (8), 1881–1901.