146
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
with respect to the proper interpretation of Landsat TM/ETM+ images and validated
with ground-truth data. Regional-scale heat fluxes were estimated with the aid of
remote sensing images and the surface energy balance algorithm for land (SEBAL)
model (Bastiaanssen et al. 1998a,b). All preparatory efforts led to the development
of TDVI and RWSI for final analysis in the context of drought monitoring. The following sections will introduce these algorithms and equations in a greater detail.
7.2.3 RetRieval of lulc PatteRnS
After a plethora of investigations of synergistic potential with regard to the use of
multisource and multisensor data, this study applies the four-, three-, and two-band
false-color composites of Landsat satellite data to characterize the LULC of the study
area. With the aid of high-resolution remote sensing images collected on May 7, 1987
(Landsat TM) and May 2, 2000 (Landsat ETM+), LULC classes were extracted
based on computer-aided manual interpretation. The projection was made possible
based on WGS_1984_UTM_Zone_50N. A large set of pixels was used for training
to optimize the representation of environmental heterogeneity in this coastal bend.
Classification accuracy was assessed with spectral and field-checked error matrices
over different categories of LULC patterns.
7.2.4 RetRieval of land SuRface heat fluxeS
With Landsat satellite images, the heat fluxes (Rn, Gn, H, and LE) in this study
were estimated by the traditional surface heat balance equations in the SEBAL
model (Bastiaanssen et al. 1998a,b). The SEBAL model can assimilate multisource
and multisensor information to estimate land water and heat fluxes, making use of
the regional advantage of remote sensing technology. These equations are based
on the theory that incoming net solar radiation drives all energy exchanges on the
earth’s surface and can be expressed as a surface energy balance equation as follows
(Bastiaanssen et al. 1998a):
R n = H + G n + LE,
(7.1)
where R n is the net radiation flux (in watts per square meter), G n is the soil heat flux
(in watts per square meter), H is the sensible heat flux (in watts per square meter), and
LE is the latent heat flux (in watts per square meter). As long as the values of R n , G n ,
and H are known, the LE value can be calculated to obtain ET.
In Equation 7.1, net radiation is the summation of soil heat flux, sensible heat
flux, and latent heat flux, as indicated, and can be calculated on the basis of the land
surface radiation as follows:
R
R
T T
n
s
s
a a
s
= −
+
−
↓
(
)
(
),
1
4
4
α
ε σ ε
(7.2)
where R s↓ is the incident solar shortwave radiation, also known as the total solar
radiation (in watts per square meter); α is the surface albedo (in percent); ε s is the
surface emissivity (dimensionless); σ is the Stefan–Boltzmann constant (5.6696
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
with respect to the proper interpretation of Landsat TM/ETM+ images and validated
with ground-truth data. Regional-scale heat fluxes were estimated with the aid of
remote sensing images and the surface energy balance algorithm for land (SEBAL)
model (Bastiaanssen et al. 1998a,b). All preparatory efforts led to the development
of TDVI and RWSI for final analysis in the context of drought monitoring. The following sections will introduce these algorithms and equations in a greater detail.
7.2.3 RetRieval of lulc PatteRnS
After a plethora of investigations of synergistic potential with regard to the use of
multisource and multisensor data, this study applies the four-, three-, and two-band
false-color composites of Landsat satellite data to characterize the LULC of the study
area. With the aid of high-resolution remote sensing images collected on May 7, 1987
(Landsat TM) and May 2, 2000 (Landsat ETM+), LULC classes were extracted
based on computer-aided manual interpretation. The projection was made possible
based on WGS_1984_UTM_Zone_50N. A large set of pixels was used for training
to optimize the representation of environmental heterogeneity in this coastal bend.
Classification accuracy was assessed with spectral and field-checked error matrices
over different categories of LULC patterns.
7.2.4 RetRieval of land SuRface heat fluxeS
With Landsat satellite images, the heat fluxes (Rn, Gn, H, and LE) in this study
were estimated by the traditional surface heat balance equations in the SEBAL
model (Bastiaanssen et al. 1998a,b). The SEBAL model can assimilate multisource
and multisensor information to estimate land water and heat fluxes, making use of
the regional advantage of remote sensing technology. These equations are based
on the theory that incoming net solar radiation drives all energy exchanges on the
earth’s surface and can be expressed as a surface energy balance equation as follows
(Bastiaanssen et al. 1998a):
R n = H + G n + LE,
(7.1)
where R n is the net radiation flux (in watts per square meter), G n is the soil heat flux
(in watts per square meter), H is the sensible heat flux (in watts per square meter), and
LE is the latent heat flux (in watts per square meter). As long as the values of R n , G n ,
and H are known, the LE value can be calculated to obtain ET.
In Equation 7.1, net radiation is the summation of soil heat flux, sensible heat
flux, and latent heat flux, as indicated, and can be calculated on the basis of the land
surface radiation as follows:
R
R
T T
n
s
s
a a
s
= −
+
−
↓
(
)
(
),
1
4
4
α
ε σ ε
(7.2)
where R s↓ is the incident solar shortwave radiation, also known as the total solar
radiation (in watts per square meter); α is the surface albedo (in percent); ε s is the
surface emissivity (dimensionless); σ is the Stefan–Boltzmann constant (5.6696