ASSESSMENT OF ADEQUACY OF DEM LEVEL-1 AS A

SUBSTITUTE FOR DEM LEVEL-2 FOR RADIOMETRIC

CORRECTION OF SATELLITE IMAGERY FOR NATIONAL LAND
COVER MAPPING OF SAUDI ARABIA
A. I. ALOMRAN § and M. J. MCCULLAGH Ф
§ General Commission for Survey, Riyadh, Saudi Arabia
Phone number: +966-555205816
Email: alomranali@hotmail.com
Ф School of Geography, University of Nottingham, Nottingham, UK.
Phone number: +44 (0)1332-874103
Email: michael.mccullagh@nottingham.ac.uk

Abstract
The Kingdom of Saudi Arabia has not yet had a full (national) land cover
mapping. The very limited areas in Saudi Arabia covered by the 30m resolution
DEM level-2 that is compatible with the fine and medium resolution (SPOT and
TM) imagery makes it unreliable as a topographic source. The validity of the
lower spatial resolution (100m) DEM level-1 of full coverage of the Kingdom as
a substitute is tested for that purpose. Radiometric (atmospheric and
topographic) correction of satellite imagery for the desert bare soil study area is
achieved by using Radiance and two Reflectance (Chavez (1996) COST and
Radiative Transfer Code in ATCOR-3) based techniques, implementing four
Lambertian and non-Lambertian topographic correction models. The results
show that the adequacy of DEM level-1 decreases with increase of terrain slope
and DEM level-1 can be an adequate alternative to DEM level-2 for areas of flat
to gently sloping (0º to 5º). The optimum performance of DEM level-1 will be on
sun-facing slopes in phase angles 46º to 90º and its worst performance in the
sun-facing-away slopes in phase angles 136º to 180º. Moreover, for better
performance of topographic correction using DEM level-1 for high sun angle
images in flat to moderate terrain (0º to 25º), C-correction should be used as
the optimum model, followed by Minnaert and last by Cosine. But for rugged
terrain (steeper than 25º), the Cosine should be used, followed by C-correction
and last by Minnaert. Using low sun angle imagery in flat to gentle terrain (0º to
5º), C-correction is the optimum, Minnaert comes as second and Cosine is the
last. For moderate to rugged terrain (steeper than 5º), C-correction should be
used, followed by Cosine and last by Minnaert. Results also demonstrates
preference of using fine resolution and high sun angle imagery with DEM level-
1.

1. Introduction
The Kingdom of Saudi Arabia (about two million square kilometres area) has
not yet had a full (national) land cover mapping survey. To perform accurate
land cover classification, satellite imagery must be radiometrically corrected for
atmospheric and topographic effects. Correction for those effects requires an
accurate DEM that is compatible with the satellite image resolution, such that

1
derived slope and aspect will match image pixel in size and location. The very
limited areas in Saudi Arabia covered by the 30m resolution DEM level-2 that is
compatible with the fine and medium resolution (SPOT and TM) imagery
received by King Abdulaziz City for Science and Technology (KACST) in
Riyadh makes it unreliable as a topographic source. DEM level-1 of full
coverage of the Kingdom as a substitute makes testing its validity for that
purpose essential. If DEM level-1 is substitutable then land cover mapping for
the Kingdom using DEM level-1 instead of level-2 for radiometric correction can
be carried out on an immediate operational basis. To reach this goal, two
questions are to be answered. The first is to what extent in terms of terrain
slope and orientation is DEM level-1 an adequate alternative to DEM level-2 in
radiometric correction and consequently in land cover mapping? The second is
what is the optimum combination of temporal (sun angle) selections, sensor
images and radiometric correction techniques that leads to the best
performance of DEM level-1? Thus, performance of DEM level-1 against level-2
is tested under the effect of the variable image sun angles and resolutions:
SPOT-XI (representing finer resolutions and higher sun elevation angles), TM8-
5 (representing coarser resolutions and higher sun angles) and TM16-1
(representing coarser resolutions and lower sun angles) acquired on 14-5-2001,
8-5-2001 and 16-1-2001, respectively, at sun zenith angles at the image
acquisition times of 17.4º, 26º and 54º, respectively, and at sun azimuth angles
of 118.7º, 102º, and 145º, respectively. For same purpose, finding the optimum
techniques for radiometric correction of atmospheric and topographic effects is
investigated by comparing three different atmospheric correction based
techniques: Radiance, Chavez (1996) COST simplified reflectance and
ATCOR-3 accurate reflectance compiled using Radiative Transfer Code (RTC),
and four different simplified topographic and anisotropic correction methods:
Lambert (Cosine) and non-Lambert Minnaert, C-correction and Cicone and
Malila’s (1972) Modified Lambert.

2. Radiometric Correction Techniques used in this Study

Before radiometric (atmospheric and topographic) corrections, all satellite
images involved in this study were ortho-rectified. Slope and aspect values
used for topographic correction were derived from the DEM level-1 and level-2
using Erdas Imagine© software (its algorithm is similar to the widely used
Sharpnak and Akin (1969) and Evans (1980) algorithms). Programs were
written by the authors to perform all the radiometric corrections of image data,
except for ATCOR3 Lambert correction. Minnaert and C topographic correction
of the satellite images were performed using single (global) K and C
coefficients.

2.1 Radiance based radiometric correction

The atmospheric effect was corrected by removing path radiance from apparent
(i.e., at satellite) radiance using the Chavez (1988) improved Dark Object
Subtraction (DOS) method. The main disadvantages of the improved DOS
method are its low accuracy and its correction only for additive scattering
effects (due to path radiance) and the assumption of full upward and downward
transmission.

2
The Lambertian (Cosine) topographic correction model has the following
formula:
LT
Ln =
cos( i )
where cos(i) is the cosine of the incidence angle between the sun and the
normal vector to the surface, calculated as follows:
cos(i) = cos(s) cos(n) + sin(s) sin(n) cos(Φn - Φs)
where i is illumination (incidence) angle; s is the sun zenith angle; n is terrain
slope angle; Φn is terrain aspect angle; and Φs is the sun azimuth angle;
Ln is the normalised radiance (i.e., corrected for topographic effect) that would
be measured when i = e (exitance angle) = 0; and LT is the radiance at tilted
surface (i.e., uncorrected).
Developed by Smith et al. (1980), the Backward Radiance Correction
Transformation (BRCT) employing the Minnaert law (Minnaert, 1941) can be
described as:
LT
Ln =
cos ( i ) cos k‫ـ‬1 ( e )
k

where k is the Minnaert coefficient, and e is terrain slope angle.

In similar fashion to Minnaert, the C-correction is used for topographic and
anisotropic correction. The formula of this model is:
cos( θ s ) + C
Ln = LT ( )
cos( i ) + C
where θ s is the sun zenith angle.
The Modified Lambertian model (Cicone and Malila, 1977) can be described as:
LT
Ln =
FRI
where FRI is the relative inolation factor and can be computed as follows:
FRI = tan( θ s ) sin( e ) cos( Φ ) + cos( e )
Where Φ is the phase angle (difference between sun azimuth and terrain
aspect).

2.2 Simplified reflectance based radiometric correction

This technique, which is supposedly correct for atmospheric effects more
accurately than the previously discussed radiance based technique, is
implemented using the Chavez (1996) COST model, in which atmospheric
downward transmittance is approximated by the cosine of the sun zenith angle,
and upward transmittance is assumed 1.0 (full transmittance).

3
The Lambertian topographically corrected surface reflectance based on Chavez
(1996) technique is expressed as:
πd 2 ( Ls L‫ـ‬ p )
ρLamb =
E o cos( i ) cos( θ s )
Where ρLamb is the target surface reflectance under the assumption of surface
Lambertian behavior; L p is the path radiance for TM and SPOT images
computed with an assignment of one-percent (0.01) reflectance to the dark
features (Song et al., 2001); E o is the sun exo-atmospheric irradiance; d is the
relative earth-sun distance to the mean distance in astronomical units at the
image acquisition day; and cos( θ s ) is the cosine of sun zenith angle
representing downward transmittance.
Similarly, the Minnaert correction model integrated into the Chavez (1996)
COST reflectance model is:
πd 2 ( Lsat L‫ـ‬ path ) cos( e )
ρMinnaert = ;
E o cos( θ s )[cos( i ) cos( e )] k
and the C-correction reflectance formula is expressed as follows:
πd 2 ( Lsat L‫ـ‬ path )[cos( θ s ) + C ]
ρC =
E o cos( θ s )[cos( i ) + C ]
The reflectance formula of the Modified Lambertian model will be similar to the
Lambertian formula, except that the incidence angle is replaced by FRI :

πd 2 ( Lsat L‫ـ‬ path )

ρmod =
E o FRI cos( θ s )

2.3 Accurate reflectance based radiometric correction

The third technique to radiometric correction used in this study is assumed to
be the most accurate technique in atmospheric correction, for that it uses
Radiative Transfer Code (RTC). This technique is implemented using the
ATCOR-3 program for atmospheric (compiled using the MODTRAN-4) and
topographic correction of rugged terrain, developed by the DLR-German
Aerospace Centre and integrated in Erdas Imagine© by Geosystems© Gmbh.
ATCOR-3 was used to estimate visibility, adjusted sensor calibration
coefficients and aerosol type and standard atmosphere and to perform the
atmospheric and topographic correction for the three satellite images used in
this study.
The three images were corrected for topographic effects under the Lambertian
assumption using ATCOR-3 and with Minnaert model implemented for ATCOR-
3 by the present authors using the formula:

4
 ρLamb
ρMinnaert =
cos ( i ) cos k‫ـ‬1 ( e )
k‫ـ‬1

where ρLamb is the output Lambertian reflectance from ATCOR-3.

3. Study area
The study area is 10 km by 10 km, located in the central part of Saudi Arabia in
an area between two small towns: Malham and Huraimla, 65 km to the
Northwest of Riyadh, the capital city. It is composed of desert bare soil of gentle
slope on top of plateau ruptured by areas of rock protrusions in steep to semi-
orthogonal rock outcrops, and torn by wide waterways (wadis) and small gullies
(figure 1). This study area was chosen to represent the overwhelming majority
of the Kingdom's terrains covered with same desert bare soil.

(a) (b)

Figure 1. The study area. (a) A picture of the desert bare soil covering the study area.
(b) A picture of one side of one of the waterways (Wadis) tearing the study area
plateau(same small Jeep in the waterway bed is illustrated for scale purpose.

4. Results and Analysis

4.1 Visual results and analysis
DEM level-1 corrected images using the four topographic correction models
have shown less shadow reduction and more artifacts compared to those of
DEM level-2. This finding is applicable to both coarse resolution (e.g. TM) and
finer resolution (e.g. SPOT). Figure 2, as an example, explains this finding by
comparing DEM level-2 corrected TM images with those corrected using DEM
level-1. Shadow effect and artifact are more pronounced in the DEM level-1
corrected images, due to the coarser resolution of level-1 DEM and its failure to
detect wadis (valleys) short sides. This indicates initial less support to the idea
of DEM level-1 being an adequate alternative to DEM level-2.

5
Figure 2. Shadow effect reduction and artifact appearance after topographic correction
using DEM level-1 and level-2. (a) the uncorrected low sun angle image (TM16-1) of
the study area. (b) and (c) same image corrected with Minnaert model using DEM
level-1 and level-2, respectively. (d) a subset (the red square in "a") of the uncorrected
high sun angle image (TM8-5). (e) and (f) same subset image corrected with "C"
model using DEM level-1 and level-2, respectively.

4.2 Semi-variogram results and analysis

Researchers (e.g. Bishop and Shroder, 2000; Bishop et al., 2003) have
conducted semi-variogram analysis to measure topographic complexity and
topographic correction. As a part of the process of evaluating the performance
of DEM level-1 against level-2 for topographic correction, semi-variogram
analysis was conducted on the two DEMs corrected radiance for a selected
transect of a length of about 1km (30 TM pixels) crossing a 500m wide wadi in
the study area.
The semi-variance, γ(h), measures the variance between points at successively
greater distances. Pixels (point values) at small lags will have lower semi-
variance than at greater lags. Smi-variance can be calculated as (Treitz and
Howarth, 2000; Bishop et al., 2003):
1 m( h )
γ( h ) = [ Z( x i ) Z‫ ( ـ‬x i + h )] 2
2 m( h ) i =1
where γ ( h ) is the average semi-variance of several pixel pairs at each lag;
Z( x i ) is the value of the variable to be tested (e.g., DN, radiance) at position
x i ; Z( x i +h) is the value of the variable at lag distance h from x i ; and m (h) is
the number of data pairs separated by same lag h.
Figure 3 illustrates the resulting semi-variograms for the Green bands of TM8-5
and TM16-1 uncorrected and corrected (with C model) radiance using the two
DEMs. The substantial decrease of semi-variance in both DEMs corrected
radiance of TM8-5 compared with those of uncorrected radiance and the small

6
differences in semi-variance between DEM level-1 and level-2 corrected
radiance indicates the effectiveness of either DEMs and the adequacy of DEM
level-1 as an alternative to level-2 for topographic correction of high sun angle
images. However, the higher semi-variance for the TM16-1 corrected radiance
using either DEMs, especially DEM level-1 compared with those for the TM8-5
indicates the low efficiency of topographic correction for low sun angle images,
and consequently a preference to use high sun angle images instead. In
contrast with the high sun angle image, the considerable differences in semi-
variance between DEM level-1 and level-2 corrected radiance for the low sun
angle TM16-1 image and at the same time the close semi-variance values of
uncorrected radiance to those of DEM level-1 corrected radiance indicates the
inadequacy of DEM level-1 as an alternative to level-2 for topographic
correction of low sun angle images.

Figure 3. Semi-variograms of the TM8-5 and TM16-1 uncorrected and corrected

radiance with the two DEMs, calculated at lag distance increments of TM pixel size
(30m).

4.3 Effect of terrain slope and phase angle on performance of DEM level-1
against level-2 for radiometric correction
Number of studies (e.g. Cicone and Malila, 1977; Stohr and West, 1985;
Thomson and Jones, 1990) have demonstrated the effect of terrain slope and
aspect and phase angle (the difference between terrain and sun azimuths, on
satellite image data, and others (e.g. Smith et al. 1980; Justice et al., 1981;
Chen et al., 2001; Falkenstrum and Ekstrand, 2002) consequently on
topographic and anisotropic corrections using techniques such as, Cosine,
Minnaert, C-Correction, etc.
As the effect of terrain aspect will vary from one image to another according to
the image sun azimuth angle, the effect of phase is investigated instead in this
study.

7
Most earth surfaces reflects unequally in all directions (anisotropy). Surface
anisotropy (BRDF effects), generally speaking, is strongly manifested in
backward and forward scattering, with the maximum in backward scattering.
Considering desert areas, however, some studies (e.g. Holben and Justice,
1980; Takemata et al., 2000) have suggested Lambertian behaviour whereas
others (e.g. Shoshany, 1993; Karnieli and Cierniewski, 2001) have suggested
anisotropic (BRDF) effects. Hence, correcting for topography and BRDF effects
in this study has involved Lambertian and simple empirical BRDF (non-
Lambertian) models, such as Minnaert and the semi-empirical C-correction.
Owing to the major effects of terrain slope and phase angle on satellite image
data and topographic correction mentioned above, performance of DEM level-1
against level-2 in topographic correction has been investigated thoroughly in
this study as described below. This investigation includes the effect of terrain
slope and phase angle in determining terrain slope and phase angle (aspect)
limitations, optimum topographic correction models, optimum sun elevation
angles and resolutions, and optimum radiometric (atmospheric) correction
techniques for optimum performance of DEM level-1 compared to level-2.

4.3.1 Effect of terrain slope on performance of DEM level-1 against level-

2 for radiometric correction
4.3.1.1 Effect of terrain slope on correlation between the two DEMs
radiometrically corrected data
Performance of DEM level-1 against level-2 for radiometric (atmospheric and
topographic) correction is measured here by the degree of correlation between
DEM level-1 and level-2 corrected radiance images. Higher correlation indicates
a greater likelihood of DEM level-1 to replace level-2. Hence, only the high
correlation values (r>0.5) were considered. For that purpose, four monochromic
(0-255; representing correlation value range of 0-1.0) images of correlation
between DEM level-1 and level-2 corrected (using Minnaert, C, Modified
Lambertian and Cosine models) TM8-5 NIR were produced using MIPS
software (Mather, 1999). To investigate the effect of terrain slope on
performance of DEM level-1 against level-2, the study area slopes were
separated into four classes: 0º-5º (flat to gentle terrain), 6º-15º (moderate
terrain), 16º-25º (steep terrain) and 25º+ (very steep terrain). DEM level-2 slope
image was used as a source for its higher accuracy compared with that of level-
1. For that purpose, Spatial Analyst Model Builder of ArcView was used to
perform the slope and correlation images classifications and the arithmetic
overlay analysis. Figure 4 illustrates the results. Performance of DEM level-1
against level-2 is expressed by what we called " Relative Area Covered
Percentage-RACP", which is the number of pixels of high correlation (i.e., r>0.5
or grey values of 127 to 255) divided by the total number of pixels in that slope
class. This is to provide unbiased results to slope classes occupying larger
area. Higher RACP indicates higher performance of DEM level-1 against level-
2. As shown in figure 4, sharp decrease in RACP starts after the first terrain
slope class 0º-5º (a drop of about 40%), then decreases asymptotically with
increase of slope. This indicates that adequacy of DEM level-1 decreases with
increase in terrain slope and DEM level-1 can not be an adequate alternative to

8
DEM level-2 for topographic correction of areas with terrain slope higher than
5º. Figure 4 also shows that the differences between the four topographic
correction models are unsubstantial, except possibly for the Modified
Lambertian model for slopes higher than 5º.

Figure 4. Effect of terrain slope on the high correlation (r>0.5) between DEM level-1
and level-2 corrected radiance of NIR of TM8-5 for the four topographic correction
models. Performance of DEM level-1 against level-2 represented by Relative Area
Covered Percentage-RACP of high correlation values in the four slope classes.

4.3.1.2 Effect of terrain slope on classification accuracy using the two

DEMs radiometrically corrected data
The effectiveness of radiometric correction in improving classification accuracy
is still arguable in literature. Some researchers (e.g. Itten and Meyer 1993;
Tokola et al., 2001; Gitas and Devereux, 2006) have reported noticeable
success in improving classification by radiometric correction, others (e.g. Teillet
et. al., 1982; Song et al., 2001), however, have reported otherwise.
The DEM level-1 and level-2 radiometrically corrected images were
unsupervised classified (using Erdas© ISODATA) into six spectral classes. This
number of classes was chosen arbitrarily, owing to the limited number of desert
bare soil types of study area and to that no limitation for the number of spectral
classes that can be derived spectrally using unsupervised classification.

9
 Figure 5. Effect of terrain slope on classification accuracy of DEM level-1 classified
radiometrically corrected high (TM8-5 and SPOT XI) and low (TM16-1) sun elevation
angle images (with reference to their corresponding DEM level-2 classified images)
with the four topographic correction models using the three radiometric correction
techniques. (a) Radiance results. (b) Chavez (1996) results. (c) ATCOR-3 results.
The effect of terrain slope on performance of using DEM level-1 instead of
level-2 in classification is investigated for low and high sun elevation angles
(TM8-5 and TM16-1, respectively), for the three level of atmospheric correction
accuracies (Radiance, Chavez (1996) and ATCOR-3 techniques) and for the

10
four topographic correction models. Each of DEM level-1 and level-2 classified
images was separated into four images based on the four slope classes (0º-5º,
6º-15º, 16º-25º and >25º) using programs written by the authors. Other
programs were also developed to calculate the Khat values of the error
matrices, produced from comparing DEM level-1 classified images with those of
level-2, as a reference, for the four slope classes. Classified DEM level-2
radiometrically corrected images were used in the error matrices as reference
for the DEM level-1 corrected images due to the absence of ground truth (i.e.,
reference data) and to the fact that the assessment of DEM level-1
performance is relative to that of level-2. Khat (estimate of Kappa) is adopted
against Overal Accuracy due to the fact that it is more informative, as it
incorporates omission and commission errors (Congalton and Green, 1999).
Performance of DEM level-1 against level-2 is evaluated through the
relationship between classification accuracy (Khat values) of DEM level-1
images (compared with those of level-2 as a reference) and terrain slope.
Figures 5(a), (b) and (c) illustrate the relationship of Khat values for the four
topographic correction models with terrain slope for Radiance, Chavez (1996)
and ATCOR-3, respectively. A similar general trend to that found in the
previously discussed correlation test (section 4.3.1.1) is also found here. The
rate of decrease in Khat values for the high sun angle images TM 8-5 and
SPOT-XI is rapid from about 80% for slope class 0º-5º to about 20% for slope
class >25º. However, the rapid decrease slows by the slope class 6º-15º for the
low sun angle image (TM16-1), then decreases moderately until slope class
>25º. The continuous sharp drop of Khat values (classification accuracy) after
slope class 0º-5º in the high sun angle images and the low Khat values in the
low sun angle image beyond slopes 0º-5º indicates limitation of DEM level-1
adequacy to replace DEM level-2 to flat to gentle terrain (0º-5º) only.
In figure 5 and based on the high sun angle SPOT-XI image using the
Radiance technique, with the exception of Modified Lambertian model, C is the
optimum model at all slope classes, followed by Minnaert and last by Cosine.
Similar results apply for the other high sun angle TM8-5 image, but Cosine
replaces C as optimum at the slope class (>25º). The higher performance of the
Modified Lambertian model compared with other models in high sun angle
images and its comparable performance with other models in low sun angle
image TM16-1 indicates to its sensitivity to sun angle rather than to topographic
effect (slope and aspect), which lowers its reliability. The Chavez (1996)
technique shows similar results for SPOT-XI, except Cosine replacing C as
optimum at the slope class (>25º). TM 8-5 is similar to SPOT-XI, except that C
is the optimum only up to 15º, whereas Cosine is the optimum above 15º. For
ATCOR-3 where only Minnaert and Cosine models are used, model ranking is
similar to that of Chavez (1996), such that Minnaert is the optimum up to 25º
and Cosine is the optimum above. For the low sun angle image (TM16-1), and
based on Radiance and Chavez (1996) results, C is the optimum model,
followed by Minnaert and least by Cosine in the flat to gentle terrain (0º to 5º).
In moderate to rugged terrain (i.e., slopes higher than 5º), C is optimum,
followed by Cosine and last by Minnaert. Using ATCOR-3, model ranking is
similar to those for Radiance and Chavez (1996), such that Minnaert is the

11
optimum below 5º and Cosine above 5º. Thus, for optimum performance of
DEM level-1, with few exceptions, if high sun angle images are used in flat to
moderate terrain (slopes of 0º to 25º) using any radiometric (atmospheric)
correction technique, one should use C as the optimum model, followed by
Minnaert and last by Cosine. However, if the terrain is rugged (slopes higher
than 25º), one should use Cosine, followed by C and last by Minnaert. For
using low sun angle imagery in flat to gentle terrain (slope of 0º to 5º), C is the
optimum, Minnaert comes as second and Cosine is the last. For moderate to
rugged terrain (slopes higher than 5º), one should use C, followed by Cosine
and last by Minnaert, whose performance with DEM level-1 deteriorates with
the increase of terrain slope.
For optimum sun angle, figure 5 shows that the performance of DEM level-1
against level-2 in TM8-5 is better (higher Khat values) than in TM16-1 for flat to
moderate slopes (i.e., 0º-15º), then the differences decrease with increase of
slope until for slopes higher than 25º where performance in TM16-1 is better
than that for TM8-5 using C and Cosine models in Radiance technique (figure
5a) and C in Chavez (1996) technique (figure 5b). The better performance of
DEM level-1 for TM16-1 in this slope class using Minnaert and Cosine in
ATCOR-3 confirms this (figure 5c). Thus, as found before, for better
performance of DEM level-1 with any topographic correction model using any
radiometric (atmospheric) correction technique, high sun angle images should
be used instead of low sun angle images. However and oddly, low sun angle
images are better used in areas with terrain slopes higher than 25º. The reason
is not clear to the authors and worthy future investigation.
For optimum image resolution, figure 6 shows that the performance of DEM
level-1 in association with the fine resolution SPOT-XI image is better than that
with coarser resolution TM8-5 image, except when using the Cosine model for
slopes between 16º and 25º for Chavez (1996) and ATCOR-3 techniques. This
exception can be ignored owing to the insignificant difference between the two
images (Khat = 23.0% compared with 23.1% for Chavez (1996)) and to the
many findings of weakness of the Cosine with DEM level-1. Hence, it can be
said that for better performance of DEM level-1, one should use finer resolution
(e.g. SPOT-XI) instead of coarser resolution (e.g. TM8-5) images for all terrain
types. Differences in DEM level-1 performance (i.e., Khat values) between the
two images, however, get smaller with the increase of terrain slope.
For optimum radiometric correction technique, figure 6 illustrates that the
differences in DEM level-1 performance between the three radiometric
correction techniques for the high sun angle image TM8-5 decrease with
increase of slope until slopes higher than 25º, after which the performances are
almost identical. For the low sun angle image TM16-1, the differences increase
continuously with the increase of slope for the favour of ATCOR-3. The reason
for this for the hazy (visibility of 12km) TM8-5 may be the increase of
dominance of atmospheric effect compared to topographic effect with the
decrease of slope in flatter slopes, which lead to higher differences in
atmospheric correction between the three techniques for the favour of
Radiance and Chavez (1996) compared with ATCOR-3. The dominance of
topographic effect in steeper slopes makes the differences smaller due to the

12
use of marginally similar topographic correction methods by the three
radiometric correction techniques. For the clearer sky (visibility of 25km) TM16-
1, the explanation may be the accurate estimation of terrain contribution in
radiation by ATCOR-3 (e.g. adjacency effect and terrain radiation) in the
absence of or very slight atmospheric effect in a clear sky compared with other
two techniques that do not consider them.

Figure 6. Performance of DEM level-1 in classification (compared with DEM level-2 as

a reference) under the effect of terrain slope using the three radiometric correction
techniques. Only Minnaert and cosine corrections are included, as they are the only

13
two corrections implemented in ATCOR-3. (a) SPOT-XI results. (b) TM8-5 results. (c)
TM16-1 results.
From the above, it can be stated that more accurate atmospheric correction
using Radiative Transfer Code-RTC (ATCOR-3) will not add substantial
improvement in performance of DEM level-1 for rugged terrain (slopes higher
than 25º) for high sun angle high atmospheric effect images (like TM-8-5
image). Despite that, the accurate atmospheric correction technique ATCOR-3,
which demonstrates the true but the weaker performance of DEM level-1, is
preferred to be used, owing to that the apparent better DEM level-1
performances in the Radiance and Chavez (1996) are due to their poorer
atmospheric effect removal and lower topographic effect influence. The
performance of DEM lvel-1 compared with level-2 for all three techniques for
the clear sky low sun elevation angle TM16-1 image is almost identical for flat
to gentle terrain (slopes of 0º to 5º), and as slope increases, its performance
using ATCOR-3 improves compared with the other two techniques. Moreover
and for same image, performance of DEM level-1 using Radiance and ATCOR-
3 techniques is considerably better than that for Chavez (1996).
Despite its higher accuracy and fidelity, using the RTC based ATCOR-3
atmospheric correction technique for operational use for the anticipated
national land cover mapping of the Kingdom is questionable, considering the
uncertainties involved in estimation of the atmospheric parameters (especially
for aerosols), the cost and practicality of collecting required information about
atmospheric condition for uninhabited areas.

4.3.2 Effect of phase angle on performance of DEM level-1 against level-2

for radiometric correction
4.3.2.1 Effect of phase angle on correlation between the two DEMs
radiometrically corrected data
The effect of phase angle on the high correlation (i.e., r > 0.5) between DEM
level-1 and level-2 corrected (using "C" model) radiance of the NIR band of the
high sun angle TM8-5 image data was investigated in a similar fashion to that
made for terrain slope effect (section 4.3.1.1).
In figure 7, RACP values of TM8-5 for all four topographic correction models
start low, increase to their highest values (optimum performance of DEM level-
1) in phase class 46º-90º, then they decrease with slopes facing away from the
sun to lowest values in phase class 136º-180º. Thus, the optimum performance
of DEM level-1 (highest RACP) for high sun angle images is expected to be
found in pixels located on sun-facing slope with phase angles 46º-90º, whereas
worst performance will be in sun-facing-away slopes at 136º-180º. Similar to
terrain slope results, the differences between the four topographic correction
models are unsubstantial (figure 7).

14
 Figure 7. Effect of phase angle on the high correlation (r>0.5) between DEM level-1
and level-2 corrected radiance of NIR band of TM8-5 for the four topographic
correction models. Higher RACP value indicates higher performance of DEM level-1.

4.3.2.2 Effect of phase angle on classification accuracy using the two

DEMs radiometrically corrected data
A similar procedure was used to that implemented for assessment of terrain
slope effect on classification accuracy of classified images (section 4.3.1.2),
except that phase angle was investigated instead of slope.
In figures 8(a), (b) and (c), relation of classification accuracy (Khat values) with
phase angle for the three radiometric correction techniques follows an identical
pattern to that found in previous experiment where Khat increases with increase
of phase angle until the optimum phase class 46º-90º, then decreases with the
increase of phase angle to lowest values in phase angles 136º-180º. The only
exception is when correcting SPOT-XI with Minnaert where Khat values at
phase class 91º-135º are just little higher than those at 46º-90º class, but can
be ignored. Thus, Optimum performance of DEM level-1 for classification is
expected with phase angles 46º to 90º.
For optimum topographic correction model, figure 8(a) based on SPOT-XI using
Radiance technique shows that C is the optimum model at all phase angle
classes, followed by Minnaert and last by Cosine, except at 0º-45º where it
performed equally to C and Minnaert. Performance of Cosine worsens as
terrain faces away from the sun. Similar results apply for TM8-5. The higher
performance of the Modified Lambertian model compared with other models, as
discussed earlier, is due to its sensitivity to sun angle rather than to
topographic effect. For the low sun angle image (TM16-1), C is also the
optimum, followed by Minnaert, Modified Lambertian and last by Cosine.
Modified Lambertian exceptionally performed better than C at 0º-45º. The
Chavez (1996) technique in figure 8(b) shows similar results, except that the

15
Modified Lambertian model performed equally to C at 46º-90º. For ATCOR-3
where only Minnaert and Cosine models are used, model ranking is similar to
previous results where Minnaert performed better than the Cosine, except with
TM16-1 at 0º-45º. Thus, same conclusion derived in the terrain slope results
can be adopted here where by using any radiometric correction technique and
any image resolution and sun angle, one should use C as the optimum model,
followed by Minnaert and last by Cosine.

Figure 8 Effect of phase angle on classification accuracy of DEM level-1 classified

radiometrically corrected high (TM8-5 and SPOT XI) and low (TM16-1) sun angle
images using the four correction models using the three radiometric correction
techniqueses with reference to their corresponding level-2 classified images.

16
Classification accuracy is represented by Khat values. (a) Radiance based results. (b)
Chavez (1996) based results. (c) ATCOR-3 based results.
For optimum sun angle, figure 8 shows that differences in Khat values between
TM 8-5 and TM16-1 images are significant, which strongly recommend the use
high sun angle images for better performance of DEM level-1 data.

Figure 9 Performance of DEM level-1 in classification (compared with DEM level-2 as a

reference) under the effect phase angle using the three radiometric correction
techniques. Only Minnaert and cosine corrections are included, as they are the only
two corrections implemented in ATCOR-3. (a) SPOT-XI results. (b) TM8-5 results. (c)
TM16-1 results.

17
For optimum radiometric correction technique, figure 9 shows that the Khat
values of DEM level-1 for the high sun angle images using the Radiance
technique are highest, followed by Chavez (1996), and last by ACTOR-3. Once
again as discussed earlier, this should not mislead one to the extent of relying
on the two simplified techniques, as the more accurate atmospheric effect
removal using ATCOR-3 reveals the poor performance of DEM level-1.
Therefore, If simplified atmospheric correction techniques are used with high
sun angle images containing high atmospheric effects (e.g. TM8-5 and SPOT-
XI), the performance of DEM level-1 using the Radiance technique will be
marginally better than Chavez (1996). However, if ATCOR-3 is used to remove
the atmospheric effects more accurately, substantially lower performance of
DEM level-1 compared with the other two techniques is expected and the
adequacy of DEM level-1 to replace level-2 is questionable.
Differences in Khat values in figure 9 between the three techniques for the low
sun angle image TM16-1 are small, especially in sun-facing-away slopes (less
than 5%). The small differences here compared with those found previously
with TM8-5 for instance would be due to the atmospheric effect for TM 8-5
image was greater than that for TM 16-1 image (visibility of 12 km and 25 km,
respectively), giving less atmospheric correction for TM16-1, especially with
ATCOR-3. Radiance and ATCOR-3 techniques using Minnaert have produced
almost identical Khat values, but these are higher than those for Chavez
(1996), especially in the first three phase classes. No clear trend has been
found using Cosine model. Thus, if correcting low sun angle low atmospheric
effect images (e.g. TM16-1), performance of DEM level-1 using one of the three
techniques will not add major improvement compared with the others.
Moreover, performance of DEM level-1 using Minnaert using the Radiance or
ATCOR-3 techniques will be better than using Chavez (1996).

5. Conclusions
The validity of using the 100m resolution DEM level-1 of full coverage as an a
adequate alternative to the 30m resolution DEM level-2 of very limited coverage
for radiometric correction of satellite imagery and consequently in land cover
mapping of the Kingdom of Saudi Arabia has been investigated in this study.
Visual results have revealed the inadequacy of DEM level-1 to act as an
alternative to level-2, but semi-variogram results have shown that DEM level-1
may be used for high sun elevation angle imagery only. Results based on the
effects of terrain slope and phase angle on performance of DEM level-1 against
level-2 for radiometric correction, which have been adopted for their more
reliably as they have thoroughly considered the effects of the sun elevation and
azimuth angles, terrain variables (slope and aspect) and image resolution, have
shown that that the adequacy of DEM level-1 as an alternative to DEM level-2
for radiometric correction decreases with the increase of terrain slope for both
high and low sun elevation angle imagery. The results have also shown that
DEM level-1 can be an adequate alternative to level-2 for areas of flat to gently
sloping (slopes of 0º to 5º). The optimum performance of DEM level-1 will be on
sun-facing slopes in phase angles 46º to 90º, and its worst performance in the
sun-facing-away slopes in phase angles 136º to180º.

18
The analysis of effects of terrain slope and phase angle on performance of
DEM level-1 against level-2 have revealed insignificant differences in DEM
level-1 performance between the four topographic correction models and
shown that if high sun elevation angle images are used in flat to moderate
terrain (slopes of 0º to 25º) using any atmospheric correction technique
(Radiance, Chavez (1996) COST simplified reflectance and ATCOR-3 accurate
reflectance compiled using Radiative Transfer Code-RTC), one should use C
as the optimum model, followed by Minnaert and last by Cosine. But for rugged
terrain (slopes higher than 25º), one should use Cosine, followed by C and last
by Minnaert. Using low sun elevation angle imagery in flat to gentle terrain
(slopes of 0º to 5º), C is the optimum, Minnaert comes as second and Cosine is
the last. For moderate to rugged terrain (slopes higher than 5º), one should use
C, followed by Cosine and last by Minnaert whose performance with DEM level-
1has been found to deteriorate with the increase of terrain steepness. The
Modified Lambertian model has been found more sensitive to sun elevation
angle rather than to topography (i.e., DEM information), which lowers its
reliability.
Results have also demonstrated the low efficiency in topographic correction of
low sun elevation angle imagery, and high sun elevation angle imagery should
be used for better DEM level-1 topographic correction. Oddly, performance of
DEM level-1 with low sun elevation angle imagery for terrain slopes higher than
25º has been found better than that with high sun elevation angle imagery. The
reason is not known to the authors and worthy future investigation. In addition,
for better DEM level-1 topographic correction, finer resolution imagery (e.g.
SPOT-XS) should be used instead of coarser resolution (e.g. TM).
Use of accurate atmospheric correction implementing the Radiative Ttransfer
Code in ATCOR-3 is recommended compared with the simplified models
employed in the Radiance and Chavez (1996) COST reflectance techniques,
for its greater fidelity in revealing the actual but weaker performance of DEM
level-1 compared with DEM level-2. However, for operational use for the
anticipated national land cover mapping of the Kingdom, the worth of using
accurate atmospheric correction is questionable, considering the uncertainties
involved in estimation of the atmospheric parameters (especially for aerosols),
the cost and practicality of collecting required information about atmospheric
condition for uninhabited areas. In case the use of the accurate atmospheric
correction technique in ATCOR-3 is not feasible, Radiance technique has more
potentiality compared with Chavez (1996) COST reflectance for better
performance of DEM level-1 radiometric correction.
The authors here see the importance of performing these types of
investigations on the new generation of high resolution satellite imagery, such
as SPOT-5, IKONOS, GeoEye, WorldView, etc., and see if this study derived
conclusions can still be generalised and adopted.

References
Bishop, M.P., and Shroder Jr ,J.F, 2000, Remote sensing and
geomorphometric assessment of topographic complexity and erosion dynamics

19
in the Nanga Parbat massif. In: Khan, M.A., Treloar, P.J., Searle, M.P. and Jan,
M.Q. (eds.) Tectonics of the Nanga Parbat Syntaxis and the Western Himalaya.
Geological Society, London, 181-200.
Bishop, M.P., Shroder Jr., J.F., and Colby, J. D., 2003, Remote sensing and
geomorphometry for studying relief production in high mountains.
Geomorphology, 55:345-361.
Chavez, P.S., 1988, An improved dark-object subtraction technique for
atmospheric scattering correction of multispectral data. Remote Sensing of
Environment, 24:459-479.
Chavez, P., 1996, Image-based atmospheric corrections-revisited and
improved. Photogrammetric Engineering and Remote Sensing, 62(9):1025-
1036.
Chen, F., Muramoto, K., and Kubo, M., 2001, Improved topographic correction
for satellite imagery. IEICE Transactions on Information and Systems, E84-D
12:1820-1827.
Cicone, R., and Malila, W., 1977, Investigation of techniques for inventorying
forested regions. Vol. II. Forestry Information System Requirements and Joint
Use of Remotely Sensed and Ancillary Data. Final Report NASA-CR-ERIM-
122700-35-F2.
Congalton, R.G., and Green, K., 1999, Assessing the Accuracy of Remotely
Sensed Data: Principles and Practices. Lewis Publishers, Boca Raton, 131p.
Evans, I.S., 1980, An integrated system for terrain analysis for slope mapping.
Zeitschrift fur Geomorphologie, 36:274-295.
Falkenstrom, H., and Ekstrand, S., 2002, Evaluation of IRS-1C LISS-3 satellite
data for defoliation assessment on Norway spruce and Scots pine. Remote
Sensing of Environment, 82:208-223.
Gitas I.Z., and Devereux, B.J., 2006, The role of topographic correction in
mapping recently burned Mediterranean forest areas from LANDSAT TM
images. International Journal of Remote Sensing, 27(1):41 - 54.
Holben, B., and Justice, C.O., 1980, The topographic effect on spectral
response from nadir-pointing sensors. Photogrammetric Engineering and
Remote Sensing, 46(9):1191-1200.
Itten, K.I, and Meyer, P., 1993, Geometric and radiometric correction of TM
data of mountainous forested areas. IEEE Transactions on Geoscience and
Remote Sensing, 31(4):764-770.
Justice, C.O., Wharton, S.W., and Hoblen, B.N., 1981, Application of digital
terrain data to quantify and reduce the topographic effect of Landsat.
International Journal of Remote Sensing, 2:213-230.
Karnieli A., and Cierniewski, J., 2001, Inferring the roughness of desert rocky
surfaces from their bidirectional reflectance data. Advances in Space Research,
28(1):171-176.

20
Mather, P., 1999, Computer Processing of Remotely Sensed Images. Wiley,
Chichester, 292p.
Minnaert, M., 1941, The reciprocity principle in lunar photometry. Astrophysical
Journal, 93(2):403-410.
Sharpnak, D., and Akin, G., 1969, An algorithm for computing slope and aspect
from elevations. Photogrammetric Engineering and Remote Sensing, 35:247-
248.
Shoshany, M., 1993, Roughness-Reflectance relationship of bare desert soil
terrain: An empirical study. Remote Sensing of Environment, 45:15-27.
Smith, J. A., Lin, T. L. and Ranson, K. J., 1980. The Lambertian assumption
and Landsat data. Photogrammetric Engineering and Remote Sensing, 46, pp.
1183-1189.
Song, C., Woodcock, C., Seto, C., Lenney, M., and Macomber, S., 2001,
Classification and change detection using Landsat TM data: When and how to
correct atmospheric effects? Remote Sensing of Environment, 75(2):49-58.
Stohr, C.J., and West, T.R., 1985, Terrain and look angle effects upon
multispectral scanner response. Photogrammetric Engineering and Remote
Sensing, 51(2):229-235.
Takemata, K., Izumiya, T., and Kawata Y., 2000, Analysis of ADEOS/ POLDER
data over land surfaces. Advances in Space Research, 26(7):1065-1068.
Teillet, P.M., Guindon, B., and Goodenough, D.G., 1982, On the slope-aspect
correction of multispectral scanner data. Canadian Journal of Remote Sensing,
8(2):84-106.
Thomson, A. G., and Jones, 1990, Effects of topography on radiance from
upland vegetation in North Wales. International Journal of Remote Sensing,
11(5):829-840.
Tokola, T., Sarkela, J., and Linden, V., 2001, Use of topographic correction in
Landsta TM-based forest interpretation in Nipal. International Journal of
Remote Sensing, 22(4):551-563.

21