4, False
2,4 and
Landsat
ath:175,
Sensor,
sis, and
(d out
rformed
-Hasana -
cultural
nponent
ve been
wadis
ounding
laries in
ervised
site 1
ith the
selected
distance
faximum
th data
rmed to
Al Arish
> Were
)n:
ir com-
) enable
) of El-
0,000).
area re-
were;
is Gebel
(890m)
ebel El-
sana
b) Synclinal lowlands (foot slopes or
hamada) between the above mentiond
anticlines (hills & domes);
c) Part of the great pidmont plain, which
dominates most of north Sinai region;
and
d Wadis which could be divided into two
types namely; main wadies, and
Tributaries. The defferentiation was
carried out based on their width,
soil depth and slope gradient.
4.2 Site Selection and Soil
Characterization
The produced soil consociation map was
used to select a number of sites in which
soil samples were collected from repre-
sentative soil profiles. The obtained an-
alytical data were presented following the
coding system of FAO (1990), table (1).
These data were used to classify the
mapped soils according to Soil Taxonomy
(1990). Torriorthents, Torrifluvents, Tor-
ripsamments, and Haplorthents were the
main soil groups in the study area.
The number and degree of limitations for
agricultural use in Wadi regions found to be
relatively low comparing to the other geo-
morphic units. Therefore, two areas cov-
ered mostly with Wadi El-Arish were se-
lected for detailed investigation using
digital image processing techniques. The
first window located east of El-Hasana
city between Lat. 300 20' and 300 30' N,
and Long. 330 50' and 340 O5' E. The second
window is located north of Nakhl city be-
tween Lat. 300 03' and 300 17' N, and Long.
330 40' and 330 55' E. The total area for
site 1 and site 2 were 637.3 km?2 and
280.3 km2 respectiviely.
4.3 Digital Image Processing:
Filtering and Principal Component Analysis
(PCA) techniques were performed to en-
hance and delineate the wadis boundaries.
The unsupervised classification technique
was performed using Thematic Mapper data
(bands 2, 3, 4 and 7) of wadi region (site1)
fig. (3). Twenty classes including soils,
rocky lands and soil covered with natural
vegetation were produced. Each class has
3
special reflectance characteristics in the
four studied bands. Five soil classes were
discriminated in Site 1, namely playa, wadi
terrace, gravely plain, flood plain, and wadi
bed. The mixed soils are grouped in a separ-
ate class. The latter includs ploughed and
cultivated fields. The total area of soil
class and its spectral mean values in each
band were presented in table (2). Each soil
class is characterized after field and la-
boratory investigations. The mapped soils
were classified according to Soil Taxonomy
(1990).
The same technique was applied to site 2,
where there is no enough soil data avai-
labe. The classified image , fig. (4), shows
the presence of five soil classes including
the mixed soil. The similarity in re-
flectance characteristics between the dis-
criminated soil classes in the two sites
was the criterion used to define the soil
class in site (2).
The main soil characteristices and the tax-
onomic classification of the soils mapping
units could be summerized as follows:
1. Soils of wadi terraces: are almost
flat,moderately deep, high lime and / or
gypsum content, moderately to highly
saline, gravely sandy clay loam or
loamy sand. These soils could be
classified as Typic Torriorthents.
2. Soils of the outwash; are deep, coarse
to moderately coarse texture, high
gypsum and/or lime content, highly
saline. The soil surface is covered
with desert pavement and scattered
natural vagetation. The coarse
texture and the absence of diagnostic
horizons or features were the
main reason to group these soils in
Typic Torriorthent subgroup.
3. Soils of the flood plain; are undulating,
having calcareous crust, deep,
sandy texture, highly saline, highly
calcareous The soil surface is
covered with dense vegetation. Typic-
Torrifluvents was the main
subgroup in this mapping units. In
some parts of the flood plain,
gypsic or calcic horizon. were noticed
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996
