in the region and the water levels on some parts of 
the chotts are dominated by this runoff. These areas 
do not, however, have spring mound or aioun fields. 
As spring mounds are formed by upward groundwater 
seepage they are frequently associated with specific 
underlying lithologies and structures, particularly 
well developed faulting and jointing Reeves (1968). 
The overall structure in this area consists of a 
series of sedimentary horizons, with known aquifers, 
dipping to the north. This structure itself is not 
conducive to groundwater upwellings despite the 
fact that groundwater is present. The Alpine folding 
and faulting of the Atlas Mountains affects the 
sedimentary strata on the northern margin of the 
Saharan Platform. Folding of these sediments on 
the Saharan Platform led to the formation of the 
Chott el Fedjadj anticline. The Djebel Tebaga is 
the prominent southern limb of this anticline and 
it plunges westward under the Chott el Djerid. The 
area still suffers from infrequent tectonic activity. 
It can be hypothesised that the occurrence of spring 
mounds are related to faulting and jointing 
associated with this folding. This can be tested to 
examining the directions of the alignments of these 
isolated features on Landsat TM imagery by lineament 
analysis. 
In addition the distribution of aioun can also be 
examined and related to the structural trends. This 
should enable the origins of aioun to ascribe to 
either deep or shallow groundwater sources. 
3. IDENTIFICATION OF SPRING MOUNDS AND AIOUN ON 
SATELLITE IMAGERY 
Difficulties exist in attempting to identify and map 
spring mounds and aioun by ground survey 
Geomorphological problems are encountered during the 
identification and mapping of aioun because of their 
lack of relief and the difficulty in mapping subtle 
variations in salt concentrations on flat playa 
surfaces. Spring mounds are far easier to identify 
in the field because of their relief and the presence 
of vegetation. Vegetation is commonly found on the 
less saline margins of playas. It mainly takes the 
form of halophytic shrubs and grasses and the centre 
of the spring mounds are also vegetated, often with 
less salt tolerant species, in response to spring- 
flow. The logistical problems of field work on playas 
are immense; they mainly concern difficulties of 
access, heat, salt glare and the location of ground 
control points. As a consequence most topographic 
maps mark playas as featureless voids! 
Remotely sensed imagery is therefore an extremely 
important tool in the identification and mapping of 
playa features such as spring mounds and aioun. 
The advantages of remotely sensed imagery in such 
studies not only lie in their ability to overcome 
the geomorphological and logistical problems but 
also because of the shape and spectral characteristics 
of the features themselves. 
A very important identification criterion for 
spring mounds and aioun is their circular shape; 
this makes them readily identifiable on remotely 
sensed imagery (Figs, 2 and 3). However it can be 
seen on Fig. 3 that the spring mounds develop long 
tails to the SW. These tails are deposits of 
gypsiferous sands deposited in the lee of the spring 
mounds in the prevailing wind directions. 
In addition there are significant variations in 
spectral responses between the features and the 
adjacent playa surfaces, Table 1. Spring mounds 
are often vegetated with either a natural vegetation 
community of tamarisk and date palms or, more 
commonly in this area, they form the nucleus of 
irrigated oases with date palms and an understory 
of smaller trees and ground crops. As they are 
Figure 2. Band 3,4,5, false colour composite of the 
Chotts el Djerid and el Fedjadj showing the playa 
facies (from Mitchell, 1982), the aioun and spring 
mound fields and the Chott el Fedjadj plunging 
anticline. 
Figure 3. Band 3 image of Seftimi showing spring 
mounds with sand tails. The resistant limestone 
beds forming the southern limb of the Chott el 
Fedjadj anticline can be seen at the bottom of the 
image. 
located in marginal playa facies they can be 
surrounded by unvegetated surfaces or halophytic 
plant communities. Where they are surrounded by 
bare surfaces identification is easy because of the 
differences in absorption between vegetation and 
bare soils; particularly in TM Bands 3 (0.63- 
0.69ym) and 4 (O. 79-0.90y.m) . It is also relatively 
easy to spectrally separate spring mound vegetation 
from the surrounding halophytic vegetation. As 
spring mounds have adequate water supplies all the 
year around their vegetation is rarely under stress 
and maintains a low reflectance in TM Bands 3 and 4 
throughout the year. Halophytic vegetation does 
however exhibit some seasonality and partially dies 
back in the summer; stressed halophytic vegetation 
is seen on TM imagery in this area. In addition the 
1 
S 
S 
 
halophyte 
than the n 
single stc 
consequenc 
and densit 
halophyte 
vegetation 
the domina 
or reddish 
date palms 
Aioun or 
the Chott 
However th 
effloresen 
salt conce 
circular p 
visible an 
detected i 
4. DIGITAL 
Digitally 
to map pla 
1986a,b; F 
el Fedjadj 
seen that 
confined t 
restricted 
This stu 
in more de 
aioun with 
processing 
on a TM in 
1983. 
The deta 
subscene c 
that it en 
Chott el F 
inspection 
area that 
tion betwe 
5 (1.55-1. 
geomorphol 
shows the 
reflective 
High corre 
all correl 
correlatic 
two middle 
bands. It 
primarily 
selected t 
of vegetat 
implicatio 
bare surfa 
an infra-r 
images pro 
narrow pix 
After co 
Band 3 (0. 
analysis. 
This was 
(by chloro 
playa surf 
of spring 
608