




































 
 
 

Asian Review of Environmental and Earth Sciences 
Vol. 4, No. 1, 20-27, 2017 

ISSN(E) 2313-8173 / ISSN(P) 2518-0134 
DOI: 10.20448/journal.506.2017.41.20.27 

 
 
 
 

 

20 
 

Discontinuous Nature of Phreatic Aquifers in Granitic Rocks at Watershed Scale: A 
Stratiform Model from Perennial Streams and Well Data 

 
Théodore Koffi YAO1

    

Olivier FOUCHÉ2     

Konan Emmanuel KOUADIO3    

Marie-Solange OGA4    

 

 
(Corresponding Author) 

 
1,3,4Department of Science and Technology of Water and Environmental Engineering, Faculty of Earth Sciences 
and Mineral Resources, University Felix Houphouet-Boigny, Abidjan, 22 BP 582, Côte d’Ivoire.  

 
2Conservatoire national des arts et métiers de Paris, Dept. ICENER – Service de Géotechnique, 2 Rue Conté, 

75003 Paris, France.  

 
Abstract 

The study was carried out within 4,000 km2 segment of Sassandra watershed located in Soubré, 
south-west of Côte d’Ivoire where coffee and cocoa production are preponderant. This work aimed 
to provide a general methodology in order to draw the flow directions and map the potential 
productive aquifer in hard-rocks. Drilling data, remote sensing, and geomorphic data in a context 
of weathered plutonic and metamorphic Precambrian were used. The regional water table was 
modeled through a linear relationship between the topographic surface of the digital elevation 
model and the base surface of the perennial streams thalwegs. As result, a map of regolith 
thickness obtained has been compared with the geology to emphasize their relationship. 
Furthermore, other correlations have been found between the hydraulic data and 
geomorphological features to get more precise stratification model. Concealed by the regolith, the 
hard-rock aquifer is made up of three layers. From top to bottom we have the saprolite (< 10 m), 
the weathering induced fissured layer (35 m mean thickness) below the base of the saprolite and 
finally, the unweathered with very low hydraulic conductivity. Each layer is characterized by a 
constant density of water-bearing fissures. This shows the impact of a stratiform weathering 
profile on the layering of the aquifer. High variability of observing yields is mainly due to 
thickness heterogeneity of the higher-storage regolith and underlying higher-conductive fissured 
saprock. Also, the wells tapping water under a regolith with medium or high thickness were the 
most productive justifying that regolith thickness mapping is a first rank tool in hydrogeology 
prospecting. The thickness map obtained from the interpolated base surface of the regolith and 
the DEM represents a useful tool for groundwater management. 

 
Keywords: Precambrian, DEM, Regolith, Saprock, Fracture. 

 
Citation | Théodore Koffi YAO; Olivier FOUCHÉ; Konan 
Emmanuel KOUADIO; Marie-Solange OGA (2017). Discontinuous 
Nature of Phreatic Aquifers in Granitic Rocks at Watershed Scale: A 
Stratiform Model from Perennial Streams and Well Data. Asian 
Review of Environmental and Earth Sciences, 4(1): 20-27. 
History:  
Received: 6 October 2017 
Revised: 27 October 2017 
Accepted: 2 November 2017 
Published: 7 November 2017 
Licensed: This work is licensed under a Creative Commons 

Attribution 3.0 License  
Publisher: Asian Online Journal Publishing Group 
 

Contribution/Acknowledgement: All authors contributed to the conception 
and design of the study. 
Funding: This study received no specific financial support. 
Competing Interests: The authors declare that they have no conflict of 
interests. 
Transparency: The authors confirm that the manuscript is an honest, 
accurate, and transparent account of the study was reported; that no vital 
features of the study have been omitted; and that any discrepancies from the 
study as planned have been explained. 
Ethical: This study follows all ethical practices during writing.   

 

 

Contents 
1. Introduction ...................................................................................................................................................................................... 21 
2. General Information of the Study Area ....................................................................................................................................... 21 
3. Method and Data ............................................................................................................................................................................. 22 
4. Results and Interpretations ........................................................................................................................................................... 23 
5. Discussion .......................................................................................................................................................................................... 26 
6. Conclusions ....................................................................................................................................................................................... 26 
References .............................................................................................................................................................................................. 26 
 

http://creativecommons.org/licenses/by/3.0/
http://creativecommons.org/licenses/by/3.0/
https://orcid.org/orcid-search/quick-search?searchQuery=Théodore Koffi YAO
https://orcid.org/orcid-search/quick-search?searchQuery=Olivier FOUCHÉ
https://orcid.org/orcid-search/quick-search?searchQuery=Konan Emmanuel KOUADIO
https://orcid.org/orcid-search/quick-search?searchQuery=Marie-Solange OGA
https://orcid.org/orcid-search/quick-search?searchQuery=Théodore Koffi YAO
https://orcid.org/orcid-search/quick-search?searchQuery=Olivier FOUCHÉ
https://orcid.org/orcid-search/quick-search?searchQuery=Konan Emmanuel KOUADIO
https://orcid.org/orcid-search/quick-search?searchQuery=Marie-Solange OGA
https://orcid.org/orcid-search/quick-search?searchQuery=Théodore Koffi YAO
https://orcid.org/orcid-search/quick-search?searchQuery=Olivier FOUCHÉ
https://orcid.org/orcid-search/quick-search?searchQuery=Konan Emmanuel KOUADIO
https://orcid.org/orcid-search/quick-search?searchQuery=Marie-Solange OGA
https://orcid.org/orcid-search/quick-search?searchQuery=Théodore Koffi YAO
https://orcid.org/orcid-search/quick-search?searchQuery=Olivier FOUCHÉ
https://orcid.org/orcid-search/quick-search?searchQuery=Konan Emmanuel KOUADIO
https://orcid.org/orcid-search/quick-search?searchQuery=Marie-Solange OGA
https://orcid.org/orcid-search/quick-search?searchQuery=Théodore Koffi YAO
https://orcid.org/orcid-search/quick-search?searchQuery=Olivier FOUCHÉ
https://orcid.org/orcid-search/quick-search?searchQuery=Konan Emmanuel KOUADIO
https://orcid.org/orcid-search/quick-search?searchQuery=Marie-Solange OGA


Asian Review of Environmental and Earth Sciences, 2017, 4(1): 20-27 

21 
 

 

1. Introduction 
Côte d’Ivoire, surface and subsurface geology of are made up of 97% of crystalline and metamorphic bedrock 

[1]. Here and all over the world, bedrock aquifers are considered among the least understood groundwater 
resources. Although these aquifers have a regional extension, they actually respond in a discontinuous model that 
often shows a wide range of yields in the well. This is generally attributed to the abundance of all scales fractures, 
which more often playing the role of water conduit. However, different types of fractures (primitive joints, 
tectonics-inherited faults, and secondary weathering-fissures) have not similar hydrogeological functions. 
Moreover, all conclusions have to be drawn from the heterogeneity conception to the vertical weathering profile. 
Observed hydraulic gradients are not able to explain the variability of yields and heterogeneous permeability due to 
fracture networks and/or weathering profiles [2]; [3]. The induced significant borehole failure rate, around 50% 
in the granitoids of Côte d’Ivoire, with little or no improvement since the first statistics published by Lenck [4] 
testifies to the difficulty of predictive exploration for groundwater in this environment. This comes from the 
limited sensitivity of current exploration techniques and above all, poor understanding of the controls on hard-rock 
aquifer occurrence and flow direction. While the most immediate need is to improve borehole’s success rate, to 
assist present actions and planning a crucial issue for longer-term sustained development in the context of human 
migrations and climatic changes. In addition, we need to evaluate overall resource and aquifer occurrence within a 
regional watershed, not only specialization of the groundwater occurrence, but also the direction of groundwater’s 
flow is problematic. The last one is often presumably controlled by topography and geological structure but the 
relative influence of these factors in crystalline rocks still raises basic interrogations [5].  

Aquifers contained into the fresh bedrock are of rare occurrence due to negligible intergranular porosity and 
low tectonic fracture network storativity. Higher porosity and permeability of hard-rock may locally arise within 
major brittle structures [6]; [7]; [8]; [9]. Tunnel engineering in saturated rock masses also provides 
opportunities to develop understanding on the subsurface hydraulic behavior of Recent studies in relation to 
mining or radioactive waste disposal, especially from borehole data in low permeability bedrock at depth, have also 
increased knowledge of the detailed hydraulics of fracture systems fractured media [10]; [11] at local scale. 
However, although it was proposed by De Dreuzy, et al. [12]; Kouame, et al. [13] and other advocates of the 
percolation theory, a regional approach cannot be mainly a geometrical extrapolation of the local hydraulic 
mechanisms of fresh bedrock by computing a regional permeability of a lineament network. Rather, it is necessary 
to focus on the granitic rocks and combine at greater scale lithology, morphology, hydrology, structural geology, 
weathering profile and well data [14]. 

Indeed, in those rocks, in the absence of any sedimentary cover, groundwater flows through discontinuous 
aquifers characterized by a vertical weathering gradient with variable porosity from the surface below to almost 
unfractured rock at depth [15]. At the top, weathered disaggregated rocks are both sandy and clay-rich and so 
have very low intergranular permeability. Clay strata, especially in the saprolite which is the bottom layer of the 
regolith, may in places support some isolated perched aquifers [16]. Between the regolith aquifer and the deeper 
unfractured bedrock, both with low permeability, it is usual to distinguish the fissure aquifer with low porosity but 
much higher permeability, almost entirely dependent on secondary fissures due to weathering. 

The work conducted here in a part of Sassandra watershed located in Soubré is intended to provide a general 
methodology for the investigation of flow directions and mapping of favorable areas for the occurrence of a 
productive aquifer in hard rocks and to contribute to the comprehension of groundwater flow in hard rock.  

The borehole dataset including regolith thickness, initial hydraulic head and well yield, is examined in relation 
to the topographic surface and the perennial streams. We focus on a 4,000 km2 area within the granitoids. The 
model took from Wyns, et al. [17]; Yao, et al. [18] allows for mapping the water table and the regolith-saprock 
interface. Then a map of regolith thickness achieved is compared to the geology of this area. 
 

2. General Information of the Study Area 
The study region comprises a segment of the Sassandra River’s watershed located in west Africa especially in 

the Southwest of Côte d’Ivoire between latitudes 5°19 - 6°34 and longitudes 6°12 - 7°08, and has a surface area of 8 
590 km2 (Fig. 1). 

 

 

Figure-1. Location and geology of the study area  
Source: Yao [19]. 



Asian Review of Environmental and Earth Sciences, 2017, 4(1): 20-27 

22 
 

 

The geomorphic setting is a vast and slightly waved plateau formed by Precambrian crystalline rocks, with 200 
m average elevation and inclined 0.1% southwards, incised by a dense pattern of tributaries. The region is drained 
by Sassandra River which offers a variety of configurations along its course and benefits from a subequatorial 
climate (hot and humid). The study region comprises a 100 km segment of the Sassandra River’s watershed. A 210 
m elevation seems to be characteristic scale of the water table along the limit of the watershed. Excluding the 
isolated peaks at an elevation exceeding 300 m around the more recent granitoids (Fig. 2), the typical elevation 
gradient in the direction transverse to the Sassandra River implies a hydraulic gradient of about 0.2%.  

It is the case of the Lobo valley, a major tributary of Sassandra River: its stream south-westwards is controlled 
by the HLFZ (Hana-Fault Zone) which plays the role of a hydraulic barrier as outlined by the great meander of 
Sassandra River where abutting on the fault. The barrier effect due to fault gouge has been reported elsewhere for 
such fault zones through Precambrian rocks in other geological contexts [20]; [21]. 
 

 
Figure-2. Digital map of the relief from the DTM of studied area 

                                              Source: Primary data analysis (2017) 
 

Actually, rains are abundant (between 1600 mm and 1800 mm) and reach a peak around June and July. During 
the year, temperatures vary between 26 °C and 32 °C. 

The great N-S Sassandra Fault belongs to a family of impressive structures which may have originated [22] as 
ductile shears or sutures during the early stage of the Eburnean orogeny with repeated subsequent movement to 
include late brittle faulting in the intrusives. These faults zones guide the course of the main continental streams as 
for instance the Sassandra River North of the main SW-NE trending crustal structure typical of the Baoule-Mossi 
domain [23] that we will refer to HLFZ, the northern part of the study area belongs to the main supracrustal 
sequences of Birimian [24]. At the south of the HLFZ, outcropping rocks associate varied intrusive granitoïd 
lithologies of the Eburnean orogeny with reworked metamorphosed Archean components (cf. Fig. 1). 

The relatively sodic early rocks underlie featureless plateau whereas the latter potassic plutons form positive 
outcrop features, locally with a thick saprock. The high potassium content and abundance of quartz prevented the 
saprock to be broken down into a thick soil cover and now allows it to be a good aquifer. Retrogressive 
metamorphism replaced biotite by other hydrous mineral phases (chlorite, amphibole) which made the host rock 
less susceptible to weathering [25]. 
 

3. Method and Data 
In order to tackle both questions, groundwater occurrence and flow direction, we referred to the conceptual 

scheme of a stratiform bedrock inherited from a single phase weathering paleoprofile, proposed by Wyns, et al. 
[17] and further developed in Dewandel, et al. [26] and Yao, et al. [18]. The layering roughly follows the 
paleotopography and presents a gently dipping sequence at regional scale. The model stands for two superposed 
aquifers, the regolith and the saprock or “fissure aquifer” with possible exchange between them and with some 
vertical tectonic fault zones allowing for a deeper flow component. The used model was demonstrated in Yao, et al. 
[18] which was being forwarded. In this model, other fracture zones being isolated from both the surface and 
fissure aquifer are therefore non-conductive. The base surface of the perennial streams’ thalwegs, denoted Sr (r for 
reference), is obtained by kriging of the values extracted from the DEM along the perennial streams. The real 
potentiometric surface Sp, including borehole data, is still unknown at this stage. Following Chilton and Foster 
[16] let us define the parameters a and b at every point by: a = St – Sr and b = Sp – Sr. (1) 



Asian Review of Environmental and Earth Sciences, 2017, 4(1): 20-27 

23 
 

 

Next to a perennial stream, the surfaces Sp and St are likely to intermingle. When increasing the distance, 
however, we can hold the following tendency for true: the higher a point between two streams (nearer to 
interfluves), the deeper the hydrostatic level. Now, considering the three surfaces Sr, Sp, and St, we could better 
write b = p x a – h with p, h, real numbers.  

Finding such a relation will allow determining the hydraulic head as a direct function of the soil elevation. The 
1-degree digital elevation model (DEM) is obtained from the Shuttle Radar Topographic Mission (SRTM), with 3 
x 3-arc-second data spacing (90 m). The drainage network and topography were extracted from the DEM and 
three paper elevation maps, 1:200,000 scaled, elaborated in 2001. 

The borehole dataset provides the depth of a mechanical interface which is the contact between saprolite, i.e. 
the disaggregated but structure-preserved lower horizon of the regolith, and underlying saprock which is still 
cohesive but weathered fissured bedrock. Below this interface, with increasing depth, the saprock to fresh bedrock 
junction is transitional or even fluctuating in banded sequences. Thus, the depth of the mechanical interface is also 
the thickness of the regolith and we denote it by Na (a for alteration). 

The depth of the hydrostatic level measured at the borehole completion is denoted Np (p for potentiometric), 
and also the terrain elevation Zt is transformed to Zp into an altimetric level. Even when the elevation of the site 
was recorded in the borehole file, it did not turn out to be more precise that by pointing the X, Y position on the 
DEM. Proceeding this way to acquire Zt, systematically, one deals with an absolute error of 10 m along the 
vertical axis. Then, the main source of uncertainty resides in X, Y positions of the boreholes: locations are 
estimated to be accurate to ± 100m. Gathering boreholes drilled at years of distance, the hydrostatic level database 
obviously is diachronous. Here we advocate that the effect of the year or month of borehole completion is negligible 
for our purpose: the diachronous database will be considered as a homogeneous source of data. 

We compile water-well yields reported from drillers’ logs. The yields are modest, typically less than 5m3/h, 
and many boreholes had run dry or been abandoned on completion. In this study, we only analyze data from wells 
that were considered as successful and are still productive to date. Detailed water level and yield evolution are not 
available, only being registered the first air lift test when the borehole has been completed. Yields were generally 
determined as the rate of water that can be air-lifted on a continuous, short-term (tens of minutes) basis. Such 
method introduces an unquantifiable variance in measured yields but is acceptable for a statewide or regional 
evaluation. 
 

4. Results and Interpretations 
4.1. Flow Directions: A Model of the Water Table 

It turned out to be sufficient to extract 220 values of Zt, each one representative of a 90 m x 90 m cell, since the 
220 points are scattered enough to make an interpolation (Fig. 3) and get a map of Sr. The extracted value is a 
superior bound of the water level in the stream. However, as soon as we will observe boreholes at a distance of a 
few hundreds of meters from a perennial stream, the value of Zt extracted at the river is a good local minimum for 
Sr and Sp. 
 

 
Figure-3. Elevation of the basal surface of the thalwegs of perennial streams (Surfer7®), kriged from 220 values pointed along low elevation 
channels of the DEM  
Source: Yao, et al. [18] 

 
Working on a population of 55 boreholes (Fig. 4), three correlations were found but a good correlation factor 

found for b is: b = 0.92 a – 7.00 m (R2 = 0.95) (2) 
We continued to improve the correlation achieved; So, with another dataset, wells taken away from the 

streams, we found this time: b = 1.00 a – 9.64 m (R2 = 0.97). (3) 
 



Asian Review of Environmental and Earth Sciences, 2017, 4(1): 20-27 

24 
 

 

 
Figure-4. Relation between the land surface, the base surface of the 
thalwegs of perennial streams, and the hydrostatic level in 55 boreholes 

                                                           Source:  Yao, et al. [18] 
 
The slope very close to unity of this relation means that Sp lies about 9.6 m in average under St in a quasi-

parallel fashion.  
Working on the b/a relation, Wyns, et al. [17] have interpreted the negative value of y-axis at the origin as 

the depth between the water surface and the banks. However, the water surface of the rivers in the study area is 
incised 1 m or 2 m below the banks. Our result is rather consistent with the mean Np value (7.33 m) and with the 
fact that most of the studied boreholes are sited away from perennial streams for safety towards flooding hazard. 
Thus, we are induced to distinguish two relations: one in the neighborhood of the thalwegs and another at longer 
distance. In the vicinity of the perennial streams, we suggest making possible the convergence of the surfaces Sp 
and Sr, i.e. with an assumption the negligible incision below the banks. To satisfy this condition, we need to force 
the relationship to respect linearity in a strict sense: then, we find b = 0.74 a without losing too much signification 
level (R2 = 0.90), valid in the range of values up to 60 m. Let notice that the value 0.74 is very close to the value 
found by Wyns, et al. [17] through the “with bank incision” assumption. 

From these linear relations, we deduce the elevation of Sp in any 90 m x 90 m cell with the help of a simple 
transformation of the base surface Sr. The relation Zp – Zr = 0.74 (Zt – Zr) gives: Zp = 0.74 Zt + 0.26 Zr. Beyond 
the value a = 60 m, the best linear fitting with a slope close to 1 should be used since Sp, despite its variability, is 
roughly parallel to Sr. We obtain the map of Figure 5 from which local flow directions are very clear. The standard 
error of the regression line (±1.85 m) is likely to include the natural variability of the water table, which gives a 
backward argument favorable to the neglecting of the diachronous character of well data at section 3. 
 

 

Figure-5.  Surface of the water table computed from the basal surface of the thalwegs of perennial streams and the DEM 
Source: Yao, et al. [18] 
 

4.2. Water Occurrence: Strata Function and Thickness 
In order to target some interesting sites for new population settlements, and given the storage function of the 

regolith, it is essential to understand the regional control on the distribution of the thickness parameter. The 
spatial distribution of the residual regolith thickness is obtained by direct kriging of data from 183 boreholes (Fig. 
6A). Despite the low resolution, significant zones emerge from this map and the distribution is compared with the 
maps of lithology and relief. The thickness, 15 m in average, displays distinct mapped domains. 
 

y = 0,92 x - 7,00

R2 = 0,95
y = 0,74 x

R2 = 0,90

y = 0,004 x
2
 + 0,503 x

R2 = 0,954

-20

0

20

40

60

80

100

120

0 20 40 60 80 100 120 140

a (m)

b
 (

m
)



Asian Review of Environmental and Earth Sciences, 2017, 4(1): 20-27 

25 
 

 

 

Figure-6. A: Raw values of the regolith thickness (183 borehole data). B: Regolith basal surface (RBS) interpolated from 105 boreholes. 
Kriging by 250 m step with a Gaussian variogram. The regolith thickness is obtained (Fig. 6C). 
Source: Primary data analysis (2017) 
 

Low thickness (< 25 m) light zones are observed on the highlands of both hillsides of the Sassandra, also 
characterized by higher values of slope giving access to the interfluve limits of the watershed. This is consistent 
with lower weather ability of the rocks around the inselbergs. Another set of low-thickness values are concentrated 
into a SW-NE alignment passing through Okrouyo and Meagui within the gneiss and intrusive syenite or binary 
granite bodies: its trend is N10 and marked as a dotted bold line on Fig. 6A. Two major gneiss enclaves within 
migmatite are the sites of main tributary confluents of the Sassandra River and display low regolith thickness. 

High thickness (> 50 m) dark zones are in an alignment trending N30, beginning within migmatite near the 
Sassandra and continued along the anatexis granodiorite: it is marked as a continued bold line on Fig. 6A. Another 
high thickness zone is well characterized around Oupoyo within migmatite, south of the two major gneiss enclaves. 

Thus, differential weathering as a function of lithology appears to be the main control for the spatial 
distribution of the regolith. The advocated influence of lithology is consistent with the results of Moore, et al. [27] 
orWalsh and Clark [28] under different climatic conditions but similar geology: migmatites display higher 
regolith thicknesses and yields than foliated plutons. 

In order to reduce the variance and the nugget effect [17] the base surface of the regolith on the 4,000 km2 
study area has been kriged by using a Gaussian variogram (Fig. 6B). This map explains why regolith thickness Na 
is not correlated with the soil elevation Zt. At the time of weathering, the basal surface of the regolith was parallel 
to the ancient topography. Differential erosion has made the present-day morphology of the landscape mainly 
independent from the inherited weathering profile. Then, an improved map of the regolith thickness is obtained 
(Fig. 6C) by subtracting the regolith’s base surface from the topographic surface (DEM). On this new map, both 
the correlation with lithology and the effect of erosion by streams are highlighted. 

Well yields generally give mediocre correlation with other parameters and it is the case here. This is caused by 
the heterogeneity of the yield values and the interferences between the various influences involved. Nevertheless, in 
our data values of regolith thickness inferior to 10 m do not allow for high yield values (> 10 m3/h) but higher 
yields arise around the average thickness (15.7 m for this borehole population). Thus, wells tapping water under 
medium or high thickness regolith are the most productive. 

As the main aquifer component is the fissure layer, the linear yield or yield by well depth must be defined with 
reference to the well height below the base of the saprolite rather than the total well depth or saturated height 
[29]. From Fig. 7, 95% (respectively 80%) of the cumulative linear yield of a well population is obtained within a 
35 m layer (respectively 25 m) under the base surface of the regolith. The first segment of the cumulative curve 
shows that half the total linear yield comes from a thin horizon (no more than 5 m thick) immediately under the 
contact and which is likely to drain the regolith. Almost all of the other 50% (second segment on the curve) comes 
from the 5 m to 35 m layer and it is noticeable, with a constant gradient. The gradient changes again between 35 m 
normalized and 40 m under the base of the regolith and drilling deeper becomes less profitable. 
 

 
Figure-7. Linear cumulated yield as a function of the effective saturated height 
under the base of the regolith, 55 boreholes 

                                                   Source: Yao [25]. 
 



Asian Review of Environmental and Earth Sciences, 2017, 4(1): 20-27 

26 
 

 

5. Discussion 
Using the base surface of the perennial streams’ thalwegs, we model the potentiometric surface in the regolith 

+ saprock aquifer. Near to the interfluves, the water table is deeper and the hydraulic gradient is lower. This may 
explain why well yield is often negatively related to local slope, i.e. the rate of change in soil elevation surrounding 
a well. Convergence and merging of the potentiometric surface with the base surface when approaching the 
perennial streams’ thalwegs is quantified by a linear relationship. In this frame, we question the respective role of 
the regolith and saprock. 

The fissure layer or saprock is the most productive part of the hard-rock aquifer, as already shown elsewhere 
Houston and Lewis [30]; Maréchal, et al. [31]; Taylor and Howard [32]. Following Wright [33] effective 
development of the saprock component requires interaction with storage available in the overlying saturated 
regolith. We observe as Barker, et al. [34] that thicker the regolith, greater the saturated thickness of regolith. 
Moreover, it is generally accepted that the frequency of clusters of subhorizontal fissures within the top of the 
saprock is greatest where the regolith is thickest. Thus the rational in borehole siting is to try to locate the deepest 
regolith on the assumption that this will maximize yields from both the regolith and the saprock [35]. 

The average regolith thickness is about 30 m in West Africa and only 15 m in the study area. The main 
phreatic aquifer occurs within the variable layer of saprock. The more fissured and productive parts of the aquifer 
have transmissivity values up to 100 m2/day. Wells in the bedrock tap groundwater from the fissure aquifer mainly 
in semi-confined conditions under the clayey regolith, or rarely unconfined. Yields range from a few m3/day in 
some upland or slope areas to 43 m3/h in wells adjacent to a perennial stream. 

However, mapping the regolith thickness [17] is still a new approach in hard rock hydrogeology and is not 
fully accepted. We advocate that the thickness map obtained from the interpolated base surface of the regolith and 
the DEM is a useful tool for groundwater management which should be integrated for vulnerability mapping [36] 
and favorability mapping [37] through GIS methodology. 

Our improvement of the stratiform model arises from the study of linear yield against depth. Decrease in linear 
yield or hydraulic conductivity is not progressive since we can distinguish three sub-zones sharply contrasted. 
Within each sub-zone, the density of fissures opened to flow is not decreasing with depth. However, dilation of 
fissures continuously reduces at depth with overburden pressure. So, in contradiction with a recurrent idea in the 
literature, the role of exhumation pressure release cannot account for these subzones. Our observation is rather in 
conformity with the leading part of mica swelling in the inherited weathering and fissuring profile of the bedrock 
[17]. For instance, small thickness of the saprolite above syenite (9 m) is mainly attributed to poor weather ability 
of this rock due to lack of biotite. The two micas or binary granite which contains some biotite has thicker regolith 
(17 m). 

The contrasted susceptibility to weathering between different lithologies is not the only control on the 
formation of the weathering profile: contrasts in tectonic joint and fracture density and in relief amplitude also play 
a part. As for the role of the local tectonic fracture network, it is often stated that high density of local fractures is 
favorable to deeper and more penetrative weathering and secondary fissuring, thus leading to higher well yields. 
We don’t think so. At the time of weathering, opened tectonic joints in uplands allowed through the flow of 
meteoric waters to the level of the water table deep beneath the hills [19]. Once in relief, the fresh bedrock masses 
tended to remain dry and tectonic fracture density had a conservative function. The late plutons as unfoliated 
syenite are representative of this case in the study area. At present in this environment, rapid infiltration through 
thin overburden and diffuse tectonic joints down to the saprock still prevents water from mineralizing. 
 

6. Conclusions 
Dealing with the concept of stratiform aquifers in relation with the acquired weathering profile in hard-rocks, 

the region of Soubré (Côte d’Ivoire, West Africa) provides a rich example of how Precambrian multiphase 
magmatism and final tectono-thermal events controlling lithology and jointing, on the other hand, Tertiary 
weathering and Quaternary erosion controlling bedrock layering and morphology, are both responsible for present 
infiltration, groundwater storage and flow within the regional regolith + saprock aquifer. 

Using the base surface of the perennial streams’ thalwegs, we model the potentiometric surface within the 
regolith + saprock aquifer. The direction of shallow groundwater flow in the study area directly deduces from the 
local gradient of the potentiometric surface in sub-catchments. 

Wells tapping water under a regolith with medium or high thickness are the most productive, which justifies 
for regolith thickness mapping as a first rank tool in hydrogeology prospecting. We conclude against the general 
statement that incremental yield continuously decreases at depth, leading to the idea of lithostatic pressure 
controlling the openness of fractures more commonly called “sheeting fractures”. At the opposite, the saprock 
under the regolith is parted into three sub-layers, each one being characterized by a constant density of water-
bearing fissures. This highlights the impact of a stratiform weathering profile on the layering of the aquifer. In this 
frame, the high variability of observed yields is mainly due to thickness heterogeneity of the higher-storative 
regolith and underlying higher-conductive fissured saprock. 
 

References 
[1] I. Yacé, "Initiation à la géologie. L’exemple de la Côte d’Ivoire et de l’Afrique de l’Ouest. Éditions CEDA (Abidjan), Côte d’Ivoire," 

p. 183, 2002. 
[2] W. F. Megahan and J. L. Clayton, "Saturated hydraulic conductivities of granitic materials of the Idaho batholith," Journal of 

Hydrology, vol. 84, pp. 167-180, 1986. View at Google Scholar | View at Publisher 
[3] J. P. Faillat, "Origine des nitrates dans les nappes de fissures de la zone tropicale humide: Exemple de la Côte d’Ivoire," Journal of 

Hydrology, vol. 113, pp. 231-264, 1990. View at Google Scholar | View at Publisher 
[4] P. P. Lenck, "Données nouvelles sur l'hydrogéologie des régions à substratum métamorphique ou éruptif. Enseignements tirés de 

la réalisation de 900 forages en Côte d'Ivoire," Comptes Rendus de l'Académie des Sciences - Series IIA - Earth and Planetary Science, vol. 
285, pp. 497-500, 1977. View at Google Scholar   

[5] P. Brunner, H. J. H. Franssen, L. Kgotlhang, P. Bauer-Gottwein, and W. Kinzelbach, "How can remote sensing contribute in 
ground water modeling?," Hydrogeology Journal, vol. 15, pp. 5-18, 2007. View at Google Scholar | View at Publisher 

https://scholar.google.com/scholar?hl=en&q=Saturated%20hydraulic%20conductivities%20of%20granitic%20materials%20of%20the%20Idaho%20batholith
http://dx.doi.org/10.1016/0022-1694(86)90050-8
https://scholar.google.com/scholar?hl=en&q=Origine%20des%20nitrates%20dans%20les%20nappes%20de%20fissures%20de%20la%20zone%20tropicale%20humide:%20Exemple%20de%20la%20Côte%20d’Ivoire
http://dx.doi.org/10.1016/0022-1694(90)90177-y
https://scholar.google.com/scholar?hl=en&q=Données%20nouvelles%20sur%20l'hydrogéologie%20des%20régions%20à%20substratum%20métamorphique%20ou%20éruptif.%20Enseignements%20tirés%20de%20la%20réalisation%20de%20900%20forages%20en%20Côte%20d'Ivoire
https://scholar.google.com/scholar?hl=en&q=How%20can%20remote%20sensing%20contribute%20in%20ground%20water%20modeling?
http://dx.doi.org/10.1007/s10040-006-0127-z


Asian Review of Environmental and Earth Sciences, 2017, 4(1): 20-27 

27 
 

 

[6] J. S. Caine and R. A. Tomusiak, "Brittle structures and their role in controlling porosity and permeability in a complex 
Precambrian crystalline-rock aquifer system in the Colorado Rocky Mountain front range," GSA Bulletin, vol. 115, pp. 1410-1424, 
2003. View at Google Scholar | View at Publisher 

[7] M. Douglas, "Mixing and temporal variations of groundwater inflow at the Con Mine, Yellowknife, Canada; an analogue for a 
radioactive waste repository," Unpubl. M.Sc. , Thesis, Dept of Geology, University of Ottawa, Canada, 1997. 

[8] O. Fouché, R. Cojean, and M. Arnould, "Geological and geometrical characterization of natural fractures in a granitic formation 
from cored boreholes," Bulletin of Engineering Geology and the Environment, vol. 60, pp. 231-240, 2001. View at Google Scholar   

[9] O. Fouché and J. Diebolt, "Describing the geometry of 3D fracture systems by correcting for linear sampling Bias," Mathematical 
Geology, vol. 36, pp. 33-63, 2004. View at Google Scholar | View at Publisher 

[10] S. B. Mabee, P. J. Curry, and K. C. Hardcastle, "Correlation of lineaments to ground water inflows in a Bedrock Tunnel," Ground 
Water, vol. 40, pp. 37-43, 2002. View at Google Scholar | View at Publisher 

[11] A. Pouya and O. Fouché, "Permeability of 3D discontinuity networks: New tensors from boundary-conditioned homogenisation," 
Advances in Water Resources, vol. 32, pp. 303-314, 2009. View at Google Scholar | View at Publisher 

[12] J. R. De Dreuzy, P. Davy, and O. Bour, "Hydraulic properties of two-dimensional random fracture networks following a power law 
length distribution: 1-effective connectivity," Water Resources Research, vol. 7, pp. 107-111, 2001.  

[13] K. F. Kouame, A. Akaffou, T. Lasm, J. R. De Dreuzy, P. Davy, and O. Bour, "Simulation des écoulements dans les réservoirs 
fracturés: Application au socle Archéen de Touba (nord-ouest de la Côte d'Ivoire)," Actes du Colloque Internationale SITIS 05 Yaoundé 
(Cameroun), pp. 39-46, 2005. 

[14] P. Lachassagne, R. Wyns, and B. Dewandel, "The fracture permeability of hard rock aquifers is due neither to tectonics, nor to 
unloading, but to weathering processes," Terra Nova, vol. 23, pp. 145-161, 2011. View at Google Scholar | View at Publisher 

[15] C. Faillace, Hydrogeology of hard rock in some eastern and western African countries. In: Krasny, J. and Sharp, J.M. Jr (Eds), Groundwater 
in fractured rocks. Selected papers on hydrogeology 9. London, UK: Taylor & Francis, 2007. 

[16] P. J. Chilton and S. S. D. Foster, "Hydrogeological characterization and water-supply potential of basement aquifers in tropical 
Africa," Hydrogeology Journal, vol. 3, pp. 36-49, 1984.  

[17] R. Wyns, J. M. Baltassat, P. Lachassagne, A. Legchenko, J. Vairon, and F. Mathieu, "Application of SNMR soundings for 
groundwater reserves mapping in weathered basement rocks (Brittany, France)," Bulletin de la Société Géologique de France, vol. 175, 
pp. 21-34, 2004. View at Google Scholar | View at Publisher 

[18] T. K. Yao, O. Fouché, K. E. Kouadio, M. S. Oga, and T. Lasm, " Modélisation de la surface piézométrique des aquifères fissurés en 
zone de socle précambrien métamorphisé: cas du bassin versant du Sassandra (sud-ouest de la Côte d’Ivoire)," Journal of Water 
Science, vol. 28, pp. 105-117, 2015. View at Google Scholar | View at Publisher 

[19] C. R. Twidale and E. M. Campbell, "Fractures: A double edged sword – a note on fracture density and its importance," Zeitschrift 
für Geomorphologie, vol. 37, pp. 459-475, 1993. View at Google Scholar   

[20] T. Gleeson and K. Novakowski, "Identifying watershed-scale barriers to groundwater flow: Lineaments in the Canadian shield," 
Geological Society of America Bulletin, vol. 121, pp. 333-347, 2009. View at Google Scholar | View at Publisher 

[21] J. Marler and S. Ge, "The permeability of the Elkhorn fault zone, South Park, Colorado," Ground Water, vol. 41, pp. 321-332, 2003. 
View at Google Scholar | View at Publisher 

[22] R. Fabre, "Les effets tectoniques des décrochements N-S sénestres dans les formations volcaniques birimiennes de la région de 
Yaouré, centre de la Côte d'Ivoire," C.R. Acad. Sc., Paris, tome 300, Série II, n°19, 1985. 

[23] B. Tagini, "Esquisse structurale de la Côte d’Ivoire. Essai de géotechnique régionale," Thèse, 1971. 
[24] A. Papon and R. Lemarchand, "Géologie et minéralisation du Sud-Ouest de la Côte d’Ivoire," Synthèse de l’Opération SASCA (1962-

1968). SODEMI, Abidjan, p. 284, 1973. 
[25] K. T. Yao, " Hydrodynamisme de l’eau souterraine dans lesaquifères de socle cristallin et cristallophyllien du Sud-Ouest de la Côte 

d’Ivoire: Cas du département de Soubré. Apports de la télédétection, de la géomorphologie et de l’hydrogéochimie," Thèse de Doct. 
Conservatoire National des Arts et Métiers,Paris, France, 2009. 

[26] B. Dewandel, P. Lachassagne, R. Wyns, J. C. Maréchal, and N. S. Krishnamurthy, "A generalized 3-D geological and 
hydrogeological conceptual model of granite aquifers controlled by single or multiphase weathering," Journal of Hydrology, vol. 330, 
pp. 260-284, 2006. View at Google Scholar | View at Publisher 

[27] R. B. Moore, G. E. Schwarz, S. F. Clark, G. J. Walsh, and J. R. Degnan, "Factors related to well yield in the fractured bedrock 
aquifer of new Hampshire," USGS Professional Paper, vol. 1660, p. 51, 2002.  

[28] G. J. Walsh and S. F. Clark, "Contrasting methods of fracture trend characterization in crystalline metamorphic and igneous rocks 
of the Windham quadrangle, New Hampshire," Northeastern Geology and Environmental Sciences, vol. 22, pp. 109-120, 2000. View at 

Google Scholar   
[29] N. Courtois, B. Dewandel, P. Lachassagne, R. Wyns, R. Blanchin, S. Somé, A. Tapsoba, and F. D. Bougaïré, "Experimental GIS 

hydrogeological mapping of hard-rock aquifers in Burkina Faso, to help for groundwater management and planning," Groundwater 
and Climate in Africa, 24-28 June 2008, Kampala, Uganda, 2008. 

[30] J. F. T. Houston and R. T. Lewis, "The Victoria Province drought relief project II. Borehole yield relationships," Ground Water, 
vol. 26, pp. 418-426, 1988. View at Google Scholar | View at Publisher 

[31] J. C. Maréchal, B. Dewandel, and K. Subrahmanyam, Characterization of fracture properties in hard rock aquifer system. In: Thangarajan 
M, 2007, Groundwater – resource evaluation, augmentation, contamination, restoration, modeling and management vol. 362: Springer, 2007. 

[32] R. G. Taylor and K. W. F. Howard, "The influence of tectonic setting on the hydrological characteristics of deeply weathered 
terrains: Evidence from Uganda," Journal of Hydrology, vol. 218, pp. 44-71, 1999. View at Google Scholar | View at Publisher 

[33] E. P. Wright, "The hydrogeology of crystalline basement aquifers in Africa," London Special Publication, vol. 66, pp. 1-27, 1992. View 

at Google Scholar  | View at Publisher 
[34] R. D. Barker, C. C. White, and J. F. T. Houston, "Borehole siting in an African accelerated drought relief project," Geological Society 

Special Publication, vol. 66, pp. 183-201, 1992. View at Google Scholar | View at Publisher 
[35] K. W. F. Howard and J. Karundu, "Constraints on the exploitation of basement aquifers in East Africa – water balance implications 

and the role of the regolith," Journal of Hydrology, vol. 139, pp. 183-196, 1992. View at Google Scholar | View at Publisher 
[36] S. C. Denny, D. M. Allen, and J. M. Journeay, "DRASTIC-Fm: A modified vulnerability mapping method for structurally 

controlled aquifers in the southern Gulf Islands, British Columbia, Canada," Hydrogeology Journal, vol. 15, pp. 483-493, 2007. View at 

Google Scholar | View at Publisher 
[37] V. Madrucci, F. Taioli, and C. César de Araújo, "Groundwater favorability map using GIS multicriteria data analysis on crystalline 

terrain, São Paulo State, Brazil," Journal of Hydrology, vol. 357, pp. 153-173, 2008. View at Google Scholar | View at Publisher 

 
 

 

 

  

Asian Online Journal Publishing Group is not responsible or answerable for any loss, damage or liability, etc. caused in relation to/arising out of the use of the content. 
Any queries should be directed to the corresponding author of the article. 
 

https://scholar.google.com/scholar?hl=en&q=Brittle%20structures%20and%20their%20role%20in%20controlling%20porosity%20and%20permeability%20in%20a%20complex%20Precambrian%20crystalline-rock%20aquifer%20system%20in%20the%20Colorado%20Rocky%20Mountain%20front%20range
http://dx.doi.org/10.1130/b25088.1
https://scholar.google.com/scholar?hl=en&q=Geological%20and%20geometrical%20characterization%20of%20natural%20fractures%20in%20a%20granitic%20formation%20from%20cored%20boreholes
https://scholar.google.com/scholar?hl=en&q=Describing%20the%20geometry%20of%203D%20fracture%20systems%20by%20correcting%20for%20linear%20sampling%20Bias
http://dx.doi.org/10.1023/b:matg.0000016229.37309.fd
https://scholar.google.com/scholar?hl=en&q=Correlation%20of%20lineaments%20to%20ground%20water%20inflows%20in%20a%20Bedrock%20Tunnel
http://dx.doi.org/10.1111/j.1745-6584.2002.tb02489.x
https://scholar.google.com/scholar?hl=en&q=Permeability%20of%203D%20discontinuity%20networks:%20New%20tensors%20from%20boundary-conditioned%20homogenisation
http://dx.doi.org/10.1016/j.advwatres.2008.08.004
https://scholar.google.com/scholar?hl=en&q=The%20fracture%20permeability%20of%20hard%20rock%20aquifers%20is%20due%20neither%20to%20tectonics,%20nor%20to%20unloading,%20but%20to%20weathering%20processes
http://dx.doi.org/10.1111/j.1365-3121.2011.00998.x
https://scholar.google.com/scholar?hl=en&q=Application%20of%20SNMR%20soundings%20for%20groundwater%20reserves%20mapping%20in%20weathered%20basement%20rocks%20(Brittany,%20France)
http://dx.doi.org/10.2113/175.1.21
https://scholar.google.com/scholar?hl=en&q=Modélisation%20de%20la%20surface%20piézométrique%20des%20aquifères%20fissurés%20en%20zone%20de%20socle%20précambrien%20métamorphisé:%20cas%20du%20bassin%20versant%20du%20Sassandra%20(sud-ouest%20de%20la%20Côte%20d’Ivoire)
http://dx.doi.org/10.7202/1032293ar
https://scholar.google.com/scholar?hl=en&q=Fractures:%20A%20double%20edged%20sword%20–%20a%20note%20on%20fracture%20density%20and%20its%20importance
https://scholar.google.com/scholar?hl=en&q=Identifying%20watershed-scale%20barriers%20to%20groundwater%20flow:%20Lineaments%20in%20the%20Canadian%20shield
http://dx.doi.org/10.1130/b26241.1
https://scholar.google.com/scholar?hl=en&q=The%20permeability%20of%20the%20Elkhorn%20fault%20zone,%20South%20Park,%20Colorado
http://dx.doi.org/10.1111/j.1745-6584.2003.tb02601.x
https://scholar.google.com/scholar?hl=en&q=A%20generalized%203-D%20geological%20and%20hydrogeological%20conceptual%20model%20of%20granite%20aquifers%20controlled%20by%20single%20or%20multiphase%20weathering
http://dx.doi.org/10.1016/j.jhydrol.2006.03.026
https://scholar.google.com/scholar?hl=en&q=Contrasting%20methods%20of%20fracture%20trend%20characterization%20in%20crystalline%20metamorphic%20and%20igneous%20rocks%20of%20the%20Windham%20quadrangle,%20New%20Hampshire
https://scholar.google.com/scholar?hl=en&q=Contrasting%20methods%20of%20fracture%20trend%20characterization%20in%20crystalline%20metamorphic%20and%20igneous%20rocks%20of%20the%20Windham%20quadrangle,%20New%20Hampshire
https://scholar.google.com/scholar?hl=en&q=The%20Victoria%20Province%20drought%20relief%20project%20II.%20Borehole%20yield%20relationships
http://dx.doi.org/10.1111/j.1745-6584.1988.tb00407.x
https://scholar.google.com/scholar?hl=en&q=
http://dx.doi.org/10.1016/j.sbspro.2013.06.643
https://scholar.google.com/scholar?hl=en&q=The%20hydrogeology%20of%20crystalline%20basement%20aquifers%20in%20Africa.%20In:%20Wright,%20E.P.%20and%20Burgess,%20W.G.%20(Eds),%20Hydrogeology%20of%20crystalline%20basement%20aquifers%20in%20Africa
https://scholar.google.com/scholar?hl=en&q=The%20hydrogeology%20of%20crystalline%20basement%20aquifers%20in%20Africa.%20In:%20Wright,%20E.P.%20and%20Burgess,%20W.G.%20(Eds),%20Hydrogeology%20of%20crystalline%20basement%20aquifers%20in%20Africa
http://dx.doi.org/10.1144/gsl.sp.1992.066.01.01
https://scholar.google.com/scholar?hl=en&q=Borehole%20siting%20in%20an%20African%20accelerated%20drought%20relief%20project.%20In:%20Wright,%20E.P.%20and%20Burgess,%20W.G.%20(Eds),%201992.%20Hydrogeology%20of%20crystalline%20basement%20aquifers%20in%20Africa,
http://dx.doi.org/10.1144/gsl.sp.1992.066.01.09
https://scholar.google.com/scholar?hl=en&q=Constraints%20on%20the%20exploitation%20of%20basement%20aquifers%20in%20East%20Africa%20–%20water%20balance%20implications%20and%20the%20role%20of%20the%20regolith
http://dx.doi.org/10.1016/0022-1694(92)90201-6
https://scholar.google.com/scholar?hl=en&q=DRASTIC-Fm:%20A%20modified%20vulnerability%20mapping%20method%20for%20structurally%20controlled%20aquifers%20in%20the%20southern%20Gulf%20Islands,%20British%20Columbia,%20Canada
https://scholar.google.com/scholar?hl=en&q=DRASTIC-Fm:%20A%20modified%20vulnerability%20mapping%20method%20for%20structurally%20controlled%20aquifers%20in%20the%20southern%20Gulf%20Islands,%20British%20Columbia,%20Canada
http://dx.doi.org/10.1007/s10040-006-0102-8
https://scholar.google.com/scholar?hl=en&q=Groundwater%20favorability%20map%20using%20GIS%20multicriteria%20data%20analysis%20on%20crystalline%20terrain,%20São%20Paulo%20State,%20Brazil
http://dx.doi.org/10.1016/j.jhydrol.2008.03.026

