Multicriteria Analysis by Codification for the Determination of Soil Landscape Units in Forest and Pre-Forest Zones of Cote D’Ivoire: The Case of the Square Degrees of M’Bahiakro and Daloa ()
1. Introduction
In Côte d’Ivoire, population growth and the need to secure the sustainability of economic (agricultural growth, energy needs), social and cultural processes mean that natural resources, particularly the soil, are increasingly being used [1] . With the importance of agricultural activities, the Daloa and M’Bahiakro degrees squares (located in the forest and pre-forest zones) contribute to Côte d’Ivoire’s economic development [2] . These areas have thus become a preferred destination for many people in search of land suitable for industrial crops (coffee, cocoa, rubber, oil palm, etc.). Cocoa production accounts for 40% of the world market [3] . Cocoa alone generates nearly 44% of export earnings, or over 15% of Gross Domestic Product [4] . Coffee and cocoa employ 60% of the working population. The sustainability of these two crops and the volume of exports are compromised by the ageing of orchards [5] , accentuated by the combined action of biotic [6] , climatic [7] and soil [8] hazards, which considerably reduce the economic life of orchards. Consequently, the sustainability of cocoa and coffee growing depends on sustainable soil management. Soil surveys are necessary to make the right choice of plants to grow and to initiate suitable cultivation techniques. Soil characterization based on pedological surveys requires knowledge of a large amount of multi-disciplinary, multi-source and often multi-format geoscientific information [1] . Processing this information requires mastery of tools and methods such as Spatial Reference Information Systems (SIRS), Remote Sensing (RS) and Multicriteria Analysis (MA). SIRS specialize in the capture, storage, spatial analysis and cartographic restitution of information. The availability and control of homogeneous, synoptic information in digital form that remote sensing provides, facilitates storage and manipulation operations, and makes it possible to cross-reference data from several sources. AM methods are designed to provide synthetic information, so that decision-makers are not overwhelmed by an excess of information [1] [9] [10] . This system is based on satellite images (Landsat Oli 8 and SRTM) and cartographic images. The aim of this work is to determine soil landscape units by applying the multicriteria analysis method by coding in these square degrees in order to reinforce soil survey tools using more suitable and efficient base maps.
2. Materials and Methods
2.1. Study Area
The study area covers two square degrees (M’Bahiakro and Daloa) according to the map of Côte d’Ivoire (Figure 1(a) and Figure 1(b)).
The M’Bahiakro square degree is located in central-eastern Côte d’Ivoire, between 7˚ and 8˚ north latitude and 5˚ and 4˚ west longitude, and covers some 12,000 km2. It has a humid tropical climate with a marked dry season and low winds [11] . The terrain is generally flat, with a maximum altitude of 200 m. Temperatures in the study area range from 14˚C to 33˚C, with humidity varying between 40% and 70% [12] . The area’s geology is composed of a complex of igneous, metamorphic and volcano-sedimentary rocks [13] . The flora is dominated in the pre-forest zone by shrubs interspersed with groves and forest islands. The Daloa square degree, with a surface area of around 12,000 km2, is located in central-western Côte d’Ivoire between 6˚ and 7˚ north latitude and 6˚ and 7˚ west longitude. The landscape is characterized by lower elevations (200 to 300 m) in a NE-SW direction, with tabular or rounded interfluves. Plains and low plateaus dominate the landscape, with a tropical climate and an average temperature of 25.6˚C. Average annual rainfall is 1317 mm/year [14] , with average monthly temperatures fluctuating between 24˚C and 28˚C. The geological bedrock of the Daloa square degree is characterized by Middle Precambrian formations, dominated by granites, migmatites and schists.
2.2. Data Used
The data used in this study concern, on the one hand, satellite images (Landsat OLI 8 and SRTM) and, on the other hand, geological maps of the areas covered.
![]()
Figure 1. Presentation of the square degrees of M’Bahiakro (a) and Daloa (b) according to the square degree map of Côte d’Ivoire.
Landsat images were used to map land use, while SRTM images were used to map relief in terms of slopes and altitudes. As for the geological maps, they were used to understand the geological structures of the areas studied. It should be noted that radiometric and atmospheric processing were used to enhance the quality of Landsat images.
2.3. Study Methodology
The methodological approach adopted in this study is based on the use of multi-criteria analysis methods to draw up soil-landscape maps [15] [16] [17] .
2.3.1. Identification of Criteria
According to Youan and et al. (2011), the identification of criteria is a decisive and delicate phase, which influences the quality of the information generated for decision-making [17] . Thus, with reference to the work of Youan (2008), a number of criteria are identified, selected and evaluated in order to achieve the fixed objective of determining cartographic units [16] . For this study, the criteria deemed most important are: slope, geology, land use and altitude.
2.3.2. Classification of Criteria
The criteria for identifying soil landscapes were classified on the basis of work carried out by Youan and et al. (2015) [18] . In this method, codes were assigned to the different criteria classes (sub-criteria). Geology was coded as 10, 20, 30, 40, taking into account the chemical composition of geological structures. Codes 100, 200, 300, 400 for the slope criteria and codes 1000, 2000, 3000, 4000 for the relief criteria. Meanwhile, codes 1, 2, 3, 4 and 5 have been assigned to land cover according to its predefined classes. Combining these codes or classes into a single code will define a soil landscape type. In this methodology, the ranking order of the criteria is not mandatory (Table 1 and Table 2).
2.3.3. Combination of Criteria
In order to identify and map soil landscape units, it was necessary to combine all these previously codified criteria. To facilitate interpretation of the codified soil landscape map, an interpretation was made by combining the codes derived from the reclassification of the following criteria: geology, relief, slope and land use. For example, code “1132” was used to define a soil landscape with codes “1000”, “100”, “30” and “1” corresponding to a type of geology (30 equals mafic rock), relief (1000 equals valley), slope (100 equals gentle slope) and land use (2 equals degraded forest).
3. Results
3.1. Analysis of the Spatial Distribution of the Major Factors Used in the Modeling Process
3.1.1. Landform Characteristics in the Square Degrees of M’Bahiakro and Daloa
Figure 2 shows that the relief is variable, consisting of a vast, monotonous
![]()
Table 1. Reclassification of soil landscape criteria by weighting (M’Bahiakro).
![]()
Table 2. Reclassification of soil landscape criteria by weighting (Daloa).
![]()
Figure 2. Relief of the square degree of M’Bahiakro (a) and Daloa (b).
peneplain with few irregularities over the whole of the study area. Altitude ranges from 50 to 649 meters. Altitudes are highest in the north, decreasing towards the center, where they are lowest. Slopes in the square degrees range from 0% to 44%. Gentle slopes cover almost all square degrees, with a proportion ranging from 77% to 93%. Medium slopes are between 5% and 8.5% and cover the territory moderately. As for steep slopes, they are located mainly approximately the study areas to the north and center. There are steep slopes to the north and south-east of the two square degrees. Low-slope areas correspond to valley relief. Medium-slope regions correspond to low- and mid-slope areas. Steeply sloping areas are characterized by plateaus, especially in the north and center of the square degrees (Figure 3).
3.1.2. Geological Characteristics of the M’Bahiakro and Daloa Square Degrees
The geological map shows a great diversity of surface formations, with four (4) formations. Sedimentary rocks (with 31.60% of surface area) in Daloa, result from the accumulation of various sediments (clasts: mineral fragments, shell fragments, etc.). Intermediate rocks account for a very small proportion (4.33% of surface area) and are composed of andesites and dacites. They are rich in aluminum, iron and silicon and are found around the Daloa area. Mafic rocks, made up of basic to intermediate lavas and pegmatite gabbro in Daloa, cover 63.93% of surface area, compared with 23.47% in M’Bahiakro, where these rocks are rich in ferromagnesium. As for acid rocks, they cover almost all (76.53%) of M’Bahiakro and are very sparsely represented in the Daloa zone (with 0.14% of surface area) (Figure 4).
![]()
Figure 3. Elevation map of the square degree of M’Bahiakro (a) and Daloa (b).
![]()
Figure 4. Geological map of the square degree of M’Bahiakro (a) and Daloa (b).
3.1.3. Land Use in the Square Degrees of M’Bahiakro and Daloa
Figure 5 shows the land cover with five (5) classes: forest/degraded forest, sa-vane, crop/fallow, habitat/bare ground and water. The savannah class is the most predominant (51.57%) throughout the M’bahiakro area. The same is true of degraded forest in Daloa (61.67%). The forest class is weakly concentrated in the south and south-east of the study areas. In addition, on average, all square degrees are covered by crop/fallow land, which is more concentrated in the North and Centre. The habitat and bare soil classes are sparsely distributed throughout the study areas. Finally, the water class is poorly represented in almost all zones.
![]()
Figure 5. Map of land use in the square degree of M’Bahiakro (a) and Daloa (b).
3.2. Soil Landscape Mapping by Coding
Figure 6 and Figure 7 show the result of combining the four (4) criteria that model the soil landscape. The result is twenty-seven (27) soil landscapes for the M’Bahiakro area.
Of these, the most dominant are acidic rocks on a moderate relief under savannah and crop/fallow land. After that, mafic rocks make their appearance with a moderate relief under forest, another type of land use. The Daloa area comprises forty-one (41) soil landscapes.
At this level, the different soil landscapes are dominated by mafic rocks with a moderate relief under degraded forests and crops/fallow land. Next come the sedimentary rocks with an equally moderate relief under forests/degraded forests and crops/fallow land. Lastly, intermediate rocks with moderate relief under forest/degraded forest.
4. Discussion
The implementation of GIS based on the multicriteria analysis method using codification has made it possible to determine soil landscape units. Four (4) criteria (geology, relief, slope and land use) that best define soil landscapes were taken into account in the GIS implementation. These criteria influence the spatial distribution of soil landscapes in both forest and pre-forest environments. Eimberck M. and Joly B. (2008) have used these four (4) soil landscape criteria in the digitization of medium-scale soil surveys in France [19] . Indeed, the influence of geology on pedogenesis has long been demonstrated [20] [21] . The authors conclude that a pedological study resulting from the alteration of rocks presents great difficulties due to their heterogeneity, responsible for the variation in facies.
![]()
Figure 6. Soil landscape map by coding of M’Bahiakro.
![]()
![]()
Figure 7. Soil landscape map by coding of Daloa.
The results obtained show a total of one hundred and three (103) soil landscapes for the M’Bahiakro square degree and one hundred and thirty (130) soil landscapes for the Daloa square degree. It was therefore necessary to reduce the number of classes by grouping them together using the most relevant variables, as was done by Eimberck M. and Joly B. (2008) [19] . The result is twenty-seven (27) and forty-one (41) soil landscapes for the square degrees of M’Bahiakro and Daloa respectively. It should be noted that each soil landscape obtained provides information on the participation of the criteria involved in modelling the soil cover of the environment. The M’Bahiakro square degree shows a soil-landscape cover dominated by acidic rocks at medium altitudes under savannah; forest/degraded forest and cultures/fallow. Mafic rocks follow this on moderate relief under savannah, water and crops/fallow. The Daloa area, on the other hand, has forty-one soil landscapes, with mafic rocks dominating the soil landscape, on low and mid-slopes under degraded forest and crops/fallow. This is followed by sedimentary rocks at moderate altitudes under forest/degraded forest and crop/fallow land. The difference in the number and characteristics of soil landscapes between zones is related to the geological formations found there.
However, difficulties may be encountered in this methodology (codification). This does not guarantee reliable (100%) prediction of results. The first difficulty concerns the choice of factors. The large number of criteria involved in soil landscape modelling makes prediction difficult. This was noted by Jourda (2005) in their study of fissured aquifers in West Africa [15] . According to this study, the coding method does not guarantee 100% results. Also, the accuracy of the results is influenced by the hydrological and hydrogeological regimes that condition erosive dynamics and landform configuration. For this reason, it is need to integrate a comprehensive harmonization methodology into digital soil mapping, in which it is essential to involve expert soil scientists and cartographers [22] .
The second difficulty concerns the choice of rating scales ranging from 1 to 9, with their corresponding reciprocal weighting of criteria, as noted by Belton (1986) [23] . Indeed, the difficulty lies in the comparison of criteria, i.e., which criteria are the most important and the most discriminating according to the scales.
Finally, the last difficulty lies in taking into account the different aspects for the orientation of decisions. Multicriteria analysis alone is incapable of taking into account all aspects of spatially referenced decision problems. For this reason, Kêdowidé (2010) and Ouédraogo and et al. (2012) suggest coupling it with GIS to obtain better results [24] [25] .
5. Conclusion
At the end of this study, based on the application of the multicriteria analysis method using coding coupled with GIS, the analyses highlighted soil landscape units in forest (Daloa) and pre-forest (M’Bahiakro) zones. Twenty-seven soil landscapes were identified in the M’Bahiakro degree square. Acidic rocks are the most dominant in combination with different types of land use (savannah, degraded forest and crops/fallow). The Daloa area, on the other hand, has forty-one soil landscapes identified throughout the zone. Mafic rocks predominate on a moderate relief under different land uses (forest/degraded forest, water and crops/fallow). These results can be used as a guide for pedological prospecting.
Acknowledgements
The authors would like to express their gratitude to the “Conseil du Café et du Cacao” - CCC (project owner) and the “Fonds Interprofessionnel pour la Recherche et le Conseil Agricole” - FIRCA (delegated project owner) for funding the “Génération de Technologies pour l’amélioration de la productivité des vergers de cacacaoyers et de caféiers (2QC)” project, the Laboratory of Soil, Water and Geomaterials Sciences (LS2EG) of the Training and Research Unit of Earth Sciences and Mining Resources (UFR STRM) of the Félix Houphouët-Boigny University of Cocody-Abidjan and other partners for their support and backing for this study.