SOTER-map (SOil and TERrain) of south-west Niger
Frieder Graef and Karl Stahr
Department of Soil Science and Land Evaluation, University of Hohenheim, 70593 Stuttgart, Germany, e-mail: graef@uni-hohenheim.de
Keywords: Soil, terrain, SOTER, Niger,
remote sensing, land evaluation
1 Aims of the research
During the past two decades soil research in SW-Niger focussed mainly on deep sandy soils, which are of major interest for cropping to the smallholders. The studies were carried out on large scales (1:100 - 1:25.000) and on only few sites. However, today the whole region including also land with marginal potential for agriculture are used for pasture or cropping. Knowledge of chemical and physical qualities of shallow sandy or clayey soils on the "Continental Terminal" (CT) and of soils on the basement is poor. The only comprehensive soil survey in the study area was carried out by Gavaud and Boulet (1967) at 1:500.000 scale.
The objective of this study was to establish
an extensive and complete SOTER (Soil and Terrain Digital Database) - database
containing on one hand geomorphological maps and on the other soil, terrain
and climatic data. A SOTER map was also established in order to provide
a detailed base map of the allocation of soil and terrain resources in
SW-Niger, that can be subsequently used as a basis for land evaluation
but also for multiple other scientific questions e.g. to determine the
soil degradation status or land management options or by extension services
e.g. to delineate terrain-specific intervention areas.
Figure 1 (annotations) presents a scheme of the activities and tools applied for SOTER mapping and data retrieval. The SOTER data structure (ISRIC, 1993) was applied, which is organised at three different levels (Figure 2, annotations):
1) Terrain units (TU): General terrain
description such as major landform, general lithology. A terrain unit comprises
one or more terrain components.
2) Terrain components (TC): Detailed terrain
description with parameters such as surface form, surface drainage, slope
form and length. A terrain component comprises one or more soil components.
3) Soil components (SC): Detailed description
of soils with parameters such as erosion degree and rootable depth. For
evaluation of soil variability, the original SOTER structure (Figure 2),
allowing only one reference profile per soil component, was modified by
Weller and Stahr (1995), by introducing the term 'profile set' (compare
with figure 2: Modified Soter structure). A profile set contains a
free number of soil profile (point data) descriptions. A soil profile is
made up of a number of horizons which are described and analysed for chemical
and physical properties.
Secondary data i.e. documents and maps
containing SOTER-relevant data were collected in order to
a) get information of the biophysical
setting in the study area;
b) collect existing data to feed the SOTER
database from two types of documents: maps with accompanying explanations
(area data) and documents with soil profile descriptions and analysis (point
data with attributes);
c) locate areas with missing data; and
d) determine transect sites.
The types of existing maps are
a) topographic maps (IGN, 1965-1996) at
different scales (1:200.000 - 1:50.000);
b) soil maps (Gavaud and Boulet, 1967)
at 1:500.000;
c) geological maps from Greigert (1961)
at 1:1000.000 and from Machens (1966) at 1:200.000 and
d) a physiographic map at 1:100.000 (INRAN,
1977).
Only soil profiles with precise geographical positioning or accompanying maps and a mandatory set of profile descriptions and analyses (ISRIC, 1997) were integrated. For the database 250 profiles from 22 documents were considered suitable. The soils were classified according to ISRIC (1994) and attributed to their respective profile set. Own soil analyses complemented the secondary data so that the soil inventory comprises 475 profile data sets.
Since in arid and semi-arid regions the
relation between vegetation, soil and geomorphologic characteristics is
very distinct (Löffler, 1994), remote sensing was considered an essential
tool for SOTER mapping in Niger. Landsat TM5 scenes were chosen and the
following digital data were processed:
TM 193/50 full scene, (07-04-85);
TM 193/51 full scene, (03-02-88);
TM 192/51 full scene, (31-12-86);
TM 192/51 standard quarter scene (SW),
(25-4-94) and
TM 192/50 full scene, band 1-4, (18-09-92).
View an example
of a Landsat TM5 scene- explanation of numbers:
1) pediment from granite
2) Inland delta of the Niger
3) Schist crest
4) High-gradient slope on Continental
terminal sediments
5) Ironstone plateau with tiger bush
6) Secondary ironstone ledges
7) Terrace of the Niger
Scenes from dry seasons were chosen to reduce interference of vegetation cover. For data processing the raster-based geographic analysis system IDRISI (Eastman, 1996) was used.
Further image treatment (e.g. equalisation, filtering, overlay...) was carried out with Adobe Photoshop. The satellite data were geometrically corrected with IDRISI both from ground control points taken with a Global Positioning System (GPS) and from distinct features (e.g. plateau edges) on topographic maps. Image processing included linear stretching, median filtering and hue saturation. From the channel combinations 7-4-3, 7-4-1 and 3-2-1 (all red-green-blue) tested for false colour composites, the combination 7-4-3 was considered most suitable for recognising geomorphologic features.The false colour images were interpreted visually on-screen and on printouts (1:50.000) by comparison with the above mentioned thematic maps and groundtruth data. Aerial photographs and several overflights with a microlight plane were used to better understand the relationships between landscape morphology, soil surface characteristics and corresponding satellite image features. On this occasion some transects already investigated in the field were checked again from the air.
Soil-geomorphic transect investigations
of several kilometres length have been chosen to obtain
a) an extensive inventory of the soils
and their distribution in terrain units and components;
b) semi-quantitative information about
soil type occurrence and
c) spatial information about present land
use and degradation features.
The transects were selected and placed
with the aid of satellite images and the above mentioned maps using the
criteria
a) locations, where differentiation between
geomorphologic features in the satellite images is feasible,
b) minimum size (> 500 m) of geomorphologic
features,
c) minimum number of at least 3-4 terrain
component samples, to cover the spatial variability,
d) sufficient accessibility,
e) wide distribution across the study
area in order to cover potential climatic effects and achieve representative
terrain information in the study area,
f) location both near and far from main
roads and villages in order to recognise former mapping errors, reduce
own potential mapping errors due to poor terrain accessibility and include
at the same time the range of land use types and intensities.
More information on transect positioning
and realisation is provided by Graef et al. (1998).
During transect investigations the terrain, soil and surface features were investigated following the SOTER manual. Terrain features, crops and vegetation were described within a 25 m distance from the auger point, whereas soil surface parameters, more closely related to their respective soil type, were restricted to a 5 m distance. In addition to the augerings, pits 30-100 cm deep were opened as profile face for adequate horizon descriptions and sampling for further soil analyses. Soil descriptions and classification were made according to the 'Guidelines for Soil Profile Descriptions' (FAO, 1990) and the FAO classification (ISRIC, 1994). A special focus was put on soil crusting and on other soil degradation features. Soil surface colours were determined with the Munsell colour charts for later matching with satellite data (Pouget et al., 1990).
The mapping of TUs and TCs with specific
characteristics was based on the criteria shown in Table
1 (annotations). The observed close relationships between soil, soil
surface and geological substrate supported the mapping by visual interpretation
of satellite images but also extensive groundtruth data from reconnaissance
tours and field transects were consulted. Mapping units of the physiographic
map (INRAN, 1977) were partly regrouped, new units were inserted and unit
boundaries adjusted. In areas not covered by that map the delineation was
based on satellite images, topographic maps (1:50.000) and more extensive
groundchecks and transect studies. The maps were digitised at 1:50.000
- 1:100.000.
Following the criteria for delineation
of SOTER units and the procedures for data retrieval 17 terrain units with
53 terrain components were distinguished (Table 2). The SOTER map shows
a landscape dominated by plateaux or hills and by large sand-covered pediments.
The landscape is divided into 4 major units: Three areas determined by
the geological substrate (Continental Terminal, granitic basement and sedimentary
rock), and one heterogeneous area dominated by recent or subrecent processes
(fluvial and eolian systems) [compare pictures in the
annotations] Among the first three areas a toposequential slope scheme
(INRAN, 1977) was used to subdivide the land into TUs: a) plateaux or hills
b) high-gradient slopes ( 3 % and c) low-gradient slopes < 3 % (Stahr
et al., 1996). The TCs were distinguished for their slope, morphogenesis,
relief, eolian sand cover, lithology and hydrological properties (Table
1, annotations). The TU and TC coverage in the study area is presented
in Table 2. More details of the SOTER map legend are given in the annotations.
Table 2: Legend and area (%) of terrain
units and terrain components in the study area
| TUs and TCs |
(%) |
TUs and TCs |
|
| Continental terminal: | (Hs2) high-gradient 8-20 % |
|
|
| (P) Plateaux |
|
(Hs3) high-gradient >20 % |
|
| (P1) flat |
|
(Hs4) Hs1-3 with rock outcrops |
|
| (P2) gently undulating (slopes <3 %) |
|
(Ls) Low-gradient slopes < 3 % |
|
| (P3) undulating (slopes 3-8 %) |
|
(Ls1) formed through abrasion |
|
| (P4) highly dissected |
|
(Ls2) with residual hills or strong relief |
|
| (P5) lower level terrace |
|
(Ls3) from alluvio-colluvial deposits |
|
| (P6) P1-5 with eolian deposits (0.1-1 m) |
|
Fluvial and eolian systems: | |
| (Hc) Higher-gradient slopes ³ 3 % |
|
(De) Depressions |
|
| (Hc1) medium-gradient <8 % |
|
(De1) on or between (P) |
|
| (Hc2) high-gradient 8-20 % |
|
(De2) between other CT units |
|
| (Hc3) very high-gradient >20 % |
|
(De3) on basement |
|
| (Hc4) Hc1-3 with rocks dominating |
|
(De4) with sand deposits ³ 1m |
|
| (Lc) Low-gradient slopes < 3 % |
|
(Da) Dallol |
|
| (Lc1) formed through abrasion |
|
(Da1) dunes and eolian deposits ³ 2 m |
|
| (Lc2) with residual hills and strong relief |
|
(Da2) zones with high ground water level |
|
| (Lc3) from alluvio-colluvial deposits |
|
(E) Eolian Systems |
|
| Basement Complex (granites) | (E1) fixed dunes |
|
|
| (B) Hills or plateaux |
|
(E2) complex of dunes with depressions |
|
| (B1) rocky ridges and inselbergs (<30m) |
|
(Cc) Combination of (E) and CT units |
|
| (B2) plateaux |
|
(Cc1) eolian deposition on "P" ³ 1m |
|
| (Hb) Higher-gradient slopes ³ 3 % |
|
(Cc2) eolian deposition on "Lc" ³ 1m |
|
| (Hb1) medium-gradient <8 % |
|
(Cc3) eolian deposition on "Hc" ³ 1m |
|
| (Hb2) high-gradient 8-20 % |
|
(Cb) Combination of (E) and Basement |
|
| (Hb3) very high-gradient >20 % |
|
(Cb1) eolian deposition on "Lb, Ls" ³ 1m |
|
| (Hb4) Hb1-3 with rock outcrops |
|
(Cb2) eolian deposition on "Hb, Hs" ³ 1m |
|
| (Lb) Low-gradient slopes < 3 % |
|
(T) Terraces |
|
| (Lb1) formed through abrasion |
|
(T1) fluviatil terraces |
|
| (Lb2) with residual hills or strong relief |
|
(T2) fluviatil islands |
|
| (Lb3) from alluvio-colluvial deposits |
|
(T3) T1-2 with sand deposits (³ 1m) |
|
| Basement (sedimentary rocks) | (A) Recent fluvial systems |
|
|
| (S) Hills or plateaux |
|
(A1) narrow valleys with sand deposits |
|
| (S1) rocky ridges (<30m) |
|
(A2) large valleys with sand deposits |
|
| (S2) small hills (30-60m) |
|
(A3) alluvial seasonally flooded riverplain |
|
| (S3) plateaux |
|
(A4) valleys with loamy or clayey deposits |
|
| (Hs) Higher-gradient slopes ³ 3 % |
|
(R) Riverbed |
|
| (Hs1) medium-gradient <8 % |
|
(R1) major riverbed |
|
Overview of TUs and TCs
41 % of the study area consists of pediments of which 22 % have deep cover sands (Cc2, Cb1). These cover sands occur on 49 % of the total area, however even more TCs such as Lc1, Hc1, P5 include patches with thin sand deposits, so that the potentially sand-influenced area may be extended to 82 %. The predominance of the eolian mantle is also reflected by the large size of sand-covered TCs such as A2, Da2, Cb1, Cc2. The proportions of geological areas are 4 : 1 : 1 for CT, granitic and sedimentary basement respectively. The plateaux surface (24 %) in the CT landscape (P and Cc1) of which 9 % is sand covered (P5, Cc1), exceeds the pediment surface (20 %). Hills and plateaux on the basement (S, B), however, account only for 1 % of the surface compared to 18 % of pediment plains (Lb, Ls, Cb1). This is due to the disappearing plateau-forming CT-sediments in the east of the study area (Machens, 1973). Areas influenced by water and sediment influx such as A, De, R, Da2 cover 9 % of the area, however only 3 % (R, De, A2, Da2) is continually supported with water during more than 2 months. The CT showed the following land surfaces: pediments (46 %), plateaux (34 %), hillslopes (13 %) and valleys (7 %). These ratios were confirmed by remote sensing results of d'Herbès and Valentin (1997).
The relief intensity of geological areas can be quantitatively expressed by the quotient of high-gradient slopes per geological area: The relief intensity sequence then is sedimentary basement (0.71) > granitic basement (0.36) > CT (0.22). Not only the area of high-gradient slopes but also pediment slopes tend to be higher on the sedimentary basement. The high slope SDs even on relatively flat TUs such as P, Lc, Da show that the landscape although quite plain has an highly undulating and sloping mesorelief. The types of slope form are mostly concave or uniform. From satellite images the CT landscape appears significantly less dissected that the basement area. This is caused by the different substrate resistances (Löffler, 1994).
Due to the climatic and relief erosion force most terrain surfaces are covered with eolian deposits, ironstone or basement material. The surface drainage is usually well to very rapid, exept for the depressions or fluvial areas receiving additional water. This is an effect of sloping relief, dissection, substrate texture and short intensive rainfalls.
With respect to data retrieval and mapping
the frequent, extended or heterogeneous TUs and TCs were investigated more
intensively (e.g. Lb2, Hc1, Cc2), compared to rare (B1, S2), small (Hc2,
Hc3) or homogeneous (P1, P5) land units. Small TCs, especially high-gradient
slopes, were probably underestimated because they were often merged with
larger TCs. Due to the lack of terrace map data, the CT terrace formations
were assigned to the CT-plateaux (P) of similar morphology and the Holocene
to Pleistocene formations were grouped together.
Overview of soils
41 soil types represented by profile sets were morphologically distinguished in the field. A detailed description of the occuring soils is available in Graef (submitted). For an overview the following shares based on estimations from the transects can be given: deep sandy soils (38 %), soils with thin sand cover (22 %), loamy soils with clay translocation (7 %), loamy soils (6 %), shallow stony soils (20 %) and clayey soils (5 %). This generalised delineation of soil types for physical properties was further specified through chemical factors such as pH, N, P, exchangeable bases and CEC. With respect to these chemical factors six major categories were presented in a fertility sequence: loamy alluvial soils > loamy basement soils > loamy CT soils > sandy alluvial soils > sandy soils overlying basement substrate > sandy soils on dunes or overlying CT substrate.
Soil chemical and physical properties were
found to vary considerably within and between SCs throughout the landscape
but also at very detailed scale (Gavaud 1977, Hammer 1994). The outcome
contradicts the prevailing view of the Sahel being a uniform region with
deep acid sandy soils. It needs to be emphasised that a large variety of
soil feature combinations exists, which was hitherto probably underestimated,
although indicated in the past e.g. Gavaud (1966, 1977).
Methodical results
Difficulties encountered with the physiographic
maps of INRAN (1977) included missing projection type (UTM31 was assumed),
poor quality of map sheets and few wrong designations of map units. Nevertheless,
the detailed terrain description and delineation was highly appreciated
and integrated into the SOTER map after corrections through ground-truthing
and comparison with other thematic maps.
Towards the south (below 12°30' latitude)
and in areas with high fallow shares, where the vegetation cover increasingly
hampered clear unit delineations via satellite images, mapping decisions
were supported by groundchecks and topographic maps (IGN 1965-1996). The
Landsat TM resolution of 30 m and the thematic maps enabled a mapping accuracy
sufficient for the 1:200.000 scale. Correlation tests and groundchecks
carried out for soil and terrain features showed that they were sufficiently
defined by their surface features for delineation via visual image interpretation.
An extensive SOTER database was created at 1:200.000 scale for an area of 24.000 km² in SW-Niger. It provides an overview of the soil and terrain resource allocations. Soil chemical and physical properties were found to vary considerably throughout the landscape but also at very detailed scale, which contradicts the prevailing view of rather uniform soils in SW-Niger. Both SOTER database and map can be updated and extended to further areas if necassary. The SOTER database and map can be used for subsequent elaboration of sustainable land use scenarios and for crop suitability evaluations. It can also be connected to economic models that refer to terrain-related production potentials.
The application of the presented methodology
and the three-level data structure in SW-Niger has proved that this is
an effective approach for retrieval and management of numerical and spatial
soil and terrain data. The SOTER map and the SOTER database offer the possibility
of relatively precise spatial estimations of terrain related factors such
as land use, soil types and soil factors in SW-Niger. It also allows extrapolation
of research results to different scales. It is noted, however, that soil
coverages are still estimations based on the transect results and that
soil variability is often higher than can be registered on the transects.
Remote sensing specially in this semi-arid region is an essential tool
for surveying of the main features relevant for SOTER.
D'Herbès, J.M. and Valentin C., 1997: Land surface conditions of the Niamey region: ecological and hydrological implications. J. Hydrol. 188-189: 18-42.6 Further ReadingsEastman J. R., 1996: Idrisi for Windows Manual. Clark Univ., Worcester, MA, USA.
FAO, 1990: Guidelines for soil profil description, 3rd edition, FAO, Rome. 70p.
Gavaud, M. and Boulet, R. 1967: Carte Pédologique de reconnaissance de la République du Niger 1:500000, Feuille Niamey, ORSTOM. Paris.
Gavaud, M., 1966: Etude pédologique du Niger Occidental. Rapport général Tome II+III. Monographie des sols. Centre ORSTOM de Hann, Dakar. 523p.
Gavaud, M., 1977: Les grands traits de la pédogenèse au Niger Méridional. Travaux et documents de l'ORSTOM. Paris. 102p.
Graef, F., van Duivenbooden, N. and Stahr, K., 1998: Remote sensing- and transect-based retrieval of spatial soil and terrain (SOTER) information in semi-arid Niger. J. Arid Environments 39: 631-644.
Graef, submitted: Evaluation of agricultural potentials in semi-arid SW-Niger - (NiSOTER) A soil and terrain study, PhD-thesis, Hohenheim University. 200p.
Greigert, J. 1961: République du Niger. Carte géologique de reconnaissance du Bassin des Iullemeden 1:1 Mio. BRGM, Niamey, Niger.
Hammer, R., 1994: Bodensequenzen und Standorteigenschaften im Südwest-Niger/Westafrika. Hohenheimer Bodenk. Hefte 22. Universitat Hohenheim. 145p.
IGN, 1965-1996: Carte de l'Afrique de l'Ouest au 1:50 000, Republique du Niger. IGN. Paris.
INRAN, 1977: Carte des unités physiographiques de la région agricole du Sud-Niger, PNUD/FAO, Niamey, Niger.
ISRIC, 1993: Global and national soils and terrain digital databases (SOTER). Procedures Manual. Wageningen. 115p.
ISRIC, 1997: Soil map of the world, revised legend with corrections. Technical Paper 20, ISRIC, Wageningen. 140p.
Löffler, E., 1994: Geographie und Fernerkundung. Eine Einfuhrung in die gograpische Inerpretation von Luftbildern und modernen Fernerkundungsdaten. Stuttgart. 251p.
Machens, E. 1966: Carte géologique du Niger Occidental 1:200000, BRGM, Paris.
Pouget, M., Madeira, J., Le Floch, E. and Kamal, S. 1990: Charactéristiques spectrales des surfaces sableuses de la région cotière nord-ouest de l'Égypte: application aux données satellitaires SPOT. In: M. Pouget (Ed.) Caractérisation et suivi des milieux terrestres en régions arides et tropicales. ORSTOM, Paris: Colloques et séminaires. 198p
Stahr, K., Bleich, K.E. and Graef, F., 1996: A geomorphological approach using satellite images for landuse planning in the Ct3 and Liptako (SW-Niger). pp. 53-82 in Standortgemäße Landwirtschaft in Westafrika. Arbeits- und Ergebnisbericht 1994-1996, SFB 308: Hohenheim University (ed.). Stuttgart.
Weller, U. und Stahr K., 1995: Eine Standortskarte fur Sudbenin - Erfassung von Geländeeigenschaften und Bodenparametern. Mitteilungen der Deutschen Bodenkundlichen Gesellschaft 76: 1221-1224.
Brabant, P., 1993: Pédologie et système d'information géographique. Comment introduire les cartes de sols et les autres données sur les sols dans le SIG? Cah. ORSTOM, sér. Pédol., vol, XXVIII, no 1: 107-135.Lopez-Blanco, J. and Villers-Ruiz, L. 1995: Delineating boundaries of environmenental units for land management using a geomorphological approach and GIS: a study in Baja California, Mexico. Remote Sens. Env. 53: 109-117.
SOTER newletter: http://www.isric.nl/sn10.html
INTERNATIONAL
SOIL REFERENCE AND INFORMATION CENTRE
Global
and National Soils and Terrain Digital Databases (SOTER)
ISRIC, P.O. Box 353, 6700 AJ Wageningen,
The Netherlands
Figure 1: Flowchart with methods and interactions of activities for SOTER mapping
Figure 2: SOTER data structure
Picture 4: 1) Schist crest, 2) Degraded soils on schist, 3) Cover-sands on schist with deep gullies
Table1: Criteria
for delineation of SOTER units in SW-Niger
| Terrain units level | Terrain components level |
| 1. Geological substrate,
lithology
2. Physiography, major landform 3. Slope 4. Type and degree of dissection 5. Eolian deposits |
1. Physiography
2. Morphogenesis 3. Type of fluvial deposits 4. Mesorelief 5. Eolian deposits 6. Lithology 7. Hydrological properties |
Detailed description of SOTER map legend
Continental terminal
(P): The level, only slightly dissected plateaux in the CT account for 22 % of the land. They consist of a gravelly ironstone carapace overlying the clayey CT sandstone strata and are covered with mostly degraded tiger bush. Their TCs are distinguished for mesorelief, dissection, origin and eolian deposits.
* (P1): Flat plateaux with almost level
surface.
* (P2): These plateaux types are gently
undulating.
* (P3): Undulating plateaux with slopes
3-8 % and strong vegetation in the depressions.
* (P4): Small, narrow and highly dissected
plateaux with sparse vegetation.
* (P5): Lower-level ironstone ledges with
escarpment.
* (P6): Plateaux (P1-P5) with thin eolian
deposits (0.1-1m), that often have a microdune relief.
(Hc): The higher-gradient slopes (3 % account for 3 % of the land. They mediate between the breakaway scarp of (P) and the lower-gradient slopes (Lc). They are concave, have ironstone debris tongues reaching out downslope and are generally severely eroded and pierced with deep gullies. The scarp, usually very steep and covered with dense bush vegetation, is often too small for separate mapping and therefore merged with the dominant slope. Sand deposits are common.
* (Hc1): Most CT slopes consist of medium-gradient
slopes 3-8 %.
* (Hc2): The high-gradient slopes 8-20
% are located at the breakaway scarp.
* (Hc3): Slopes >20 % occur along the
deeply intersected Niger valley and the breakaway.
* (Hc4): Slopes (Hc1-3) with dominating
ironstone rock cover.
(Lc): The low-gradient slopes <3 % account for the large pediment areas (9 %) located between valleys (A) and higher-gradient slopes (Hc). They have an undulating mesorelief with intercalated ironstone ledges. Sand cover (<1m) is common, but surficial ironstone and other CT substrates are dominating. The TCs were distinguished with respect to their origin and morphology.
* (Lc1): Relatively flat pediments formed
through hydraulic abrasion.
* (Lc2): Slopes with residual hills (plateaux)
and strong relief by colluvial ironstone outcrops.
* (Lc3): Undulating slopes formed from
alluvio-colluvial deposits with ironstone.
Basement complex (granites)
(B): The hills or plateaux are small and rare granite outcrops.
* (B1): The rocky ridges of rounded blocks
and inselbergs (<30m) show a convex relief.
* (B2): Plateaux with an ironstone carapace
from granitic alteration products on top of granite have a highly dissected
escarpement and a slightly undulating relief.
(Hb): The higher-gradient slopes (3 % are
found adjacent to the plateaux (B2). Their steep ironstone scarp is merged
with the overall slope. They are concave and have an undulating mesorelief.
Similar to (Hc) their often colluvial and sand-covered substrate is highly
dissected with gullies. The TC features of Hb1, Hb2, Hb3 and Hb4 correspond
to those of Hc.
(Lb): The low-gradient slopes <3 % represent
large, often undulated pediments located between the valleys and higher-gradient
slopes. Sand-cover is common, but granite or pre-weathered colluvial material
is dominating. The TC features of Lb1, Lb2 and Lb3 correspond to Lc. The
undulating relief of the Lb3 slopes reflects different substrate mixtures
of pre-weathered granitic and lateritic deposits.
Basement complex (sedimentary rocks)
(S): The hills or plateaux from shist are more distinct, higher and frequent than those from granite (B). They may extend to several km, often showing a spectrum of differing sedimentary rocks. Their TCs are distinguished concerning relief intensity, morphology and substrate.
* (S1): Lower rocky shist ridges (<30m)
or outcrops.
* (S2): Higher rocky hills (30-60m).
* (S3): Plateaux with ironstone carapaces
from shist alteration products overlying the bedrock have dissected escarpements
and undulating mesoreliefs.
(Hs): The higher-gradient slopes (3 % are
located between the plateau escarpments or rocky outcrops (S) and the pediment
plains (Ls). They are concave, undulating and dissected by gullies. Their
often colluvial and sand-covered substrate is mixed with varying amounts
of ironstone leading to mosaic-like soil patches. The TC features of Hs1,
Hs2, Hs3 and Hs4 correspond to those of Hc.
(Ls): The low-gradient slopes <3 % represent
large rather flat pediments, located between the valleys and higher-gradient
slopes. Sand cover may occur, but shists and colluvial lateritic material
are dominating. The TC features of Ls1, Ls2 and Ls3 correspond to Lc.
Fluvial and eolian systems
The following TUs are characterised and distinguished by more recent deposits and processes.
(De): Depressions seasonally flooded and accompanied with dense vegetation occur rarely (0.5 %). They are distinguished for their sediment texture and origin: clayey-loamy, sandy or both alternating.
* (De1): Depressions between CT plateaux
are filled with clayey-silty sediments. They may extend from few metres
(pseudodolines) to several km.
* (De2): Depressions between Hc and Lc
are located on the pediments and filled with fine sediments. Their drainage
is often interrupted by eolian sand deposits.
* (De3): Depressions on basement filled
with clayey sediments are often created by recent eolian sand barriers.
* (De4): Depressions with sand deposits
>1m depth occur between sand dominated TUs.
(Da): The Dallol Bosso, filled predominantly
with sandy alluvium and low dunal sand deposits has spatially alternating
hydromorphic conditions. The TCs are specified by morphology and groundwater
level.
* (Da1): Undulating eolian deposits >2m
partly from redistributed alluvium.
* (Da2): Zones with high groundwater level
are often ancient clayey water channels.
(E): The Eolian Systems are marked by high ((5-45 m) dunal formations that prevail in the north of the study area. Their TCs are distinguished by their morphology and relief.
* (E1): The fixed dunes are typical longitudinal
(NNW-SSE) dune bands.
* (E2): Large dunal complexes with small
depressions and low relief.
(Cc): The Combination of (E) and CT is common (20 %), where eolian sand overlies CT substrate to (1m depth sometimes with microdunes. The CT substrate may influence the upper soil properties. Their TCs are specified by the morphology of the underlying TU.
* (Cc1): Eolian undulating deposits on
plateaux (P), coexist with sand deposits in lower terrain positions.
* (Cc2): Eolian deposits on CT-pediments
(Lc) with undulating or low dunal relief.
* (Cc3): Eolian deposits on CT-slopes
(Hc), usually highly eroded, are referred to as sandy skirts.
(Cb): A Combination of (E) and Basement
(9 %) is given when eolian sand overlies the basement to (1m depth. The
basement substrate often influences the upper soil properties. Their TCs
are specified by the morphology of the underlying basement TU.
* (Cb1): Eolian deposits on pediments
(Lb, Ls) have an undulating to low dunal relief.
* (Cb2): Eolian deposits on high-gradient
slopes (Hb, Hs), usually highly dissected and eroded, are referred to as
sandy skirts.
(T): Flat fluvial Pleistocene to Holocene Terraces occuring along the Niger river are often covered with recent sand deposits. Their TCs are distinguished by terrain position, sediment texture and eolian superposition.
* (T1): The fluvial terraces are composed
of silty-clayey sediments.
* (T2): The fluvial islands, composed
of sand deposits, often have a basement rock core.
* (T3): Eolian sand deposits (1m depth
on T1 and T2.
(A): The Recent fluvial systems account for 7 % of the land. They refer to the river Niger, its tributaries and the valleys including slopes. The TCs are distinguished with respect to different inundation periods and sediment types.
* (A1): Narrow valleys with sand deposits
are most common (3 %). They are inundated only for some hours after rainfall.
* (A2): Large (0.5-2 km) sand-covered
valleys are located between the wide CT-pediments (Lc, Cc2). They are inundated
only for a few hours after rainfall except for local depressions bearing
water during the entire rainy season.
* (A3): The alluvial seasonally inundated
Niger riverplain with clayey-silty deposits.
* (A4): Valleys in the basement area only
have loamy or clayey deposits, where eolian sand influence is weak. They
have almost continually flowing water during the rainy season.
(R): The Riverbed of the Niger is flooded
for 3-5 months of the year. One TC: (R1).
Includes the following data (if available) compressed into one "zip" file:
- GIS: layers (ArcView, *.shp), projects (ArcView project files, *.apr) and legends (ArcView legends *.avl)
- source data: tab delimited text (*.txt), excel5 (*.xls), rich text format (*.rtf)
metadata: text file (*.txt)