During the late Pleistocene and the Holocene the eas-
 tern Sahara was subject to dramatic climatic changes 
oscillating from very arid to humid and finally increa-
 sing arid conditions again (Pachur 1987; Kröpelin 1987; 
1993; Kuper and Kröpelin 2006). Holocene pluvial con-
 ditions began about 9800 BP, at the end of the Younger 
Dryas cold interval. During the climate optimum the re-
 constructed annual precipitation was between 400 mm 
(Kröpelin 1993) and 600 mm (Vernet 1995). As a conse-
 quence, between 9500–3500 BP the southeastern Saha-
 ra was characterized by a raised groundwater table and 
numerous lakes (Kröpelin 1993). Analysis of about 320 
charcoal samples from prehistoric sites in the eastern 
Sahara furnishes evidence for a fundamental change 
of vegetation during the Early and Middle Holocene. 
Around 7000 BP the Sahelian vegetation zones were 
500–600 km north of their present range (Neumann 
1989). A savanna-type vegetation existed that attracted 
Sudano-Sahelian fauna and therefore human groups to 
the region until about 5000 BP, when the current epi-
 sode of hyperaridity ensued (Haynes 2001; Kuper and 
Kröpelin 2006).
 There is evidence that the prehistoric inhabitants 
adapted their economic systems to the changing cli-
 matic conditions in the Wadi Howar region, one of the 
main research areas of the ACACIA-project of the Uni-
 versity of Cologne (»Arid Climate Adaption and Cultu-
 ral Innovation in Africa«). Here, within 3 000 years the 
prehistoric inhabitants developed from gatherers and 
hunters to pastoralists with intensive cattle herding. 
In response to increasingly arid conditions goats and 
sheep were finally added to the herds (Keding 1998; 
Keding and Vogelsang 2001; Jesse 2006a; 2006b). To-
 day, the Wadi Howar region is very inhospitable and 
only a few nomads with their camels thrive there, but 
traces of the prehistoric occupants are visible in many 
places (e. g. Jesse 2005; 2006a; 2006b). Not only ar-
 chaeological material such as pottery sherds, lithic ar-
 tefacts and faunal remains can be found but also more 
subtle traces which can be detected for example by 
pedological analysis. The study of soils does not only 
give indications on climatic changes but can also show 
the effects of human impact on the landscape. In this 
paper, the results of pedological studies made during 
the 2006 archaeological field season of the ACACIA 
project in Lower Wadi Howar are presented.
 Study area
 The Wadi Howar is located in the north-west of the Su-
 dan and traverses over a length of about 1 000 km the 
southern margin of the Sahara between Jebel Marra in 
the west and the Nile Valley in the East. Once it was 
a tributary of the river Nile and joined it north-west of 
Ed Debba (Pachur 1987; Pachur and Kröpelin 1987; see 
fig. 1). Today, the Wadi Howar region is located within 
the two 100 mm rainfall isohyets at 17° N and 31° N 
(Ritchie et al. 1985) and is part of the hyperarid core of 
the eastern Sahara desert with partly less than 25 mm an-
 nual rainfall and a potential evaporation of 5–6 m (Petit-
 Maire and Guo 1998). Vegetation in the eastern Sahara 
is very sparse and mainly confined to a few oases and 
to wadi channels, where single Tundub-bushes (Cappa-
 ris decidua), acacia trees and transient communities of 
annuals and herbaceous perennials are present. The area 
of Wadi Howar is mainly characterized by sedimenta-
 ry rocks (Nubian sandstone), aeolian sediments, fluvial 
and lacustrine deposits. Partly volcanic (basalt), plutonic 
(granite) and metamorphic (quartzite) rocks were found.
 During the last years, research of the ACACIA pro-
 ject in the Wadi Howar region focused on the lower 
section of Wadi Howar which comprises the 400 km 
between Jebel Rahib in the west and the Nile Valley in 
the east. Intense occupation during the Neolithic period, 
especially the 4th and 3rd millennia BC, could be evi-
 denced in the area of Abu Tabari and Conical Hill: more 
than a hundred sites were found and mapped. The sites 
are either located on slightly elevated former sandbanks 
or on top of up to 15 m high, artefact covered dunes 
of parabolic shape (Jesse 2003; in press; Lange 2005). 
Two sites in the area of Abu Tabari were selected for 
pedological studies: S02 / 52 and S02 / 28. Further inves-
 tigations took place near the fortress Gala Abu Ahmed, 
situated about 100 km west of the Nile (fig. 1).
 Material and Methods
 Sampling
 At each study location an assessment of topographic 
units and potential interesting soil profiles was perfor-
 med by an intense survey. Soil profiles were described 
Desert soils in Lower Wadi Howar (Northwest Sudan)
 Evidence of climate change and human impact during the Holocene
 ANDREAS GUNDELWEIN
according to the US Soil Taxonomy (Soil Survey Staff 
1999). Disturbed as well as partly undisturbed soil sam-
 ples were taken from each horizon. The description and 
sampling in the field was completed by laboratory ana-
 lysis of the soil samples.
 Carbon quantification
 Bulk density was determined according to Schlichting 
et al. (1995) by sampling two replicates of undisturbed 
soil material from each horizon in 850 cm3 stainless 
steel cylinders. The soil was air-dried and weighed with 
a precision balance (Sartorius, Göttingen).
 Total carbon (TC) of air-dried soil samples (< 2 mm), 
milled to < 63 µm, was determined with a CHN-Analy-
 ser (Vario EL III, Elementar, Hanau). To determine in-
 organic carbon (TIC) the samples were heated to 550° C 
for 6 h to destroy organic carbon before measurement. 
Organic carbon (TOC) was determined as the difference 
of TC and TIC:
  TOC = TC – TIC   (1)
 The size of the soil organic carbon stocks on a spatial 
basis was determined based on the carbon contents of the 
different horizons:
 	 ∑ C
 i 
= ∑ c
 i
  x l
 i
  x d
 i 
x 10  (2)
 where C
 i
  is the TOC content of a soil layer in kg m-2; 
c
 i
  is the TOC concentration in g 100g-1soil (< 2mm); 
l
 i
  is the depth of the layer in cm; and d
 i
  is the bulk den-
 sity in g cm-3.
 To determine total P concentrations, 10 ml of concentra-
 ted HNO
 3
  (13 M) were added to 0.5 g soil and heated to 
185° C for 20 min in a closed PTFE container. P concen-
 trations were measured photometrically by a continuous 
flow analyzer (Skalar, Erkelenz) with a detection limit of 
0.55 µM P, according to Murphy and Riley (1962).
 Conductivity
 10 g of soil was mixed with 25 ml of de-ionized water and 
stirred for 30 min by hand. The conductivity was determined 
at 20° C with a Cond.315 (WTW instruments, Weilheim).
 Andreas Gundelwein2
 Fig. 1: Study area eastern Sahara with selected sites.
3
 margin were studied. The results of the two profiles in 
the extreme upper and lower position (# 8 and # 10) 
are as follows (tab. 2a and 2b).
 Profile 8 on the hilltop showed a weakly develo-
 ped soil with a small accumulation of organic carbon 
between 2 and 65 cm below the surface (fig. 2). The 
sediment was well rooted and showed large phospho-
 rus contents, which decreased with depth. The soil was 
characterized as a Typic Torripsamment, according to 
US soil taxonomy.
 The thickness of the upper soil horizon with a mean 
organic carbon content of 0.12 % and a bulk density of 
1.63 g cm-3 was 0.6 m. It covered an area of approxima-
 tely 4 000 m2. So, the site contained an estimated amount 
of 4.7 t of organic carbon or 9.4 t organic material.
 OSL-Dating
 The sample for Optically Stimulated Luminescence da-
 ting (OSL) was collected from the centre-part of an un-
 disturbed soil layer during night and sealed in light-safe 
black plastic bags. Sample preparation involved drying 
and sieving, followed by the removal of acid soluble 
carbonates and organic material in dilute hydrochloric 
acid (10 %) and hydrogen peroxide (10 %) and Na-oxa-
 late (0.01 %) according to Aitken (1998). Sodium poly-
 tungstate at 2.58 g cm-3 specific gravity for potassium­
 feldspars and 2.68 g cm-3 specific gravity for quartz 
grains was used to enrich the quartz- and feldspar-frac-
 tion. The quartz­fraction was etched in hydrofluoric 
acid (40 %) for 40 min to remove any residual feldspars 
and the outer 10 µm of quartz grains. Following another 
treatment with hydrochloric acid to remove soluble fluo­
 rides and final dry sieving a monolayer of grains was 
mounted with silicon on 1 mm diameter steel discs for 
analysis.
 Optically stimulated luminescence was determined 
by SAR-BLSL (Single Aliquot Regenerative Dose Blue 
Light Stimulated Luminescence, according to Murray 
and Wintle 2000) for quartz and SAR-IRSL (Single Ali-
 quot Regenerative Dose Infrared Stimulated Lumines-
 cence, according to Wallinga et al. 2000 and Preusser 
2003); for details see table 1.
 Results and discussion
 Paleo-environmental conditions and soil formation at 
Neolithic settlements in the region of Abu Tabari
 The site S02 / 52 was located on a former sandbank. The 
sandbank was elevated up to 2 m above the plain ground 
of the Wadi Howar and covered an area of about 100 m 
by 400 m. The surface was covered by a dense layer of 
artefacts, comprising pottery sherds, lithic objects and 
animal bones. Via the pottery an attribution of the site to 
the 4th or the first half of the 3rd millennium BC can be 
supposed (Lange 2005). The ground of the wadi was co-
 vered by fluviatile sediments and pebbles. To get an idea 
of the former environmental conditions, soil profiles, 
lined up from the centre and top of the site to the lower 
Desert soils in Lower Wadi Howar (Northwest Sudan)
 Feldspars Quartz
 Radiation
 Sr / Y-90 beta source
 0.1426 Gy / s
 Sr / Y-90 beta source
 0.1085 Gy/s
 Stimulation
 IR-diodes (880 ± 80 nm)
 300 s @ 50° C
 Blue diodes (470 ± 30 nm)
 50 s @ 125° C
 Detection filter
 Schott BG39
 Schott GG400
 Corning 7–59
 U­340­filter (7.5 mm, transmission at 290–370 nm)
 Pre-heating 230° C, after test dose 250° C 240° C, after test dose 200° C
 Tab. 1: Parameters for OSL-dating of quartz and feldspars.
 Fig. 2: Weakly developed soil at the Neolithic site S02/52, 
Lower Wadi Howar, profile 8.
4
 Profile 10 at the wadi ground showed cemen-
 ted, lacustrine sediments (»Playa sediment«) with 
high contents of inorganic carbon and also distinct 
contents of organic carbon. The phosphorus content 
was distinctly smaller than in the centre of the site. 
These Playa sediments covered an area of about 0.5 
km2. The sediments contained relicts of a land mol-
 lusc, most probably Subulina isseli (Jickeli 1874; 
pers. comm. N. Pöllath).
 In both soil profiles we found large concentrations of 
salt in 20–40 cm depth, which is typical for arid and semi-
 arid soils due to an upward groundwater movement. With 
up to 500 µS cm-1 the conductivity was within the normal 
range of arid soils, as reported by Gabriel (1986).
 Site S02 / 28 was also located on a relatively flat former 
sandbank with a maximum height of about 1 m above 
ground level. The site consists of a large scatter of archaeo-
 Andreas Gundelwein
 Location: Lower Wadi Howar
 Parent material: Aeolian sediments covered by artefacts (rock tools, bones, etc.)
 Horiz.
 Depth
 [cm]
 Text.
 C
 inorg
 [%]
 C
 org
 [%]
 P
 t
 [mg kg-1]
 other
 0–2 sand – – –
 50 vol. % rock fragm. > 2 mm
 (artefacts)
 7.5YR7 / 6
 single grain
 no roots
 A 2–23
 loam. 
sand
 0.05 0.12 353
 < 5 vol. % rock fragm. > 2 mm
 10YR4 / 3
 coherent single grain
 6–20 roots 10 cm-2
 conductivity 99 µS cm-1
 bulk density 1.63 g cm-3
 AC 23–65 sand 0.03 0.12 208
 < 5 vol. % rock fragm. > 2 mm
 10YR6 / 4
 6–20 roots 10 cm-2
 single grain
 conductivity 490 µS cm-1
 C > 65 sand – – –
 < 5 vol. % rock fragm. > 2 mm
 10YR7 / 6
 no roots
 single grain
 Tab. 2a: Typic Torripsamment (profile 8) on the top of the settlement site S02 / 52, Lower Wadi Howar; Munsel colour was 
determined with moist soil.
 Location: Lower Wadi Howar
 Parent material: Lacustrine sediments (»Playa«)
 Horiz.
 Depth
 [cm]
 Text.
 C
 inorg
 [%]
 C
 org
 [%]
 P
 t
 [mg kg-1]
 Conduct.
 [µS cm-1]
 other
 0–1 sand – – – –
 50 vol. % rock fragm. > 
2 mm
 single grain
 no roots
 Ccm
 1–20
 20–> 42
 silty 
sand
 1.28
 1.57
 0.17
 0.17
 –
 153 
142
 608
 2.5Y6 / 3
 massive (petrocalc.hor.)
 no roots
 Tab. 2b: Calcareous sediments (profile 10) at the margin of the settlement site S02 / 52, Lower Wadi Howar; Munsel colour was 
determined with moist sediment.
5
 logical finds stretching over an area of about 500 m 
x 450 m. Radiocarbon dates point to an occupation of 
the site around 3000 BC (Jesse in press). A geomagne-
 tic survey in 2006 showed anomalies near the western 
edge of the sandbank.1 In the test trenches distinct redo-
 ximorphic features and concentrations of iron nodules 
in 30–80 cm depth were found (fig. 3, table 3). The 
boundary between the upper, oxygen rich and well ae-
 rated zone and the groundwater zone with reducing soil 
conditions followed the relief of the slope (fig. 4).
 The lacustrine sediments and the redoximorphic features 
prove, that during the Holocene humid phase this land-
 scape was most probably at least temporarily covered by 
the shallow waters of an extended lake with several flat 
islands such as the sandbanks of the sites S02 / 52 and 
S02 / 28. So, most probably both settlements were ori-
 Fig. 3: Nodules of iron-oxide at the western edge of the Neo-
 lithic settlement S02/28, Lower Wadi Howar, profile 17.
 1 The geophysical survey was done by Carsten Mischka of the Institute for Prehistoric Archaeology at the University of 
Cologne.
 Location: Lower Wadi Howar
 Parent	material:	Aeolian-fluviatile	sediments,	covered	by	artefacts	(stone	tools	etc.)
 Horiz.
 Depth
 [cm]
 Text. other
 0–1 sand
 35 vol. % rock fragments > 2 mm
 7.5YR6 / 6
 single grain
 no roots
 Ak 1–20
 weak 
silty sand
 < 2 vol. % rock fragments > 2 mm
 2.5YR5 / 3
 coherent single grain
 6–10 roots
 B 20–40
 weak 
silty sand
 < 2 vol. % rock fragments > 2 mm
 2.5YR5 / 3
 coherent single grain
 6–10 roots 10 cm-2
 10 % redoximorphic features, single ironoxide nodules
 Bc 40–50
 weak 
silty sand
 2–5 vol. % rock fragments > 2 mm
 10YR6 / 4
 coherent single grain
 10–20 roots 10 cm-2
 50 % redoximorphic features, 15 % ironoxide nodules
 BC 50–> 100
 weak 
silty sand
 < 2 vol. % rock fragments > 2 mm
 2.5YR5 / 3
 coherent single grain
 6–10 roots 10 cm-2
 10 % redoximorphic features, single ironoxide nodules
 Tab. 3: Typic Torripsamment with distinct relictic redoximorphic features (profile17) at the margin of the settlement site 
S02 / 28, Lower Wadi Howar; Munsel colour was determined with moist soil.
 Desert soils in Lower Wadi Howar (Northwest Sudan)
6
 Tab. 4: Paleosoil (Typic Torriorthent, profile 1) at the basis of the tumulus S06 / 1 near the fortress Gala Abu Ahmed, Lower 
Wadi Howar; Munsel colour was determined with moist soil.
 sample
 paleodose 
(Gy)
 error (Gy)
 dose rate 
(Gy / ka)
 error (Gy / ka) depth (cm)
 water 
content (%)
 age (ka) Error (ka)
 S106 Q 1,09 0,11 0,777 0,031 25 ± 5 2 1,40 0,15
 S106 KF 1,18 0,09 1,402 0,214 25 ± 5 2 0,84 0,14
 Tab. 5: OSL-ages of the covering sand layer of profile 1, site S06 / 1, Lower Wadi Howar. Fractions: – Q quartz; – KF potassium-
 feldspar; – D
 e
  paleodose; – D
 o
  dose rate. Rel. SD: 43 % for quartz and 38 % for potassiumfeldspar.
 Location: Ground of Wadi Howar; Parent material: lacustrine sediments under a thin aeolian sand cover
 Horiz.
 Depth
 [cm]
 Text.
 C
 inorg
 [%]
 C
 org
 [%]
 other
 0–5 sand – –
 < 5 vol. % rock fragments > 2 mm
 7.5YR7 / 6
 single grain
 no roots
 ACb 5–45
 silty 
loam
 0.13 0.13
 10YR4 / 3
 blocky
 6–10 roots 10 cm-2
 conductivity 99 µS cm-1
 – 45–83                                                             no samples
 C 83–134
 Silty 
loam
 0.09 0.15
 10YR5 / 4
 blocky
 no roots
 conductivity 490 µS cm-1
 Cg > 134
 Silty 
loam
 – –
 7.5YR5 / 2
 blocky
 Tab. 6: Typic Torriorthent (profile 7) on the wadi ground near the fortress Gala Abu Ahmed, Lower Wadi Howar; Munsel colour 
was determined with moist soil.
 Location: South slope (20°) of Lower Wadi Howar; Parent material: rocky debris under recent aeolian sand cover
 Horiz.
 Depth
 [cm]
 Text.
 C
 inorg
 [%]
 C
 org
 [%]
 other
 0–20 sand 0.02 0.03
 < 2 vol. % rock fragments > 2 mm
 7.5YR8 / 8
 single grain
 no roots
 Ab 20–34 silty sand < 0.02 0.05
 60–80 vol. % rock fragments > 2 mm
 7.5YR4 / 4
 coherent single grain
 3–10 roots 10 cm-2
 Bb 34–66 silty sand 0.04 0.05
 > 90 vol. % rock fragments > 2 mm
 7.5YR5 / 6
 coherent single grain
 1–5 roots 10 cm-2
 BCb 66–82 silty sand 0.03 0.04
 > 90 vol. % rock fragments > 2 mm
 7.5YR6 / 8
 coherent single grain
 1–5 roots 10cm-2
 Andreas Gundelwein
7
 ginally located adjacent to open water surfaces. Similar 
lacustrine sediments were found in the Oyo Depression 
in NW Sudan (19° N, 26° E) and were dated by 14C 
to 6 500–2 500 BC (Ritchie et al. 1985). Considerable 
sand contents were lacking. This indicates carbonate 
deposition in a stratified lake with stagnant bottom wa-
 ters, according to Ritchie et al. (1985). A comparison 
with actual sedimentation cycles allows the reconstruc-
 tion of a humid tropical climate with annual monsoonal 
rainfall of at least 400 mm (ibid.). This humid period 
with formation of lacustrine sediments ended between 
4500 BP (Nicoll 2001) and 3700 BP (Petit-Maire and 
Guo 1998).
 The study of the faunal remains of site S02 / 52 in-
 dicates a subsistence strategy of the Neolithic dwellers 
based on cattle herding and especially fishing. It seems 
that site S02 / 52 was used as a preferred and specialized 
fishing place (N. Pöllath, pers. comm.). The exposed po-
 sition within a shallow lake offered the best conditions 
for such a specialized, maybe seasonal use. Fish bone 
finds from early to middle Holocene sites in the Middle 
Wadi Howar showed a rich ichthyofaunal spectrum, 
which allows to reconstruct the main freshwater bioto-
 pes that once characterized the Wadi Howar area (Pöl-
 lath and Peters 2003). The input of organic residues re-
 sulted in an enrichment of phosphorus and carbon in the 
upper soil layers as could be evidenced at site S02 / 52. 
The large amounts of phosphorus prove an intense and 
long term use of the site. This anthropogenic input of 
organic residues favoured and initiated the formation 
of raw soils. Such enrichment was exclusively found 
at settlement sites (for example also at the nearby site 
S02 / 01). This could be explained by the special conser-
 vation conditions for soils under the dense artefact cover 
of the settlements and an erosion of the upper soil lay-
 er elsewhere. But it is also plausible that soil formation 
took place exclusively at former settlements, and was 
triggered and initiated by the human input of organic re-
 sidues. That would mean that not only the environment 
determined human behaviour and living conditions but 
also that man influenced his environment.
 Archaeological sites as climate archives
 About 1.5 km east of the fortress Gala Abu Ahmed a 
number of tumuli were found along the southern shore-
 line of Wadi Howar (see Kröpelin 1993; Jesse and Ku-
 per 2004). One tumulus was excavated, site S06 / 1. It 
was located in the middle of the slope. Instead of the 
expected burial we found distinctly weathered soil mate-
 rial at the base of the tumulus in 0.2 m depth (profile 1;  
fig. 5; tab. 4). Organic carbon contents were very low, 
even if there was a weak enrichment in 0.2 to 0.66 m 
depth. The upper soil layer (horizon Ab) was well 
rooted and the minerals were distinctly weathered. The 
soil was classified as Typic Torriorthent. The OSL­age 
of the covering sand layer was 0.8 to 1.36 ka (tab. 5).
 Even for an arid soil the organic carbon contents 
were very low compared to mean organic carbon con-
 tents of other extra arid desert soils, which are normal-
 ly between 0.2–1.5 kg C m-2 (Lioubimtseva 1997). In 
contrast to the low carbon contents the red-yellowish 
colours indicate a distinct weathering of soil minerals, 
which requires humid climate conditions. Due to the 
protection by the tumulus this soil was preserved from 
erosion. Remnants of well developed fossil soils were 
also found under the dense artefact cover of the Neo-
 lithic settlements near Abu Tabari (see above) and also 
reported from other sites in the Sahara, e. g. in Egypt 
(Blume et al. 1984), Niger (Vogg 1986; Völkel 1987; 
1988), and Wadi Shaw in north-western Sudan (Gabriel 
1986). These paleo-soils verify more humid climate con-
 ditions during the middle Holocene (6000–5000 BC). 
The covering sand layer of profile 1 at site S06 / 1 de-
 veloped later as the result of erosion processes of the 
southern slope of Wadi Howar. The resultant OSL-age 
can therefore only be used as a terminus ante quem for 
the soil formation and the tumulus.
 A well in the Wadi Howar
 Near the fortress Gala Abu Ahmed a circular embankme-
 nt with a diameter of 5 m was found on the wadi ground, 
which is typical for former wells (Kröpelin 1993). 
The sediments were studied by drilling up to a maxi-
 Fig. 4: Stratigraphic situation at the margin of the Neolithic 
settlement S02 / 28, Lower Wadi Howar, soil profiles 16–18. 
The horizon Bc with high concentration of iron nodules marks 
the former shore line.
 Fig. 5: Distinctly weathered soil material at the basis of the 
tumulus S06 / 1, Lower Wadi Howar, profile 1.
 Desert soils in Lower Wadi Howar (Northwest Sudan)
mum depth of 134 cm below surface (profile 7; fig. 6;  
tab. 6). Below the aeolian sand cover, we found silty-
 loamy sediments with high contents of organic carbon 
and a distinct enrichment of salt in 83–134 cm depth. 
The soil was characterized as a Typic Torriorthent. The 
sediments prove the existence of a small lake at the dee-
 pest point of the wadi for a longer period. After the dry-
 ing up of the lake, the existing groundwater reservoir on 
this spot was obviously still used for water supply. The 
age of the well is unclear. But it is conceivable that it 
was used by the habitants of the fortress, which is dated 
by several finds and 14C­ages to the first half of the first 
millennium BC (Napata-time; Jesse and Kuper 2004). 
Maybe, the water supply even was the reason for the 
maintenance of a large fortress more than 100 km west 
of the River Nile. Finds of pottery prove that already in 
former times the Wadi Howar was used as an important 
trading route from the Nile to the west. Controlling the 
few water supplies with punctual checkpoints would 
have meant controlling the whole border line.
 Conclusions
 The analysis of desert soils near archaeological sites 
allowed new insights in the environmental and living 
conditions of the early inhabitants of the Wadi Howar 
and their interactions with their environment. It also 
provided us with more information about the soil gene-
 sis in arid regions. Further studies should concentrate 
on dating the described indicators of a more humid cli-
 mate in the eastern Sahara during the Holocene.
 Acknowledgements
 The field work was possible within the scope of the 
Sonderforschungsbereich 389 ACACIA (Arid Climate 
Adaption and Cultural Innovation in Africa) of the 
Andreas Gundelwein8
 University of Cologne, sub-project A2 »Wadi Howar: 
Siedlungsraum und Verkehrsweg am Südrand der Li-
 byschen Wüste.« This project owes its funding to the 
Deutsche Forschungsgemeinschaft (DFG).
 Special thanks to the ACACIA­Project for financial 
support of the laboratory work and the organisation of 
the expedition’s logistics. The chemical analysis of the 
soil samples was conducted by J. Siemens and his team 
from the Institute of Soil Science, Technical Universi-
 ty of Berlin. OSL-dating was done by the Department 
of Geography, University of Cologne. N. Pöllath from 
the Institut für Paläoanatomie und Geschichte der Tier-
 medizin, Ludwig-Maximilian-University of Munich, 
determined the aquatic Fauna and F. Jesse from the 
Institut für Ur- und Frühgeschichte, Forschungsstel-
 le Afrika, University of Cologne, gave strong support 
concerning the details of Sudanese archaeology.
 References
 Aitken 1998
 M. J. Aitken, An introduction to optical dating (Ox-
 ford 1998).
 Blume et al. 1984
 H. P. Blume, F. Alaily, U. Smettan and J. Zielinski, 
Soil types and associations of South-West-Egypt. 
Berliner Geowiss. Abhandl. A 50, 293–302.
 Gabriel 1986
 B. Gabriel, Die östliche Libysche Wüste im Jung-
 quartär. Berliner Geogr. Stud. 19 (Berlin 1986).
 Haynes 2001
 C. V. Haynes, Geochronology and climate change 
of the Pleistocene-Holocene transition in the Darb 
el Arba’in desert, eastern Sahara. Geoarchaeology 
16, 2001, 119–141.
 Jesse 2003
 F. Jesse, New Archaeological Work in The Lower Wadi 
Howar (Northern Sudan) – A Preliminary Report on the 
field season 2002. Nyame Akuma 60, 2003, 43–48.
 – 2005
 F. Jesse, Rock art in the Lower Wadi Howar, north-
 west Sudan. Sahara 16, 2005, 27–38.
 – 2006a
 F. Jesse, A permanent link? – The Wadi Howar 
Region and the Nile. In: I. Caneva and A. Roccati 
(eds.), Acta Nubica. Proceedings of the Xth Interna-
 tional Conference of Nubian Studies, Rome 9–14 
September 2002 (Rome 2006) 73–80.
 – 2006b
 F. Jesse, Dune Habitats in the Wadi Howar, North-
 west Sudan. In: Préhistoire en Afrique / African 
Prehistory. Acts of the XIVth UISPP Congress, Uni-
 versity of Liège, Belgium, 2–8 September 2001. 
BAR Internat. Ser. 1522 (Oxford 2006), 101–107.
 – in press
 F. Jesse, A time of experimentation? – The 4th and 3rd 
millennia BC in Lower Wadi Howar, Northwestern 
Fig. 6: Calcarous sediment of the former well near the Fortress 
Gala Abu Ahmed, profile 7.
9
 Sudan. In: Acts of the 11th International Conference 
of Nubian Studies, Warsaw University, August 27th 
– September 2nd 2006.
 Jesse and Kuper 2004
 F. Jesse and R. Kuper, Gala Abu Ahmed – Eine Fes-
 tung am Wadi Howar. Der Antike Sudan. Mitt. Su-
 danarch. Ges. 15, 2004, 137–142.
 Jickeli 1874
 C. F. Jickeli, Fauna der Land- und Süßwasser-Mol-
 lusken Nord-Ost-Afrikas (Dresden 1874).
 Keding 1998
 B. Keding, The Yellow Nile: new data on settlement 
and the environment in the Sudanese Eastern Sa-
 hara. Sudan and Nubia 2, 1998, 2–12.
 Keding and Vogelsang 2001
 B. Keding and R. Vogelsang, Vom Jäger-Sammler 
zum Hirten – Wirtschaftswandel im nordöstlichen 
und südwestlichen Afrika. In: B. Gehlen, M. Hei-
 nen and A. Tillmann (eds.), Zeit-Räume. Gedenk-
 schrift für Wolfgang Taute. Arch. Ber. 14 (Bonn 
2001) 257–282.
 Kröpelin 1987
 S. Kröpelin, Paleoclimatic Evidence from Early to 
Mid-Holocene Playas in the Gilf-Kebir (South-West 
Egypt). Paleoecol. Africa 18, 1987, 189–208.
 – 1993
 S. Kröpelin, Zur Rekonstruktion der spätquartären 
Umwelt am Unteren Wadi Howar (Südöstliche Sa-
 hara / Nordwest Sudan). Berliner Geogr. Abhandl. 
54 (Berlin 1993).
 Kuper and Kröpelin 2006
 R. Kuper and S. Kröpelin, Climate-Controlled Ho-
 locene Occupation in the Sahara: Motor of Africa’s 
Evolution. Science 313, 2006, 803–806.
 Lange 2005
 M. Lange, More archaeological work in Lower 
Wadi Howar (Northern Sudan) – a preliminary re-
 port on the 2003 field season. Nyame Akuma 63, 
2005, 15–19.
 Lioubimtseva 1997
 E. Lioubimtseva, Impacts of climatic change on car-
 bon storage variations in African and Asian deserts. In: 
R. Lal, J. M. Kimble, R. F. Follet and B. A. Stewart 
(eds.), Soil processes and the carbon cycle (Boca Ra-
 ton 1997) 561–576.
 Murphy and Riley 1962
 J. Murphy and J. P. Riley, A modified single solution 
method for the determination of phosphate in natural 
waters. Analytica Chimica Acta 27, 1962, 31–36.
 Murray and Wintle 2000
 A. S. Murray and A. G. Wintle, Luminescence dating 
of quartz using an improved single-aliquot regene-
 rative-dose protocol. Radiation Measurements 32, 
2000, 57–73.
 Neumann 1989
 K. Neumann, Holocene vegetation of the eastern 
Sahara: charcoal from prehistoric sites. African 
Arch. Rev. 7, 1989, 97–116.
 Nicoll 2001
 K. Nicoll, Radiocarbon chronologies for prehis-
 toric human occupation and hydroclimatic change 
in Egypt and Northern Sudan. Geoarch. 16, 2001, 
47–64.
 Pachur 1987
 H. D. Pachur, Vergessene Flüsse und Seen der Ost-
 sahara. Geowiss. in unserer Zeit 5, 2, 1987, 55–64.
 Pachur and Kröpelin 1987
 H. D. Pachur and S. Kröpelin, Wadi Howar: Paleo-
 climatic evidence from an extinct river system in the 
southeastern Sahara. Science 237, 1987, 298–300.
 Petit-Maire and Guo 1998
 N. Petit-Maire and Z. T. Guo, Mid-Holocene clima-
 tic change and man in the present-day Sahara desert. 
In: A. S. Alsharhan, K. W. Glennie, G. L. Whittle 
and C. Kendall (eds.), Quaternary Deserts and Cli-
 matic Change (Rotterdam 1998) 351–356.
 Pöllath and Peters 2003
 N. Pöllath and J. Peters, Paleoenvironmental interpre-
 tation of fish remains from the Wadi Howar region, 
north-west Sudan. In: G. Grupe and J. Peters (eds.), 
Decyphering ancient bones – the research potential of 
bioarchaeological collections. Documenta Archaeo-
 biologiae 1 (Rahden / Westf. 2003) 257–264.
 Preusser 2003
 F. Preusser, IRSL dating of K-rich feldspars using 
the SAR-protocol: comparison with independent 
age control. Ancient TL 21, 2003, 17–23.
 Ritchie et al. 1985
 J. C. Ritchie, C. H. Eyles and C. V. Haynes, Sedi-
 ment and pollen evidence for an early to mid-
 Holocene humid period in the eastern Sahara. 
Nature 314, 1985, 352–355.
 Schlichting et al. 1995
 E. Schlichting, H.-P. Blume and K. Stahr, Boden-
 kundliches Praktikum (Berlin 1995).
 Soil Survey Staff 1999
 Soil Taxonomy8. USDA Agriculture Handbook 436 
(Washington / DC 1999).
 Vernet 1995
 R. Vernet, Les Paléoenvironments du nord de l’Af-
 rique depuis 600 000 ans. CNRS Dossiers et Re-
 cherches sur l’Afrique 3, 1995, 149–271.
 Völkel 1987
 J. Völkel, Geomorphologische und pedologische 
Studien in den Dünenhabitaten der südöstlichen 
Sahara und des Sahels in der Republik Niger. Göt-
 tinger Geogr. Abhandl. 84, 1987, 109–125.
 – 1988
 J. Völkel, Bodenkundliche Sicht des spätquartären 
Klimawandels in der Sahara und im Sahel des öst-
 lichen Nigers. Würzburger Geogr. Arb. 69, 1988, 
255–276.
 Vogg 1986
 R. Vogg, Relief und Böden der westlichen Zen-
 tral- und Südsahara (Süd-Algerien, Nordost Mali). 
Darmstädter Geogr. Stud. 7, 1986, 7–43.
 Desert soils in Lower Wadi Howar (Northwest Sudan)
10
 Wallinga et al. 2000
 J. Wallinga, A. S. Murray and A. G. Wintle, The 
single-aliquot regenerative dose (SAR) protocol ap-
 plied to coarse-grain feldspar. Radiation Measure-
 ments 32, 2000, 529–533.
 Dr. Andreas Gundelwein
 Council for the Lindau Nobel Laureates Meetings
 Ludwigstr. 68
 88131 Lindau (B)
 Germany
 Andreas Gundelwein