Morphologic Characteristics and Migration Rate Assessment of Barchan Dunes in the Southeastern Western Desert of Egypt

Geomorphology 257 (2016) 57–74

Contents lists available at ScienceDirect

Geomorphology

journal homepage: www.elsevier.com/locate/geomorph

Morphologic characteristics and migration rate assessment of barchan dunes in the Southeastern Western Desert of Egypt

M.A. Hamdan, A.A. Refaat ⁎, M. Abdel Wahed

Geology Department, Faculty of Science, Cairo University, Giza, Egypt article info abstract

Article history: This work explores the morphologic characteristics of aeolian dune sand in the southeastern part of Western De- Received 3 May 2015 sert of Egypt. It aims to assess the movement of barchan dunes and evaluate their environmental inﬂuence on the Received in revised form 6 December 2015 Toshka Project. Morphometric investigation of barchan dunes in the Toshka area revealed that most barchans Accepted 30 December 2015 have high length/width (a/c) ratios (fat to pudgy), while one-ﬁfth of the studied barchans have lower a/c ratios Available online 31 December 2015 and so appear normal in their morphologic forms. Statistical analysis of the main parameters of barchan dunes in Key words: Toshka and other desert regions in the Kharga (Egypt), Kuwait, Southern Morocco, California and Southern Peru Barchan dune demonstrates that barchans of the Toshka area are distinctive in their appearance. They are characterized by dis- dune migration tinct aspect with higher values of length and width and greater growth in height. The high-energy wind environ- dune encroachment hazard ment in addition to the large amount of drifting sand are principal factors responsible for the unique shape of Western Desert of Egypt Toshka barchans. The migration rate of barchan dunes in four chosen test locations, within the central and western Toshka area, ranges from about 3 to 10.82 m/year. The calculated average migration rate of these dunes is about 6 m/year in a SSW direction. Sand encroachment is more extensive in the central and western parts of the investigated Toshka area. Risk evaluation of sand dune movements in the southeastern part of the Western Desert points to medium to high sand encroachment risk values. These may represent serious hazards to the newly-established Toshka Project, threatening roads, as well as cultivated lands in the area. © 2015 Elsevier B.V. All rights reserved.

1. Introduction characterized by high dune sand accumulation. The area has been chosen to serve as a model for the assessment of dune migration because it Many areas in the Western Desert of Egypt are characterized by high is affected mainly by strong trade winds which exert a strong control on accumulations of wind-blown sand (Fig. 1). Modern aeolian sand of the the mechanism of dune formation as in the eastern range of the Great Western Desert of Egypt has been the subject of scientiﬁc interest since Sahara Desert (Besler, 2008). It also represents the focal area of many the beginning of the 20th century (e.g. Beadnell, 1910; Ball, 1927; dune streams and sand accumulations that migrate from other regions Bagnold, 1931, 1941). Numerous geologic, geomorphologic, textural, of the Western Desert of Egypt into the northern Toshka area (Fig. 1). mineralogical, geochemical and source-area genetic studies of the Hence, the Toshka area is considered to be an ideal region for dune mea- sand dunes followed (e.g. Embabi, 1970; El-Baz et al., 1979; Besler, surements and can be used as a model for other regions in the Western 1986, 1998, 2000, 2008; Haynes, 1989; Stokes et al., 1998; Hamdan Desert of Egypt and North Africa. and Refaat, 1999; Hamdan, 2003; El Gammal and Cherif, 2006; Abou Sand dunes are a common feature of many desert regions all over El-Magd et al., 2013; Khedr et al., 2013a, 2013b; Refaat and Hamdan, the world and they may exist in many different types. Barchan dunes 2015; Hamdan et al., 2015). Several studies have been concerned with are the simplest and most common dune type as it maintains a crescent estimating the rate of desert dune movement in many areas of Egypt, shape and occurs in areas characterized by steady winds coming from a especially in the Western Desert (Table 1). Sand dune movements are similar direction throughout the year, providing there is not enough considered to be a speciﬁc threat to roads, irrigation networks, water re- sand to cover the entire regional surface (Sauermann et al., 2000). Bar- sources, urban areas, agriculture and infrastructure of the Toshka pro- chan dunes are well developed in many parts of the Egyptian Western ject (Wahby, 2004). Desert, including the Toshka area, as well as in many deserts areas not The southeastern part of the Egyptian Western Desert is largely rep- only on the surface of the Earth (e.g. Bagnold, 1941; Finkel, 1959; resented by the Toshka area; it covers about 50,000 km2 and is Long and Sharp, 1964; Hastenrath, 1967; Hesp and Hastings, 1998; Sauermann et al., 2000; Parteli et al., 2007) but also on the surface of ⁎ Corresponding author. Mars (e.g. Goudie and Bourke, 2008; Bourke and Balme, 2008). More- E-mail address: [email protected] (A.A. Refaat). over, several types and forms of barchan exist in these deserts,

Fig. 1. Map of the Western Desert of Egypt showing the distribution of the major sand dune ﬁelds (modiﬁed after El-Baz, 1979). The study area is located in the southeastern part of the Western desert of Egypt. Numbers inside the circle refer to dune migration rates of the Toshka area and other regions in the Western Desert of Egypt where arrows refer to the direction of dune migration at each area. The number at the head of each arrow refers to the corresponding reference: 1 = Beadnell (1910);2=Ashri (1973);3=Embabi (1979);4=Sharaky et al. (2002); 5 = El Gammal and Cherif, (2006); 6 = Abou El-Magd et al. (2013); 7 = Present study.

depending upon several factors, such as wind strength and direction, Western Desert of Egypt. Finally, the degree of sand dune encroachment amount of drifting sand, topography and vegetation. The current and its effect on the Toshka Project is evaluated. paper is a comparative study that highlights the main morphometric characteristic features of Toshka barchans and other barchan shapes from different deserts of the world. Evaluation of the main factors 2. Geomorphologic features that control the morphology and existence of these dunes are discussed. The study area is located in the southeastern part of the Western De- A further aim of this study is to investigate the distribution of aeolian sert of Egypt between 22° and 24° 00′ Nand30°15′ and 33° E (Fig. 2). dune sands in the Toshka area and to identify their types using remote With the exception of Sinn el Kadab Plateau and its pediments, which sensing and GIS techniques. The study assesses the rate of barchan lie more than 300 m above sea level and bounding the Toshka Depres- movement in the Toshka area in comparison with other regions in the sion from the north, the study area is relatively ﬂat and most of it is M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 59

Table 1 Rate of dune movement in the east of the Nile Delta and different locations in the Western Desert of Egypt, with their references.

Location area Time period (year) Rate (m/year) Reference

East of the Nile Delta 1898–1990 4.5 Cornish (1900) Kharga Depression 1908–1909 15 Beadnell (1910) Kharga Depression 1971 12 Ashri (1973) Kharga Depression 1978 20–100 Embabi (1979) Dakhla Depression 1990 0.5–14 Sharaky et al. (2002) Abu Moharik dunes 1974, 1984–2001 17.5 El Gammal and Cherif, (2006) –northern of the Bahariya oasis 7.5 –middle part of the dune ﬁeld 11.5 –southern end near the Kharga depression Toshka area 2000–2006 1.3–19.3 Abou El-Magd et al. (2013) 1987–2007 6.024 Present study

occupied by the Toshka Plain (Depression), ranging in elevation from el Kadab Plateau and Lake Nasser from the east, while the Egyptian- 150 to 250 m above sea level. Sudanese Border represents its southern boundary. The evolution of the Toshka Plain is mainly controlled by the existence of a series of 2.1. Sinn el Kadab Plateau folds and faults (El Shazly et al., 1977). The Toshka Plain is characterized by outcrops of isolated Nubian sandstone hills and short, wide wadies The extensive Sinn el Kadab plateau, which is underlain mainly by that in the past drained into the Nile Valley. Precambrian igneous and limestone, consists broadly of horizontal beds of late Cretaceous to metamorphic rocks, early Cretaceous Nubian sandstone and Quaternary early Eocene age. It is bounded eastwards and southwards by a precip- Nile sediments are all exposed in the Toshka area (Fig. 3). itous escarpment, overlooking the Nile Valley from its eastern side, while northwards the plateau extends beyond the area under consider- 2.3. Lake Nasser ation to the Kharga Depression (Fig. 1). The escarpment is mainly controlled by major faults that are generally trending in N-S and E-W Lake Nasser, the world's largest man-made lake, is a reservoir in the directions (Fig. 3). course of the River Nile, formed by construction of the Aswan High Dam in 1960. It is located at the border Egypt and Sudan between latitudes 2.2. The Toshka Plain (Depression) 21.8°N to 24.0°N and longitudes 31.3°E to 33.1°E. Its surface area is about 5200 km2 with a maximum water capacity of 165 km3 and a This plain occupies the central part of the study area and attaining an mean depth of 25 m; its surface elevation is 175 m above sea level. area of about 13,142 km2 with a lowest point located at 150 m above sea The lake is about 550 km long (more than 350 km in Egypt while the level (Wadi Toshka). It is limited to the north by the pediments of Sinn rest lies in Sudan) and 35 km across at its widest point.

Fig. 2. Satellite image of the Toshka area shows the selected locations for dune migration calculations. 60 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

Fig. 3. Geological map of the Toshka area showing the exposed rock units and the main structural elements in the study area (modiﬁed after Labib and Nashed, 2013).

One of the most interesting features of Lake Nasser is the complex of the two horns perpendicular to the wind direction along the dune geometry of its shoreline; which changes enormously according to crest. even minor changes in the water level. The shoreline possesses numer- Comparative investigations of morphological shapes of barchans ous khors, which are flooded valleys (where the valleys have been cut from different deserts in the world in the Atlantic Sahara region in by streams). Most of these khors are narrow and meander into the de- southern Morocco (Sauermann et al., 2000); Kuwaiti Desert (Khalaf sert for long distances. A vast number of islands of various sizes, and Al-Ajmi, 1993); Imperial Valley in California, USA (Long and representing the tops of former hills, are scattered throughout the Sharp, 1964) and La Pampa de la Joya in southern Peru (Finkel, 1959) lake. The number, location and size of these islands vary greatly with were also evaluated. The four parameters; length, width, height and a/ fluctuations in the lake-level. c ratio were also determined for each barchan in different regions; Cal- ifornia (27 dunes); Peru (44 dunes); Kuwaiti Desert (10 barchans) and 3. Materials and methods Morocco (8 dunes). The obtained parameters of the total of 244 studied samples are represented as variables that reach 976 data points (four Field studies were carried out using topographic maps (scale parameters for each sample). Statistical analysis of such large dataset 1:100,000 and 1:50,000), Landsat images and available Google Earth is not easy without using a convenient statistical software package satellite images, to study the aeolian sand distribution in the southeast- such as SPSS. This extensively used program for statistical analysis in so- ern region of the Western Desert of Egypt. The desert dunes in the study cial science is widely used by many geology researchers. Descriptive sta- area constitute four main categories; barchan, linear, lee and shadow tistics for each variable (including the mean, standard deviation, dunes. The distribution of different dune types with in the Western De- maximum, minimum and range values) were calculated. The interrela- sert investigation area is illustrated in Fig. 4.Thebirthofnewlyformed tionships between the different variables of the studied dunes are dem- barchan dunes and barchanoids from lee dunes downwind of topo- onstrated in Figs. 7-8. SPSS provides a number of statistical procedural graphic obstacles is shown in Fig. 5. One hundred and eleven barchan programs for doing correlation and a wide variety of analyses such as dunes in the Toshka area have been subjected to morphological shape T-test and One-way ANOVA (Analysis of Variance), which enables the investigation according to the classification given by Long and Sharp hypothesis that the means of two or more groups are not significantly (1964). In order to assess variation of barchan dune shapes, morpho- different to be tested. One-way ANOVA procedure using SPSS was metric examination was undertaken for 44 barchan dunes in the Kharga used for barchan dune parameters of Toshka area and other desert re- Depression belonging to the same dune stream of Toshka area but gions in order to specify exactly if the means of two or more variables representing its northern extension. For each selected barchan dune, are significantly different. Once differences between the means of dif- four parameters were determined; length of the windward slope (a), ferent barchan dunes were established, post hoc tests and Dunnett's width (c), height (H) and the a/c (length/width) ratio (Fig. 6). We mea- T3 pairwise multiple comparisons provided by SPSS were employed to sured the barchan dune width as the greatest distance between the tips determine which means differ. These pairwise multiple comparisons M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 61

Fig. 4. Satellite image of the Toshka area, showing the distribution of aeolian dune types in the Toshka area. test the difference between each pair of means and yield a matrix where of statistical analysis of morphological parameters of barchans in the asterisks indicate signiﬁcantly different group means at an alpha level studied Kharga and Toshka regions compared with other barchans in (probability) of 0.05 (95% conﬁdence level). Table 2 shows the results different deserts across the world. It should be noted that only variable

Fig. 5. The birth of newly-formed barchan dunes and barchanoids from lee dunes downwind as the topographic barrier has no effect on the accumulation of dunes. 62 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

Fig. 8. The length/width (a/c) ratio vs. height (H) of Toshka barchan dunes compared with dunes from different desert regions in Kuwait, Southern Morocco, California and Southern Peru. This relationship clearly demonstrates the separation of Toshka, Kharga and the Kuwaiti barchans from other dunes, as these areas enclose more fatty barchan dunes (greater a/c ratio). Variable values that lie within the estimated 95% conﬁdence interval (CI) for means are only used in the diagram.

maximum resolution ranging between 8 to 4 m/pixel and cover a time-span from 1987 to 2007. The extracted images are geometrically registered (their coordinates are identified) using the ArcGIS software. The registration process is carried out by locating 8 control points of known coordinates on the Google Earth image, saved as a shape file and used as a layer of reference points in the ArcGIS database. Barchans often migrate long distances with only minor changes in their form (e.g. Long and Sharp, 1964; Hastenrath, 1967), and barchan migration rates Fig. 6. Schematic diagram showing the dimensions of a barchan measured in the field. have been investigated by many different authors (e.g. Finkel, 1959; Haynes, 1989; Sauermann et al., 2000). Outlines of the barchan dunes values that lie within the estimated 95% confidence interval (CI) and in the two images are traced on two different layers in the database demonstrate a normal distribution are used in the illustrated diagrams and are overlain on each other in the final product. To determine the av- (Figs. 7–8), thus the extreme values were excluded. The different pa- erage amount and direction of migration of a group of dunes, points of rameters and types of barchan dunes and their locations in the Toshka well-defined locations were visually selected on the dune boundaries area are given in Appendix A. in both the older and younger images. The distance between two corre- Determination of the migration trend and rate of dune migration in sponding points indicates the point movement during the specified the present study is essentially based on remote sensing and GIS tech- time-span. The average value of distance moved for selected points is niques through accurate comparison of the dune location in satellite im- considered as the average migration of the dune group. Considering ages of two different dates separated by a considerable time span. the time-span between the two satellite images, the yearly rate of Previous studies (e.g. Abou El-Magd et al., 2013) that dealt with the dune migration is then calculated. The distance between two corre- dunes movement in the southern parts of the Western Desert were sponding points and the direction of the line connecting them was ob- based mainly on ETM images which considerably reduce the accuracy tained automatically within the ArcGIS environment by creating a line of results. Indeed, the identification of dunes and tracing their bound- feature between corresponding points on a separate layer that is aries with significant accuracy is difficult for ETM Landsat images of saved as a shape file and exporting the data into an Excel spreadsheet. 15 m/pixel resolution, since its maximum mapping scale does not ex- The exported data include the direction and length of each line ceed 1:50,000. Considering that the barchan dunes within the study connecting corresponding points. area are generally less than 400 m in width, higher resolution images Seventeen barchan dunes were selected from 111 of Toshka bar- (at least twice that of ETM) should be used. chans that have been examined for their morphologic shape investiga- In the present work, Google Earth historical satellite images are used tion to perform dune movement measurements. Four test locations as a basis for the dune movement calculations. The available Google (Fig. 2) enclosing the selected barchan dunes were chosen within the Earth images of the barchan dunes within the area of interest are of a central and western parts of Toshka area near the Toshka Lakes through visual interpretation of satellite images. The amount and direction of migration of all selected points for the test areas in the studied satellite images are illustrated in Table 3. The minimum and maximum amounts of movement together with the average migration rate of the different test locations are summarized in Table 4. The satellite images of dates 1987 and 2007 and the drawn interpreted maps of the different locations in the study area are shown in subsequent figures (Figs. 9–12).

4. Results and discussion

4.1. Distribution of the aeolian sand and dune types

Satellite images of the southeastern Western Desert indicate that ae- Fig. 7. Height (H) vs. width (c) of barchan dunes from the Toshka area and other desert olian sand covers about 26.8% of the whole study area (Fig. 4). The esti- sites in the world. The dune width was measured as the longest distance between the 6 tips of the two horns perpendicular to the wind direction along the dune crest. Note the mated amount of transported aeolian sand is about 210 × 10 kg good linear relationship between these two variables. annually (Altorkomani, 1999). In fact, these sands show variable M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 63

Table 2 Mean and range values (measured in meters), ANOVA test (P-values) and Post Hoc Test P-values (Dunnett's T3 pairwise comparisons) for Toshka and Kharga barchans and comparative dunes from different sites around the world.

Location N Length (a) Width (c) Height (H) a/c ratio

Mean values, range in parentheses

Toshka 111 183.24 (31.2–608.2) 175.10 (31.2–476) 16.61 (2.6–45.8) 1.06 (0.47–1.97) Kharga 44 54.4 (18–145.5) 45.4 (7.5–110) 7.8 (1.3–17.4) 1.8 (0.2–8.6) Kuwait 10 46.69 (18.2–91) 39.92 (15.7–65) 3.31 (1.8–7) 1.19 (0.74–1.77) (Khalaf and Al-Ajmi, 1993) Southern Morocco 8 37.75 (17.8–79) 57.48 (28.5–96.3) 4.66 (2.3–8.5) 0.65 (0.48–0.88) (Sauermann et al., 2000) California, USA 27 67.62 (33.5–111.3) 108.32 (41.2–254.5) 5.89 (2.7–12.2) 0.74 (0.23–1.53) (Long and Sharp, 1964) Southern Peru 44 22.55 (9–40) 36.87 (11.4–66) 3.40 (1.4–5.8) 0.64 (0.43–1.53) (Finkel, 1959)

ANOVA test (P-values) ⁎ ⁎ ⁎ ⁎ b0.001 b0.001 b0.001 b0.001

Source (I) Source (J) Post hoc Test P-values ⁎ ⁎ ⁎ TKhb0.001 b0.001 b0.001 .108 ⁎ ⁎ ⁎ K b0.001 b0.001 b0.001 .944 ⁎ ⁎ ⁎ ⁎ M b0.001 b0.001 b0.001 b0.001 ⁎ ⁎ ⁎ ⁎ C b0.001 b0.001 b0.001 b0.001 ⁎ ⁎ ⁎ ⁎ P b0.001 b0.001 b0.001 b0.001 ⁎ ⁎ Kh K .993 .998 b0.001 .001 M .552 .962 .099 .430 ⁎ ⁎ C .374 b0.001 .163 .001 ⁎ ⁎ ⁎ P b0.001 .596 b0.001 .003 ⁎ K M .997 .754 .904 .008 ⁎ ⁎ ⁎ C .253 b0.001 .012 .033 ⁎ P .075 1.000 1.000 .007 ⁎ M C .061 .029 .924 .957 P .553 .479 .846 1.000 ⁎ ⁎ ⁎ CPb0.001 b0.001 b0.001 .832

N: number of samples; T: Toshka; Kh: Kharga; K: Kuwait; M: Morocco; C: California; P: Peru. ⁎ P-values b 0.05 indicate significant differences among sources (bold face type). distribution patterns across the study area and their accumulation pat- 40 m, in width from 5 to 30 m and from 1 to 4 m in heights; while the tern is lowest at the top of the Sinn el Kadab Plateau. The high sand ac- latter vary in length from 140 to 420 m and from 150 to 490 m in cumulation of dunes is recorded in the central part of the Toshka Plain to width at heights of 15–50 m, and with a dune cover density varying the south of Sinn el Kadab escarpment and partly along the western from 5.4 to 10.9 per km2. Barchans tend not to remain isolated but can shore of Lake Nasser (Hamdan et al., 2015). join up and form complexes ranging from train-like successions of bar- The free-moving dunes of the southeastern part of the Western De- chans to real dune massifs. This process has been explained in a recent sert exist mainly in the Toshka Plain. However, a few small barchan theoretical work by Melo et al. (2012). dunes are also recorded at the top of Sinn el Kadab Plateau. The anchored dunes (lee dunes and sand shadows) exist over both the Sinn 4.2. Morphological classification of barchan dunes el Kadab Plateau and the Toshka Plain. Sand shadows (falling dunes) were accumulated at the escarpments of the Sinn el Kadab Plateau The data of the barchans in the southeastern part of the Western De- and in the leeward sides of the Nubian sandstone hills in the Toshka sert provide information that concerns latitudinal variations in bar- Plain that may also lead to the development of lee dunes. Where the to- chans. The investigated barchan dunes in the Toshka area are pography no longer has a significant effect, some of the linear lee dunes characterized by a wide range of their morphologic parameters. They often split downwind into individual separate barchan dunes and vary in length from 31 to 608 m, averaging 183.2 m (Table 2), whereas barchanoids (Fig. 5). The Toshka dunes primarily originated from an the mean width of these dunes is 175 m and they may reach 46 m in older (late Pleistocene) dune sand to the north of study area (Hamdan height. To the north, the barchan dunes in the Kharga Depression, et al., 2015). which represents the northern extension of the same barchan stream The eastern part of the Toshka area is characterized by a relatively that traverses the Toshka area, also exhibit the same wide range in di- little sand accumulation in the form of individual barchans and an- mension, but tend to be much smaller. The dunes located in the Kharga chored dunes parallel to the western coast of Lake Nasser (El Shazly et range in length from 18 to 145 m, averaging 54 m, and between 7.5 and al., 1977). In the western part of the Toshka Plain, linear dunes that 110 m in width, averaging 45 m; they are notably lower in height than are about 1 km long with sinuous slip faces exist together with solitary Toshka dunes, averaging 7.8 m. small barchans (Geofizika, 1966). The central and western parts of the Barchan dunes can be classified as simple, compound or complex, as study area is characterized by high sand accumulations in the form of suggested by McKee (1979). Based on a ratio of the length of the wind- barchan and linear dune fields, which are 10 to 45 km in length and ward slope (a) and the width (c), barchan dune shape has been catego- 2–8 km in width, and extended in a north–south direction. The linear rized by Long and Sharp (1964) as: Fat (≥1), Pudgy (0.75), Normal (0.5) dune types occupy the northern part of the dune field, while the bar- and Slim (0.25). From 111 barchan dunes investigated in Toshka area chan types exist in its southern part. The observed barchans are of two for their morphologic classification based on the a/c ratio, only 22 bar- types; small, very rapidly formed dunes without slipfaces, probably chans were normal in their morphologic forms, representing attaining more accurately termed dome dunes (McKee, 1966), and moderate to about 20% of the recognized barchans. However, most of the investi- large crescent mega barchans. The former vary in length from 10 to gated barchans were fat (54%) or pudgy (24%), characterized by a high 64 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

Table 3 Depression, representing 65% and 23.7%, respectively, whereas normal Dune migration calculations of the four test locations in the Toshka area, southeastern part and slim dunes (11.3%) are less prominent. of the Western Desert of Egypt. Comparing the studied barchans with other dunes in different sites Test Dune Migration Azimuth of Migration in Average dune indicates that the Toshka dunes seem to be much larger than other bar- area no. points the migration meters migration (m) chans not only in the same dune stream of the same desert (the Kharga – (in D.DD) (1987 2007) dunes) but also compared with barchans in other deserts: in Kuwait, the Area 11–1 189.88 102.32 138.48 Atlantic Sahara in southern Morocco, and La Pampa de la Joya (southern – 1 2 2 189.04 167.54 Peru). Larger barchans occur in California but still their dimensions are 3–3 201.95 145.59 24–4 189.91 142.87 166.96 smaller than barchans in Toshka area (Table 2). The barchan dimensions 5–5 182.09 192.21 in the Toshka area as well as in other regions are linearly related to each 6–6 183.64 165.79 other. An example is the relationship between the dune height and its 37–7 186.25 112.88 145.81 width displayed in Fig. 7, which shows a clear linear (r = 0.76) signifi- 8–8 185.59 126.12 cant correlation. Most of the Kharga barchan and some of California's 9–9 184.57 198.42 410–10 182.50 211.30 211.60 samples deviate from the trend line through the majority of points, 11–11 184.20 215.96 reflecting that a fair relationship exists between width and height of 12–12 184.73 207.53 barchan dunes in these areas. Although Tsoar (1985) provides an expla- – 51313 186.58 91.89 118.20 nation for the steady state response of dunes such as barchans, no expla- 14–14 178.02 152.24 15–15 210.58 110.46 nation states why a linear relationship should exist between the height Average 187.97 156.21 of barchan dunes and the width of the horns (Hesp and Hastings, 1998). Area 61–1 187.18 99.24 76.23 Performing one-way ANOVA test for the dune dimensions shows signif- 2 2–2 199.95 66.66 icant differences between means of dune parameters (P b 0.001) for 3–3 176.22 62.79 both barchans of the Toshka area and other dunes from different regions 74–4 186.21 76.53 86.04 5–5 173.72 94.55 (Table 2). Among the different considered dunes, the Toshka barchans 6–6 192.35 87.05 are comparatively characterized by a larger dune shape. Post hoc tests Average 185.94 81.14 and Dunnett's T3 pairwise multiple comparisons differentiated the – Area 811 186.03 90.19 113.16 Toshka dunes as a separate group among the other barchan dunes 3 2–2 193.14 125.00 3–3 179.45 124.30 (Table 2). With the exception of a/c ratio of barchans in the Kuwaiti de- 94–4 180.00 101.23 112.43 sert (Khalaf and Al-Ajmi, 1993), the Toshka dunes are significantly dif- 5–5 182.79 97.50 ferent in their parameters than barchans from other sites. The Kharga 6–6 182.94 138.57 barchans, like other barchans from different deserts, are not signifi- – 10 7 7 180.98 123.03 114.70 cantly different from each other in their parameters. An exception is 8–8 174.38 107.58 9–9 186.59 113.51 barchans from California (Long and Sharp, 1964), which may be shape Average 182.92 113.43 differentiated from barchans in other regions, especially those from Area 11 1–1 195.78 119.39 113.05 southern Peru (Table 2). The previous results obtained from statistical – 4 2 2 193.23 72.43 approach of the considered dunes can be clarified by plotting the a/c 3–3 186.00 110.22 4–4 198.93 150.16 ratio vs. height of barchans from the Toshka and the Kharga with 12 5–5 191.74 115.27 115.27 other sites (Fig. 8). This diagram clearly shows the complete separation 13 6–6 194.45 138.49 142.96 of the investigated barchans of Toshka and Kharga and those from 7–7 179.26 140.11 Kuwait from the rest of the dune samples, indicating that they are sig- – 8 8 184.87 136.36 nificantly different in their length and width, but the Toshka barchans 9–9 199.49 156.86 14 10–10 188.68 155.50 155.50 still increase more in height. However, the Kharga samples are relatively 15 11–11 186.45 138.45 119.96 more dispersed in Fig. 8 than are other dunes, implying that they con- 12–12 196.59 99.68 tain barchans with a wide range of a/c ratios. Dunes from other locations – 13 13 192.85 121.74 (California, southern Morocco and southern Peru) overlap, which sig- – 16 14 14 199.57 123.91 133.69 fi 15–15 198.33 143.48 ni es that they are quite similar in their dimensions (Table 2; Fig. 8). 17 16–16 189.79 118.67 130.81 17–17 193.89 126.07 4.3. Hazards of sand encroachment on the Toshka Project 18–18 191.70 147.70 Average 192.31 128.58 4.3.1. Dune movement The first attempt to estimate the rate of movement of Egyptian de- a/c ratio (Appendix A). Fat dunes occur in areas where there is a sub- sert dunes was done by Cornish (1900), who measured a rate of stantial sand influx and lower shear velocity (Parteli et al., 2007). Few 4.5 m/year in the dune crests east of the Nile Delta (Table 1). Many of the recognized barchans are slim. The fat and pudgy barchans are areas in the Western Desert were the subject of evaluation of sand also common in the investigated dune streams of the Kharga dune movements by different authors (e.g. Beadnell, 1910; Ashri,

Table 4 Summary of results of dunes migration computations of aeolian dune sands of the four test areas in the southeastern part of the Western Desert of Egypt.

Test area No. of Time period Min. Max. Average Min. migration rate Max. migration rate Average migration rate dunes (years) movement movement movement (m) (m/year) (m/year) (m/year) (m) (m)

Area 1 5 19.97 91.89 215.96 156.21 4.60 10.82 7.82 Area 2 2 19.97 62.79 99.24 81.14 3.15 4.97 4.06 Area 3 3 19.97 90.19 138.57 113.43 4.52 6.94 5.68 Area 4 7 19.71 72.43 156.87 128.58 3.67 7.96 6.53 Toshka area 17 19.90 79.33 152.66 119.84 3.99 7.67 6.02 average M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 65

Fig. 9. Google Earth satellite images of the ﬁrst location of dates 1987 (a) and 2007 (b) together with constructed maps illustrating the location of dunes (c) and the points used in migration calculations (d) during this time period.

1973; Embabi, 1979; Sharaky et al., 2002; El Gammal and Cherif, 2006) In the study area, the investigated barchans were concentrated into who assigned diverse dune rates ranging from 0.5 m/year to be as high four test areas along a narrow N-S dune stream, located immediately as 100 m/year. More recently, Abou El-Magd et al. (2013) estimated the west of the Toshka lakes in the central and western parts of the Toshka rate of movement of barchans and transverse dunes in south Western area. Seventeen barchans were picked from this dune stream through Desert which ranged from 1.3 to 19.3 m/year. visual interpretation of satellite images of the four chosen test locations. 66 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

Fig. 10. Satellite images of the second location of dates 1987 (a) and 2007 (b) in addition with the drawn maps that show the location of dunes (c) and the points used in movement calculations (d).

The dunes were carefully selected to be of well-defined boundaries and migration determinations (Fig. 9). Final results of the point to point maintaining their shape in as many successive images as possible. Also, correlation between the two images are listed in Table 3,inwhich they were considered to be free to move without the presence of any the amount and direction of movement in each point over the hindrance, such as surrounding bedrock exposures that may prevent assigned time period and the average migration of each dune are or limit their movements. presented. This analysis reveals a minimum and maximum point The first test area is located to the west of the northernmost lake of migration of 91.89 and 215.96 m, respectively, with an average mi- the Toshka region. The satellite images were about 7 m/pixel resolution gration value of 156.21 m. Consequently, the minimum and maxi- with dates 29/9/1987 and 12/9/2007 and a time-span, between the two mum migration rates are estimated at 4.60 and 10.82 m/year, images, of 19.97 years. Five barchans dunes with sharp boundaries and respectively, with an annual rate of 7.82 m/year (Table 4). The aver- similar morphology on these two images were recorded in for this loca- age direction of movement of these dunes in this area was SSW tion. Fifteen well-defined points on the dune boundaries are selected for along an azimuth of 187.97°. M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 67

Fig. 11. Satellite images of both dates 1987 (a) and 2007 (b) together with the constructed maps representing the dune distribution in the third test location (c) and the points used in migration measurements (d).

The second location was at a distance of about 13 km southwards of area. Four barchan dunes are recorded; three of them were suitable the first test area. The satellite images are of resolution of about 5 m/ for analysis, while the fourth had a poorly defined boundary (Fig. 11). pixel, and have dates and a time-span between images similar to Nine selected points were projected along the dunes borders. Point to those of the first location. The location encloses only two barchan point correlation of the two images indicates the greatest and least dunes typically suitable for migration measurements, i.e. well defined value of point migration as 138.57 and 90.19 m, respectively, with an outlines and maintenance of shape in the two images (Fig. 10). Six average migration of 113.43 m. The calculated minimum, maximum points were located on the dune boundaries. Processing estimates a and average migration rates are 4.517, 6.9 and 5.68 m/year, respectively lowest and highest point movement of 62.79 and 99.24 m, respectively, (Table 4). The average direction of dune movement is SSW (azimuth of led to an average movement of 81.14 m. Computation indicates 3.15, 182.92°). 4.97 and 4.06 m/year as minimum, maximum and average rates of The fourth test area is located to the south of Toshka Lakes at a dis- dune migration, respectively. The trend of dune migration is due SSW tance of about 55 km to the southwest of the third test area. The satellite along an average azimuth of 185.94°. images are of about 8 m/pixel resolution and have dates 31/12/1987 and The third test area is located at a distance of about 11.5 km south- 12/9/2007. The time-span between the two images is 19.71 years. This west of the second location. Satellite images of this location are of location enclosed 7 barchan dunes with well-defined boundaries. Eigh- 6 m/pixel resolution and have also the same dates as the first test teen points were chosen for migration determination (Fig. 12) and the 68 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

Fig. 12. Satellite images of a successive row of barchans in the fourth location of dates 1987 (a) and 2007 (b) and the drawn interpreted maps showing the distribution of dunes in the fourth location (c) and the points used in migration calculations (d). results of dune migration analyses are shown in Table 3. The minimum nature of theses dunes (Appendix A), whereas the ﬁrst location area and maximum point movement were 72.43 and 156.87 m, respectively, contains the smaller normal dunes that move more speedily. By averag- with an average value of 128.58 m. The minimum and maximum rates ing dune migration measurements of the 17 representative barchan of dune migration were 3.67 and 7.96 m/year. The average migration dunes in the central and western parts of Toshka area during the of these dunes is 6.53 m/year, along a SSW direction (azimuth of 19.9 year time-span, the minimum point movement is 79.33 m and 192.31°). the maximum is 152.66 m, with an average point movement of The minimum and maximum values of dune movement together 119.84 m (Table 4). The annual rate of migration of the studied sand with their rates and the average rate of migration representing all the dunes varies from about 4 to 7.67 m/year. The calculated average migra- test areas are presented in Table 4. This table shows that the ﬁrst test tion of these dunes is 6.02 m/year along a SSW direction (azimuth of area exhibits the highest migration of barchan dunes in the Toshka 187.29°). This calculated average migration rate is very similar to that area of a maximum point movement of about 216 m and a maximum given by Besler (1986), who measured the rate of migration of 5 m/ migration rate (10.82 m/year), averaging 7.82 m/year. This contrasts year to the dunes of the Toshka area. with dunes in the second test area which show the lowest dune move- In order to investigate the relationship between dune morphology ment, with only 4 m/year as an average rate of migration. This second and dune movement, a correlation of the length/width (a/c) ratio with location contains the largest studied barchans as well as the ‘fatty’ the dune migration of the selected barchans is illustrated in Fig. 13. M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 69

against their height (Fig. 14) shows an inverse relationship. Some points depart from the linear relationship which may be due to the shape of these dunes as most of them are of the ‘fatty’ type. This implies that dunes which are greater in their height move much slowly than smaller ones.

4.3.2. Sand drift potential In recent years, a major project developed in the southeastern part of the Western Desert (the Toshka project) and has provided many bene- fits to Egypt (Wahby, 2004). It helped in doubling the cultivated and reclaimed land for Upper Egypt, attracting workforce, constructing an efficient network of main and side roads, facilitating power generation and utilizing the massive amounts of water sorted in the Lake Nasser. Fig. 13. The length/width (a/c) ratio vs. dune migration of the investigated barchans of the Toshka area in the southeastern Western Desert shows clearly that small dunes migrate However, aeolian sand encroachment may cause serious hazards to more rapidly than larger ones. the sustainable development of the Toshka Project (Abdelmoaty et al., 2011). The southeastern division of the Western Desert is located This diagram shows clearly the inverse relationship between these two within the hyper-arid zone where precipitation occurs once every two variables, where normal barchans (the smallest a/c ratio) migrate more or three decades. Data obtained from the Tropical Rainfall Measuring rapidly than fat barchan dunes. It is obvious that small dunes (e.g. Mission (TRMM) shows that the Toshka area received b20 cm of total Table 2; Dunes No. 2, 4 and 14) move faster than large fat ones rainfall from September 1999 to September 2006 (Bastawesy et al., (Table 2; Dunes No. 5, 6, 7, 10, 15 and 16). Similar observations are 2008). Meteorologically, the mean air pressure in this area varies be- also reported by previous studies (e.g. Pye and Tsoar, 2009). Some ob- tween 940 and 1001 mbar, whereas the mean air temperature ranges served barchans are also characterized by strongly elongated horns from 16 to 40.5 °C (Table 5). The mean humidity is low and ranges be- and asymmetric progress of their plan shapes (e.g. Table 2; Dunes No. tween 15.6 and 48.8%, whereas the mean wind speed ranges between 8, 10, 14 and 16). These could be as a result of superimposing of two suc- 3 and 7.1 m/s (Khedr et al., 2013a). El-Baz and Wolfe (1982) studied cessive barchans on one another. Wind direction variations and asym- the wind regime of the Western Desert of Egypt and revealed that two metric sand influx due to a complex distribution of upwind sand major wind regimes exist; a narrow band of predominantly westerly sources (complex dune patterns upwind of the dune) may also cause along the Mediterranean coast and generally a north northwesterly dune asymmetry (Bourke, 2010; Parteli et al., 2014). In fact in the inves- though the Western Desert. They also noticed some local variations in tigated images (e.g. Fig. 12) some asymmetric barchans not directly par- the general wind trend and attributed these to topographic effects ticipating in collision may have the origin of their asymmetry associated such as that at the Farafra, Kharga and Aswan areas. At Aswan (east of with the presence of upwind dunes. Therefore the influx asymmetry the study area), the general wind trend is imposed by the north–south may be a possible origin of the asymmetry of some dunes in the Toshka topographic highs of the Eastern Desert and Sinn el Kadab Plateau. region. Based on the surface wind data of 9 years (2000–2008) of the Kharga Many factors affect dune displacement, such as differences in local and East Uwinat meteorological stations; the wind in the Toshka area wind regime, sand supply, topography, and vegetation but size of the blows mainly from the north (about 42%) and the northeast (nearly dunes, especially height of the slip face, is generally regarded as the 35%). The Kharga Depression has dominant winds mostly from the most important factor (Long and Sharp, 1964). In evaluating the influ- northwest (34.2% of time) and from the north (32.6% of time) direc- ence of the a/c ratio on movement, it appears important to know tions. Wind data also reveal a considerable variation of strength. Effec- whether the barchan is growing or is in a steady-state condition. As- tive wind strength (12 knots/h) is maximal in this area and blows for suming a uniform sand flux, a growing fat barchan should move more 46.7% of time (Hereher, 2010). Sand drift potential (DP) refers to the slowly than a growing barchan that is slim because a greater volume amount of sand moved by the surface wind that includes values above of sand is required to produce corresponding increase in height of the the threshold velocity. It is calculated using Fryberger (1979) equation, fat dune. On the other hand, in a steady-state condition, a fat barchan as follows: might move more rapidly since most of the sand causing advance is de- rived from the windward slope of the dune itself and a fat barchan has a DP ¼ V2ðÞV−V t: long windward slope as a source of supply. The sand coming to a steady- t state dune from upwind serves principally to set the dune's sand into motion and to replace the loss from the tip of the horns (Long and where, DP refers to the sand drift potential in vector units (VU); V is the

Sharp, 1964). Plotting dune migration of each of the studied 17 barchans average wind speed in knots at 10 m height; Vt is the threshold velocity (the velocity required to entrain sediment) in knots; and t is the percent of time during which wind velocity is greater than the threshold velocity required to entrain the sediment. Based on the average grain size of Toshka sands which is 0.35 mm (Hamdan et al., 2015), the threshold velocity is estimated to be 12 knots (6.2 m/s) under dry conditions (Fryberger, 1979).

Table 5 Summary of meteorological data of the area west of Lake Nasser (adopted from Khedr et al., 2013a).

Meteorological parameter Min. Max. Average

Pressure (mbar) 940 1001 989.55 Temperature (°C) 16 40.5 27.05 Fig. 14. Height of slip face vs. dune migration of the investigated barchans of the Toshka Humidity (%) 15.6 48.5 29.59 area in the southeastern Western Desert shows clearly that higher elevated barchan Wind speed (m/s) 3 7.1 4.79 dunes move more slowly than smaller ones. 70 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

Resultant drift potential (RDP) represents the magnitude of DP, width. These dune streams begin at the foot of Sinn el Kadab Plateau whereas the resultant drift direction (RDD) refers to the direction in as lee dunes then change to be linear dunes and finally become barchan which sand will be transported. The ratio of the resultant drift potential and barchanoid types. Because of the variance in the rate of dune migra- (RDP) to the drift potential (DP) is known as the directional variability tion, two or more barchan dunes may merge to form a small linear dune of surface wind (RDP/DP). Higher RDP/DP values indicate that the connecting the main dune stream. wind comes from the same direction (a narrowly unidirectional wind regime) whereas lower values indicate a multidirectional wind regime 4.3.3. Aeolian sand risk assessment of the Toshka area with multiple significant drift directions. Stopping dune movement is extremely costly and it is only a practi- Sand drift potential is always coincident with both the strength and cal and economic solution in areas likely to be affected with dune en- direction of the effective winds. Drift potentials are measures of the encroachment (Tag El Din, 1986). Sand dune movement risk evaluation ergy of surface winds in terms of sand movement. Fryberger (1979) is important for countries that have migrating dunes that threaten classified energy of surface winds in arid regions, based upon the aver- their towns, transportation routes, or agricultural fields, like the West- age annual drift potential, into three groups: (1) low-energy environ- ern Desert of Egypt (Effat et al., 2011). Remote sensing data has facili- ment with drift potential (DP) less than 200 vector units (VU); tated desert exploration and modeling risk and vulnerability of sand (2) intermediate-energy wind environment with DP ranges from 200 encroachment. The rate of desertification, especially formation and ex- to 399 VU; and (3) high-energy wind environment with DP greater pansion of dune fields, depends mainly on six parameters; these are: than 400 VU (Table 6a). The calculated potential sand transport param- terrain surface elevation, dune slope angle, dune slope direction (as- eters showed that the potential sand drift of the Kharga area attains pect), relative moisture index, prevailing wind direction, and prevailing about 173 VU (low energy environment), which is much lower than wind speed. These six parameters were obtained from Shuttle Radar To- that of the Toshka area which reaches 571 VU (Table 6b). The greatest pographic Mission data (SRTM) and the land cover map produced by DP value of the Toshka area is classified as high-energy wind environ- the Food and Agricultural Organization (FAO), which were used by ment. Fryberger et al. (1984) estimated the amount of sand creep as Effatetal.(2011)to model potential dune migration in the Western De- each VU represents 0.07 m3 of sand moving across 1 m width of land. sert of Egypt. Based on risk scale given by Effatetal.(2011), the risk as- The corresponding amount of sand drifting per 1 m width of land in sessment of aeolian sand dune encroachment in the southeastern the Toshka area, based on Fryberger's rate, is about 40 m3/year. In the Western Desert can be evaluated (Table 7). The elevation of the desert Kharga area, the amount of sand drift is estimated at about 12 m3/ area is represented as elevation risk index, where higher elevation year, which is less than that in the Toshka region. High directional var- areas are more at risk to sand dune activity. According to the SRTM iability index is recorded in both the Kharga and Toshka regions (RDP/ data and topographic maps of the Toshka area, altitude ranges from DP = 0.82 and 0.78, respectively); this indicates that wind blows from 114 to 478 m above sea level. When it is represented as a dune elevation a narrow wind direction (from the north and northwest) and where risk index, medium risk is indicated. On the other hand, dune slope in barchans form chains extending parallel to the RDD. The resultant the desert area is represented as a slope risk index, where sand accumu- drift direction (RDD) is due S and SSW (azimuth of 181.2° and 192.3°) lated on steeper slope dunes is more risky. The dune slope value of the for the studied Kharga and Toshka areas, respectively (Table 6b). Toshka area varies between 40 and 60°. Consequently, the dune slope These results reveal that the Toshka area is characterized by the highest risk index here is medium to high risk. The third measured index is sand activity and drifting in the Western Desert of Egypt. Such a high the dune slope direction (aspect), where dunes facing the prevalent amount of transported sand every year will result in hazards to the sus- wind direction are most likely to experience dune movement and thus tainable development with in the Toshka Project. have the highest risk aspect. The greater aspect angle the greater is The adverse impacts of sand encroachment and mobile dunes are a the risk. The azimuth of the dune migration in the study area ranges great threat, especially to the agricultural activity areas (Fig. 15a). Ef- from 173.72° to 210.58° (Table 3). Hence, the aspect risk index of fects extend to the roads (Fig. 15b), irrigation canals and to the Toshka Toshka area is measured to be low to medium risk. The average wind Lakes (Fig. 15c) causing significant economic losses and considerable speed of Toshka area varies between 9.5 and 11 m/s (Effat et al., expense for mitigation measures. What happens to a moving dune 2011), so its wind speed risk value is thus high. The spatial distribution when it reaches a lake shore is that it stops shortly after entering the of wind directions, especially the dominant directions, are more effec- lake. Lake Nasser itself may also be influenced by the sand encroach- tive in generating a sand movement risk. Meteorological data ment that can be expected to diminish its water storage capacity (Hereher, 2010) indicate that the annual average prevailing wind direc- (Abdelmoaty et al., 2011). The Google Earth images and field investigation of the Toshka area varies between 0° and 45° (NNE) and its wind tions of the present study show that the central and western parts of direction risk is thus, medium. Finally, moisture is measured as a rela- Toshka area are highly affected by sand encroachment compared to tive moisture index, where sand dunes are more active when the rela- the eastern part. The Toshka lake area is threatened by about 10 dune tive moisture is low (dry). As the study area is hyper-arid, the sand is streams, each stream is about 5–10 km in length and 0.3–1kmin very dry and the relative moisture risk index will be the highest.

Table 6 a) The classiﬁcation of wind energy environments using drift potential (DP) and directional variability (adopted from Fryberger, 1979). b) The potential sand transport parameters (DP, RDP, and RDD) as well as the directional variability index (RDP/DP) of the studied Kharga and Toshka regions.

DP (VU) Wind energy environment RDP/DP Directional variability Wind regime

200 Low 0.3 High Complex or obtuse bimodal 200–399 Intermediate 0.3–0.8 Intermediate Obtuse or acute bimodal 400 or more High 0.8 Low Acute Unimodal

DP Area Sand drift (m3/m-w/year) RDP RDP/DP RDD (VU)

Kharga 173 12.11 142 0.82 181.2 Toshka 571 39.97 447 0.78 192.3 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 71

Fig. 15. a) Sand dunes encroaching on the cultivated land of the Toshka area. b) Dune encroachment on Aswan-Abu asphaltic road in the Toshka area. c) A series of barchan dunes crossing the Toshka Lakes from the north in a dune stream trending in N–Sdirection.

Table 7 Sand dune risk evaluation of the aeolian sand dunes of Toshka area in the southeastern Western Desert of Egypt.

Parameter Risk scale rangea Toshka area Risk evaluation

Elevation 4–1033 114–478 Medium (m) Dune slope angle 0–90 40–60 Medium–High (degrees) Slope aspect 0–360 173–210 Low–Medium (azimuth degrees) Relative Moisture Index −4.5–12.6 −4.5 to −3.2a High Prevailing wind speed 0–23 18.5–21.6aa High (knots/h) Prevailing wind direction 0–360 0–45 Medium (azimuth degrees)

a Data from Effat et al. (2011). 72 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

Accordingly, the sand dune risk for the Toshka area is about a me- whereas the central and western parts of the study area are distin- dium to high risk and the road vulnerability for sand encroachment guished by high sand accumulations in the form of barchan and linear risk is high. These may cause serious hazards to the Toshka project dune streams in a north–south direction. Linear dune types occupy and agricultural lands in this area. the northern part of the dune field, while the barchan types exist in its southern part. 4.4. Factors controlling dune occurrence, morphology and movement The barchan dunes of the Toshka area have been studied for their morphometric analysis in the form of length, width and height and Factors controlling the occurrence and morphology of sand deposits their relationships in order to identify their morphologic characteristics. are complex (Gifford et al., 1979). They include the wind direction, Most of the recognized barchans in the Toshka area are fat and pudgy; strength and duration; the nature, extent and rate of erosion at the sed- while normal barchan dunes are less common. The main parameters iment source; the distance from the source; the grain and fragment size; of Toshka barchans are compared statistically with barchans from the the underlying and surrounding topography; the nature of the surface Kharga Depression (Western Desert of Egypt) and dunes in different de- (rough or smooth); the amount and type of vegetation and the amount sert regions in Kuwait, Southern Morocco, California and Southern Peru. of rainfall. It is important to stress that we do not take into consideration The Toshka barchans are unique in their dimensions; characterized by in this discussion all these factors, but only those that we have consid- greater aspect (higher values of length and width and increase more ered in our field studies and laboratory investigations for barchan in height). dunes in the southeastern Western Desert of Egypt and from compari- The potential for sand drift in the Toshka area is considerable as it is a sons with barchans in other deserts of the world. The morphological dif- high-energy wind environment. The area is characterized by the maxi- ference between barchan in the Toshka area and other regions in the mum sand activity and drifting in the Western Desert of Egypt. This may world could be due to increased wind activity, to increased sand supply, explain the unique large form of barchan dunes in the Toshka area. Field or both. Meteorological data from many sources proved that the average investigations show that the central and western parts of this area are wind speed of Toshka area varies between 9.5 and 11 m/s, which is highly affected by sand encroachment. Risk assessment of the south- greater than that of Kharga in the Western Desert of Egypt (2.9 m/s) eastern Western Desert indicates a medium to high sand dune risk and California, USA that reach 4 m/s, and both of the Pampa de la Joya and high sand encroachment risk to vulnerable areas. This may repre- in Peru and in the Kuwaiti Desert (5.4 m/s), while the average wind sent serious hazards to the newly-established Toshka Project, roads, as speed in southern Morocco varies from 7 to 8.5 m/s. Increased sand sup- well as the reclaimed agricultural lands in this area. ply seems likely to be the most important factor. In the Kharga Depres- The trend and rate of migration of barchan dunes is essentially based sion, the sand supply is limited due to the presence of the northern on remote sensing and GIS techniques through the accurate comparison escarpment of the depression, which prevents much sand from of dune location in satellite images of two different dates (nearly reaching the depression floor (Hereher, 2010). Limited supply of sand 20 years interval). The mean movement of 17 representative barchan is funneled, forming the barchan dunes. On the other hand, large bar- dunes in the central and western parts of Toshka area varies from chans in the Toshka area were formed due to the great availability of about 4 to 7.67 m/year. The calculated average mean of migration of sand, i.e. the Great Sand Sea, located to the north. Considering these fac- these dunes is nearly 6 m/year along SSW direction (azimuth of tors that affect the barchan dunes in the Toshka area and the results 187.29°). discussed in the preceding sections, we can conclude that the high The high-energy wind environment combined with the high wind speed (9.5–11 m/s) and the considerable quantity of drifting amount of drifting sand are principal factors that may facilitate the prev- sand (40 m3/year across 1 m width of land) in these dunes comes alence of large fat and pudgy barchan dunes in the Toshka area. The from dune streams crossing the north of the Bahariya Oases through great shape and fatty nature of the Toshka barchans together with the the Farafra, Dakhla and Kharga Oases to reach the Toshka area in the rugged bedrock topography may explain the reduction of the barchan southeastern Western Desert (Fig. 1) and corresponds with a high- migration rate compared with dunes from other desert regions in the energy wind environment (571 VU). All these factors together have Western Desert of Egypt. assisted barchan dunes in the Toshka area to increase in size significantly compared with barchan dunes in other desert areas. Acknowledgements Sauermann et al. (2000) showed that barchan dunes move proportionally to wind velocity and inversely proportionally to their height. How- Field work of this paper was supported by the Ministry of State For ever, the lowest movement of Toshka dunes compared with other areas Environmental Affairs. We would like to express our thanks to Prof. in the northern and central Western Desert of Egypt (Fig. 1) may be at- Roger J. Flower, Department of Geography, UCL — University College tributed to several topographic and morphologic elements. From these London for carefully reading the manuscript and his valuable notes, features (Figs. 3-4) are: 1) the existence of Sinn el Kadab plateau, on which have greatly improved the text. The authors are also indebted the northern margin of the Toshka Plain; 2) the rugged ground surface to anonymous reviewers and the editor for their helpful comments of the western part of the studied Toshka area; 3) the occurrence of that improving the final version of the manuscript. Toshka Lakes in the central part of the Toshka area, which intercept barchan streams crossing from the north to south; 4) frequent small hills of Appendix A. The morphologic parameters; length (a), width (c) and Nubian sandstone that are scattered in different parts of the Toshka height (H) measured in meters and the a/c ratios and the locations area; and 5) large dune shape (‘fatty’ nature) and elevated height of of the investigated barchans in the southeastern part of the Western the Toshka barchans. This could support the conclusion that the Toshka Desert of Egypt. The first 17 barchans have been investigated for the dunes are unique in their morphometric characteristics. dune movement study

5. Conclusions

The southeastern part of the Western Desert represented mainly by Dune no. Latitude Longitude Length Width Height a/c Dune the Toshka area is characterized by considerable aeolian sand accumu- (a) (c) (H) ratio typea lations. These sands can be categorized into free-moving barchan and 1 23.36933549 30.46212531 180.55 214.05 20.39 0.84 P linear dunes and anchored (lee and shadow) dunes. A relatively low 2 23.35829092 30.4590697 150.89 161.09 15.25 0.94 P sand accumulation in the form of solitary barchan, lee dunes and 3 23.35921976 30.46244395 215.25 297.43 28.49 0.72 P shadow dunes are scattered in the eastern part of the Toshka area, M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74 73

(continued) (continued)

Dune no. Latitude Longitude Length Width Height a/c Dune Dune no. Latitude Longitude Length Width Height a/c Dune (a) (c) (H) ratio typea (a) (c) (H) ratio typea

4 23.34696085 30.46887841 139.02 207.45 19.75 0.67 N 79 22.25961191 30.30182966 231.55 175.5 16.65 1.32 F 5 23.33753278 30.46370905 407.56 405.04 38.94 1.01 F 80 22.3551144 30.2959085 208.17 176.12 16.71 1.18 F 6 23.23588047 30.45889261 413.26 380.14 36.52 1.09 F 81 22.57901379 30.37153469 128.13 100.69 9.39 1.27 F 7 23.22761538 30.45503355 402.86 290.22 27.79 1.39 F 82 22.58024132 30.37251815 58.29 57.3 5.17 1.02 F 8 23.20138828 30.55587043 252.59 245.16 23.41 1.03 F 83 22.57369245 30.37036733 46.55 59.21 5.36 0.79 P 9 23.19046799 30.5637644 255.1 244.53 23.35 1.04 F 84 22.57151289 30.37182659 113.86 75.4 6.93 1.51 F 10 23.1840701 30.56082285 269.58 285.41 27.32 0.94 P 85 22.5711963 30.36772501 127.01 92.76 8.62 1.37 F 11 22.78776306 30.84440371 241.57 145.05 13.69 1.67 F 86 22.57681215 30.36736598 172.1 87.34 8.09 1.97 F 12 22.77895616 30.84367206 175.3 156.7 14.83 1.12 F 87 22.57891424 30.36920683 44.81 31.21 2.64 1.44 F 13 22.77575491 30.84404791 151.5 237.82 22.70 0.64 N 88 22.61329469 30.36759474 138.88 121.38 11.40 1.14 F 14 22.76828231 30.84396202 128.96 189.6 18.02 0.68 N 89 22.61128786 30.36310305 77.23 60.71 5.51 1.27 F 15 22.7659679 30.84541657 219.54 113.95 10.67 1.93 F 90 22.60882584 30.36510272 132.7 94.75 8.81 1.40 F 16 22.76011405 30.84404791 274.68 322.77 30.95 0.85 P 91 23.23241428 30.68558973 111.79 126.65 11.91 0.88 P 17 22.7542602 30.84321989 143.53 199.1 18.94 0.72 N 92 23.19400162 30.61439429 187.23 165.2 15.65 1.13 F 18 23.44493897 30.46785953 203.05 192.54 18.30 1.05 F 93 23.24501744 30.59414593 166.65 165.77 15.71 1.01 F 19 23.44012016 30.47124002 151.14 145.23 13.71 1.04 F 94 23.23552556 31.55812087 382.22 293.3 28.09 1.30 F 20 23.43767169 30.46868864 100.05 124.27 11.68 0.81 P 95 23.20599913 31.55815218 141.72 95.28 8.86 1.49 F 21 23.4428341 30.47682367 217.35 146.58 13.84 1.48 F 96 22.6277727 30.44665095 286.08 222.34 21.20 1.29 F 22 23.44053177 30.48173063 105.14 146.22 13.81 0.72 N 97 22.62887239 30.39196837 130.89 165.05 15.64 0.79 P 23 23.41673741 30.46969831 202.26 173.42 16.45 1.17 F 98 23.05951844 30.46294336 68.35 70.67 6.47 0.97 P 24 23.391083 30.46513303 278.59 298.55 28.60 0.93 P 99 23.06196491 30.46279297 51.33 105 9.81 0.49 S 25 23.37662353 30.47225545 242.17 233.48 22.28 1.04 F 100 23.06634194 30.46344588 56.25 45.57 4.04 1.23 F 26 23.36032157 30.4691588 152.84 173.28 16.43 0.88 P 101 23.21205156 30.48616649 198.7 264 25.24 0.75 P 27 23.33067584 30.42639765 404.26 383.75 36.87 1.05 F 102 23.20643583 30.48986848 290.99 346.35 33.24 0.84 P 28 23.28885417 30.4239268 151.47 200.35 19.06 0.76 P 103 23.20002991 30.4784155 120.12 96.58 8.99 1.24 F 29 23.27937621 30.47587728 405.29 324.15 31.08 1.25 F 104 23.19599443 30.49193214 202.35 122.89 11.54 1.65 F 30 23.27482882 30.47133054 360.11 217.33 20.71 1.66 F 105 23.17765626 30.49023068 365.62 291.3 27.89 1.26 F 31 23.26903 30.47997856 358.33 208.38 19.84 1.72 F 106 22.7383956 30.84338745 175.7 176.6 16.76 0.99 F 32 23.28869836 30.46565101 608.15 475.98 45.82 1.28 F 107 22.73311866 30.843468 156.25 218.95 20.87 0.71 N 33 23.30002953 30.47940546 234.13 196.55 18.69 1.19 F 108 22.7240155 30.84188888 189.87 211.6 20.16 0.90 P 34 23.23034323 30.38731255 363.19 284.27 27.21 1.28 F 109 22.68641578 30.84281142 161.25 218.2 20.80 0.74 N 35 23.24960017 30.39157826 100.05 211.04 20.10 0.47 N 110 22.70459028 30.87324951 124.7 165.1 15.64 0.76 N 36 23.24996503 30.38661708 153.84 156.33 14.79 0.98 P 111 22.71723548 30.87405127 115.77 172.27 16.34 0.67 N 37 23.21923155 30.48913429 271.36 213.75 20.36 1.27 F a F = fat; P = pudgy; N = normal; S = slim. 38 23.21806763 30.45270948 231.84 125.24 11.77 1.85 F 39 23.2060776 30.4560242 245.28 185.28 17.60 1.32 F 40 23.20057696 30.4668068 74.61 153.67 14.53 0.49 S References 41 23.2014249 30.46801553 31.15 57.14 5.16 0.55 N 42 23.19892236 30.46824438 82.88 143.99 13.59 0.58 N Abdelmoaty, M.S., Ibrahim, H.M., El Samman, T.A., 2011. Protection of open channels from 43 23.1945989 30.46729698 179.35 206.95 19.70 0.87 P sand dunes movements (case study — Toshka Project). Fifteenth International Water 44 23.19058559 30.46659568 54.93 83.24 7.69 0.66 N Technology Conference, IWTC-15 2011. Egypt, Alexandria. 45 23.18815604 30.46827496 44.75 62.77 5.71 0.71 N Abou El-Magd, I., Hassan, O., Arafat, S., 2013. Quantiﬁcation of sand dune movements in 46 23.17831012 30.46506728 120.35 141.86 13.38 0.85 P the South Western part of Egypt, using remotely sensed data and GIS. J. Geogr. Inf. – 47 23.16904693 30.46547547 87.52 150.82 14.25 0.58 N Syst. 5, 498 508. Altorkomani, G., 1999. The geomorphology of Toshka area and its development potenti- 48 23.16455303 30.47193746 61.21 87.63 8.12 0.70 N alities (in Arabic). Geogr. Research Series 4, p. 218. 49 23.15850649 30.46474155 48.26 89.09 8.26 0.54 N Ashri, A.H., 1973. The movement of the Sand Dune at Kharga Oasis. Egypt. Journal of Geol. 50 23.15310341 30.47606925 172.18 186.21 17.69 0.92 P 17, 37–46. 51 23.14301148 30.51678317 71.97 123.98 11.65 0.58 N Bagnold, R.A., 1931. Journeys in the Libyan Desert 1929 and 1930. Geogr. Journal 78 (13– 52 23.14038973 30.51570579 109.06 67.23 6.14 1.62 F 19), 524–535. 53 23.1376941 30.51633254 61.52 60.02 5.44 1.02 F Bagnold, R.A., 1941. The Physics of the Blown Sand and Desert Dunes. Chapman and hall, 54 23.13522808 30.51597731 62.9 81.23 7.50 0.77 P London, p. 265. 55 23.04762236 30.69678934 312.53 238.99 22.81 1.31 F Ball, J., 1927. Problems of the Libyan Desert. Geogr. Journal 70, 21–28, 105–129, 209–225. 56 23.06075424 30.72914359 132.89 156.58 14.81 0.85 P Bastawesy, M.A., Khalaf, F.I., Arafat, S.M., 2008. The use of remote sensing and GIS for es- 57 22.69149307 30.74341194 145.06 124.91 11.74 1.16 F timation of water loss for Toshka lakes, south Western Desert, Egypt. J. Afr. Earth Sci. 58 22.65410361 30.74218445 123.59 129.11 12.15 0.96 P 52, 73–80. – 59 22.64846431 30.74323677 222.97 159.18 15.07 1.40 F Beadnell, L., 1910. The sand dunes of the Libyan Desert. Geogr. Journal 35, 379 395. 60 22.62234727 30.73930558 166.44 172.85 16.39 0.96 P Besler, H., 1986. The Toshka-Canal dune: analysis of development and dynamics. In: 61 22.60134139 30.73885185 146.19 228.57 21.80 0.64 N Nickling, W.G. (Ed.), Aeolian Geomorphology. The Binghamton Symposia in Geomor- phology. International Series 17. Allen & Unwin, Boston, pp. 113–130. 62 22.59001816 30.73902934 212.54 331.25 31.77 0.64 N Besler, H., 1998. Periods of aeolian activity and stability in the Great Sand Sea in Egypt. 63 22.32164098 30.73150679 118.59 128.5 12.09 0.92 P Zentralblatt für Geologie und Paläontologie, Teil 1 (1997), 23–39. 64 22.31747065 30.73184094 200.09 157.37 14.89 1.27 F Besler, H., Petit-Maire, N., 2000. Modern and Paleo-modeling in the Great Sand Sea of 65 22.27448457 30.71713632 181.91 171.17 16.23 1.06 F Egypt (initial results from the Cologne Cooperative Research Project 389). In: 66 22.3021565 30.4502637 238.35 160.24 15.17 1.49 F Kröpelin, S. (Ed.), Palaeomonsoon Variations and Global Change during the Late Qua- 67 22.30055202 30.45501669 256.63 156.78 14.83 1.64 F ternary. Global and Planetary Change 26. Elsevier, Amsterdam, pp. 13–24. 68 22.3028744 30.4600076 355.73 258.27 24.69 1.38 F Besler, H., 2008. The Great Sand Sea in Egypt. Developments in Sedimentology 59. Elsevier 69 22.31941233 30.27394258 189.99 106.9 9.99 1.78 F Publ, Amsterdam, p. 250. 70 22.31269298 30.27870335 241.38 166.31 15.76 1.45 F Bourke, M.C., 2010. Barchan dune asymmetry: observations from Mars and Earth. Icarus 71 22.30738183 30.27368256 209.91 111.75 10.46 1.88 F 205, 183–197. 72 22.29964921 30.27650778 202.35 231.11 22.05 0.88 P Bourke, M.C., Balme, M., 2008. Megabarchans on Mars. Planetary Dunes Workshop: A Re- 73 22.30365692 30.28116411 51.62 49.5 4.42 1.04 F cord of Climate Change. p. 2. ﬁ 74 22.27499235 30.27352602 180.73 154.7 14.63 1.17 F Geo zika Co., Zagreb-Yugoslavia, 1966. Regional geological and geophysical explorations and topographic mapping of south Kharga and Tushka area. New Valley Project, Final 75 22.27852508 30.29780967 103.5 164.51 15.58 0.63 N report. Cairo, Geology and Geophysics, E.G.D.D.O. 76 22.27356442 30.30240768 190.42 127.15 11.96 1.50 F Cornish, V., 1900. On desert sand dunes bordering the Nile Delta. Geogr. Journal 43, 1–32. 77 22.2699087 30.301921 92.1 169.05 16.02 0.54 N Effat, H.A., Hegazy, M.N., Haeck, B., 2011. Mapping sand dunes risk related to their terrain 78 22.26495231 30.30112851 77.3 103.24 9.63 0.75 P characteristics using SRTM data and cartographic modeling. J. Land Use Sci. 6, 231–243. 74 M.A. Hamdan et al. / Geomorphology 257 (2016) 57–74

El Gammal, E.A., Cherif, O.H., 2006. Use of remote sensing for the study of the hazards of Hesp, P.A., Hastings, K., 1998. Width, height and slope relationships and aerodynamic Ghard Abu Muharik Sand Dune Field, Western Desert, Egypt. The 2nd International maintenance of barchans. Geomorph. 22, 193–204. Conf. on Water Resources and Arid Environment. NARS, Cairo, pp. 1–20. Khalaf, F.I., Al-Ajmi, D., 1993. Aeolian processes and sand encroachment problems in El Shazly, E.M., Abdel Hady, M.A., El Kassas, I.A., El Amin, H., El Shazly, M.M., Abdel Megid, Kuwait. Geomorph. 6, 111–134. A., Mansour, S.I., Tamer, M.A., 1977. Geology and groundwater conditions of Toshka Khedr, E., Abu El-Magd, K., Halfawy, M., 2013a. Factor analysis of metrological data of ae- basin area, utilizing Landsat satellite images. Remote Sensing Center and Academy olian sands in arid area as a geo-environmental clue: a case study from Western Bank of Sci. Res. and Tech, Cairo, p. 75. of Lake Nasser, Egypt. International Journal of Civil & Environmental Engineering 13, El-Baz, F., 1979. Eolian features in the Western Desert of Egypt and some applications to 21–37. Mars. J. Geophys. Res. 84, 8205–8221. Khedr, E., Abu El-Magd, K., Halfawy, M., 2013b. Rate and budget of blown sand movement El-Baz, F., Wolfe, R.W., 1982. Wind patterns in the Western Desert. In: El-Baz, F., Maxwell, along the western bank of Lake Nasser, southern Egypt. Arab J. Geosci. http://dx.doi. T.A. (Eds.), Desert Landforms of Southeast Egypt: A Basis for Comparison with Mars. org/10.1007/s12517-013-1006-2. NASA CR-3611, pp. 119–139. Labib, M., Nashed, A., 2013. GIS and geotechnical mapping of expansive soil in Toshka re- El-Baz, F., Slezak, M.H., Maxwell, T.A., 1979. Preliminary analysis of color variations of gion. Ain Shams Engineering Journal 4, 423–433. sand deposits in the Western Desert of Egypt. In: El-Baz, F., Warner, D.M. (Eds.), Long, J.T., Sharp, R.P., 1964. Barchan-dune movement in Imperial Valley, California. Geol. Apollo-Souez Test Project — Earth Observations and Photography. NASA, Soc. Am. Bull. 75, 149–156. Washington, D.C., USA, pp. 237–262. McKee, E.D., 1966. Structure of dunes at White Sands National Monument, New Mexico Embabi, N.S., 1970. Structures of barchan dunes at the Kharga Oases Depression, the (and a comparison with structures of dunes from other selected areas). Sedimentol- Western Desert, Egypt (and a comparison with structures of two aeolian microforms ogy 1, 1–69. from Saudi Arabia). Bull. Soc. Geogr. d' Egypte 43/44, 53–71. McKee, E.D., 1979. A Study of Global Sand Seas. Prof. Pap. U.S. Geol. Survey, Washington Embabi, N.S., 1979. Barchan dune movement and its effect on economic development at D.C., 1052. p. 1. the Kharga Oasis depression (in Arabic). Journal of Middle East Research Centre 11, Melo, H.P.M., Parteli, E.J.R., Andrade, J.S., Herrmann, H.J., 2012. Linear stability analysis of 141–155. transverse dunes. Physica A 391, 4606–4614. Finkel, H.J., 1959. The barchans of southern Peru. J. Geol. 67, 614–647. Parteli, E.J.R., Durán, O., Herrmann, H.J., 2007. Minimalsizeofabarchandune.Phys.Rev.E Fryberger, S.G., 1979. Dune forms and wind regime. In: McKee, E.D. (Ed.), A Study of 75, 1–10. Global Sand Seas. U.S.G.S. Prof. Paper 1052, pp. 137–170. Parteli, E.J.R., Durán, O., Bourke, M.C., Tsoar, H., Pösche l, T., Herrmann, H.J., 2014. Origins Fryberger, S.G., Al-Sari, A.M., Clisham, T.J., Tizvi, A.R., Al-Hinai, K.G., 1984. Wind sedimen- of barchan dune asymmetry: insights from numerical simulations. Aeolian Res. 12, tation in the Jafurah sand sea, Saudi Arabia. Sedimentology 31, 413–431. 121–133. Gifford, A.W., Warner, D.M., El-Baz, F., 1979. Orbital observation of sand distribution in the Pye, K., Tsoar, H., 2009. Aeolian Sand and Sand Dunes. Springer-Verlag, Berlin Heidelberg, Western Desert of Egypt. In: El-Baz, F., Warner, D. (Eds.), Apollo-Soyuz Test Project p. 458. Summary Science Report, Vol. II. National Aeronautics and Space Administration, Refaat, A.A., Hamdan, M.A., 2015. Mineralogy and grain morphology of the aeolian dune Washington DC, pp. 219–236 (NASA SP-412). sand of Toshka area, southeastern Western Desert, Egypt. Aeolian Res. 17, 243–254. Goudie, A.S., Bourke, M.C., 2008. Varieties of barchan form in the Namib Desert and on Sauermann, G., Rognon, P., Poliakov, A., Herrmann, H.J., 2000. The shape of the barchan Mars. Planetary Dunes Workshop: A Record of Climate Change, p. 2. dunes of Southern Morocco. Geomorph. 36, 47–62. Hamdan, M.A., 2003. Textural, mineralogical and geochemical characteristics of the aeo- Sharaky, A.M., Labib, T.M., Philip, G., 2002. Sand dune movement and its effect on culti- lian dune sand of the Gilf-Uwinat area, South Western Desert. 5th International vated lands in Africa: case study: Dakhla Oasis, Western Desert, Egypt. Land Degrada- Conf. of the Geology of the Middle East, pp. 171–187. tion in Egypt and Africa (23–24 March 2002), Cairo University, pp. 1–15. Hamdan, M.A., Refaat, A.A., 1999. Textural and mineralogical characteristics and distribu- Stokes, S., Maxwell, T.A., Haynes, C.V., Horrocks, J., 1998. Latest Pleistocene and Holocene tions of aeolian sand deposits at Qattara-Siwa area, northwestern desert, Egypt. Int. sand-sheet construction in the Selima Sand Sea, Eastern Sahara. In: Alsharhan, A.S., Conf. on Geol. of the Arab World, GAW4, 1998. Cairo Univ., Egypt, pp. 684–706. Glennie, K.W., White, G.L., Kendall, C.G. (Eds.), Quaternary Desert and Climate Hamdan, M.A., Refaat, A.A., Abu Anwar, E., Shallaly, N.A., 2015. Source of the aeolian dune Changes. Balkema, Rotterdam and Brookfield, pp. 175–183. sand of Toshka area, southeastern Western Desert, Egypt. Aeolian Res. 17, 275–289. Tag El Din, S.S., 1986. Some aspect of sand stabilization in Egypt. In: El-Baz, F., Hassan, Hastenrath, R.L., 1967. The barchans of the Arequipa region, southern Peru. Z. Geomorph. M.F.A. (Eds.), Physics of Desertification. Martinus-Nijhoff, Dordrecht, pp. 118–126. 11 (3), 300–331. Tsoar, H., 1985. Profile analysis of sand dunes and their steady state signification. Geogr. Haynes, C.V., 1989. Bagnold's Barchan: A 57-Yr record of dune movement in the Eastern Ann. 67A (l-2), 47–59. Sahara and implications for dune origin and paleoclimate since Neolithic times. Quat. Wahby, W.S., 2004. Technologies applied in the Toshka Project of Egypt. J. Technol. Stud. Res. 32, 153–167. 30, 86–91. Hereher, M.E., 2010. Sand movement patterns in the Western Desert of Egypt: an environmental concern. Environ. Earth Sci. 59, 1119–1127.