The Relationship between Groundwater, Landuse, and Demography in Dakhla Oasis, Egypt

Transcription

1 The Relationship between Groundwater, Landuse, and Demography in Dakhla Oasis, Egypt Hiroshi KATO 1, Salwa ELBEIH 2, Erina IWASAKI 3, Ahmed SEFELNASR 4, Adel SHALABY 2, and Elsayed ZAGHLOUL 2 Abstract: Understanding of the continuous dynamic relationship between water availability and land use /land cover change (LUCC) is an essential step in urban development planning. This is particularly crucial in dryland environments, where water and fertile soil are very limited and nonrenewable. Dakhla Oasis is located in the heart of the Western Desert of Egypt, 190 km to the west of Kharga Oasis, and is the oasis furthest from the main settlements of Egypt. Dakhla Oasis contains highly fertile lands and is rich in groundwater. It supports a relatively high population compared to Kharga Oasis with about 100,000 inhabitants. Groundwater is the only water source for irrigation and domestic use. The economy of Dakhla depends on several industries including agriculture, handicrafts and tourism. This paper presents the main results of an interdisciplinary research project in Dakhla Oasis with special emphasis on Rashda Village within the Oasis. The study is a model investigation of groundwater resources, land use/land cover and their link with demographic characteristics. One 1972 Landsat Multispectral Scanner System image, two 1984 Landsat Thematic Mapper images and Six 2011 SPOT4 satellite images were used for LUCC detection. In addition, a detailed database of groundwater extraction from 1960 to 2005 was available. Supervised image classification and visual interpretation were integrated to achieve more accurate LUCC maps. The results show that the study area has undergone very severe land cover changes with significant increases in urban settlements and in agricultural land. Local aquifer recharge and recovery fails to compensate for artificial groundwater extraction. The overall discharge increased between 1976 and 2006 by 54% to 62%, in different areas. This was associated with increases in population ranging from 8.5% to 150% in the same period. If extraction continues at current rates, pumping with current technology will become uneconomical in the next 90 years. Keywords: Remote Sensing, groundwater resources, land use, Rashda Village, Dakhla Oasis 1 Hitotsubashi University, Japan 2 National Authority for Remote Sensing and Space Sciences, Egypt 3 Sophia University, Japan 4 Geology Department, Faculty of Science, Assiut University, Egypt 3

2 1. Introduction 1.1 Research concern Food scarcity and the continuous loss of agricultural land are issues of global concern. The government of Egypt has adopted policies aimed at self-sufficiency in food production, e.g. extension of cultivated areas and maximization of production from the existing agricultural land. The principal purpose is to counteract Egypt's predominantly unfavorable population to agricultural land ratio. Urbanization is an inevitable process due to economic development and rapid population growth. Encroachment of urban settlements on agricultural lands may entail dire consequences. The continuous increase in population density causes increasing pressure on areas already inhabited and has caused a decrease in area per capita from 0.12 ha in 1950 to 0.06 ha in 1990 (Suliman 1991) and to 0.04 ha in 2009 (CAPMAS 2009). Desert and uninhabited lands represent about 95% of the total area of Egypt. However, the majority of the population is concentrated around the River Nile. This unbalanced distribution causes serious social and economic problems. Since the 1980s, the Egyptian government has initiated plans to adjust this situation through redistribution of population growth by restricting horizontal urban expansion to the desert areas near the fringes of the Nile Delta. This policy aims to reduce pressure on old and highly productive agricultural land, decrease population density in the inhabited areas and decrease pollution sources by establishing industrial areas outside the Nile Valley and Delta. Therefore, determination of the trend and rate of land cover conversion is necessary for development planning in support of rational land use policy (Shalaby and Tateishi 2007). There are many techniques available to detect and record differences in satellite images (e.g. image differencing, ratios or correlation) that might be attributed to land use change (Singh 1989; Stow et al. 1996; Yuan et al. 1999). However, the simple detection of change is rarely sufficient in itself: information is generally required on the initial and final land cover types or land uses from-to analysis (Khorram et al. 1999). Post-classification comparison of derived thematic maps goes beyond simple change detection and attempts to quantify the different types of change. The degree of success depends upon the reliability of the maps made by image classification. Broadly speaking, large-scale changes such as widespread logging or major urban development can be mapped reasonably easily. With evolutionary changes such as erosion, succession, colonization, or degradation, the boundaries may be indistinct, and class-labels uncertain (Foody and Boyd 1999; Khorram et al. 1999; Shalaby and Tateishi 2007). 1.2 Study area The site with which the paper is concerned is Dakhla Oasis. It is located in the heart of the Western Desert of Egypt between longitudes and east and latitudes and north, 190 km to the West of Kharga Oasis. Dakhla extends 155 km from Tenieda Village in the East to Mawhoub Village in the West (Figure 1) with an area about 155 km long and 60 km wide suitable for agriculture. It contains highly fertile lands, it is rich with water and it supports a higher population than Kharga Oasis. Dakhla is the biggest oasis in Egypt s Western Desert and lies farthest away from the main cities. About 100,000 inhabitants live in 17 different settlements. The whole depression of the Dakhla Oasis comprises a number of smaller oases, separated by hills or desert. Among the 17 settlements, only Mut, Qasr and Balat are qualified as towns. Tenieda and Balat are the main settlements in the east. Mut, in the middle, is the largest town in the Oasis with more than 20,000 inhabitants, while Qasr and Mawhoub are located in the west. The climate of Dakhla Oasis is hot and dry, with a high rate of evaporation, a high level of solar radiation (sunshine), and no rainfall. The aridity index used by the United Nations Environment Program (UNEP 1997), indicates that this district lies in a hyper-arid region, with an annual rainfall close to 0.0 mm. Originally, Dakhla was fed by about 520 springs and ponds, but in modern times many have dried out, and others only yield water through use of electric pumps (Sefelnasr et al. 2014). The economy of Dakhla is based on agriculture, production of handicrafts, and some tourism. Before the road was constructed, Dakhla must have felt like a planet of its own. Only a few inhabitants ever travelled as far as the neighboring oases Kharga and Farafra. Rashda is one of the villages located in the Dakhla Oasis. Interdisciplinary research has been conducted in this village since 2005 (Kato and Iwasaki 2008; Kato et al. 2010, 2012). Figure 1. Location map of Dakhla Oasis (overlaid over a 2011 SPOT4 satellite image) In summary, Dakhla Oasis is typical dryland, where water and fertile soil are very limited and nonrenewable. In such environments, the ever-increasing population and urban sprawl, in addition to land degradation, put huge pressure on the limited agricultural land with consequent decreases in area per capita. The amount, rate and intensity of loss of agricultural land need to be analyzed. Hence, changes in water resource distribution and land cover, along with the environmental impact of these changes, should be carefully considered. 4

3 1.3 Hydrogeology of Dakhla Oasis The Nubian Sandstone Aquifer System is composed of different water bearing strata laterally and/or vertically interconnected. Dakhla Basin is the largest and the most important basin in Egypt. It consists of fine to coarse-grained continental sandstone from the Lower Cretaceous with good permeability and thickness reaching 3,500 m (Fig. 2). Unmanaged groundwater extraction in Dakhla Oasis has led to several changes that can be seen in the hydraulic head patterns and flow directions. Continued uncontrolled pumping will lead to shallow wells becoming completely dry and increases in the depth to groundwater, which will become too large to for extraction to be economic. Figure 3. Map for the land use checkpoints distributed throughout Dakhla Oasis (4) Groundwater data A groundwater database was prepared for water wells in Dakhla villages (as shown in Figure 4) for the years from 1960 to 2005 with discharge in cubic meters per day (Table 1). Table 1. Part of the groundwater database of Dakhla Oasis; discharge in cubic meters per day (1960 to 2008) Year Tenieda Balat Ismant Maasara Mut Hindaw Oweina Rashda Qalamon Figure 2. Simplified cross sections of selected locations in the east, middle and west of the Dakhla oasis based on the well logs (modified after Sefelnasr, 2002, 2007) The objective of this study was to understand the on-going dynamic relationship between groundwater resources, land use /land cover and demographic trends in Dakhla Oasis. By integrating remote sensing and geographic information system (GIS) manipulation, the spatial distribution of land cover change and its impact on agricultural land were determined. 2. Data and methods 2.1 Data (1) Population census The statistical data for population were collected from the national censuses, which were executed from 1897 to 2006, roughly every 10 years (CAPMAS, various years). (2) Satellite data The satellite data used were one Landsat Multispectral Scanner System (MSS) image acquired in Nov. 1972, two Landsat Thematic Mapper (TM) images acquired in November and December 1984 and six SPOT4 satellite images acquired in August and September (3) Site investigation data A total number of 217 ground truth points for different land use and land cover classes were collected during fieldwork in April These check points are distributed among all the villages of Dakhla Oasis. The assigned land use classes were urban, bare land, perennial crops, annual crops, water bodies, land under reclamation, and sabkhas (salt flats) (Figure 3). Figure 4. Location map of the Dakhla Oasis showing the distribution of pumping-well fields (Sefelnasr et al. 2014) (5) Topographic and hydrogeological maps Four 1: topographic maps centered on Rashda, Mut, Balat and Qasr respectively and issued in 1986, one 1:25000 historical topographic map of Rashda issued in 1932, and one 1: hydrogeological map of the Dakhla oasis issued in 1998 were used, in addition to field data and images, to support mapping of land cover changes in Dakhla Oasis. 2.2 Methods: image processing Geometric correction of the SPOT4 satellite images of 2011 was carried out using ground control points from the topographic maps. This image was then used to register the images of 1972 and The root mean square error between the images was less than 0.4 pixels, which is acceptable. 5

4 Image enhancement was carried out to improve the visual interpretability of the images by increasing the apparent distinction between the features. Contrast stretching was applied on the three images and three false color composites (FCC) were produced. These FCC were visually interpreted using on-screen digitizing in order to delineate land cover classes that could be easily interpreted, such as urban and sabkha. Land cover classes are typically mapped from digital remotely sensed data through the process of supervised digital image classification (Campbell 1977; Thomas et al. 1987; Shalaby and Tateishi 2007). The overall objective of the image classification procedure is to automatically categorize all pixels in an image into land cover classes or themes (Lillesand and Kiefer 1994). The maximum likelihood classifier quantitatively evaluates both the variance and covariance of the spectral response patterns of the image classes when classifying an unknown pixel. It is considered to be one of the most accurate classifiers since it is based on statistical parameters. Supervised classification was done using ground checkpoints and digital topographic maps of the study area. Then accuracy assessment was carried out using 300 field data points and existing land cover maps. In order to increase the accuracy of land cover mapping of the three images, ancillary data and the results of visual interpretation were integrated with the classification result using GIS. A post-classification change detection technique was applied. The most obvious method of change detection requires the comparison of independently produced classified images. Post-classification comparison proved to be the most effective technique, because data from two dates are separately classified, thereby minimizing the problem of normalizing for atmospheric and sensor differences between the two dates. Cross-tabulation analysis was carried out to study the spatial distribution and area of different land cover changes. The methodology adopted in the study is shown in Figure 5. more detail). From this demographic analysis, the settlements in the oases are characterized by a rapid increase between 1960 and The figure shows that the population growth in the two oases was below the national average until In fact, in Dakhla Oasis the population increased only very slowly from in 1897 to in However, its population increased suddenly to in This population growth was three times higher than the national average, and is obviously due to population inflow from outside the oases. The principal reason was population inflow enhanced by national settlement projects, such as New Valley Project, during the 1960s and 1970s Egypt Kharga Dakhla Figure 6. Demographic Trend in Kharga and Dakhla Oases and the whole of Egypt (annual population growth %) (Source: Population Census [various years]) The demographic trends, at the level of the smallest administrative unit (shiaykha [town] or village) (Table 2, Figs. 7a, 7b, 7c), reveal that tremendous population growth occurred within certain village administrative units. Although population data at village level is not available for 1960, the data in 1947 and 1976 suggest that the greatest population growth occurred within the administrative boundaries of Mut, Rashda, Bedkhulu, Qasr, Tenieda, and Hendaw during the 1960s up to After 1976, it was Mut, Qasr, Hendaw, and Balat that experienced above average population growth, especially between 1996 and These villages that experienced population inflow were not newly created. They existed before the New Valley Project was launched. In Balat and Qasr, many new administrative units were created between 1996 and 2006, but they became administratively independent from these two villages (Figure 7d). Therefore, the demographic explosion in Dakhla occurred within the preexisting village boundaries and was not accompanied by the new creation of villages Figure 5. Methodology flowchart 3. Results and discussion 3.1 Demographic trends of Dakhla Oasis Figure 6 shows the annual population growth of the whole of Egypt, Kharga Oasis and Dakhla Oasis from 1907 to 2006, based on Egyptian population censuses (cf. Kato et al for 3.2 Land use and land cover The false color composites generated from bands 4, 3 and 2 of the three images were visually interpreted through on-screen digitizing. The visual interpretation gave a general idea about land cover changes over the study period. Many urban settlements were erected in the oasis during the study period. A noticeable change is detected in areas of land reclamation in the study area where part of the desert was converted to cultivated lands. On-screen digitizing was carried out for urban, sabkha 6

5 and salt marsh land cover classes as well as the road network and integrated with the result of classification, using GIS to enhance the accuracy of the classification results. Supervised classification using all reflective bands of the three images acquired in 1972, 1984 and 2011 was carried out using a maximum likelihood classifier. In order to increase the accuracy of land cover mapping of the three images, ancillary data and the result of on-screen interpretation were integrated with the supervised classification results using GIS. Figure 7d. Newly created administrative units between 1996 and 2006 Source: Population Census [1947], [1976], [1996], [2006] Table 2. Demographic trend in the administrative units (village and town) in Dakhla Oasis ( ) Figure 7a. Population by administrative unit (1947) Population New shiyakha created between Population increase (%) Mut 2,513 8,122 11,890 16,252 20, Ismant 1,277 2,169 2,737 2,690 2, Gedida 2,789 2,991 3,369 3,142 3, Rashda 1,739 4,398 5,574 4,364 5, Qalamon 1,587 2,272 2,184 2,128 1, Sheikh Wali 1,721 2,054 1,990 2, Oweina 1,659 1,884 Qasr 2,155 5,813 5,257 3,723 10, Maasara 1,202 2,374 3,031 3,282 3, Moshiya 1,146 1,571 1,778 2,146 2, Hindaw 1,038 2,174 2,774 3,062 3, Bedkhulu 537 1,113 1,216 1,523 1, Balat 2,668 4,952 5,492 5,126 6, Tenieda 1,020 2,261 2,774 3,114 3, Ezab Qasr 1,711 3,210 3, Mawhoub 1,871 1,678 2 Mawhoub West 3,879 6,278 7, Total 21,382 45,141 57,881 62,350 79, (Source: Population Census [various years]) Figure 7b. Population by administrative unit (1976) This overlay of the visual interpretation on the result of the classification led to increases in the overall accuracies by about 10% for the three images. The benefits of adding the on-screen interpretation flows from the fact that the urban settlements are spectrally confused with bare land class because these building are made from the surrounding local materials. Post-classification change detection was carried out through a cross-tabulation GIS module, for the classification results of 1972, 1984 and 2011 images, to produce change images (Figure 8) and statistical data (Table 3). Figure 7c. Population by administrative unit (2006) Table 3. Change of agricultural land and urban settlements in Dakhla Oasis villages between 1984 and 2011 and population between 1986 and 2006 Agric. Urban Population % increase between (km 2 ) (km 2 ) 1984 and Agric Urban Pop Balat & Tenieda 16,983 36,734 0,427 1,527 8, ,29 257,61 18,55 Mut 36,846 69,254 3,486 6,679 11, ,95 91,59 70,67 Qalamon & Rashda 55, ,31 1,369 3,175 7, ,49 131,92-6,74 Total 109, ,298 5,282 11,381 27, ,10 115,46 33,72 7

6 Table 4: Total Extraction in cubic meters per day in Balat Tenieda, Mut and Qalamon Rashda Villages Extraction (m3/day) % change in extraction Population Pop increase% ( ) ( ) ( ) ( ) Balat & Tenieda 114, , , ,213 8,266 9, Mut 68, , , ,122 11,890 20, Qalamon & Rashda 42,982 71,504 69, ,670 7,758 7, ,74 Total 225, , ,109 22,005 27,914 37,328 Figure 8: Change in agricultural areas distribution in Dakhla Oasis villages from 1972 to 2011 Table 3 compares the rates of land cover changes with the rates of population changes for Mut, Balat Tenieda and Qalamon Rashda for the periods and respectively. For example, the total agriculture land cover increased from km 2 in 1984 to km 2 in In other words, it increased by about 90%. The total urban land cover increased from 5.28 km 2 in 1984 to km 2 in 2011 a growth of about 115% during this period. In addition, the table shows that the total population increased in the period from 27,914 to 37,328 with a growth rate of about 34%, which accompanied the even greater increase in urban land cover and agricultural land. 3.3 Groundwater extraction Dakhla Basin is the largest and the most important structural geological unit in Egypt. It consists of fine to coarse-grained continental sandstone ranging in ages from Paleozoic to Lower Cretaceous with good permeability and thickness reaching 3,500 m. The area of the Nubian Sandstone Aquifer System (NSAS) is composed of different interconnected water-bearing strata. Groundwater mining of the reserves of the NSAS is presently taking place and is increasing yearly. The unmanaged groundwater extraction in Dakhla Oasis has led to several changes: 1. Altered hydraulic head patterns and flow directions indicated by the development of local flow patterns. 2. Noticeable cones of depression, as a result of continuous groundwater extraction, in the last 50 years (since the beginning of the New Valley Project in the 1960s). 3. Increasing groundwater salinity. It is obvious that the aquifer in Dakhla Oasis is non-renewable, this is clear from the continuous lowering of the groundwater level in wells within the oasis and the change of the hydraulic behavior of some wells, from being artesian flowing to non-flowing especially on the eastern side of the oasis. Dozens of shallow wells have dried out already. The aquifer is still being stressed by pumping from wells that double in number from one decade to another. The development ambitions of both the government and residents lead to increasing utilization of the groundwater resources. Figs. 9a, b and c show that extraction pattern had the same profile from 1960 to 2005 with extraction concentrated near some main villages, especially Moshiya Gedida, (characterized by high extraction rates). This almost constant extraction profile also reflects that the development in Dakhla Oasis has not extended to additional new villages from 1960 to On the other hand, extraction rates has increased in the same areas, starting from the 1960s. From Table 4, it is clear that the overall discharge increased from 1976 to 2006 by 55% to 63% (depending on area), which is associated with an increase in population ranging from 8% to 150% in the same period. By contrast, from 1986 to 2006 the discharge decreased by between 2% and 14% while population continued to increase by 19% to 71%. This decrease in discharge in the last 20 years could be attributed to some combination of: Figure 9a: Change in the extraction pattern, in cubic meters per day, in Dakhla Oasis villages in 1960 Figure 9b: Change in the extraction pattern, in cubic meters per day, in Dakhla Oasis villages in

7 Acknowledgments The authors are grateful for the National Authority for Remote Sensing and Space Sciences (NARSS) for supplying the recent satellite images and for facilitating the field trips to the research site. References Figure 9c: Change in the extraction pattern, in cubic meters per day, in Dakhla Oasis villages in Using modern methods in irrigation of cultivated lands including (sprinklers, drip irrigation). 2. Lowering of the groundwater table. 3. Drying of some wells. 4. Conclusions This study showed that, from a methodological viewpoint: Combination of statistical data with geographical information is significant for the study of oasis societies. Integrating visual interpretation with supervised classification leads to an increase in the overall accuracy of land use/cover classification. Integrating GIS and Remote Sensing provides valuable information on the area and spatial distribution of urban expansion. This analysis showed that Dakhla Oasis has undergone very significant land cover changes, increases in urban settlements and in agricultural land, because of land reclamation and population growth. The total population increased in the period from 27,914 to 37,328 with growth rate of about 34%, which is a clear basis for an even greater increase in the total urban land cover and area of agricultural land. In Dakhla Oasis, the relative abundance of water enabled this rapid increase in population in the latter half of the 20th century. The overall discharge of water continuously increased from 1976 to 1986 associated with an increase in population ranging from 8.5% to 150%, in different areas, in the same period. Then, the discharge started to decline for several reasons including lowering of the water table. Local recharge and aquifer recovery cannot compensate for the present level of artificial water extraction. If extraction continues at current rates, pumping will become uneconomic in the next 90 years with current technology, as the depth to groundwater will exceed the so-called "100 m limit" of lifting depth. This situation may occur sooner if any increase in extraction occurs. This dark scenario for Dakhla Oasis is realistic given the population increase and excess use of land. CAPMAS Statistical Year Book. Central Agency for Public Mobilization and Statistics (CAPMAS), annual report, Cairo, Egypt. CAPMAS. Various Years. Population Census. Campbell, J.B., Introduction to Remote Sensing. New York: Guilford. Foody, G.M. and Boyd, D.S Detection of Partial Land Cover Change Associated with the Migration of Inner-Class transitional zones, Int. J. Remote Sensing, 20, pp Kato, H. and Iwasaki, E Rashda: A Village in Dakhla Oasis, Mediterranean World, 19, pp Kato, H., Iwasaki, E., Nagasawa, E., Anyoji, H., Matsuoka, N. and Kimura, R Rashda: System of Irrigation and Cultivation in a Village in Dakhla Oasis, Mediterranean World, 20, pp Kato, H., Kimura, R., Elbeih, S.F., Iwasaki, E., and E.A. Zaghloul Land Use Change and Crop Rotation Analysis of a Government Well District in Rashda Village Dakhla Oasis, Egypt Based on Satellite Data, Egypt. J. Remote Sensing Space Sci. 15 (2), pp Kato, H., Tsumura, H., and Iwasaki, E GIS as a Tool for Researching the Socioeconomic History of Modern Egypt, Journal of Asian Network for GIS-based Studies (JANGIS), 1, pp Khorram, S., Biging, G.S., Chrisman, N.R., Congalton, R.G., Dobson, J.E, Ferguson, R.L., Goodchild, M.F., Jensen, J.R. and Mace, T.H Accuracy Assessment of Remote Sensing-Derived Change Detection, American Society of Photogrammetry and Remote Sensing, Bethesda. Lillesand, T.M. and Kiefer, R.W Remote Sensing and Image Interpretation, 4th edition, New York, John Wiley and Sons. Sefelnasr, A Hydrogeological Studies on Some Areas on the New Valley Governorate, Western Desert, Egypt. MSc thesis, Assiut University. Sefelnasr, A Development of groundwater flow model for water resources management in the development areas of the Western Desert, Egypt, Ph.D. thesis, Martin Luther University. Sefelnasr, A., Gossel, W. and Wycisk, P Three Dimensional Groundwater Flow Modeling Approach for the Groundwater Management options for the Dakhla Oasis, Western Desert, Egypt, Environmental Earth Sciences, DOI: /s Shalaby, A. and Tateishi, R Remote Sensing and GIS for Mapping and Monitoring Land Cover and Land-Use Changes in the Northwestern Coastal Zone of Egypt, 9

8 Applied Geography, 27, pp Singh, A Digital Change Detection Techniques Using Remotely Sensed Data, International Journal of Remote Sensing, 10, pp Stow, D. A., Chen. D.M. and Parrott, R Enhancement, Identification and Quantification of Land Cover Change in Morain, S.A., Lopez Barose, S.V. (Eds.), Raster Imagery in Geographical Information Systems, One Word, Santa Fe, pp Suliman, M. K Universities and Development of the Desert Land in the ARE, The Second Annual University Conference, Cairo, 2-5 November Thomas, I.L., Benning, V.M. and Ching, N.P Classification of Remotely Sensed Images, Bristol: Adam Hilger. UNEP (United Nations Environment Programme), World Atlas of Desertification 2ED. UNEP, London. Yuan, D., Elvidge, C.D. and Lunetta, R.S Survey of Multi-Spectral Methods for Land Cover Change Analysis in Lunetta, R.S., Elvidge, C.D. (Eds.), Remote Sensing Change Detection: Environmental Monitoring Methods and Applications. Taylor & Francis, London, pp