Abandoned Oases represented by El-Arag, El-Bahrein, El-Numessa, and Setra depressions are located in the northern part of the Western Desert of Egypt to the ...
LAND RESOURCES PROCESS MODELING OF ABANDONED OASES, WESTERN DESERT, EGYPT M. M. Ibrahim1, A. H. El Nahry2 and A. M. Saleh2 Soil and Water Department, Faculty of Agriculture, Tanta University 2 National Authority for Remote Sensing and Space Sciences (NARSS) 1
ABSTRACT Abandoned Oases represented by El-Arag, El-Bahrein, El-Numessa, and Setra depressions are located in the northern part of the Western Desert of Egypt to the South of ElQattara Depression. They lie between longitudes 26° 15´ 00” & 27° 18´ 45” E and Latitudes 28° 37´ 30” & 29° 00´ 00” N. The study area represents small and inconspicuous oases characterized by scattered natural vegetation and some wild palm tress. These oases lie within the arid zone. The aim of this study is to apply the powerful tools of advanced Remote Sensing (RS) and Geographic Information System (GIS) techniques through managing and integrating spatial modeling data for capability evaluation and agricultural land use priorities. Terrain units were identified using draped satellite ETM+ image over Digital Terrain Mode (DTM) to express the landscape and the associated landforms. The major landforms of the area under consideration could be grouped and descried as basins, sandy flats, sandy plains, peneplains, bolsons, alkali flats, footslopes, plateau, sand dunes, inter-dunal sand strips, isolated hills, mesas, rock out crops, salt efflorescence and corrosion forms. Using US Taxonomy bases, soils of the abandoned oases could be classified into Typic Aquisalids, Typic Torrifluvents, Gypsic Haplosalids, Lithic Torripsamments, Lithic Haplocalcids, Typic Torripsamments, Typic Haplocalcids, Typic Haplosalids, Calcic Aquisalids, and Typic Psammaquents. Land capability evaluation was performed using CERVATANA capability model. Three capability classes could be recognized as follows (S2) Land with good use capability, (S3) Land with moderate use capability and (N Marginal or non-productive land. Site Selection Model (SSM) was designed with the aid of geomorphologic and soil units, capability classes, slope, and slope gradient for getting the available grades of agricultural land use priorities. INTRODUCTION Abandoned Oases lie in the northern part of the Western Desert of Egypt, south of Qattara Depression and situated about 168 km to the east of Siwa Oasis. It is bounded to the north by the southern borders of Qattara Depression, while towards the south; it is limited by the dunes of the Great Sand Sea and extends about 41.8 km to the south direction. The study area is bounded by longitudes 26° 15´ 00” & 27° 18´ 45” E and latitudes 28° 37´ 30” & 29° 00´ 00” N. The availability of advanced technology for managing large sums of data should help the planners and decision makers to organize the information, understand their spatial association, and provide a powerful means for analyzing and synthesizing relevant information. Moreover, the launching of space-borne satellite for gathering information about the state of land over time allows the review of changes in land use and land cover during different periods. These advanced and powerful tools include Remote Sensing (RS) and Geographic Information System (GIS). Applying the capabilities of these information systems through managing and integrating spatial data enabled us to analyze terrain and associated soils of Abandoned Oases for producing digital geomorphologic soil, and capability maps as a base of site selection and defining agricultural land use priorities.
MATERIALS AND METHODS Digital image processing was performed using ENVI software version 4.1 to achieve the objectives of the current study. Image processing included images calibration to reflectance, enhancement, rectification, and sub-setting. For enhancing the ground resolution from 30 m to 15m, the fusion methodology was applied according to Ranchin and Wald (2000). A rapid reconnaissance survey was made throughout the investigated area in order to identify the major landforms and gain an appreciation of the broad soil patterns and characteristic landscape. The primary mapping units were verified based on the pre-field interpretation and the information gained during the reconnaissance survey. Twenty five soil profiles were dug to fulfill the requirements of the digital soil maps and 220 augers were executed for the purpose of recognizing the boundary among the different mapping units. A detailed morphological description of soil profiles were noted based on the basis outlined by FAO (1990). A land capability evaluation procedure was applied using the MicroLEIS- CERVATANA model (De La Rosa et al., 2004). The land capability evaluation applied by linking MicroLEIS-CERVATANA model with ArcGIS 9 (Environmental Systems Research Institute, ESRI 2004) spatial modeling environment using relational database files which have identifier key attribute property. Site Selection Model (SSM) was designed with the aid of geomorphologic and soil units, capability classes, slope, and slope gradient to produce grades of land use priorities for agriculture purposes. ArcMap 9.0 (ESRI, 2004) was used to display and produce geomorphologic, soil, land capability, and site selection maps. RESULTS AND DISCUSSION An image data fusion procedure was applied to have a high differential accuracy and desired resolution quality. The satellite ETM+ multispectral bands (28.5 m resolution) were sharpened using ETM+ panchromatic band (14.25 m resolution). The hue and saturation of the multispectral bands were merged with the Value of the panchromatic band to produce a new enhanced multispectal image with 14.25 m resolution. The Digital Terrain Model (DTM) is defined as continuous variation of relief over space, (Burrough, 1986). DTM is consisting of a sampled array of regularly spaced elevation values referenced horizontally to a geographic coordinate system (USGS, 1998). Extracted DTM of abandoned oasis has a profile of 30 m square grid spacing along and between each profiles, grid columns and rows. The 30 m spatial resolution was essential in order to coincide with that of the Landsat ETM+ imagery to identify the geomorphology and terrain analysis. Satellite ETM+ image was draped over the Digital Terrain Model (DTM) to get the feel of natural 3D terrain and a better understanding of the geomorphologic units and to facilitate extracting of these units. Geomorphologic units encountered could be categorized as follows: plateau, sand dunes, alkali flats, depressions, bolsons, peneplains, basins, mesas, footslopes, sandy plains, inter-dunal sand strips, sandy flats, rock out crops, mantle overlain bedrock, efflorescence forms, isolated hills, and corrosion forms. According to the Soil Survey Staff (1999), soils of the area could be classified as follows: Soils of sandy plains: The depth of soils is around 150 cm, texture is sandy, pH is between 7.2 and 7.65, EC ranges between 1.4 and 2.1 dS/m, and the CaCO3 content is about 5.0. The soils are classified as Typic Torripsamments Soils of sand flats: The soil depth varies around 150 cm, texture is sandy, pH is between 7.4 and 8.4, EC ranges between 1.1 and 2.6 dS/m, and the CaCO3 content ranges between 7.4 and 17.2%. The soils are classified as Typic Torripsamments.
2
Soils of hummocks: The depth ranges between 80 and 150 cm with different pattern of sedimentation having sandy to loamy sand textures, the pH ranges between 8.2 and 8.5, the EC ranges between 1.5 and 90 dS/m, and the CaCO3 content ranges between 6.2 and 11.7%. The soils are classified as Typic Torripsamments, Typic Psammaquents, and Typic Haplosalids. Soils of footslopes: The soil depth ranges around 35 cm, the texture fluctuates between sandy to loamy sand, the pH ranges between 7.2 and 8.2, EC ranges between 1.4 and 104 dS/m, and the CaCO3 content ranges between 12.9 and 40.1%. The soils are classified as Lithic Haplocalcids. Soils of pediments: The depth ranges between 30 and 40 cm with a lithic contact, the texture is sandy, the pH ranges between 7.15 and 7.8, EC ranges between 1.7 to 4.8 dS/m, and the CaCO3 content ranges between 2.8 and 9.0%. The soils are classified as Lithic Torripsamments. Soils of peneplains: The depth of soils ranges between 50 and 55 cm with occasional lithic contact, the texture is sandy, the pH ranges around 7.2 and 7.7, the EC ranges between 7.4 to 12.4 dS/m, and the CaCO3 content ranges between 10.0 and 17.4 %.. The soils are classified as Lithic Haplocalcids. Soils of overflow basins: The depth is between 95 and 150 cm, the texture is sandy loam, the pH is between 7.4 and 7.8, the EC ranges between 0.74 and 1.2 dS/m, and the CaCO3 content ranges between 4.6 and 14.0 %. The soils are classified as Typic Torripsamments and Typic Torrifluvents. Soils of decantation basins: The depth is between 80 and 150 cm, the texture ranges between loamy and sandy loam and sandy, the pH is between 7.7 and 8.3, the EC ranges between 3.6 and 52 dS/m, and the CaCO3 content ranges between 0.9 and 8.9 %. The soils are classified as Typic psammaquents and Gypsic Haplosalids. Soils of old lake beds: The depth soils is around 150 cm, the texture class ranges between loamy sand to silty loam, the pH is between 7.7 and 7.9, the EC ranges between 3.45 and 5.75 dS/m, and the CaCO3 content ranges between 24.6 and 50.4%. The soils are classified as Typic Haplocalcids. Soils of wet sabkhas: The depth is between 35 and 50 cm, the texture ranges between loamy sand and silty loam, the pH is 7.5 to 7.8, the EC ranges between 38.3 to 117.0 dS/m, and the CaCO3 content ranges between 9.3 and 18.4%. The soils are classified as Typic Aquisalids. Soils of dry sabkhas: The depth of soils is around 65 cm, the texture is sandy, the pH ranges between 8.0 and 8.4, the EC ranges from 43.2 to 72.5 dS/m, and the CaCO3 content ranges between 12.0 and 16.0%. The soils are classified as Calcic Aquisalids. Land Capability The CERVATANA model works interactively, comparing the values of the characteristics of the land-unit to be evaluated with the generalization levels established for each use capability class. Following the generally accepted norms of land evaluation (FAO, 1976; Dent and Young, 1981; and Verheye, 1986), the CERVATANA model forecasts the general land use capability for a broad series of possible agricultural uses. The methodological criteria refer to the system designed earlier by De La Rosa and Magaldi (1992) and modified for computing purposes by De La Rosa et al. (2004). The prediction of general land use capability is the result of a qualitative evaluation process or overall interpretation of the following biophysical factors: relief, soil, climate, and current use or vegetation.
3
According to CERVATANA model, three capability classes were recognized. Class S2 includes land with good use capability, Class S3 includes land with moderate use capability, whereas Class N includes marginal or non-productive land. Site Selection Model To find the best priorities of land use for agriculture of abandoned oases, a site selection spatial model was designed and processed in ArcGIS spatial modeling environment. The site selection model was based on the following parameters: geomorphologic units, soil units, capability classes, slope, and slope gradient. The first step in the model is to input the needed data to the model, and then in the second step the entered data were derived to gain new information. The new information was then classified to common scale where the higher values were given to the more suitable particular locations for agriculture. In the last step, the classified data weighted according to a percentage influence where the higher the percentage, the more influence a particular location will have in the model. After weighting, the data combined to produce a map of grades of land use priorities for agriculture purposes. REFERENCES Burrough, P.A. 1986. Principles of Geographical Information Systems for Land Resources Assessment. Oxford University Press. New York. De la Rosa, D., and D, Magaldi. 1992. Rasgos metedologicos de un sistema de evaluacion de tierras para regions mediterraneas. Soc. Esp. Cien. Suelo. Madrid. De la Rosa, D., J.A. Moreno, L.V. Garcia, and J. Almorza. 2004. MicroLEIS: A micro computer-based Mediterranean land evaluation information system. Soil Use and Management. 8: 89-96. Dent, D., and A. Young. 1981. Soil survey and land evaluation. Allen and Unwin, London. FAO. 1976. A framework for land evaluation. Soils Bulletin 32. Rome. FAO. 1990. Guidelines for soil profile description. 3rd edition. FAO, Rome. Ranchin T., and L. Wald. 2000. Comparison of different algorithms for the improvement of the spatial resolution of images. Proceeding of the third conference "Fusion of Earth data: merging point measurements, raster maps and remotely sensed images", Sophia Antipolis, France. Rowell, D.L. 1995. Soil Science Methods & Applications” Library of Congress Catalogingin- Publication data. New York, NY 10158. U.S.A. Soil Survey Staff. 1999. Soil Taxonomy, A Basic System of Soil Classification for Making and Interpreting Surveys, second edition. Agriculture Handbook No. 436.U.S.D.A., Nat. Res. Cons. Service. U. S. Geological Survey. 1998. USGS Fact Sheet. No. 102-96. Verheye, W. 1986. Land evaluation and land use planning in EEC. CEC-DG. VI. Draft. Rep., Brussels.
4
