Automatic Mapping of Lineaments Using Shaded Relief Images Derived from Digital Elevation Model (DEMs) in the Maran Sungi Lembing Area, Malaysia

Transcription

1 Automatic Mapping of Lineaments Using Shaded Relief Images Derived from Digital Elevation Model (DEMs) in the Maran Sungi Lembing Area, Malaysia Anwar Abdullah School of Environmental and Natural Resource Sciences, Faculty of Science and Technology, Universiti Kebangssan Malaysia, UKM Bangi, Selangor, Malaysia; Juhari Mat Akhir School of Environmental and Natural Resource Sciences, Faculty of Science and Technology, Universiti Kebangssan Malaysia, UKM Bangi, Selangor, Malaysia; Ibrahim Abdullah School of Environmental and Natural Resource Sciences, Faculty of Science and Technology, Universiti Kebangssan Malaysia, UKM Bangi, Selangor, Malaysia ABSTRACT The area of this study is one of the important places in Malaysia, which contains many ore deposits, plantation fields, landslides, and villages. Because of this, study of geological structures (lineaments) is one of the most important factors, which gives indications about the locations of ore deposits, landslides, and groundwater movements. Shaded relief images created from digital elevation models (DEMs) are helpful in identifying lineaments in different distinct relief and topography. The method can enhance lineaments at different orientations by simulating topographic illumination under varied light directions. Combining four shaded relief images into a final image leads to enhancing lineaments within the four sun azimuth directions. Two shaded relief images with multi-directional light were created, the first image with the four azimuth angles of the light sources that are 0, 45, 90 and 135, and the second image with the four azimuth angles of the light sources that are 180, 225, 270,and 315. Both images have been used for automatic lineament extraction using PCI Geomatica software. The extracted lineaments of the first image show that the most automatic lineaments represent positive lineaments; and the automatic lineaments of second image represent negative lineaments of the study area. The directional analysis of the automatically extracted lineament maps have been done with the reference of the fault map of the area. The results verify the similarity in the directional behaviour of the lineaments, the fault lines and the structures measured in the field. The main trends of the extracted lineaments of both images are NW-SE, N-S, E-W and NE-SW. KEYWORDS: DEM; Shaded relief; PCI Geomatica software; Automatic lineaments

2 Vol. 15 [2010], Bund. J 950 INTRODUCTION DEM data are digital elevation data set recording the topographic surface expression of any area. Topographic attributes can be extracted from digital elevation data by applying special computer algorithms. These include slope, aspect and shaded relief algorithms. The most widely used is the shaded relief algorithm, because perception of elevation can be attained and when combined with other type of images, new products can be produced. DEM, unlike Landsat images, represent true reap projections with no distortion; tonal variations are entirely identified with relief, the sun position can be placed anywhere above the terrain surface, and in most cases, DEM has better resolution than satellite images (Batson et al. 1975; Schowengerdt and Glass, 1983).Techniques of shaded relief have been used in cartography to produce the illusion of a 3- dimensional relief map (Yoeli, 1965). During the last decades, digital image processing techniques have been developed to display DEMs as shaded relief images (Yoeli, 1967 ; Batson et al. 1975).In the case of DEM's, shaded relief images and terrain derivative maps (slope, aspect and curvatures) have largely demonstrated their usefulness for lineaments and fault extraction (Ganas et al. 2005). Linear features that can be extracted from DEMs often provide useful information for scientists. The term lineament is one of the most commonly used terms in geology. A lineament is any extensive linear surface on a planet, as a fault line or fracture line. Hobbs (1904) first used the term lineament to define a significant line of landscape of Basement rocks. O Leary et al. (1976) described the term lineament as a mappable, simple or composite linear feature of a surface whose parts are aligned in a rectilinear or slightly curvilinear relationship and which differs from the pattern of adjacent features and presumably reflects some sub-surface phenomenon. Lineaments can be defined as linear topographical or tonal features on the terrain representing zones of structural weakness (Williams, 1983). Gupta (1991) summarized the definition of lineament in different geological features, such as (1) shear zones/faults; (2) rift valleys; (3) truncation of outcrops; (4) fold axial traces; (5) joint and fracture traces; (6) topographic, vegetation, soil tonal changes alignment etc. Lineaments are natural crustal structures that may represent a zone of structural weakness (Masoud and Koike, 2006). Positive straight lineaments, interpreted as linear ridges, scarps, ridges, troughs and crater (light toned lineaments) and negative straight lineaments, representing joints, faults, and shear zones(dark toned lineaments). The purpose of this paper is to clarify the ability of shaded relief images with multidirectional of the light sources (different sun azimuth angles) derived from digital elevation module for lineament extraction (positive and negative lineament extraction). MATERIALS AND METHODS Study Area The study area is situated in the eastern part of Pahang, Malaysia, about 2340 sq. km, between 3 39'6" and 4 04' 1" N and longitudes '5"and '5" E (Fig.1). According to Lee s (1990) classification, the rocks in the study area divided into three groups as follows: 1- Upper Paleozoic sediments ( Kambing beds and Seri Jaya beds)

3 2- Mesozoic sediments( Buluh sandstones and Hulu Lepar beds) 3- Granitoid intrusives and volcanic rocks. The geological map of the study area (Fig. 2) has been digitized from six geological sheet maps at 1:63,360 scales prepared by Minerals and Geoscience Department Malaysia (JMG). Figure 1: Location of the study area Figure 2: Geological map of the study area Data Processing The nature of elevation data is continuous, not discrete. However, due to the limitation of recent technology, DEM data can be displayed in forms i.e. grid, contour, profile, and TIN (Triangulated Irregular Network) which are not completely continuous. In this study, these data are displayed in the form of grid. The DEM was acquired from contour map and was originally generated by digitize the contour lines and give the contour lines the height value they represent, then converted it into grid form. The DEM was geo-rectified using topographic maps with points collected at road intersections and other landscape features throughout the area. Shaded relief images were derived from a digital elevation model (DEM) with a resolution of 30 meters are used for linear extractions. In order to identify linear topographic features from the DEM, eight shaded relief images were generated. The first step is the production of eight separate shaded relief images with light sources coming from eight different directions. The first shaded relief image created had a solar azimuth (sun angle) of 0, and a solar elevation of 30. An ambient light setting of 0.20 was used, which produces a good contrast. The ambient light setting is simply a scaling factor in the imagine topographic program (Erdas, 1998). The other seven shaded relief images were created with seven contrasting illumination directions 45, 90,135, 180, 225, 270,and 315 (Fig. 3). The second step is to combine four shaded relief image to produce one shaded relief image, for this purpose, the combinations of the four shaded relief maps are computed by using GIS overlay technique, where the first four shaded relief images are overlaid to produce one image with multi - illumination directions (0,45,90,and135 ) and

4 Vol. 15 [2010], Bund. J 952 the second overlay is to produce one image with multi - illumination directions (180, 225, 270, and 315 ) ( Fig. 4-A and 4-B). Finally, those two images have been used for automatic lineaments extraction over the study area. Sun angle 0 Sun angle 45 Sun angle 90 Sun angle 135 Sun angle 180 Sun angle 225 Sun angle 270 Sun angle 315 Figure 3: Eight shaded relief images derived from DEM with illumination directions( sun azimuth),0,45,90,135, 180,225, 270,and 315 and a solar elevation of 30. An ambient light setting of 0.20

5 4-A 4-B Figure 4: Two shaded relief images created by combining different shaded relief images. A: Combining four shaded relief images with sun angle of 0,45,90 and 135.And B: Combining four shaded relief images with sun angle of 180,225,270 and 315 Edge enhancement leads to image sharpening, in which the geometric details of an image may be modified and enhanced (Richards, 1986). Features such as lineaments, drainages and certain landforms are often characterized by abrupt changes in radiometric responses. Edge enhancement is a useful and efficient technique to enhance these features. Edge enhancement has been applied to the two shaded relief images to increase frequency and contrast in the images. Lineament Extraction The lineament extraction algorithm of PCI Geomatica software consists of edge detection, thresholding and curve extraction steps (PCI Geomatica, 2001). These steps were carried out over two shaded relief images under the default parameters of the software as follows: Name Description Values RADI Radius of filter in pixels 12 GTHR Threshold for edge gradient 90 LTHR Threshold for curve length 30 FTHR Threshold for line fitting error 10 ATHR Threshold for angular difference 30 DTHR Threshold for linking distance 20 RESULTS AND DISCUSSION The results of automatic lineament extractions are shown in Fig. 5-A and 5-B. A comparison between fault map of the study area and the lineaments extracted from enhanced shaded relief images in terms of number of lineaments and the total length of lineaments is shown in Table 1. This technique identified the highest number of lineaments and the highest

6 Vol. 15 [2010], Bund. J 954 total length of lineaments compared with the number and the total length of faults in the study area. 5-a 5-b Figure 5: Automatic lineament maps of shaded relief images. A: Automatic lineament map of combining four shaded relief images with sun angle of 0,45,90 and 135. And B: Automatic lineament map of combining four shaded relief images with sun angle of 180, 225, 270 and 315 Table 1: Basic statistics of the automatic lineaments maps and fault map Fault map Automatic lineaments map of Multi-directional light (0, 45, 90, and135 ) Automatic lineaments map of Multi-directional light (180, 225, 270, and 315 ) Count Min. length (km) Max. length (km) Mean (km) Total length (km) The directional analysis of the automatically extracted lineament maps have been done with the reference of fault map of the study area. The results verify the similarity in the directional behaviour of the lineaments. The major trends of lineaments in the study area are NW-SE, N-S, E-W and NE-SW (Fig. 6). The dominant lineaments tend to run in the NW-SE direction. Lineament extracted from both images reveal the significance of NW-SE, N-S, E- W and NE-SW oriented lineament sets which are interpreted as same orientations of fault lines and field data orientations. Based on the pattern of the lineament including the length, linking characteristics spatial relationships, and visual interpretation of automatic lineament maps with topographic map, drainage pattern map, three dimensions of DEM and aerial photos, the extracted lineaments of shaded relief image with sun angle of 0, 45, 90, and 135 are mostly resemble a positive system lines in the area. The most important feature in the area is the presence of topographic relief pattern. It is clearly to issue that, there is good relationship between the lineament and topographic relief pattern distribution. On the other hand, the extracted lineaments of shaded

7 relief image with sun angle of 180, 225, 270, and 315 are mostly resemble a negative system lines in the area. The most important feature in the area is the presence of drainage lines patterns and fault zones. It is clearly to issue that, there is good relationship between these lineaments and drainage pattern system distribution especially with third and fourth river orders. Moreover, Lineaments in other parts especially in the central part of the area display a typical pattern of the faults as already reported in the literature. a b c d e Figure 6: (a) Rose diagram of automatic lineament map of combining four shaded relief images with sun angle of 0, 45, 90 and 135. (b) Rose diagram of automatic lineament map of combining four shaded relief images with sun angle of 180, 225, 270 and 315. (c) Rose diagram of fault map. d: Rose diagram of positive field data. And (e) Rose diagram of negative field data Pattern of the lineament map (Fig. 5a) suggest that some faults that belong to the Lepar Faults Zone (within Lepar granite body) are not properly identified particularly around Hulu Lepar area. On the contrary, Pattern of the lineament map (Fig. 5b) suggest that some faults that belong to the Lepar Faults Zone are slightly identified particularly around Hulu Lepar area. Frequency of the lineaments is higher within granite bodies and Jurassic sandstones rocks. South East, North Northeast, and North Northwest of area are the areas with the least lineaments. Generally the frequency of automatically extracted lineaments is greater more than number of lineaments in the previous fault map of the study area. The most important factor for this is that the lineaments in the automated one are shorter in length compared with the length of fault lines. An automatic lineament extraction method is one of the methods to extract negative and positive lineaments. The automatic lineament extraction method approach does not discriminate man made features and non geological features during the analysis. This is could lead to an increase the total number and total length of the lineaments

8 Vol. 15 [2010], Bund. J 956 CONCLUSION Variations of sun angle, azimuth and height exaggeration in DEM data and shaded relief image assist in geological feature recognition. DEM data can be used as source data to enhance some features for increasing interpretation accuracy. DEM-based hill shading has led to the discovery of positive and negative lineaments in the study area, which had not been found by other satellite data interpretation. Creation and interpretation of shaded-relief images can contribute significantly to solving various problems in geomorphology and geology researches. ACKNOWLEDGEMENT The first author would like to give a special thanks to supervisors and staff of Geology Department in UKM for their support in this work. REFERENCES 1. Batson, R.M., Edwards. K. and Eliason, E.M. (1975) Computer generated shadedrelief Images, Journal Research U.S. Geological Survey 3 (4): ERDAS (1998) Using ArcView Image Analysis, Atlanta: GA:ERDAS Inc. 3. Ganas, A., Pavlides, S. and Karastathis, V. (2005) DEM-based morphometry of range-front escarpments in Attica, central Greece, and its relation to fault slip rates, Geomorphology 65(3-4): Gupta, R.P. (1991) Remote Sensing Geology, Berlin, Heidelberg: Springer-Verlag. 5. Hobbs, W. H. (1904) Lineaments of the Atlantic Border Region, Geological Society.American Bulletin 15: Lee. A. K. (1990) The geology and mineral resources of the Hulu Lepar area, Pahang, Geological Survey of Malaysia. District Memoir Masoud, A. and Koike, K. (2006) Tectonic architecture through Landsat-7 ETM+/ SRTM DEM- derived lineaments and relationship to the hydrogeologic setting in Siwa region, NW Egypt, Journal of African Earth Sciences 45: PCI Geomatica (2001) PCI Geomatica user s guide version 9.1, Ontario. Canada: Richmond Hill. 9. O Leary, D. W., Friedman, J. D., and Pohn, H. A. (1976) Lineament, linear, lineation: Some proposed new standards for old terms, Geological Society America Bulletin 87: Richards, J.A (1986) Remote Sensing Digital Image Analysis, New York: Springer-Verlag.

9 11. Schowengerdt, R.A., and Glass, C.E. (1983) Digitally processed topographic data for regional tectonic evaluation, Geological Society of America Bulletin 94: Williams, R. S. (1983) Geological applications, In. Colwell, R. N. (eds). Manual of Remote Sensing, Falls Church, VA: American Society of Photogrammetry. 13. Yoeli, P. (1965) Analytical Hill Shading, Surveying and Mapping 25(4): Yoeli, P. (1967) The Mechanisation of Analytical Hill Shading, Cartographic Journal 4: ejge