1. INTRODUCTION The objective of this research was to explore the potential of hand held Space Shuttle photographs for remote sensing analysis of meteorite impact craters. Functionality of the existing on line databases was also studied and results obtained from other sensors (SIR- C/X-SAR) are discussed and compared. 1.1. METEORITIC IMPACT CRATERS Meteoritic craters are circular depressions formed by the impact of a meteorite on a planetary body. Explosive impact craters are formed by large meteorites (HODGE, 1994). Smaller meteorites (up to few metres in diameter) do not have a sufficient energy to cause an explosion. They produce dug craters with diameters of less than 10 metres (HODGE, 1994). Morphologically impact craters can be divided into two main groups: simple craters (Fig. 1) and complex structures (craters) (Fig. 2) (GRIEVE, 1991). Simple craters consist of a bowl shaped depression with an uplifted rim. Complex structures (Table 1) form at crater diameters above 2 km in sedimentary rocks and greater than 4 km in magmatic rocks. They consist Remote Sensing Analysis of Selected Terrestrial Impact Craters and a Suspected Impact Structure in South Korea using Space Shuttle Photographs Neven TRENC 1, Ken M. MORGAN 2, R. Nowell DONOVAN 2 and Arthur B. BUSBEY2 of a central peak, an annular trough and outer uplifted rim (GRIEVE, 1991). As a result of his study of lunar craters MELOSH (1989), defined multiring basins as a separate class of features characterized by at least two asymmetrical scarped rings (one of which may be the original rim of the crater). In 1996, 153 terrestrial impact craters were known (GRIEVE et al., 1995). Their diameters vary from 10 m to 300 km (since pits smaller than 10 m are not included in the record) (HAMILTON, 1996). The earth’s mete- oritic record, when compared to other terrestrial planets shows a deficit of smaller and older features. This phe- nomenon reflects the processes of erosion and sedimen- tation that are active on the earth surface and tend to “heal” impact effects (GRIEVE, 1990). Shock metamorphic effects are exclusively associat- ed with meteoritic impacts. No other natural process generates such extreme pressures and temperatures in such short time (GRIEVE, 1990). They include shatter cones, formation of planar deformation features in quartz and feldspar grains, formation of diaplectic glass and high-pressure polymorphs of some minerals. 1.2. USE OF REMOTE SENSING IN THE ANALY- SIS OF METEORITIC IMPACT CRATERS Remote sensing has been successfully used for the analysis of meteorite craters on the Earth as well as on the other planets and the Moon. G EOL. CROAT. 52/2 203 - 215 10 Figs. 4 Tabs. ZAGREB 1999 Key words: Shuttle photographs, meteoritic craters, Manicouagan, Aorounga, Roter Kamm, Meteor Cra- ter, South Korea. KljuËne rijeËi: snimci sa Space Shuttle-a, meteoritski krateri, Manicouagan, Aorounga, Roter Kamm, Me- teor Crater, Juæna Koreja. 1 Idrijska 35, HR-10000 Zagreb, Croatia. 2 Texas Christian University, Dep. of Geology, Box 298830, Fort Worth, Texas 76129, USA. Abstract Typical morphological elements for impact craters were detected on all analyzed Shuttle photographs. Features with diameters smaller than 2 km could not be successfully analyzed. The search for suitable Shuttle scenes was successfully performed using on-line Internet databases. Image enhancement and lineament analysis were also performed on the acquired Shuttle, Landsat and SPOT images of a suspected impact feature in South Korea. A fault bounded polygonal rim and a central uplift were identified on all images. A distinct system of annu- lar and radial faults was not detected. The results of the study do not exclude the possibility of a meteoritic origin of this structure. Saæetak Morfoloπki elementi terena tipiËni za udarne strukture identifici- rani su na svim prouËavanim fotografskim snimcima sa Space Shut- tle-a. Strukture manje od 2 km nije bilo moguÊe uspjeπno analizirati. Identifikacija odgovarajuÊih fotografija sa Shuttle-a ostvarena je pre- traæivanjem mreænih baza podataka s Interneta. Analizirani Shuttle, Landsat i SPOT snimci moguÊe meteoritske strukture u Juænoj Koreji pokazuju rasjedima okruæeni poligonalni rub te centralno uzdignuÊe, no jasan sustav anularnih i radijalnih rasjeda nije bio registriran. Rezultati analize nisu osporili moguÊnost da je meteoritski udar uzrokovao postanak ove strukture. 204 Geologia Croatica 52/2 Typical morphological indicators of impact craters on satellite images include: circular raised rims, polyg- onal shape (in complex craters), central uplifts (within the craters), radial drainage patterns (GARVIN et al., 1992), and dissected topography (BUTLER, 1994). Lineament analysis of impact features often reveals radial or annular fault and fracture patterns (Figs. 3 and 4) (INNES, 1964; BUTLER, 1994) in the crater vicini- ty. SABINS (1996) summarized some remote sensing characteristics of impact craters on Venus and their spe- cific signatures on radar imagery (Table 2). GARVIN et al. (1992) suggested that the impact induced change of the spectral signatures, characteristic for the mineral quartz in the thermal infrared region, could be used for the recognition of impact features on the multispectral imagery. Impact breccia deposits (GARVIN et al., 1992) showed characteristic signature on the third, fourth and fifth principal component colour imagery of the Zhamanshin crater (Russia). The proceses of erosion and sedimentation impede recognition of terrestrial crater and they can distort and even totally remove the morphological imprint of a crater. A number of other geological processes (volcan- ism, doming, folding etc.; ROLAND, 1976) can create ring structures (OLUI∆, 1983) that may mimic mete- oritic craters. 2. METHODS 2.1. SPACE SHUTTLE PHOTOGRAPHS Since 1982 Space Shuttle astronauts have used hand held and mounted cameras to take photographs during their stay in space (EROS, 1996). Each Shuttle mission produces from 4,000 to 5,000 photographs and approxi- mately 125,000 photographs have been taken since the beginning of the project (EROS, 1996). Over eighty percent of the photographs are views of the Earth (EROS, 1996) and display large variations of angle, lighting and season (MUEHLBERGER, 1996). Three main types of cameras used in the project are: Hassel- blad 500/M 70 mm, Linhof Aero Technika 45 127 mm and Large Format Camera (LFC) (EROS, 1996) (Table 3). An LFC camera was used only on the mission 41-G and these photographs provide stereo coverage over certain areas (EROS, 1996). A majority of these images are regular RGB photographs, though some black and white and colour-infrared scenes have been taken (EROS, 1996). Spatial resolution of the photographs is dependent on the camera and the lens used and ranges between 150 and 20 m. 2.2. SPACE SHUTTLE ARCHIVES AND DATABASES After each Shuttle mission, rolls of film are sent to the Johnson Space Center’s Image science division in Houston. Photographs are processed and classified in three groups: earth observation images, experiment images and public release images (JSC, 1994). Datasets are digitized at the capture station and TARGA files of approximately 1.1 M (756 by 486 pixels with 8 bits per colour) are produced (JSC, 1994). A portion of files is displayed to the public over the Internet or in the form of videodisc (JSC, 1994). However, because of relative- ly poor resolution, scenes from a disc and the Internet are not suitable for the remote sensing analysis. Fig. 1 Schematic presentation of a simple crater (modified from GRIEVE, 1991). Sl. 1 Shematski prikaz jednostavnog kratera (prema GRIEVE, 1991). Fig. 2 Schematic presentation of a complex strucuture (modified from GRIEVE, 1991). Sl. 2 Shematski prikaz kompleksne strukture (prema GRIEVE, 1991). Group Central One Few peak Ring Rings Central peak craters yes no no Central peak basins yes yes no Peak ring basins yes yes no Multi ring basins yes no yes Table 1 Groups of complex craters (PIKE, 1985). Datasets with a higher resolution must be obtained directly from JSC. In this study, the initial search for suitable scenes of impact craters was performed through the online data- bases at the Earth Science Branch of the Lyndon B. Johnson Space Center web site (http://www.jsc.nasa. gov). The database, Space Shuttle Earth Observations Project Database of Photographic Information and Images (JSC, 1996) http://eol.jsc.nasa.gov/sseop) was accessed through the Internet (WWW) and a text search was performed. Space Shuttle Earth Observations VideoDisc (1981- 1991), released to the public by Johnson Space Center, with approximately 91,500 still images of the Earth tak- en during the Space Shuttle missions STS-1 through STS-44 was used to identify optimal scenes. The film rolls with selected scenes were later locat- ed in the Shuttle Archive at the Lockheed Martin Build- ing in Houston and transported to the Lyndon B. John- son Space Center. A Limhof digital camera, Adobe PhotoShop software and Macintosh hardware were used in the center to digitize the photographs. Nominal resolution was around 500 DPI and the size of created files varied between 5 and 10 M. The original resolution of Shuttle photographs was inevitably reduced during the process. The main limit- ing factors were mounting and the optical characteris- tics of the Limhof camera. 205Trenc, Morgan, Donovan & Busbey: Remote Sensing Analysis of Selected Terrestrial Impact Craters... Fig. 3 Rings and faults associated with Haughton impact structure (Canada) (modified from ROBERTSON, 1980). Sl. 3 Prstenovi i rasjedi vezani za Haughton strukturu (prema ROBERTSON, 1980). Fig. 4 Schematic map of prominent lineaments around Deep Bay impact structure (Canada) (modified from INNES, 1964). Sl. 4 Shematski prikaz znaËajnijih lineamenata u okolici Deep Bay strukture (prema INNES, 1964). Material Description Radar Signature Lava plain Lava flows impacted by the bolids. Many plains are Dark to intermediate depending on flat and featureless. Others are faulted and folded to roughness. Bright linear features are various degrees. fault scarps. Rim Ejecta deposited up to three crater radii from the Bright near rim, becoming intermediate centre. Coarse hummocky material near the centre outward with bright patches. becoming somewhat finer outward. Wall Rough terraced deposits on steep inner slopes. Bright to intermediate. Annular zone Slump blocks and talus are common. surrounding the crater floor. Floor Level to smooth interior plains ranging from smooth Bright to dark, circular to irregular to hummocky surfaces. Materials include impact melt, outline. shocked rock, fallback or volcanic fill. Very bright with characteristic Peaks and rings Peaks and ridges within the craters that rise above morphology. the floor materials. Uplifted, shocked, crushed and sheared rock. Outflow Fluidized material formed by impact that resembles Bright to intermediate with a lava flow. At some larger craters outflows extend flowerlike outline. for hundreds of kilometres. Table 2 Impact affected materials on radar imagery (SABINS, 1996). 206 Geologia Croatica 52/2 Data were stored on an external hard drive and transported to the Texas Christian University Center for Remote Sensing and Energy Research for further pro- cessing. 2.3. SIR-C/X-SAR IMAGES AND DATABASES SIR-C/X-SAR (Spaceborne Imaging Radar-C/X- Band Synthetic Aperture Radar) was flown on two Shuttle missions in 1994 (EVANS & PLAUT, 1996). It combines two radar devices: SIR-C and X-SAR. The instrument acquires quad polarized (L, C band) as well as single polarized (X-band) impulses (FREEMAN, 1996). The resolution varies between 10 and 50 m (EVANS & PLAUT, 1996). To create SIR-C/X-SAR false colour images red, green and blue colours are assigned to different band/polarization combinations (JPL, 1996). Radar scenes used in this study were downloaded from JPL’s Space Radar Images of the Earth web site (http://www.jpl.nasa.gov/radar/sircxsar) This web page displays over 150 SIR-C/X-SAR false colour images classified in eight thematic categories and listed by geo- graphic locality. Each image can be downloaded over the World Wide Web as a JPEG file with a low (size around 140 K) and high (file size around 10 M) resolu- tion. 2.4. IMAGE PROCESSING AND LINEAMENT DETECTION Image processing was performed at the Texas Christian University Center for Remote Sensing and Energy Research on a Power Macintosh 9500/132 with 50 M RAM. The Dimple image-processing program was utilized for relative rectification of the Shuttle Pho- tographs to SIR-C/X-SAR image. A linear GCP model and nearest neighbour resampling method were used. True north was determined on all the images. Image datasets were either geographically corrected (radar) or fairly undistorted (Shuttle). Since only the relative posi - tion of the detected features in respect to the impact structures was significant in this study, strict geographi - cal rectification was not performed. Adobe PhotoShop was used for image enhance- ment. Contrast stretching was applied on all Shuttle photographs and four custom directional convolution kernels were used for edge enhancement: north, east, northwest and southwest illumination. The original image and four derived filtered images were placed in separate layers of a single Adobe Illus- trator document. A lineament is defined as a “mappable, simple or composite linear feature of a surface whose parts are aligned in a rectilinear or slightly curvilinear relation- ship and which differ distinctly from the patterns of adjacent features” (O’LEARY et al., 1976). An addi- tional layer in the Adobe Illustrator was used for tracing of the lineaments. The results obtained from various image layers were compared and an attempt was made to eliminate all lineaments that were not structurally related (e.g., sand features caused by prevailing winds). Mapict software (developed by Dr. Arthur BUS- BEY at TCU) was used to digitize traced lineaments and create text files, which were imported into Rosy (Rockware) software for production of rose diagrams. 3. RESULTS AND DISCUSSION Pictures of several impact structures were retrieved from the Shuttle database. Those with the best charac- teristics for remote sensing were selected (Table 4, Fig. 5). They represent various size categories of mappable impact craters on Earth. Diameters of the studied fea- tures ranged between 1.2 and 100 km. 3.1. THE MANICOUAGAN The Manicouagan (Canada) impact structure (Figs. 5 and 6) is located at 51°23’N 68°42’W (FLORAN & DENCE, 1976). The age is estimated to be around 214 Ma (FLORAN & DENCE, 1976). The impact area is in the Grenville province of the Canadian Shield (OR- Camera Lens Spatial Sensor With a Resolution Comparable Resolution Hasselblad 500 EL/M 100 mm 150 m MSS (80 m) Hasselblad 500 EL/M 250 mm 30 m TM (30 m) Linhof Aero Technika 250 mm 30 m TM (30 m) Large Format Camera N/A 20-30 m Spot HRV* (20 m) - TM (30 m) Table 3 Spatial resolution of Space Shut- tle photographs (EROS, 1996). *) Multispectral mode. Feature Size Size Rank Manicouagan (Canada) 100 km 5 Aorounga (Chad) 17 km 40 Roter Kamm (Namibia) 2.5 km 119 Meteor Crater (Arizona) 1.2 km 130 Suspected structure 13 km - in South Korea Table 4 Impact structures with the best characteristics for remote sensing. 207Trenc, Morgan, Donovan & Busbey: Remote Sensing Analysis of Selected Terrestrial Impact Craters... PHAL, 1978). Petrological Grenville units recognized in the Manicouagan area (Fig. 7) include meta-gabbros, meta-anorthosites and granitic gneisses (ORPHAL, 1978). Long periods of erosion and glaciation have sub- stantially affected the appearance of this structure (FLORAN & DENCE, 1976). Remote sensing analysis of the Manicouagan crater shows the typical pattern of annular lineaments that define the polygonal rim of other complex impact struc- tures. Orientation of these faults coincides with the typ- ical fault and fracture patterns in the area (NE, N) and supports the assumption (FLORAN & DENCE, 1976) that these faults follow preimpact structures and zones of weakness. Radial drainage and other characteristics typical for such features (central uplift annular trough) were also identified on the photograph. 3.2. AOROUNGA The Aorunga impact structure (Figs. 8 and 9) is located at 19°06’N and 19°15’E near the Tibesti massif in the Borkou region of eastern Chad (BECK-GIRA- DUON et al., 1992). The diameter of the crater is around 17 km (JPL, 1996) The age of the structure was estimated as late Devonian on the basis of the overlying sediments (OCAMPO, 1996). A Shuttle photograph of the feature was analyzed and compared with the SIR-C/X-SAR (Fig. 9) dataset that was downloaded from the JPL Internet site. All the elements of the structure are well developed and easily recognized. Crater rim, peak ring and central uplift were identified on both images. On the Shuttle image, sand covered areas are lighter coloured while exposed bedrock is displayed as a dark gray. Central uplift can be identified as a dark patch of the exposed Fig. 5 Locations of selected craters. Sl. 5 Poloæaj odabranih kra- tera. Fig. 6 Schematic geologic map of the Manicouagan structure (modi- fied from ORPHAL, 1978). Sl. 6 Shematska geoloπka mapa strukture Manicouagan (prema ORPHAL, 1978). Fig. 7 Lineament map of the Manicouagan impact crater based on Shuttle photograph (photograph credit: JSC, 1996). Sl. 7 Karta lineamenata oko kratera Manicouagan temeljena na Shut- tle fotografiji (izvor snimka JSC, 1996). 208 Geologia Croatica 52/2 bedrock in the centre of the structure. Radar penetrates up to several feet of dry sand (McHONE et al., 1996) showing more accurate morphology of the structure than the Shuttle image. The polygonal shape of the fault-bounded crater and annular system of lineaments, typical for an impact structure, were observed on both images. A comparison of the obtained results has shown that a larger number of lineaments was detected on the radar image. The fre- quency of N-S lineaments was also higher on this image. One of the main problems in the analysis was distinction between the structurally related lineaments and typical eolian features (orientation NE-SW) These masking effects, that were more pronounced on the Shuttle photograph, are probably the main cause for the mentioned differences of the results obtained from the analysis of two images. 3.3. SMALLER FEATURES: ROTER KAMM AND METEOR CRATER The Roter Kamm impact crater is located at 27°46’S and 16°18’E in the southern part of the Namib desert (Namibia) (FUDALI, 1973). The crater has a diameter of 2.5 km and its age has been estimated around 3.7 Ma (GRIEVE et al., 1995). Meteor Crater is located at the 35°02’N and 111°01’W in the Canyon Diablo region of the north central Arizona (USA) (SHOEMAKER, 1977). The age of the impact is around 49 Ka (GRIEVE et al., 1995). The structure has a diameter of 1,220 m (GRIEVE et al., 1995). On the Shuttle photographs of these two smaller features, both located in the arid region, only the polyg- onal fault bounded rim and very few lineaments could be identified. The main problems were the small size of the features, low resolution of the photographs and sand cover. 3.4. SUSPECTED IMPACT STRUCTURE IN SOUTH KOREA A suspected impact structure in south Korea is cen- tred at 37°49’N and 127°03’E (JONES et al., 1995). It is located 20 km north of Seoul near the town of Uijongbu in the vicinity of the demilitarized zone (JONES et al., 1995). In 1994, the crew of the Space Shuttle Endeavor (STS-59) took the first photograph of the structure. JONES et al. (1995) have described the structure and discussed its formation. A circular rim with a 13 km diameter and a central uplifted region characterize the crater like feature (JONES et al., 1995). Relief from the bottom to the crest is greater than 400 m (JONES et al., 1995). Hilly uplifted terrain encircles the structure, which is situated in the Jurassic granites and surrounded by the 2 GA old metamorphic complex of gneiss and schist. The north- south oriented Tongducheon fault cuts the structure in half (JONES et al., 1995). KANG et al. (1985) suggested that fracturing along the circular joint during cooling of the granite pluton induced formation of the structure. JONES et al. (1995) questioned the possibility of a magmatic origin for the structure and suggested that it was caused by a mete- oritic impact. Future study of rock samples from the site is expected to give definite information about its formation. The analysis has shown that the identification of lin- eaments was equally successful on Shuttle and Landsat TM images. A somewhat bigger number of lineaments were detected on the SPOT image due to its better spa- tial resolution Fig. 8 Lineament map of the Aorounga impact structure based on Shuttle photograph (image credit: JPL, 1996). Sl. 8 Karta lineamenata oko kratera Aorounga temeljena na Shuttle fotografiji (izvor snimka JPL, 1996). Fig. 9 Lineament map of the Aorounga structure based on SIR-C/X- SAR image. Sl. 9 Karta lineamenata oko kratera Aorounga temeljena na SIR- C/X-SAR snimci. 209Trenc, Morgan, Donovan & Busbey: Remote Sensing Analysis of Selected Terrestrial Impact Craters... The possible crater rim and uplifted peak were rec- ognized on the Shuttle photograph (Fig. 10) as well as on studied Spot and Landsat TM scenes. However, the eastern part of the rim is much better defined on the images than the western half. Two large lineaments that cut the structure (one of them corresponding to the Tongduchen fault) are clearly visible. Lineaments identified on the images mark a charac- teristic fault-bounded polygonal rim. Besides these well-pronounced lineaments of the “crater rim” numer- ous shorter lineaments with various orientations were detected on the scenes. They represent linear ridges and valleys. Though some of them have annular and radial orientations with respect to the structure, this informa- tion is not significant in the view of the large overall dispersion of lineament azimuths. In lower elevation areas covered by agricultural fields detection of linea- ments was not possible. The structure does show some typical elements that were identified on the Shuttle photographs of other proven impact features but they are not completely developed and the results of the analysis do not provide definite arguments for either hypothesis about the for- mation of the structure. 4. CONCLUSION Space Shuttle photographs have a good remote sensing potential but their use for the remote sensing analysis of impact craters was limited. Limiting factors include relatively low photographic resolution, nondigi - tal format and restricted areal coverage. Identification of the scenes with craters can be successfully per- formed from online and videodisc databases and a text search by the key word or geographic location is the fastest way to obtain the required data. Image enhancement and edge enhancement tech- niques were successfully applied on the scanned Shuttle photographs. However, limited mapping could be per- formed. Some elements of characteristic fault and frac- ture patterns for the impact features were identified on all studied images but results obtained from the pho- tographs of features smaller than 2 km were significant- ly inferior. Improved resolution, digital format and a mission dedicated to the photography of impact craters are needed to improve the potential of Space Shuttle pho- tographs for analysis of impact craters. Acknowledgments This research would not be possible without the support of the following organizations: Texas Christian University Research Foundation, NASA, Adkins Res- earch Fund and Lyndon B. Johnson Space Center. Dr Arthur EHLMANN (Texas Christian University) pro- vided valuable advises regarding literature and Internet resources. The authors would especially like to thank Lt. Col. Ned FLEMING and Sue RUNCO from JSC. We would also like to thank Dr Ivo VELI∆ and Dr Marinko OLUI∆ for their extremely useful suggestions and editing of the paper. 5. REFERENCES BAJI∆, M. (1985): An approach to simulation of the imaging by synthetic aperture radar.- Bilt. daljin. istraæ. fotointerpret., 1/11, 1-13, Zagreb. BECK-GIRADUON, J.F., ROUZEAU, O., GOA- CHET, E. & SOLAGES, S. (1992): Impact hyper- veloce d’une meteorite geante a l’origine de la depression circulaire d’Aorounga au Tchad (Afri- que).- Comptes Rendus de l’Academie des Scien- ces, Serie 2, 315/1, 83-88. BUTLER, H.R. (1994): Lineament analysis of the Sud- bury multiring impact structure.- In: DRESSLER, B.O., GRIEVE, R.A.F. & SHARPTON, V.L. (eds.): Large Meteorite Impacts and Planetary Evolution. Geological Society of America Special Paper, 319- 329. EROS (1996): Space Shuttle Earth Observation Pro- gram, EROS Data Center Web Site, http:// edcwww. c r . u s g s . g o v / W e b g l i s / g l i s b i n / g u i d e . p l / g l i s / h y p e r / g u i d e/shuttle. EVANS, D.L. & PLAUT J.J. (1996): Science results Fig. 10 Lineament nap of the Korean structure based on Shuttle pho- tograph (photograph credit: JSC, 1996). Sl. 10 Karta lineamenata oko Korejske strukture temeljena na Shuttle fotografiji (izvor snimka JSC, 1996). 210 Geologia Croatica 52/2 from the Spaceborne Imaging Radar-C/X-Band Synthetic Aperture (SIR-C/X-SAR): Progress Re- port Radar.- http://southport.jpl.nasa.gov/Progress- Reports0496/ TableofContents.html. FLORAN, R.J. & DENCE, M.R.M. (1976): Morpholo- gy of the Manicouagan Impact structure, Quebec, and some comparison with Lunar basins and cra- ters.- 7th Lunar. Sci. Conf, Proceedings, 2845-2865, Houston. FREEMAN, T. (1996): What is Imaging Radar.- Jet Propulsion Laboratory Web site, http://southport.jpl. nasa.gov/index.html. FUDALI, R.F. (1973), Roter Kamm, Evidence for an impact origin.- Meteoritics, 3, 245-257. GARVIN, J.B., SCHNETZLER, C.C. & GRIEVE, R . A . F . (1992): Characteristics of large terrestrial impact structures as revealed by Remote sensing Studies.- Tectonophysics, 216, 45-62. GRIEVE, R.A.F. (1990): Impact cratering on Earth.- Scientific American, April, 66-73. GRIEVE, R.A.F. (1991): Terrestrial impact: the record in rocks.- Meteoritics, 26, 175-194. GRIEVE, R.A.F., RUPERT, J., SMITH, J. & THERRI- AULT, A. (1995): The record of terrestrial impact cratering.- GSA Today, 5/10, 191-196. HAMILTON, C.J. (1996): Terrestrial impact craters.- h t t p : / / w w w . h a w a s t s o c . o r g / s o l a r / e n g / t e r c r a t e . h t m #intro HODGE, P. (1994): Meteorite craters and impact struc- tures of the Earth.- Cambridge University Press, Cambridge, 124 p. INNES, M.J.S. (1964): Recent advances in meteorite crater research at the Dominion Observatory, Ottawa, Canada.- Meteoritics, 2/3, 219-241. JPL (1996): Jet Propulsion Laboratory Public Informa- tion Office, Imaging Radar Homepage http://www. jpl.nasa.gov/radar/sircxsar/. JONES, T.D., FLEMING, N., HATHEWAY, A.W., McINTOSH, W. & SHELLUM, C.J. (1995): The Seoul impact structure - discovery and assessment.- Unpubl. JSC (1994): Johnson Space Center, NASA JSC Digital Image Collection, http://images.jsc.nasa.gov/iams/ html/index.html. J S C (1996): Johnson Space Center, Space Shutle Earth Observation Project (SSEOP) http://eol.jsc.nasa. gov/sseop/. KANG, P.C., CHI, K.H., MIN, J.C. & YOUNG, S.C. (1985): A geologic study of the Seoul-Dongducheun lineament using digital image processing techniques of Landsat data.- Journal of Korean Society of Remote Sensing, 1/1, 39-51. McHONE, J.F., GREENLY, R. & BLUMBERG, D. ( 1 9 9 6 ) : SIR-C/X-SAR Radar studies: Impact and Eolian Features, Borkue Region, Northern Chad.- Lunar and Planetary Science Conference, Abstracts, 27/2, 849-850, Houston. MELOSH, H.J. (1989): Impact cratering.- Oxford Uni- versity Press, Oxford, 245 p. MUEHLBERGER, W.R. (1996): Environmental appli- cations of hand held photography taken from the Space Shuttle.- Proceedings of the 11th Thematic Conference on Geologic Remote Sensing, Environ- mental Research Institute of Michigan, 1, 227. OCAMPO, A.C. (1996): Shuttle Imaging Radar (SIR- C) images reveal multiple impact craters at Aoroun- ga, Northern Chad.- Lunar and Planetary Science Conference, Abstracts, 27, 2, 977-978. O’LEARY, D.W., FRIEDMAN, J.D. & POHN., H.A. (1976): Lineament, linear, lineation: some proposed new standards for old terms.- Geological Society of America Bulletin, 87, 1463-1469. OLUI∆, M. (1983) Daljinska istraæivanja u geologiji.- In: DONASSY, V, TOMASEGOVI∆, Z. & OLUI∆, M.: Daljinska istaraæivanja u geoznanostima. JAZU, 141-268, Zagreb. OLUI∆, M. (1985): Globalni program snimanja i istraæivanja zemlje iz svemira do 2000. Godine.- Bilt. daljin. istraæ. fotointerpret., 1/11, 49-57, Zag- reb. ORPHAL, D.N. (1978): An alternative model for the Manicouagan impact structure.- 9th Lunar. Planet. Sci. Conf., Proceedings, 2695-2712, Houston. PIKE, R.J. (1985): Some morphology systematics of complex impact structures.- Meteoritics, 20/1, 49- 68. ROBERTSON, P.B. (1980): Haughton impact struc- ture.- Lunar Planet. Sci. Conf., Proceedings, 895- 896. ROLAND N.W. (1976): Die Ringstruktur Aorounga (Borkou, Sud-Sahara).- Geologische Jahrbuch, A.33, 117-131. SABINS, F.F. (1996): Remote sensing principles and interpretation.- W.H. Freeman and Co., New York, 447 p. SHOEMAKER, E.M. (1977): Penetration mechanics of high velocity meteorites, illustrated by Meteor Crater Arizona.- In: McCALL, G.J.H. (ed.): Mete- oritic craters. Dowden, Hutchinson and Ross, 170- 186. 211Trenc, Morgan, Donovan & Busbey: Remote Sensing Analysis of Selected Terrestrial Impact Craters... Manuscript received July 26, 1999. Revised manuscript accepted November 12, 1999. 1. UVOD Cilj ovog rada bio je istraæiti potencijalnu vrijednost foto- grafija snimljenih iz letjelice Space Shuttle za analizu struktu- ra meteoritskog podrijetla primjenom metoda daljinskih istra- æivanja. Testirana je i funkcionalnost postojeÊih baza podata- ka na Internetu te moguÊnost brze identifikacije odgovaraju- Êih Shuttle-ovih fotografija ovim putem. Rezultati dobiveni na temelju fotografija sa Shuttle-a us- poreeni su s onima dobivenima pomoÊu drugih senzora pa su tako pri analizi strukture Aorounga koriπteni i radarski snimci sa letjelice Shuttle (radi se o ureaju SIR-C/X-SAR, Space- borne Imaging Radar-C/X-band Synthetic Aperture Radar (FREEMAN, 1996) - Slikovni radar za snimanje iz svemira- C/X-radar sa sintetiËkom antenom (BAJI∆, 1985; OLUI∆, 1985)). Pri obradi potencijalnog meteoritskog kratera u Juænoj Koreji, uporabljeni su i satelitski snimci dobiveni pomoÊu tematskog kartografa (TM-Thematic Mapper) sa satelita Landsat te senzora HRV (High Resolution Visible) postavlje- nog na satelitu SPOT. 1.1. METEORITSKI KRATERI Meteoritski krateri su kruæne depresije nastale djelova- njem meteorita na planetarno tijelo. Eksplozivni krateri nasta- ju djelovanjem veÊih meteorita, dok manji meteoriti (do neko- liko metara u promjeru) nemaju dovoljnu energiju da izazovu eksploziju. Oni stvaraju iskopne kratere s promjerima manjim od 10 m (HODGE, 1994). Morfoloπki, meteoritski krateri mogu se podijeliti u dvije glavne skupine: jednostavni krateri (sl. 1) i kompleksne struk- ture (krateri) (sl. 2) (GRIEVE, 1991). Jednostavni krateri su depresije u obliku zdjele na Zemlji- noj povrπini s uzdignutim rubom. Kompleksne strukture (tab- lica 1) nastaju kod radijusa veÊih od 2 km u sedimentnim sti- jenama i 4 km u magmatskim stijenama. One se sastoje od centralnog uzdignuÊa kruænog jarka i vanjskog uzdignutog ruba (GRIEVE, 1991). Na temelju studija mjeseËevih kratera MELOSH (1989) je definirao bazene s viπe prstenova kao posebnu skupinu struk- tura karakteriziranih sa barem dva asimetriËna prstena (od kojih jedan moæe biti i originalni rub kratera). Godine 1995. bilo je poznato 153 meteoritskih kratera (GRIEVE et al., 1995). Njihov promjer varira od 10 m do 300 km (udubine manje od 10 m nisu kategorizirane) (HAMIL- TON, 1996). Kad se broj i veliËina meteoritskih kratera na povrπini Zemlje usporedi s drugim planetima sliËnih karakte- ristika, uoËljivo je da se na Zemlji manje i starije strukture javljaju razmjerno rjee. Ovaj fenomen odraz je procesa ero- zije i sedimentacije koji djeluju na njezinoj povrπini i “briπu” tragove djelovanja meteorita (GRIEVE, 1990). Metoritski udari uzrokuju ekstremno visoke pritiske i temperature tijekom izuzetno kratkog vremenskog intervala te su oni jedini prirodni proces koji moæe izazvati pojavu speci- fiËnih “udarnih” metamorfnih efektata (shock metamorphic e f f e c t s - GRIEVE, 1990), koji se smatraju kljuËnim dokazom meteoritskog podrijetla neke strukture. Neki od tipiËnih efe- kata su nastanak konusnih pukotina (shatter cones ) u udarom zahvaÊenim stijenama, nastanak planarnih deformacijskih pojava u zrnima kvarca i feldspata te nastanak dijaplektiËkih stakala i visokotemperaturnih polimorfa nekih minerala (GRI- EVE, 1990). 1.2. PRIMJENA DALJINSKIH ISTRAÆIVANJA U ANALIZI METEORITSKIH KRATERA Daljinska se istraæivanja uspjeπno primjenjuju u analizi meteoritskih kratera na Zemlji, drugim planetima i Mjesecu. TipiËni morfoloπki indikator meteoritskih kratera na satelit- skim snimcima su kruæni uzdignuti prstenovi, poligonalni ob- lik (kod kompleksnih kratera) centralna uzdignuÊa (unutar kratera), radijalni povrπinski tokovi (GARVIN et al., 1992), te isprekidana topografija (BUTLER, 1994). Analize lineamenata uz meteoritske kratere obiËno poka- zuju anularne i radijalne sustave rasjeda (sl. 3 i 4) (INNES, 1964; BUTLER, 1994). SABINS (1996) je prikazao karakteristike kratera na Ve- neri i njihove tipiËne odraze na radarskim snimcima (tablica 2). GARVIN et al. (1992) navode da se udarom izazvane pro- mjene u spektralnim vrijednostima tipiËnim za mineral kvarc, u termalnom infracrvenom podruËju, mogu koristiti za pre- poznavanje meteoritskih struktura na multispektralnim snim- cima. Naslage udarnih breËa kod kratera Zhamanshin (Rusija) mogle su se uspjeπno izdvojiti digitalnim procesiranjem sate- litskih snimaka (transformacijom 3., 4. i 5. osnovne kompo- nente) (GARVIN et al., 1992). Procesi erozije i sedimentacije oteæavaju prepoznavanje struktura nastalih meteoritskim udarom i mogu promijeniti ili Analiza odabranih meteoritskih kratera i moguÊe meteoritske strukture u Juænoj Koreji na temelju Space Shuttle-ovih fotografija Neven TRENC , Ken M. MORGAN , R. Nowell DONOVAN & Arthur B. BUSBEY Vrsta Centralno Jedan Viπe uzdignuÊe prsten prstenova Krateri sa centralnim uzdignuÊem da ne ne Bazeni sa centralnim uzdignuÊem da da ne Bazeni sa uzdignuÊem i prstenom da da ne Bazeni sa viπe prstenova da ne da Tablica 1 TabliËni prikaz vrsta kompleksnih kratera prema PIKEU (1985). 212 Geologia Croatica 52/2 Ëak potpuno uniπtiti morfoloπki otisak kratera. Po svojim karakteristikama mnogi se dokazani meteorit- ski krateri mogu svrstati u skupinu prstenastih struktura (OLUI∆, 1983). Ova kategorija obuhvaÊa sve morfostruktur- ne elemente reljefa polukruænog ili eliptiËnog oblika, i takve forme se vrlo dobro uoËavaju na razliËitim snimcima koji se koriste u daljinskim istraæivanjima (OLUI∆, 1983). Prema is- tom autoru uzrok nastanka ovih struktura (uz meteoritske uda- re) mogu biti vulkanska ili magmatska djelatnost, tektonski poremeÊaji ili dijapirizam. Meteoritske strukture je Ëesto vrlo teπko razluËiti od prste- nastih struktura drugaËijeg podrijetla (ROLAND, 1976) no ako su opisani specifiËni indikatori dobro oËuvani, te ako je prisutno nekoliko karakteristiËnih elemenata (npr. rub kratera i centralno uzdignuÊe), rezultati analize metodama daljinskih istraæivanja mogu s dosta velikom vjerojatnoπÊu ukazati na postanak putem meteoritskog udara. 2. METODE 2.1. FOTOGRAFSKI SNIMCI NA»INJENI SA SPACE SHUTTLE-A Od 1982. godine astronauti na Space Shuttle-u se sluæe ruËnim foto kamerama i kamerama na stativu prilikom snima- nja iz svemira (EROS, 1996). Svaka Shuttle-ova misija naËini nekoliko tisuÊa fotografija, pa je do sada snimljeno oko 125.000 snimaka (EROS, 1996). Viπe od 80% fotografija su pogledi na planet Zemlju i naËinjene su iz razliËitih kutova, pod raznovrsnim osvijetljenjima i tijekom svih godiπnjih doba (MUEHLBERGER, 1996). Tri glavna tipa kamera koriπtenih tijekom Space Shuttle- ovih misija su: Hasselblad 500/M 70 mm, Linhof Aero Tech- nika 45-127 mm i Large Format Camera (LFC) (tablica 3; EROS, 1996). Kamera velikog formata je koriπtena samo u misiji 41 G i tom su prilikom naËinjene stereo-fotografije nekih podruËja (EROS, 1996). VeÊinom su to normalne kolor fotografije (RGB), premda su napravljene i neke crno-bijele i kolor-infracrvene snimke. Prostorna rezolucija fotografija ovisi o kameri, visini snima- nja i upotrebljenoj leÊi, te varira izmeu 150 i 20 m. 2.2. SPACE SHUTTLE ARHIVSKI SNIMCI I BAZE PODATAKA Nakon svake Shuttle-ove misije eksponirani filmovi se πa- lju u Image Science Division Johnson Space Centra u Housto- nu. Fotografije se nakon obrade svrstavaju u tri grupe: slike promatranja Zemlje, slike pokusa, i slike za javnu uporabu (JSC, 1994). Fotografije se zatim digitaliziraju te nastaju TARGA datoteke od oko 1.1 megabajta (Mb) (756 x 486 pik- sela s 8 bita). Dio datoteka se predstavlja javnosti putem Interneta te u formi videodiska (JSC, 1994). Kako snimci s video diska i Interneta nisu namijenjeni za znanstveno-struËna istraæivanja, oni imaju razmjerno slabu rezoluciju te je odgo- varajuÊe datoteke sa viπom rezolucijom potrebno dobaviti direktno iz Johnson Space Center-a. Za ovaj rad inicijalna potraga za odgovarajuÊim snimka- ma meteoritskih kratera je izvrπena preko Web stranice Earth Science Branch of the Lyndon B. Johnson Space Center (http://www.jsc.nasa.gov/). Mreæna baza podataka “Space Shuttle Earth Observations Project Database of Photographic Information and Images” (JSC, 1996) http://eol.jsc.nasa.gov/ sseop/ kontaktirana je preko WWW (World Wide Web) te je izvrπeno pretraæivanje putem kljuËnih rijeËi. Za identifikaciju optimalnih scena koriπten je takoer i video disk, Space Shuttle Earth Observations Video Disc (1981-1991), u izdanju Johnson Space Center-a s oko 91.500 snimaka Zemlje naËinjenih tijekom misija STS-1 do STS-44. Kolutovi filmova s odabranim snimkama su zatim identi- ficirani u Shuttle arhivu u Lockheed Martin Building (Hous- ton) i preneπeni u Lyndon B. Johnson Space Center. Limhof digitalna kamera, Adobe PhotoShop program i Macintosh ra- Ëunalo su koriπteni za digitalizaciju snimaka sa filmova. Materijal Opis Orbitalni radarski snimak Lavna ravnica Tokovi lave s djelovanjem bolida. Ravne i monotone Tamne do srednje tamne ovisno o ili zahvaÊene rasjedanjem i boranjem. stupnju hrapavosti povrπina. Svijetle linearne strukture su rasjedne povrπine. Rub IzbaËeni materijal sedimentiran do tri kraterska radijusa Svijetao u blizini ruba, postaje srednje daleko od centra. Grubi humËasti materijal postaje finiji svijetao s vanjske strane sa svijetlim prema rubu. mrljama. Zid Grube terasaste naslage na strmim unutarnjim zidovima. Anularna zona okruæuje dno kratera. Prevaljeni blokovi i lepeze su Ëesti. Dno Ravne do glatke unutarnje ravnice koje variraju od Svijetao do taman, kruæan do glatkih do humËastih povrπina. Materijal ukljuËuje nepravilan obris. rastaljene “udarene” stijene, pali materijal te vulkansko ispunjenje. UzdignuÊa i prsteni UzdignuÊa i prsteni unutar kratera koji se diæu iznad materijala dna. Uzdignute udarene i napregnute stijene. Vrlo svijetli s tipiËnom morfologijom. Izljevi Fluidizirani materijal koji podsjeÊa na tok lave . Kod Svijetao do srednje svijetao s obrisom nekih veÊih kratera proteæe se na stotine kilometara. u obliku cvijeta. Tablica 2 Prikaz udarom zahvaÊenih materijala na orbitalnim radarskim snimcima (SABINS, 1996). 213Trenc, Morgan, Donovan & Busbey: Remote Sensing Analysis of Selected Terrestrial Impact Craters... Nominalna rezolucija je bila oko 500 DPI (dots per inch - toËaka po inËu), a veliËina nastalih datoteka je varirala izme- u 5 i 10 Mb. Originalna rezolucija Shuttle-ovih fotografija bila je neizbjeæno reducirana tijekom ovog procesa. Glavni ograniËavajuÊi faktor bili su stativ i optiËke karakteristike Limhof kamere. Podaci su zatim pohranjeni na prenosivi tvrdi disk i pre- neseni u Texas Christian University Center for Remote Sens- ing and Energy Research na obradu. 2.3. RADARSKI SNIMCI S I R - C / X - S A R (Spaceborne Imaging Radar-C/X-Synthet- ic Aperture Radar - Slikovni radar za snimanje iz svemira- C/X-radar sa sintetiËkom antenom) kombinira dva zasebno razvijena radarska ureaja SIR-C i X-SAR te prima “quad” polarizirane (L, C kanali) impulse kao i VV polarizirane (X- kanal) impulse (FREEMAN, 1996). Ovaj instrument je bio koriπten tijekom dvije Shuttle-ove misije u 1994. Godini, a rezolucija prikupljenih podataka varira izmeu 10 i 50 metara (EVANS & PLAUT, 1996). Crvena, zelena i plava komponenta na SIR-C/X-SAR umjetnoj kolor snimci zapravo predstavljaju razliËite kombi- nacije radarskih kanala i polarizacija signala (JPL, 1996). Radarska snimka (scena) koja je koriπtena u ovoj studiji presnimljena je preko mreæe s JPL-ove Space Radar Images of the Earth web stranice (http://www.jpl. nasa.gov/radar/sir- cxsar/). Ondje je predstavljeno 150 SIR-C/X-SAR snimaka (umjetni - kolor), klasificiranih u osam tematskih kategorija s navedenim geografskim lokalitetima. Svaki snimak moæe biti kopiran preko WWW kao JPEG datoteka s niskom rezoluci- jom (oko 140 Kb) ili visokom rezolucijom (oko 10 Mb). 2.4. PROCESIRANJE SNIMAKA I ANALIZA LINEAMENATA Snimke su obraene u Texas Christian University Center for Remote Sensing and Energy Research, na Power Macin- tosh 9500/132 raËunalu s 50 Mb RAM-a. Dimple program za obradu snimaka koriπten je za relativnu rektifikaciju Shuttle fotografija spram SIR-C/X-SAR snimaka. Primijenjen je line- arni transformacijski model i metoda “najbliæeg susjeda” (nearest neighbour ). Geografski smjer sjevera utvren je na svim snimcima, a kako su oni bili ili geometrijski ispravljeni (radar) ili priliËno nedeformirani (Shuttle), precizno geograf- sko poboljπanje nije raeno jer je tijekom ovog istraæivanja bio znaËajan samo relativan poloæaj morfoloπkih i drugih ele- menata spram struktura nastalih djelovanjem meteorita. Aplikacija Adobe Photoshop je koriπtena za poboljπanje snimaka. Tako je kontrast podeπen na svim Shuttle fotografi- jama, a primijenjena su i Ëetiri filtra za pojaËanje linearnih elemenata. Smjerovi njihove iluminacije su bili sjever, istok, sjeverozapad i jugozapad. Originalni snimak i Ëetiri izvedene filtrirane snimke pohranjene su kao aplikacijski slojevi u jedinstvenom Adobe Illustrator dokumentu. Lineament se definira kao jednostavni ili kompleksni line- arni element Ëiji su dijelovi rasporeeni pravocrtno ili blago zakrivljeno i koji se bitno razlikuje od susjednih struktura (O’LEARY et al., 1976). Opaæeni lineamenti su izvuËeni na dodatnim aplikacijskim slojevima u sklopu Adobe Illustrato- ra. Rezultati dobiveni pomoÊu razliËitih filtera su usporeeni pa su linearni elementi uzrokovani ljudskim aktivnostima ili nejasnog podrijetla bili odstranjeni kad god je to bilo moguÊe. Mapict program (dr A. BUSBEY, Texas Christian Uni- versity) koriπten je prilikom digitalizacije lineamenata koji su zatim uneπeni u aplikaciju “Rosy” za izradu rozetnih dijagra- ma. 3. REZULTATI I DISKUSIJA Snimci viπe meteoritskih kratera pronaeni su u Shuttle- ovoj bazi podataka te su odabrani oni najprikladniji za obradu (tablica 4, sl. 5). Promjeri istraæivanih struktura variraju izmeu 1,2 i 100 km, a zastupane su razliËite veliËinske kate- gorije meteoritskih kratera na Zemlji. 3.1. MANICOUAGAN KRATER Manicouagan krater u Kanadi (sl. 5 i 6) smjeπten je u istoËnom dijelu poluotoka Labrador. Starost mu je procijenje- na na 214 milijuna godina (FLORAN & DENCE, 1976). Ovo podruËje pripada Grenville provinciji kanadskog πtita te su u okolici kratera Manicouagana utvrene razliËite petroloπke Grenville jedinice (sl. 6), zastupane metagabrima, metaanor- tozitima i granitskim gnajsovima (ORPHAL, 1978). Dugotra- jna izloæenost procesima erozije i glacijacije znatno je promi- jenila ovu strukturu (FLORAN & DENCE, 1976). Analiza kratera Manicouagan metodama daljinskih istra- æivanja pokazuje tipiËni sustav anularnih (kruæno rasporee- nih) rasjeda koji definiraju poligonalni oblik kompleksnih udarnih struktura (sl. 7). Orijentacija spomenutih rasjeda koincidira s najzastupljenijim pravcima lineamenata na ovom podruËju, sjeveroistok-jugozapad i sjever-jug, πto podupire Kamera LeÊa Prostorna Senzor s usporedivom rezolucija rezolucijom Hasselblad 500 EL/M 100 mm 150 m MSS (80 m) Hasselblad 500 EL/M 250 mm 30 m TM (30 m) Linhof Aero Technika 250 mm 30 m TM (30 m) Large Format Camera N/A 20-30 m Spot HRV (20 m) - TM (30 m) Tablica 3 Usporedni prikaz znaËajki razliËitih kamera i senzora za sni- manje iz satelita (EROS, 1996). Struktura VeliËina Mjesto po vel. Manicouagan (Kanada) 100 km 5 Aorounga (»ad) 17 km 40 Roter Kamm (Namibija) 2.5 km 119 Meteor Crater (Arizona) 1.2 km 130 MoguÊa struktura 13 km - u Juænoj Koreji Tablica 4 Promjer i mjesto po veliËini analiziranih meteoritskih struktura. 214 Geologia Croatica 52/2 pretpostavku koju su iznijeli FLORAN & DENCE (1976) da je njihov postanak bio uvjetovan strukturama i zonama slabo- sti koje su na terenu postajale prije meteoritskog udara. Radi- jalni sustav povrπinskih tokova i druge karakteristike tipiËne za takve strukture (centralno uzdignuÊe, kruæna depresija) su takoer identificirani na snimku. 3.2. AOROUNGA STRUKTURA Aorounga stuktura (sl. 5, 8 i 9) smjeπtena je u blizini Tibesti masiva u Borkou regiji istoËnog »ada (BECK-GIRA- DUON et al., 1992). Promjer joj iznosi oko 17 km (JPL, 1996), a starost strukture je procijenjena na temelju pokrovnih sedimenata kao kasnodevonska (OCAMPO, 1996). Shuttle fotografija ove strukture je analizirana i usporee- na sa SIR-C/X-SAR (sl. 9) snimkom koja je preuzeta preko mreæe s JPL-ove Internet stranice. Reljefne forme tipiËne za meteoritske strukture na krateru Aorounga su dobro izraæene i uoËavaju se na oba koriπtena snimka. Na Shuttle fotografiji pijeskom prekrivena podruËja su svijetlije obojena, dok je otkrivena stijena tamno siva. Svi ele- menti strukture, rub, prsten i centralno uzdignuÊe su jasno prepoznatljivi. Centralno uzdignuÊe moæe se identificirati na snimci kao tamna mrlja otkrivene stijene u srediπtu strukture. Radarski zraci prodiru kroz nekoliko stopa suhog pijeska (McHONE et al., 1996), te se spomenuti tipiËni elementi izvrsno uoËavaju. Dva izvora podataka, koriπtena tijekom ovog istraæivanja (Shuttleove RGB fotografije i SIR-C/X- SAR snimci) meusobno su se nadopunjavali. Pri tome je di- gitalizirana RGB fotografija dala kvalitetnu informaciju o ton- alnim karakteristikama materijala na tlu, a na radarskom su snimku morfoloπki elementi strukture bili bolje izraæeni. Usporedbom rezultata analize lineamenata utvreno je da ih je veÊi broj detektiran na radarskoj snimci, te da je na njoj veÊa zastupljenost lineamenata sa pravcem sjever-jug. Veliku je poteπkoÊu pri obradi predstavljalo razluËivanje izmeu lin- eamenata koji su vezani uz udarnu strukturu i eolskih pojava (orijentacija jugozapad-sjeveroistok). Ovi maskirajuÊi efekti pjeπËanog pokrova bili su snaænije izraæeni na Shuttle fotogra- fiji, pa se time mogu objasniti i razlike u analizama dvaju sni- maka. 3.3. KRATERI ROTER KAMM I METEOR CRATER Roter Kamm (slika 5) udarni krater nalazi se u juænom dijelu pustinje Namib, Namibija (FUDALI, 1973). Ovaj rela- tivno mali krater ima promjer 2,5 km, a starost mu je procije- njena na oko 3,7 milijuna godina (GRIEVE et al., 1995). Meteor Crater (sl. 5) je smjeπten u podruËju kanjona Dia- blo, sjeverne-centralne Arizone (SAD) (SHOEMAKER, 1977). Udar meteorita prije pribliæno 49 tisuÊa godina uzroko- vao je nastanak te strukture Ëiji promjer iznosi oko 1.220 m (GRIEVE et al., 1995). Na Shuttle-ovim fotografijama navedenih manjih struktu- ra, smjeπtenih u aridnom podruËju, bilo je moguÊe registrirati samo poligonalni rub i nekoliko lineamenta. Glavni problem su predstavljali mala veliËina strukture, niska rezolucija foto- grafija i pjeπËani pokrov. 3.4. MOGU∆I METEORITSKI KRATER U JUÆNOJ KOREJI MoguÊi meteoritski krater u Juænoj Koreji smjeπten je oko 20 km sjeverno od Seoula, kod grada Uijongbu u blizini demilitarizirane zone (JONES et al., 1995). Posada Space Shuttle-a Endaveor (STS-59) snimila je fotografije ove struk- ture tijekom jedne misije izvedene 1994. godine. JONES et al. (1995) su prvi opisali taj moguÊi meteoritski krater i raspravili njegov postanak. Prema JONES et al. (1995) polukruæni greben s promje- rom od 13 km i centralno uzdignuÊe karakteriziraju krateru sliËnu strukturu. Visinska razlika od dna do vrha grebena je veÊa od 400 m a struktura je smjeπtena unutar jurskih granita, okruæenih s 2 milijarde godina starim gnajsovima i πistovima. Rasjed Tongducheon s orijentacijom sjever-jug presjeca strukturu na dva dijela. KANG et al. (1985) navode da je pucanje duæ pukotine kruænog oblika tijekom hlaenja granitskog plutona uvjetova- lo nastanak polukruæne forme, meutim JONES et al. (1995) ne smatraju da je struktura magmatskog podrijetla te pret- postavljaju da se radi o ostatku meteoritskog kratera. BuduÊe studije uzoraka stijena iz kratera trebale bi dati relevantne informacije o njezinom podrijetlu. Analiza navedene strukture je provedena na Shuttle fotografijama te Landsat TM i Spot HRV snimcima. Na sva tri snimka bilo je moguÊe podjednako dobro registrirati mor- foloπke elemente pretpostavljenog meteoritskog kratera. Broj registriranih lineamenata bio je podjednak na Shuttle fotogra- fijama i Landsatovim snimcima dok je bolja prostorna rezolu- cija SPOT snimka omoguÊila zapaæanje neπto veÊeg broja lin- eamenata (osobito oni kraÊeg pruæanja). MoguÊi rub kratera i centralno uzdignuÊe na snimcima se lako uoËavaju, no istoËni dio ruba kratera je znatno bolje ocr- tan od njegove zapadne strane. Dva snaæna lineamenta (jedan od njih odgovara Tongducheon rasjedu) jasno se uoËavaju na snimci, a imaju pravac pruæanja sjever-jug i sjeveroistok- jugozapad. Osim dobro izraæenih lineamenata ruba kratera, na snim- kama su registrirani brojni kraÊi lineamenti s razliËitim ori- jentacijama. Oni se manifestiraju kao linearne doline i gre- beni. Iako neki od njih imaju anularnu i radijalnu orijentaciju u odnosu prema strukturi, ovaj podatak nije znaËajan ako se uzme u obzir velika sveukupna disperzija pravaca ovih linea- menata. U niæim predjelima prekrivenim poljoprivrednim povrπinama identifikacija lineamenata nije bila moguÊa. Razmatrana prstenasta struktura pokazuje neke tipiËne elemente identificirane na Shuttle-ovim fotografijama drugih dokazanih meteoritskih struktura (centralno uzdignuÊe i dio ruba kratera), no oni nisu potpuno razvijeni i rezultati analize ne daju definitivne argumente za bilo koju od hipoteza o nje- zinom postanku. 4. ZAKLJU»AK Koriπtene Space Shuttle fotografije imaju ograniËenu pri- mjenu u daljinskoj analizi meteoritskih kratera. OgraniËavaju- Êi faktori su relativno slaba prostorna rezolucija, nedigitalan oblik podataka i nepotpuno prostorno prekrivanje. Ipak, na analiziranim fotografskim snimkama sa Shuttle-a kao i na koriπtenim radarskim i satelitskim snimkama izvrπena je ana- liza lineamenata, te je bilo moguÊe vrlo dobro registrirati ele- mente prstenastog oblika bilo meteoritskog ili drugog podri- jetla. Uspjeπna identifikacija snimaka s kraterima moguÊa je putem mreænih (WWW) baza podataka i baza podataka na vi- deo diskovima. Tekstualno pretraæivanje pomoÊu kljuËne rije- Ëi ili geografskog lokaliteta nabræi je put do traæenih scena. Tehnike poboljπavanja snimka i pojaËavanja lineamenata su uspjeπno primijenjene na skaniranim Shuttle fotografijama. Neki morfoloπki elementi te karakteristiËni sustavi pukotina i rasjeda identificirani su na svim snimcima, no rezultati dobi- 215Trenc, Morgan, Donovan & Busbey: Remote Sensing Analysis of Selected Terrestrial Impact Craters... veni na temelju snimaka kratera manjih od 2 km bili su bitno slabiji. Bolja rezolucija snimaka, njihov digitalni format, te misi- ja posveÊena snimanju meteoritskih struktura bitno bi unapri- jedili vrijednost fotografskih snimaka sa Shuttle-a pri istraæi- vanju meteoritskih kratera. Zahvale Ovo istraæivanje ne bi bilo moguÊe bez pomoÊi slijedeÊih organizacija: Texas Christian University Research Foundati- on, NASA, Adkins Research Fund, te Lyndon B. Johnson Space Center-a. Dr Arthur EHLMANN (Texas Christian Uni- versity) dao je izuzetno vrijedne savjete vezane uz literaturu i stranice na Internetu. Lt. Col. Ned FLEMING i Sue RUNCO iz Johnson Space Centra svojom ljubaznom pomoÊi omoguÊi- li su ovo istraæivanje. Takoer bismo æeljeli zahvaliti dr. Ivi VELI∆U i prof. dr. Marinku OLUI∆U na vrijednim savjetima i izuzetno korisnim sugestijama. 216 Geologia Croatica 52/2