Printer Friendly

Hyperspectral remote sensing and zone of degasification in part of Sabatayn Basin-Yemen.


Hyperspectral imagery has been particularly effective for mapping the alteration minerals associated with hydrothermal economic deposits [1, 2, 3]. ASTER multispectral images can reveal successfully mineralogical alterations induced by hydrocarbon seepages such as bleached red bed and secondary carbonates, which represent an open window to petroleum system and provide indirect evidence for the presence of hydrocarbon system at depth. Therefore, the presence of seepages documents the first element of a hydrocarbon system and reduces exploration risks. Spectral Mapping such as, Spectral Feature Fitting technique were applied by many researchers in remote sensing studies [4-15]. These algorithms were implemented based on the comparison of a pixel spectrum with the spectra of known pure materials. Multispectral remote sensing can be used to detect changes in lithology, while hyperspectral imagery can potentially be used to identify minerals and differentiate alteration products. Some of the mineralogic changes associated with oil and gas seeps have been identified in remotely sensed imagery. Bleaching and iron loss in sandstones caused by seeps have been mapped along the crests of anticlines [16- 20]. Seeps may alter gypsum to carbonate [18] and also, through the reaction of hydrocarbons with carbon dioxide, produce secondary carbonate or delta carbonate [16]. Another potential indicator of hydrocarbons is enrichment of kaolinite due to the alteration and depletion of other clays [3, 16, 18, 20]. Finally, certain minerals found predominantly in altered areas, including halloysite, siderite and alunite, [3, 17], have been used to identify subsurface reservoirs.Interaction between electromagnetic radiation and the atoms and molecules that make up the rock or minerals produces vibrational-rotational or electronic processes that create spectral features.

Reflectance spectra of minerals and rocks in the visible to short-wave infrared wavelength (0.5-2.5 [micro]m) region of the EM spectrum are characterized by absorption features, caused primarily by electronic transitions and vibrational transitions [21].These diagnostic spectral features can provide a way to detect or identify minerals to distinguish lithologic units and to determine wall rock alteration products [22] by using remote sensing. The purpose of this study is to employ spectral Mapping techniques such as Spectral Feature Fitting(SFF), algorithm on the ASTER data to discriminate possible mineral alteration zones due to oil microseeps in the study area. Analysis focus on the spatial distribution of the main types of altered minerals including kaolinite, alunite, calcite, siderite and halloysite.

Geological Review

The geological evolution of Yemen was driven by the plate motions that broke Pangaea apart in the Mesozoic and formed the Gulf of Aden,Red Sea, and the Arabian Peninsula in the Cenozoic. The stratigraphic and regional geology of Yemen was established by detailed work of [23-29]. Hydrocarbon exploration activity became extensive after 1990 and provided considerable amount of subsurface data, which revised synthesis of basin evolution in Yemen, such as the work of [30-32]. The petroleum geology was summarized in [33]. The Sabatayn Basin is a NW-SE trending rift basin comprising a series of asymmetric half grabens [33] which are thought to have inherited the Precambrian Najd trend of strike-slip faults. Rifting was initiated in the Late Jurassic(Kimmeridgian to Tithonian) and was followed by thermally driven (post-rift) subsidence and a period of uplift and erosion during the Early Cretaceous.


A second rift phase (Hauterivian to Parremian) was associated with local reactivation of some graben-bounding fault systems. A number of large intra-basinal highs were active and controlled sedimentation throughout the Late Jurassic and Cretaceous.the Mesozoic rifting and sedimentary basin evolution is well constrained [30,32], while the complex, polyphase tectonics in the Tertiary [31] is much less understood. At the end of the syn-rift phase, the Sabatayn basin became isolated from the sea maintaining a periodically opened marine passage, which supplied saline water into the basin. The geographic separation and warm climate gave rise to massive evaporation. The deposited salt (Sabatayn Formation) produced various halokinetic features during the Cretaceous and Cenozoic time. The assessment of the hydrocarbon perspectives at the previous stage of investigations [34] was based on direct indicators of the oil presence such as oil seepage from the salt (Sabatayn Fm.) Late Jurassic-Early Cretaceous beds developed in Ayad, Shabwa, Mil Khirwa, and Mil Mqah Domes, (Fig.1). Other zones of oilseeps were shown in the area of Hajar Trough, located to the southern East part of the study area such as Mintaq, Wadi Gudah, and Wadi Jadad. However seepage of asphalt indicates the formation of Petrol. Generally, the Marls of the Madbi Formation sediment in local depressions and Troughs can be considered as the main source rocks [23].

Material and Methods

The ASTER data used in this study are level 1B data, four scenes were used in this study. The images have been pre-georeferenced to geographical Lat/Long. With WGS -84 datum. The VNIR and SWIR data were then resampled so that all 9 bands have the same 15x15 m pixel size. Atmospheric correction was applied to each scene separately, then mosaic applied to the four scenes after that a subset corresponding to the study area with (5057.0000x4407.0000) pixels was derived for analytical procedures. The procedures of spectral mapping were applied to the study area shown in the following chart, (Fig.2)


Results and Discussion

Atmospheric Correction: FLAASH is a first-principles atmospheric correction modeling tool for retrieving spectral reflectance from hyperspectral and multispectral radiance images. With FLAASH, you can accurately compensate for atmospheric effects. FLAASH corrects wavelengths in the visible through near-infrared and shortwave infrared regions, up to (2.5 [micro]m). Unlike many other atmospheric correction programs that interpolate radiation transfer properties from a pre-calculated database of modeling results, FLAASH incorporates the MODTRAN4 radiation transfer code. The first step in pre-processing was to convert the image radiance data to apparent reflectance to facilitate comparison with the library reflectance spectra. This process normalizes for solar illumination and suppresses the effect of the atmosphere, including spectral absorption and scattering by diffuse gases and particles. Each image was processed with FLAASH [35],an add-on program for ENVI that uses MODTRAN radiation transfer code and the image spectra themselves to estimate the spectral reflectance conversion factors. To begin, a scale factor was required for the input radiance data. The Aster metadata provided this information in an ASCII radiance scale factor file.

The individual image navigation files provided the image and sensor information for each image. Because no local radiosonde data were available, the Tropical (T) model atmosphere was chosen with default value (4.11) in the water column multiplier field. The navigation file listed the weather condition at the time of flight as clear, so the scene visibility was set to the default value of (40 km). Aerosol model were performed as Rural. The Areosol Retrival kept as None. Once the radiance was converted to reflectance, Internal Average Relative Reflectance IARR calibration tool were then applied in terms of obtaining the optimal atmospheric correction technique.

The IARR is used to normalize images to a scene average spectrum. It is found to be effective for reducing image data to relative reflectance in an area where no ground measurements exist and little is known about the scene. It has proven to work best in arid areas with little or no vegetation. The algorithm is designed to calculate an average spectrum from the entire scene and use it as a reference spectrum, which is then divided into the spectrum at each pixel of the image. The method found to be the most useful and give the most accurate (similar to the resampled library) results, minerals were mapped using three spectral matching methods, described in more detail below.

Endmember selection: Spectra of the image could extract from the "spectral end-member selection" procedure, including minimum noise fraction (MNF), pixel purity index (PPI) and n-dimensional visualization.

Minimum noise fraction transformation: The minimum noise fraction (MNF) transformation is used to determine the inherent dimensionality of image data, to segregate noise in the data, and to reduce the computational requirements for subsequent processing [5]. The inherent dimensionality of the data is determined by examination of the eigenvalues and associated eigenimages, The decreasing eigenvalue with increasing MNF band as shown in the eigenvalue plot in, (Fig.3a).

The data space is divided into two parts: one associated with large eigenvalues and coherent eigenimages, and a second with near-unity eigenvalues and noise-dominated images. By using only the coherent portions in subsequent processing, the noise is separated from the data, thus improving spectral processing results. In a common practice, MNF components with eigenvalues less than 1 are usually excluded from the data as noise in order to improve the subsequent spectral processing results [13], The MNF procedure was applied on the IARR data, since the eigenvalues of all MNF eigenimages of the data were greater than 1, so all the 9 bands were retained for subsequent data processing, color composite MNF Band1, 2, 3 shown in, (Fig.3b).


Pixel purity index: The Pixel purity index (PPI) is a means to determine automatically the relative purity of the pixels from the higher order MNF eigenimages [5].The Pixel Purity Index is computed by repeatedly projecting n-dimensional scatter plots onto a random unit vector. The extreme pixels in each projection are recorded and the total number of times each pixel is marked as extreme is noted. A 'pixel purity image' is created in which the digital number (DN) of each pixel corresponds to the number of times that pixel was recorded as extreme.


Pixel Purity Index was implemented on the MNF images. To select the most pure pixels a 10000 projection of the scatter plot and the threshold factor of 3.5 were applied on the data, PPI plot shown in, (Fig.4a). Density slice thresholds were used to determine pixels with high DN or pure pixels. These values were applied to computing the region of interest (ROI) shown in, (Fig.4b), being used for n-dimensional visualization.

n-Dimensional visualization and Extracted Endmember Spectra: The n-D visualization was used in conjunction with the MNF and PPI tools to locate, identify and cluster the purest pixels and the most extreme spectral responses in a data set. If spectral signatures are recorded properly and the curve shape is accurate they could be used for remote sensing applications [36].Spectra can be thought of as points in a dimensional scatter plot, where n is the number of bands [5].The coordinates of the points in n-space consists of "n" values that are simply the spectral radiance or reflectance values in each band for a given pixel. The distributions of these points in n-space were used to estimate the number of spectral endmembers (5highlighted segmentations), as shown in, (Fig.5a), producing 5 pure spectral signatures, which were extracted and plotted in an n-D visualizer plot representing the selected endmembers, as shown in, (Fig.5b).


These spectra could be derived from IARR images based on the spatial locations or can be taken from MNF images by inversing the MNF plots to the spectra. The extracted spectra were used as reference for subsequent processing. Spectral analyses and consequently targeted mineral identification could be obtained by matching the unknown spectra extracted from the 3-D visualizer to pre-defined (library) spectra, providing scores with respect to the library spectra. Three weighting methods, i.e. Spectral Feature Fitting (SFF), Spectral Angle Mapper (SAM) and/or Binary Encoding (BE), were used to identify mineral type, producing a score between 0 and 1, where 1 equals a perfect match. As is known, some minerals are similar in one wavelength range, yet very different in another. For the best results, a wavelength range that contains the diagnostic absorption features was used to distinguish among the minerals. The output of the spectral analysis is a ranked score or weighted score for each of the materials in the input spectral library, as shown in, (Table.1).

The highest score indicates the closest match and shows higher confidence in the spectral similarity, where calcite, halloysite, and alunite scored high values of 0.581, 0.549, and 0.497 respectively, while siderite and kaolinite scored 0.268 and 0.0, respectively, using SFF weighting. The same minerals recorded scores of 0.778, 0.556, 0.556, 0.889, and 0.778 respectively, using BE weighting. On the other hand, the SAM did not recognize any kind of minerals (zero score).The spectra of several clay/carbonate minerals have been taken from the IGCP Spectral Library and these spectra have been resampled to ASTER VNIR and SWIR bandpasses for comparison purposes. The results of five of these comparisons are shown below, (Fig.6).

The spectra of advanced argillic mineral alunite characterized by a diagnostic absorption feature near 2.165um.When resampled to Aster bandpasses the gross curve shape curve is preserved and the diagnostic features are depicted by a significant spectral absorption in band 5 (centred 2.165um). Whilst Aster probably can not separately identify these minerals species, it should identify the low pH/acid environments in which these minerals occur. IGCP Library spectra of kaolinite characterised by a doublet shaped diagnostic absorption feature near 2.175/2.210um.When resampled to Aster bandpasses the gross shape of the curve is preserved and the diagnostic features are depicted by an asymmetric absorption in band 6 (centred 2.210um).

Whilst Aster probably can not identify these minerals species directly, kaolinite minerals as a group should be identifiable. The spectra of calcite characterised by a single diagnostic absorption feature near 2.350um. When resampled to Aster bandpasses the gross shape of the curve is preserved and the diagnostic features are depicted by a symmetric absorption in band 8 (centred 2.320um).Siderite (FeCO3) found extensively in sedimentary beds and is frequently contaminated with clay or organic matter. It is also commonly deposited in veins by hydrothermal solution. Both manganese and magnesium substitute from iron. Siderite is characterized by a strong, board [Fe.sup.2+] band at 1.1um and some activity in the longer wavelength due to the carbonate, but due to being an opaque mineral there is an absence of features in the visible region of the EM spectrum. Siderite is roughly the equivalent of calcite but with Fe replacing the calcium. When resampled to Aster bandpasses the gross shape of the curve is preserved and the diagnostic features are depicted by a symmetric absorption in band 8 (centred 2.320um). The mineral halloysite displays a similar to structure to kaolinite and occur in a hydrated variety.

Aluminum in kaolinite has found to substitute for Fe, which causes an infection on the long wavelength shoulder of the 2.2 um absorption. the same has been noticed to halloysite, When resampled to Aster bandpasses the gross shape of the curve is preserved and the diagnostic features are depicted by a symmetric absorption in band 6 (centred 2.20um).


Spectral Feature Fitting: Spectral Feature Fitting (SFF), [37-39], is another spectral library matching technique for classifying unknown image spectra. A particular strength of SFF is that it isolates individual absorption features for comparison, and only the shapes of the features are compared, not the depth of those features. The first step in the SFF analysis is the removal of the overall shape of the spectrum, known as the continuum, from the image and reference spectra. The continuum is formed by connecting the local maxima of the spectrum with straight line segments[39]. Without removing the continuum, it is difficult to define distinct absorption features because illumination and particle size differences tend to dominate the spectra.

The image and reference spectra are therefore normalized by dividing the radiance or reflectance values by the estimated continuum values [37,39]. A constant is added to the library continuum-removed spectrum to provide a scaling factor in comparing the library and image data. This scaling is needed because the absorption features in the library data typically have greater depth than in the image spectra. Next, a least-squares-fit is calculated band-by-band between each reference endmember and the unknown spectrum, using standard statistical methods. Three types of images are produced with SFF: scale, RMS, and fit image. The scale image, produced for each endmember, is the scaling factor used to fit the unknown spectra to the library spectra. The total root-mean-square (RMS) error is a measure of the average difference between the image spectrum and the library reference spectrum. Low RMS values are equivalent to good spectral matches. The fit image is the ratio of the scale image to the RMS image.

The fit image can be used to provide an overall perspective of how well the unknown spectrum matches the reference spectrum on a pixel-by-pixel basis. The same endmembers and 0.5-2.5[micro]m spectral region were chosen for the SFF classification as were used for the SAM classification. While running SFF, this was the only parameter able to be modified. SFF does not produce a color-coded map, so post classification was required to generalize the classes. The ENVI program Rule Classifier [35], was used to create a new classified image based on thresholds from the histograms for each endmember. The thresholds, chosen subjectively, represent a scaling factor for comparing the fit values. These thresholds varied between endmembers and even between images.

Spectra Feature Fitting is a more statistical absorption matching technique than other spectral Mapping. However, for this study, SFF was found to be well suited. Without expert knowledge of the distribution of the different minerals in the area, a satisfactory threshold for each mineral could not be determined for the rule classifier by which the mineral maps are combined to form a final classification. Consequently, the final maps appear to be well classified, (Fig.7). Calcite mineral was limited only to the area of the carbonate plateau, calcite dominated the Formation of Umm Er Duma, it is observed in the area of Jawl Duqm Rahba, Wadi Dahum, and Al hujayrah. Kaolinite observed in limited zones following the streams such as the area of Salmun, Wadi Hawal, and Wadi Yabrum. Siderite mineral observed in areas associated with alunite mineral, these areas in the carbonate plateau known as Jeza Formation of Limestone component, while in the area of Recent deposits Siderite located around Alunite, this case could be observed in the area of Wadi Kharwa. Hyperspectral data from Aster were used successfully.


Mapping alteration zones due to possible oil microseeps, image processing techniques were applied and possible altered minerals such as alunite, kaolinite, calcite, siderite, and halloysite were mapped using Spectral Feature Fitting (SFF). SFF classification map shows best interpreted image among the other mapping techniques. Different layers were overlaid over SFF classification map of alerted minerals such as zones of hydrocarbon generation and expulsion, zones of interpreted traps (Leads and prospects), wellbores locations, and Zones of Faults. The result of this integration was satisfied and possible zones of minerals alteration due to oil microseeps were detected in the northern part of the study area. Hence, the research study finally concludes that alteration mapping using Aster images when applied to oil and gas exploration in area such as Sabatayn Basin in Yemen, successfully shows possibility of determining Traps and reservoirs and gives advantage to reduce exploration Ricks, cost effective and time.



[1] Hunt, G.R., 1979, Near-infrared (1.3-2.4 um) spectra of alteration minerals--Potential for use in remote sensing: Geophysics, v. 44:12, pp. 1974-1986.

[2] Crosta, A.P., Sabine, C. and Taranik, J.V., 1998, Hydrothermal Alteration Mapping at Bodie, California, Using AVIRIS Hyperspectral Data: Remote Sensing of Environment, v.65:3, pp. 309-319.

[3] Buckingham, W.F. and Sommer, S.E., 1983, Mineralogical characterization of rocksurfaces formed by hydrothermal alteration and weathering Application to remotesensing: Economic Geology, v. 78, pp. 664-674.

[4] Kruse, F.A., Lefkoff, A.B., Boardman, J.W., Heidebrecht, K.B., Shapiro, A.T., Barloom, Landes, K.K., 1973, Mother nature as an oil polluter: AAPG Bull, v.57:4, pp. 637-641.

[5] Boardman, J.W. and Kruse, F.A., 1994. Automated spectral analysis: a geologic example using AVIRIS data, North Grapevine Mountains, Nevada. In Proceedings of the Tenth Thematic Conference on Geologic Remote Sensing, 9-12 May, San Antonio, Texas (Ann Arbor, MI: Environmental Research Institute of Michigan), P. 407-418.

[6] Van Der Meer. F 1995, Spectral unmixing of Landsat Thematic Mapper data. INT.J.Remote Sensing, V. 16NO.16, P.3189-319

[7] Van Der Meer. F, 1996, Spectral mixture modeling and spectral statigraphy in carbonate lithofacies mapping. ISPRS Journal of Photogrammetry & Remote Sensing v.51, P.150-162

[8] Van Der Meer.F, 1999, Iterative Spectral Unmixing. INT.J. Remote Sensing, V.20, NO.17 P.3431-3436.

[9] Van Der Meer. F, Dejong S.M, 2001, Remote sensing and digital image processing. imaging spectroscopy Kluwer Academic publited in the Netherlands V.4, 306P.

[10] Galvao L.S, Filho R.A, Vitorello. C, 2001.Spectral discrimination of hydrothermal altered materials using ASTER Short--wave infrared bands: Evaluation in a tropical Savannah Environment. International Journal of Applied Earth Observation and Geoinformation, V.7, P. 107-114.

[11] Rown. L.C, Hook. S.J, Abrams. M.J, Mars. J.C, 2003, Mapping hydrothermally altered rocks at Cuprite, Nevada, using the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) a new satellite imaging system" V.95, P.1019-1027

[12] Rown. L.C, Schmidt. R.G, Mars. J.C, 2006, Distribution of hydrothermally altered rocks in Reko Diq, Pakistan mineralized area based on spectra analysis of ASTER data "Remote Sensing of Environment V.104, P.74-87

[13] Qui, F. Abdelsalam, M. Thakka, P. 2006, Spectral analysis of ASTER data covering part of the Neoproterozoic Allaqi-Heiani suture, Southern Egypt. Journal of African Earth Sciences., Vol. 44, pp169-180.

[14] Tangestani. M.H, Mazhari. N, Agar. B, 2005, Mapping the porphyry copper alteration zones at the Meiduk areas, SE Iran, using the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER)data" SPIE, V.59830.59830, S P.1-10

[15] Tangestani M.H, Mazhari M, Agar, Moore F. 2008, Evaluating Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data for alteration zone enhancement in a semiarid area, northern Shahr-e-Babak, SE Iran. International Journal of Remote Sensing. V. 29, No. 10, P. 2833-2850.

[16] Yang, H, Zhang, J., van der Meer, F. and Kroonenberg S.B., 1998, Geochemistry and field spectrometry for detecting hydrocarbon microseepage :Terra Nova, v. 10, pp. 231

[17] Everett, J.R., Staskowski, R.J. and Jengo, C., 2002, Remote sensing and GIS enable future exploration success: World Oil, v. 223:11, pp. 59-60, 63-5.

[18] Donovan, T.J., 1974, Petroleum microseepage at Cement Oklahoma: Evidence and mechanism: AAPG Bull, v. 58:3, pp. 429-446.

[19] McCoy, R.M., Blake, J.G. and Andrews, K.L., 2001, detecting hydrocarbon microseepage using hydrocarbon absorption bands of reflectance spectra of surface soils: Oil and Gas Journal, May 28, pp. 40-45.

[20] Berger, Z., (1994).Satellite Hydrocarbon Exploration. Interpretation and Integration Techniques. Berlin, Heidelberg, New York, London, Paris, Tokyo, Hong Kong: Springer-Verlag. Geological Magazine, vol. 133, No.1, pp 319.

[21] Van der Meer, F.D., van Dijk, P.M., Schetselaar, E.M., Little, M., Podlaha, O., Yang Hong and Biggert, E. (2000). An integrated geosciences approach for hyperspectral hydrocarbon microseepage detection. In: 2000 Proceedings of the 15th international conference on applied geologic remote sensing, Environmental Research Institute on Michigan (ERIM), 2000. Pp.81-88.

[22] Abrams, M.J., Ashley, R.P., Rowan, L.C., Goetz, A.F.H. and Kahle, A.B., (1977). Mapping of hydrothermal alteration in the Cuprite Mining District, Nevada, using aircraft scanner images for the spectral region 0.46 to 2.36pm. Geology, Vol.5, pp. 713-718.

[23] Beydoun, Z.R., (1964).The stratigraphy and structure of the eastern Aden Protectorate. Overseas Geology and Mineral Resources Supp. Ser., Bull.Supp.v.5. HMSO, London.

[24] Powers, R.W., Ramirz, L.F., Redmond, C.D., and Elberg, E.L., (1966).Sedimentary Geology of Saudy Arabia in Geology of the Arabian Peninsula. United States Geological Survey Professional Paper 560-D.

[25] Beydoun, Z.R. and Greenwood, J.E.G.W., (1968).Aden Protectorate and Dhufar, in Lexique Stratigraphique International. Vol. III., Asie (L. Dubertret, ed.), CNRS, Paris, Fasc.1062 p.

[26] Hughs, G.W., and Beydoun, Z.R., (1992). The Red Sea-Gulf of Aden: Biostratigraphy, Lithostratigraphy and Paleoenvironments. Journal of Petroleum Geology, v.15, p. 135-156.

[27] Paul, S.K., (1990). People's Democratic Republic of Yemen: a future oil province. in Brooks, J., ed., Classic Petroleum Provinces. London, Special Publications of the Geological Society, v. 50, p. 329-339.

[28] Schlumberger, (1992). Looking for Yemen's hidden treasure. Schlumberger Technical Services Dubai, Middle East Well Eval. Rev., v.12, p.12-29.

[29] Beydoun, Z.R., Bamahmaud, M.O., and Nani, A.S.O., (1993). The Qishn Formation, Yemen: lithofacies and hydrocarbon habitat. Marine and Petroleum Geology, v.10, p. 364-372.

[30] Redfern, P., and Jones, J.A., (1995).The interior rifts of the Yemen--analysis of basin structure and stratigraphy in a regional plate tectonic context. Basin Research, v.7, p.337-356.

[31] Ellis, A.C., Kerr, H.M., Cornwell, C.P., and Williams, D.O., (1996). A tectonostratigraphic framework for Yemen and its implications for hydrocarbon potential. Petroleum Geoscience, v. 2, p. 29-42.

[32] Beydoun, Z.R., As-Saruri, M., and Baraba, R.S., (1996). Sedimentary basins of the Republic of Yemen: their structural evolution and geological characteristics. Revue Institute Francais Petrole, v.51, p. 763-775.

[33] Csato, I., Habib, A., Kiss, K., Koncz, Zs. Lorinz, K., Milota, K., (2001).New Views of the Subsurface play concepts of oil exploration in Yemen. Oil and Gas Journal, V.99, No.23, pp36-47.

[34] Nilolenko V., and Goulyaeva M., (1985). Geological Structure and Petroleum Potential in PDRY. (Contract 55-189/13600), Moscow. (Unpublished report).

[35] Research Systems, 2001, FLAASH User's Guide, ENVI FLAASH version 1.0. Research Systems, Inc., pp. 58.

[36] Salisbury, J.W., L.S. Walter, N. Vergo and D.M. D'Aria, 1991. Infrared (2.1-25 micrometers) Spectra of Minerals O Johns Hopkins University Press, Baltimore.

[37] Clark, R.N., Swayze, G.A., 1991, Mapping with imaging spectrometer data simultaneously fit to multiple spectral features from multiple materials: in Proceedings of the Third Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) Workshop, JPL Publication 91-28,pp. 2-3.

[38] Clark, R.N., T.V.V. King, M. Klejwa, G. Swayze, and N. Vergo, (1990). High Spectral Resolution Reflectance Spectroscopy of Minerals: J. Geophys Res. 95, 12653-12680.

[39] Research Systems, 2002. ENVI User's Guide, ENVI 3.6. Research Systems, Boulder CO., pp. 1042.

Arafat Mohammed (1), K. Palanivel (2), C.J. Kumanan (3) and S.M. Ramasamy *

(1,2,3) Centre for Remote Sensing, Bharathidasan University Khajamalai Campus, Tiruchirappalli-23, Tamil Nadu, India

* Gandhigram Rural University, Dindigul, Tamil Nadu, India

Correspondence Author E-mail:
Table 1: Weighting methods and mineral type/
score of the extracted spectra.

Weighting Method (Score 0-1.0)

Library Spectrum/Mineral type SAM score SFF score BE score

1 Kaolinite-KL500 0.0 0.0 0.778
2 SideriteCOS2002 0.0 0.268 0.889
3 Calcite-CO2004 0.0 0.581 0.778
4 Alunite-AL705 0.0 0.497 0.556
5 Halloysite-KLH503 0.0 0.549 0.556
COPYRIGHT 2011 Research India Publications
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Mohammed, Arafat; Palanivel, K.; Kumanan, C.J.; Ramasamy, S.M.
Publication:International Journal of Petroleum Science and Technology
Date:Jan 1, 2011
Previous Article:Simulation study of surfactant flooding process in an Iranian fractured oil reservoir.
Next Article:The chemistry of chemical scale inhibitors and the mechanism of interactions with carbonate reservoir rock.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |