Mapping the Piute Mountains, California, with thermal ...eps different_1994_Hooketal... · JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 99, NO.B8, PAGES 15,605-15,622, AUGUST 10, 1994 Mapping

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 99, NO. B8, PAGES 15,605-15,622, AUGUST 10, 1994

Mapping the Piute Mountains, California, with thermal infrared multispectral scanner (TIMS) images

Simon J. Hook,' Karl E. Karlstrom,2 Calvin F. Miller,3 and Kenneth J. W. McCaffrey4.5

Abstract. Thermal infrared multispectral scanner (TIMS) data were acquired in 1990 Y over the Piute Mountains, California, to evaluate their usefulness for lithologic mapping

in an area of metamorphosed, structurally complex, igneous and sedimentary rocks. The data were calibrated and atmospherically corrected, and emissivity variations in

3 the form of alpha residuals were extracted from which color composite images were made. There was an excellent visual correlation between the units revealed in the color composite image and the lithologic units mapped in the eastern side of the area. It was also possible to correct, improve, and extend the recent map. For example, several areas mapped as granodioritic gneiss had TIMS alpha residual spectra consistent with mafic rocks and were subsequently mapped as amphibolite. The presence of a swarm of mafic dikes, of which only a few had previously been identified, was also revealed. The images also showed color variations in granitoid plutons that correlated with compositional variations previously determined by extensive field and geochemical work. The western Piute Mountains is an area that has proven to be especially difficult to map with field and air photographic methods, due to heterogeneous nature of certain units and various levels of desert varnish. The images permitted the extremely heterogeneous Proterozoic schists, gneisses, and granites to be easily mapped because the small-scale compositional variability was averaged to the TIMS pixel size (12 m x 12 m square). Areal measurements from the images show that the Proterozoic rocks in the Piute Mountains consist of granitoids (-50%), biotite gneiss (-15%), pelitic gneiss (-IS%), quartzite (-lo%), and amphibolite (-10%). TIMS data can dramatically increase the efficiency and the quality of geologic mapping in well-exposed heterogeneous areas with minimal vegetation cover where detailed mapping of lithologic contacts by traditional methods is unusually difficult.

Introduction

The majority of geologic remote sensing studies continue to focus on data from the 0.4- to 2.5-pm region, which includes the visible range between 0.4 and 0.7 pm. This is

1 due primarily to the availability of high Spatial resolution data from the Landsat and SPOT series of satellites over much of Earth's surface. These data are useful for lithological mapping. They are not ideal, however, because most spectral features that permit discrimination arise principally

9 from the presence of iron oxide minerals, hydroxyl-bearing minerals (chiefly clays), and carbonate minerals. Iron oxide

I and clay minerals typically arise through weathering and e only indirectly reflect the bulk composition of the rock.

' ~ e t Propulsion Laboratory, California Institute of Technology, Pasadena.

'Department of Geology, University of New Mexico, Albuquer-

In the thermal infrared region (8-12 pm), spectral variations typically relate to differences in the Si-0 bonding of silicate minerals. Multispectral thermal infrared images may provide a means for discriminating rocks based on their silicate mineralogy, which is an important criterion in some classification schemes for igneous [Streckeisen, 19761 and sedimentary rocks [Pettijohn et al., 19731.

Most of the limited number of studies that have evaluated the utility of multispectral thermal infrared data for lithological mapping have been in structurally simple, unmetamorphosed terranes, e.g., alluvial fans [Gillespie et al., 19841, lava flows [Abrams et al., 19911, hydrothermal alteration [Hook et al., 19921, sedimentary basins [Lung ef al., 19871, and igneous rocks [Lahren et al., 1988; Sabine et al., 19941.

Few studies have examined the use of these data for mapping structurally and lithologically complex areas of metamorphosed sedimentary and igneous rocks. This paper presents the results from a study undertaken to evaluate

que. thermal infrared multispectral images for lithological map- 3~epartment of Geology, Vanderbilt University, Nashville, Ten-

nessee. ping in a structurally complex area of metamorphic rocks, -- - - - - - - 4~ep-ent of ~ a ~ t h and Planetary Sciences, ~ o h n s Hopkins the Piute Mountains, California (Figure 1). The study area is

University, Baltimore, Maryland. particularly suited to this evaluation because a lithologically S ~ o ~ at the Department of Geology, Trinity College, University diverse suite of igneous, metamorphic, volcanic, and sedi- of Dublin, Ireland.

mentary rocks is well exposed in extensive, sparsely vege- This paper is not subject to U.S. copyright. Published in 1994 by the tated outcrops. These rocks provide an unusually complete American Geophysical Union. record from more than 1.7 Ga to the Cenozoic [Miller et al.,

I Paper number 94JB00690. 19821. The eastern side of the area was recently mapped in

15,605

HOOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNIA

Figure 1. Location of the study area. Centerline of TIMS flights also shown. Large box indicates area evaluated in this study. Smaller box indicates position of air ghotosraph (Plate 3).

detail [Karlstrom et al., 19931, while the western side of the pm, an instantaneous field of view of 2.5 mrad, and a total area has been difficult to map conventionally and is being field of view of 76.56'. The center positions, in micrometers, remapped now. of the six TIMS channels in this study were 8.407, 8.801,

The thermal infrared images utilized in this study were 9.204, 9.933, 10.703, and 11.625. Three flight lines of data acquired on July 27, 1990, with NASA's airborne thermal were obtained over the Old Woman-Piute Mountains. The infrared multispectral scanner (TIMS) under cloud-free con- locations of these flight lines and the area described in this ditions. TIMS has six channels located between 8 and 12 study are shown in Figure 1.

HOOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNLA

Theoretical Framework The reflectance and emissivity spectra of minerals exhibit

diagnostic features at various wavelengths which provide a means for their remote discrimination and identification. These features are produced by electronic or vibrational processes resulting from the interaction of electromagnetic energy with the atoms and molecules which compose the minerals that make up a rock. These different processes require different amounts of energy to proceed and therefore are manifest in different wavelength regions. Electronic processes require the most energy and result in spectral features in the visible and near-infrared wavelength regions. Fundamental vibrational processes require less energy and evidence for them occurs in the thermal infrared beyond 2.5 pm. Between 0.5 and 2.5 pm there is an overlap of features due to electronic processes and the excitation of overtone and combination-tone vibrations [Hunt, 19801.

Iron- hydroxyl-, water-, and carbonate-bearing minerals display reflectance spectral features in the 0.4- to 2.5-pm wavelength region. By contrast, emission spectra of silicate minerals, which compose most crustal rocks, exhibit spectral features in the thermal infrared between 8 and 12 pm (Figure 2). Collectively, these features compose the reststrahlen band. The emissivity minimum of the reststrahlen band occurs at relatively short wavelengths (8.5 pm) for framework silicates (quartz, feldspar) and progressively longer wavelengths for silicates having sheet, chain, and isolated Si04 tetrahedra [Hunt, 19801 (Figure 2). The only other feature in silicate mineral spectra between 8.5 and 12 pm results from the H-0-Al bond and occurs near 11 pm; it is characteristic of aluminum-bearing clay minerals [Hunt, 19801.

Other nonsilicate molecular units also give rise to spectral features in the thermal infrared. These include carbonates, sulphates, phosphates, oxides, and hydroxides, which typically occur in sedimentary and metamorphic rocks.

Geological Setting The Piute Mountains are in the eastern Mojave Desert,

east of the town of Essex. They are part of a relatively stable, unextended terrane between two moderately to highly extended terranes of mid-Tertiary age, the Colorado River extensional corridor [Howard and John, 19871 to the east and the Central Mojave Extensional complex [Dokka, 19861 to the west. Thus this has been a key area for unraveling the pre-Tertiary history of the eastern Mojave Desert [Miller et al., 19821. The Piute Mountains are bounded to the east by the NE-SW trending Little Piute fault (Figure 1). Immediately to the east of the Little Piute Mountains, buried by alluvium in the Ward Valley, is the inferred breakaway fault at the western margin of the Colo- rado River extensional corridor. To the west of the Piute Mountains is a little studied area which grades into the Central Mojave Extensional complex. The Old Woman Mountains, a prominent 40-km-long north trending range (Figure I), are just south of the Piute Mountains. The Piute and Little Piute Mountains consist of Proterozoic su-

8 9 10 11 12 Wavelength

Figure 2. Laboratory hemispherical reflectance spectra converted to emissivity using Kirchoff s law (E = 1 - R) for several pure minerals and linear mixtures of the pure minerals. Q, quartz; A, albite; M, muscovite; B, biotite; MI, linear mixture of 50% Q and 50% A; M2, linear mixture of 50% Q and 50% M; M3 linear mixture of 50% Q and 50% B. Spectra are offset for clarity; solid horizontal lines on Y axis denote an emissivity interval of 0.45. Y intercept value of Q = 0.84, A = 0.98, M = 0.99, B = 0.92, M1 = 0.91, M2 = 0.91, M3 = 0.88. X axis denotes wavelength in micrometers.

,?

1987; Bender et al., 1990; Hileman et al., 1990; Miller et al., 1990; Karlstrom et al., 1993; Gerber et al., 19941. Thermo- chronologic studies indicate that the pre-Tertiary rocks were at midcrustal levels (1&20 km) in the Late Cretaceous, then ascended to the upper crust by the end of the Cretaceous [Foster et al., 19891. The area has remained in the upper crust (generally C5-6 km) since the early Tertiary [Hileman et al., 1990; Foster et al., 19911.

Early Proterozoic supracrustal rocks are the oldest rocks in the area (> 1.7 Ga). They include strongly deformed pelitic and quartzofeldspathic schists, gneisses, and quartzite with minor amphibolite [Fletcher and Karlstrom, 19901. These are often found interleaved within a group of granitic orthogneisses emplaced between 1710 and 1725 Ma [Wooden et al., 1988; Wooden and Miller, 19901. Later in the Proterozoic,

K-feldspar megacrystic granite to granodiorite [Wooden et al., 1988; Bender et al., 19901, two-mica granites dated at 1674 Ma [Fletcher and Karlstrom, 19901, and the 1419 + 7 Ma Barrel Springs Pluton, a megacrystic syenite and granite [Gleason et al., 19881. This package was intruded by scat- tered diabase dikes believed to be between 1.1 and 1.2 Ga based on similarities with dikes dated elsewhere [Burchj?el and Davis, 198 1 ; Fitzgibbon, 19881.

The Proterozoic sequence is unconformably overlain by a cratonal Paleozoic sequence of metasedimentary rocks. The base of this sequence consists of quartzite and schist, correlated with the Cambrian Wood Canyon, Zabriskie and Bright Angel Formations, which pass upward to a banded marble correlated with the lower part of the Cambrian Bonanza King Formation [Fletcher and Karlstrom, 19901. The banded marble is overlain by a thick sequence of carbonate rocks which includes a lower massive tan-grey dolomitic marble and upper coarse-grained white calcitic marble. The lower tan-grey marble is correlated with the Middle Cambrian Bonanza King Formation, the Upper Cambrian Nopah Formation, andlor the Devonian Sultan formation [Stone et al., 1983; Brown, 19841. The upper white marble has been correlated with the Mississipian Redwall Limestone [Brown, 1984; Fletcher and Karlstrom, 19901.

During Mesozoic time the region underwent thrust fault- ing, magmatism, and metamorphism [Miller et al., 1982; Howard et al., 1987; Foster et al., 19921. The grade of Mesozoic metamorphism ranges from greenschist facies (<450°C) in the Piute Mountains to the upper sillimanite zone of the low pressure amphibolite facies (>650°C) in the Old Woman Mountains, with pressures ranging from 2.5 to 4 kbar [Hoisch et al., 1988; Foster et al., 19921. Two plutons were intruded into the Proterozoic and Paleozoic rocks of the study area during the Late Cretaceous: the 85 Ma East Piute pluton, which was emplaced syntectonically along a thrust fault [Karlstrom et al., 19931, and the Lazy Daisy Pluton, a 72 & Ma two-mica granite that crystallized after the peak deformation and induced the highest metamorphic grades in the region along its margin [Hoisch et al., 1988; Fletcher and Karlstrom, 1990; Foster et al., 1992; Kingsbury et al., 19941.

The Proterozoic, Paleozoic, and Mesozoic rocks are unconformably overlain by a sequence of unmetamorphosed Tertiary volcanic and sedimentary rocks. The lower part of this sequence is dominated by siliciclastic sediments, basalt, and basaltic andesite [Hileman et al., 19901. This is capped by the 18.5- Ma Peach Springs Tuff, which is highly variable in thickness [Hileman et al., 1990; Miller and Miller, 19911. The Peach Springs Tuff is overlain by unconsolidated Qua- ternary alluvium in the Piute Mountains. East-southeast trending, near-vertical dikes ranging from basalt to rhyolite in mineralogy are abundant in the eastern portion of the Piute Mountains [Hileman et al., 19901.

Data Processing and Analysis

values for the atmospheric correction based on an input atmospheric profile, which may be obtained from default profiles in LOWTRAN 7, or the profile may be modified or replaced with local atmospheric data. In this study, no local atmospheric data were available, and the default midlatitude winter profile was used. The winter profile was used rather than.the summer profile because the latter is far moister than would be expected for a desert area in the summer. The midlatitude winter profile contains 27% of the total water vapor of the midlatitude summer profile in the first 13 km. Atmospheric correction of TIMS data is discussed in detail by Hook et al. [1992].

After atmospheric correction the ground radiance values are a function only of surface temperature and emissivity. In this study we were interested in variations in surface emissivity, because these relate to differences in mineralogy. The emissivity information was extracted with the alpha residual technique [Hook et al., 1992; Kealy and Hook, 19931. The resultant alpha residual spectra have a shape similar to emissivity spectra; however, the mean of each spectrum is zero. This method was chosen over other methods since emissivity variations can examined in all six TIMS channels (most other methods force the emissivity to a constant in one channel), and the method is less susceptible to noise than the reference channel and emissivity normalization techniques [Kealy and Hook, 19931.

Two techniques were used to analyze the alpha residual data. First, color composite images were produced in which the alpha residual data from TIMS channels 1,2, and 3 were displayed in red, green, and blue, respectively. This color combination was selected because the majority of the rocks in the area have felsic mineralogies and therefore show the greatest variation in emissivity in TIMS channels 1 4 . A statistical technique that lumps the greatest variation from all six channels into three channels such as principal com- ponents analysis was not used because it is difficult to relate color variations in the principal component images to known physical variations in emissivity. By contrast the color of an area in the alpha residual images can be readily related to the alpha residual values of the surface in each channel using the standard principles of color additive mixing [Lillesand and Kieffer, 19791. It should be noted that in order to create an image with good contrast, the alpha residual data from each TIMS channel used for display were independently linearly scaled to the byte range (G255). The minimum and maxi- mum values for each channel were taken as the mean minus 3 standard deviations (s.d.) and the mean plus 3 s.d., respectively (a 3-sigma stretch).

In the second technique used to analyze the alpha residual data, spectra were extracted from each unit that could be discriminated in the images. These were compared with high-resolution laboratory spectra obtained from field samples collected from each unit. The image spectra were extracted from the unscaled alpha residual data. The samples were not always from the exact location of the image spectra due to access problems. The spectra are presented in

The TIMS instrument measures the energy radiated from three Figures 3a, 3b, and 6. Each figure consists of three the surface and modified by the atmosphere in six discrete panels. The left panel shows the high-resolution laboratory channels located between 8 and 12 pm. Initially, the TIMS hemispherical reflectance spectra converted to emissivity data were calibrated to radiance at the sensor [Palluconi and with KirchhoFs law ( E = 1 - R). The middle panel shows Meeks, 19851, then the atmospheric component to the total the laboratory emissivity spectra convolved to the TIMS radiance was removed using the LOWTRAN 7 radiative wavelengths and converted to alpha residuals. The right


8 9 10 11 12 8 9 I 0 11. I 2 8 Q 10 11 12 Wavelength Wavelength Wavelength

Figure 3a. (left) Laboratory hemispherical reflectance spectra converted to emissivity using KirchofPs law ( E = 1 - R) of field samples collected from areas representative of those used for extracting image spectra. Areas used for extracting image spectra are labeled A-G in Plate 1. (middle) Spectra in left column convolved to TIMS filter functions and converted to alpha residuals. (right) Average TIMS alpha residual spectra extracted from areas labeled A-43 on Plate 1. Each three line set indicates the mean +1 s.d. Number of spectra used to produce average spectra were A, 12; B, 9; C, 8;6D, 64; E, 3 6 F, 4; G, 36. Geologic units for each of the sites are A, Precambrian pelitic and aluminous psammitic metasediments (schists and paragneisses); B, Precambrian amphibolite; C, Precambrian quartzite; D, PrecarntPrian orthogneiss; E, Precambrian biotite granite; F, Precambrian NE-SW amphibolite dikes; G, Palaeozoic quartzite. Spectra are offset for clarity; solid horizontal lines on Y axis of lei? column denote an emissivity interval of 0.35. Y intercept values of left column spectra are A = 0.90, C = 0.85, D = 0.94, E = 0.95, F = 0.97, G = 0.83. Solid horizontal lines on middle and left columns indicate an alpha residual interval of 2. Y axis intercept values for center column are A = -0.3, C = -1.3, D = -0.17, E = 0.28, F = 0.84, G = -1.09. Y axis intercept value of right column spectra are A = -0.25, B = -0.03, C = -0.32, D = 0.02, E = -0.04, F = 0.11, G = -0.3. X axis denotes wavelength in micrometers. No sam~les were collected from locality B.

panel shows the image alpha residual spectra. Each image spectrum presented represents the mean r 1 s.d. for the group of pixels extracted from each unit. The number of pixels extracted depended on the size of the unit that could be discriminated on the images and is noted in the figure caption. The mean +1 s.d. is shown in order to provide an indication of the homogeneity of the spectra from each locality. The laboratory TIMS-convolved alpha residual spectra are plotted on the same scale as the image alpha residual spectra to help facilitate comparison. Generally, the

image alpha residual spectra show less contrast than the laboratory spectra presumably due to soil development. Henceforth the subscripts L, M, and R are used to refer to the spectra from the left, middle, and right panels, respectively, for each unit. For example, spectrum CL, Figure 3a, is the laboratory emissivity spectrum from a field sample of quartzite; this is similar to the emissivity spectrum of pure quartz (Figure 2). Both spectra show the well-developed doublet that indicates the surface consists of silicon and oxygen bonded together in a framework silicate. Figure 3a

r- - v

15.61 0 HOOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNIA

8 9 10 11 12 8 9 10 11 12 8 g 10 11 12 Wavelength Wavelength Wavelength

Figure 3b. (left) Laboratory hemispherical reflectance spectra converted to emissivity using Kirchoff s law ( E = 1 - R) of field samples collected from areas representative of those used for extracting image spectra. Areas used for extracting image spectra are labeled H-N in Plate 1. (middle) Spectra in left column convolved to TIMS filter functions and converted to alpha residuals. (right) Average TIMS alpha residual spectra extracted from areas labeled H-N on Plate 1. Each three line set indicates the mean & 1 s.d. Number of spectra used to produce average spectra were H, 50; I, 100; J, 64; K, 100; L, 16; M, 6; N, 25. Geologic units for each of the sites are H, Palaeozoic pelitic schist; I, Palaeozoic carbonate; J, Cretaceous granodiorite (east Piute pluton); K, Cretaceous two mica granite (Lazy Daisy pluton); L, Tertiary basalt rich conglomerate; M, Tertiary tuff (lower); N, Tertiary tuff (upper). Spectra are offset for clarity; solid horizontal lines on Y axis of left column denote an emissivity interval of 0.35. Y intercept values of left column spectra are H = 0.90, I = 0.95, K = 0.95, L = 0.96, N = O.%. Solid horizontal lines on middle and left columns indicate an alpha residual interval of 2. Y axis intercept values for center column are H = -0.03, I = 0.12, K = -0.17, L = 0.19, N = -0.1. Y axis intercept values for right column are H = -0.04, I = -0.03, J = -0,01, K = 0.05, L = 0.02, M = 0.01, N = -0.07. X axis denotes wavelength in micrometers. No samples were collected from localities J and M.

also shows the laboratory spectrum convolved to the TIMS wavelengths and converted to an alpha residual spectrum and the image alpha residual spectrum from an area of quartzite (spectra CM and CR, Figure 3a). Both spectra have low emissivity values in the wavelength range covered by the first three TIMS channels and higher values in the wavelength range covered by the last three TIMS channels. At the resolution of TIMS it is only possible to say that the spectrum is consistent with quartz, rather than indicative of quartz. In addition, since most rocks are not monomineralic, their spectra result from a mixture of the individual spectra

of the minerals that make up the rock making it more difficult to identify the mineralogical cause of changes in the TIMS alpha residual spectra without additional high-resolution laboratory or field spectra.

Data Interpretation Initially, the interpretation of the TIMS data acquired over

the eastern side of the study area is discussed and contrasted with the recently completed geological map. This interpretation is then extended to the images acquired over the

HOOK ET AL. : MAPPING THE PIUTE MOUNTAINS, CALIFORNIA

western side of the study area and used to improve the existing geologic map. The spectra, geologic map, image, and interpretation map for the eastern side of the study area are shown in Figures 3a, 3b, and 4; Plate 1; and F i e 5, respectively. The spectra, geologic map, image, and interpretation map for the western side of the study area are shown in Figures 6 and 7, Plate 2, and Figure 8, respectively. Plate 3 shows an air photograph for the western side of the study area. Modal analyses for the majority of units used in the spectral analysis are shown in Table 1.

Eastern Area A variety of Precambrian rocks are shown in the images

from the eastern side of the study area. These include deformed pelitic and mica-rich quartzofeldspathic schists and gneisses, amphibolite, quartzite, granitic orthogneiss, and the Fenner Gneiss, a relatively mafic granite with abundant K-spar phenocrysts. The pelitic and mica-rich quartzofeldspathic schists and gneisses appear orange in the images (Plate 1, locality A). The composition of this unit is very variable at the outcrop scale with well-developed garnet and staurolite in places. The sky-facing surfaces are dominated by muscovite since the unit readily cleaves along these surfaces. Examination of the image spectrum from locality A (Figure 3a, spectrum AR) indicates that the values in the first three channels decrease to a minimum in channel 3 with relatively high values in channels 1 and 2; hence the orange appearance in the images (Plate 1). The short-wavelength doublet of quartz is apparent in the laboratory spectrum for locality A, causing lower values in channel 2 than in channel 1 (Figure 3a, spectrum AL). The much lower values in channel 3 result from quartz and muscovite; the small shoulder at 9.5 pm in the laboratory spectrum results from the abundant muscovite in this unit. This effect can be seen in the mixture of the pure minerals quartz and muscovite (Figure 2, spectrum M2). It should be noted that the areas of quartz- and biotite-rich schists do not appear orange, which suggests the orange areas are fairly aluminum-rich. Areas of amphibolite appear white in the images and have high values in channels 1, 2, and 3 (Figure 3a, spectrum BR; Plate 1, locality B). The image spectrum from locality B has low values in channel 4 consistent with a more maiic composition. The Precambrian quartzite appears very dark brown to black (Plate 1, locality C). The image spectrum from this area has very low values in TIMS channels 1-3 and high values in TIMS channels 4-6 typical of quartz-rich units (Figure 3a, spectrum CR). The quartz doublet is clearly apparent in the laboratory spectrum (Figure 3a, spectrum CL). The areas of granitic orthogneiss appear red or pink (Plate 1, locality D). The image spectrum from locality D has a high values in channel 1 and low emissivity values in channels 2, 3, and 4 ( F i r e 3a, spectrum DR). The short- wavelength doublet of quartz is apparent in the laboratory spectrum, but generally, the spectrum shows the broad emission minimum characteristic of quartz, feldspar, and mica mixtures. By contrast, the Fenner Gneiss appears turquoise (Plate 1, locality E). The image spectrum has similar values in channels 1 and 2 and lower values in channels 3 and 4 (Figure 3a, spectrum ER). The lower values in channel 4 probably result from biotite. This unit is unusually rich in biotite for a granite (Table 1). The shift to longer wavelengths in the minimum is clearly apparent in the

artificial mixture of quartz and biotite (Figure 1, spectrum M3). Biotite is clearly not the only mineral present in the rock; however, it is typically exposed on the sky-facing surfaces since any biotite-rich layer provides a weak inter- face along which the rock readily splits. The Fenner Gneiss spectrum is not typical of a granite (see Figure 3a, spectra D; Plate 1, locality D). The Fenner Gneiss is cut by a series of northeast-southwest trending mafic dikes which appear white on the images (Figure 3a, locality F). The standard deviation of this spectrum is slightly larger than those of other units. This was because it was difficult to find a group of homogenous pixels, since the dike outcrops are small compared to the pixel size. The laboratory spectrum for this locality shows a well-developed minimum around 10 pm in the region of TIMS channel 4 consistent with a mafic mineralogy.

The Precambrian sequence is overlain by a Paleozoic sequence of quartzite, pelitic schists and carbonate (Plate 1,

' localities G, H, and I). The quartzite appears blackdin the color composite image (Plate 1, locality G), and it is diacult to distinguish its spectrum from that of the older Precam- brian quartzite (Figure 3a, spectra GR and CR), just as it is difficult to distinguish the Precambrian supracrustal amphibolite from the crosscutting mafic dikes (Figure 3a, spectra BR and FR). This indicates that any mineralogical variations which would manifest as differences in the alpha residual data are insufficient at the s~ectral resolution of TIMS to permit discrimination of the two quartzites and two mafic units. In addition, in places the areal extent of the quartzite on the geologic maps appears far less than the area on the images (Figure 4 and Plate 1). This is caused by scree from the quartzite obscuring other underlying units; care should b e taken in estimating the abundance of a particular unit based on the images alone.

The quartzite is overlain by a thin pelitic schist (Bright Angel Formation) that appears bright orange in the images (Plate 1, locality H) in spite of its very limited outcrop extent with respect to the TIMS pixel sizes. The orange color of the Bright Angel Schist is difficult to distinguish from similarly colored Precambrian pelitic rocks; however, the well- developed feature at 9.5 pm in its laboratory spectrum indicates that it is more micaceous (Figures 3a and 3b, spectra AL and HL). The quartzite-schist sequence is overlain by a carbonate sequence consisting of a lower banded marble, middle dolomitic marble, and upper massive calcitic marble. The entire carbonate sequence appears blue on the TIMS images (Plate 1, locality I). The blue appearance of the carbonate results from its having high emissivity values in TIMS channel 3 compared' to other materials in the scene (Figure 3b, spectrum IR). The diagnostic feature of calcite is a small sharp feature around 11.2 pm seen in the laboratory spectrum (Figure 3b, spectrum IL) associated with C-0 bonding. The wavelength minimum of this feature moves slightly as the mineralogy changes from calcite to dolomite, but TIMS does not have sufficient spectral resolution to detect this shift.

During Mesozoic time the Precambrian and Paleozoic sequence was metamorphosed and then intruded in the Late Cretaceous by two granitoids, the East Piute Pluton and the Lazy Daisy Pluton (Plate 1, localities J and K). The image spectra from these areas are very similar to the Precambrian granite image spectrum from locality D (Figures 3a and 3b, spectra DR, JR, and KR). Parts of the western margin of the


QUATERNARY DEPOSiTS

PEACH SPRINGS TUFF

PEQMATITEILEUCOCRATIC GRANITE MONZOORANITE

MAIN UNIT- BIOTITE GRANITE 340 42' M"

LOWER PALAEOZOIC ROCKS

MARC DYKES ,680 f 7 Ma

FENNER QNElSS 1683 * 5 Ma

PROTER,OZOlC SUPRACRUSTAL SCHIST AND GNEISS

a50 SHEAR ZONE

0 TERTIARY FAULT

--- DIRT ROAD

Figure 4. Geology of the eastern side of the study area based on work by Karlstrom et al. [1993].

HOOK ET AL.: MAPPING THE PLUTE MOUNTAINS, CALIFORNIA 15,613

Plate 1. Color composite image for the eastern side of the study area created by displaying the alpha residual data from TIMS channels 1, 2, and 3 in red, green, and blue, respectively. Labels A-N indicate regions used to extract average spectra shown in Figure 3. Label 0 identifies additional areas described in text.

. . l L r . , - 8

OOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNIA

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + ' & * & A L + +

',+++++,-EAST PIUTEb++++ +++++++- . PLUTPN . +++++

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

COLORlGEOLOGlC KEY

I[[g GREY TERTIARY ROCKS UNDIVIDED" LIGHT BLUE "PALAEOZOIC MARBLE" BLACK "WARTIITE SUPRACRUSTAL

AND CAMBRIAN" WHITE "MAFIC DVKES"

ea BLUE GREEN "FENNER GNEISS- RED "LEUCOCRATIC MESOZOIC

GRANITES" PINK "PRECAMBRIAN AND MESOZOIC

GRANITIC ROCKS" ORANGE "PROTEROZOIC AND CAMBRIAN

PHYLLITElSCHISTIQNEISS" --- DIRT ROAD

Figure 5. Mineralogical map of the eastern side of the study area interpreted from the TIMS alpha residual image (Plate 1) and the TIMS alpha residual spectra (Figures 3a and -3b).


8 9 10 11 12 8 9 10 11 12 8 g 10 11 12 Wavelength Wavelength Wavelength

Figure 6. (left) Laboratory hemispherical reflectance spectra converted to emissivity using Kirchoff s law ( E = 1 - R) of field samples collected from areas representative of those used for extracting image spectra. Areas used for extracting image spectra are labeled A-E in Plate 2. (middle) Spectra in left column convolved to TIMS filter functions and converted to alpha residuals. (right) Average TIMS alpha residual spectra extracted from areas labeled A-E on Plate 2. Each three-line set in the right column indicates the mean +1 s.d. Number of spectra used to produce average spectra were A, 100; B, 100; C, 25; D, 16; E, 36. Geologic units for each of the sites are A, Precambrian quartz syenite (Barrel Springs); C, Precambrian sediments; D, Precambrian quartz augen gneiss; E, Precambrian quartz augen gneiss. Spectra are offset for clarity, solid horizontal lines on Y axis of left column denote an emissivity interval of 0.35. Y intercept values of left column spectra are A = 0.97, D = 0.94, E = 0.95. Solid horizontal lines on middle and left columns indicate an alpha residual interval of 2. Y axis intercept values for center column are A = 0.14, D = 0.21, E = 0.19. Yaxisintercept valuesforright column areA = 0.01, C = -0.12, D = 0.04, E = -0.1. X axis denotes wavelength in micrometers. No samples were collected at localities B and C.

East Piute Pluton are extremely felsic due to fractionation [Karlstrom et al., 19931, and these appear much redder on the images (locality 0, Plate 1). No samples were collected from this locality, and this increase in red was not attributed to a particular mineral, but it probably results from a decrease in quartz and increase in feldspar. This subtle difference was defined by geochemical studies and is important in documenting the emplacement history of the pluton. Sabine er al. [I9941 also demonstrate the use of TIMS data for mapping mineralogical dBerences in granitoid plutons.

The Precambrian, Paleozoic, and Mesozoic rocks are unconformably overlain by Tertiary volcanic and sedimentary rocks. The lower part of the sequence is dominated by

basalt- and basaltic-andesite-rich conglomerates and basaltic andesite, both of which appear greylwhite on the images (Plate 1, locality L). The image spectra from this lithology are similar to the amphibolite and mafic dike spectra (Figures 3a and 3b, spectra BR, FR, and LR), consistent with their similar mineralogies. The basalt-rich conglomerates are overlain by the Peach Springs tuff, which has a maroon color at its base and purple above in the images (Figure 3b, spectra MR and NR; Plate 1, localities M and N). Generally, the tuff becomes more altered and devitrified upwards and the clear difference in color suggests that TIMS data may be useful in deciphering subtle internal variations within single ash flow units.

, tm- - r - - - - -

I&E6 HOOK ET AL.: MAPPING THE PIUTE.MOLJ~AINS, CALIFORNIA

[7 QUATERNARY DEWSITS

PEACH SPRINGS TUFF

TERTIARY ROCKS UNDIVIDED

LAZY DAISY QRANITE 72 * 2 Ma

UPPER PALAEOZOIC ROCKS

LOWER PALAEOZOIC ROCKS

BARREL SPRINGS QUARTZ SYENITE

BARREL SPRINQS MAFIC SYENITE

BARREL SPRINGS K-SPAR PORPHYRY

QRANITIC ROCKS 1790 - 1700 Ma

PROTEROZOIC SUPRACRUSTAL SCHIST AND QNEISS

SHEARZONE

/ TERTIARY FAULT

--- DIRT ROAD

Figure 7. Geology of the western side of the study area based on an unpublished compilation of maps from K. E. Karlstrom.

HOOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNIA 15,617

Plate 2. Color composite image for the western side of the study area created by displaying the TIMS alpha residual data from channels 1, 2, and 3 in red, green, and blue, respectively. Labels A-E indicate regions used to extract average pixel spectra shown in Figure 6. Label T indicates a river terrace.

15,618 HOOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNIA

APPROXIMATE SCALE

KILOMETERS

LOWGEOLOGIC KEY

PURPLEtRED "PEACH SPRINGS TUFF AND TERTIARY RHYOLITE"

GREY TERTIARY ROCKS"

LIGHT BLUE "PALAEOZOIC MARBLE

BLACK ',QUARTZITE SUPRACRUSTAL AND CAMBRIAN

PEACH "BARREL SPRINGS QUARTL SYENITE"

PEACIWGREY "BARREL SPRINGS K. SPAR PORPHRY

PEACWBLUE "BARREL SPRINGS MAFlC/FELSIC INJECTION ZONE

BLUE "PRECAMBRIAN BIOTITE RICH METASEDIMENTS"

WHITE "MAFIC DYKES''

BLUE GREEN TENNER AN0 BIOTITE RICH SUPRACRUSTAL GNEISS

PINK "PRECAMBRIAN AND MESOZOIC GRANITIC ROCKS"

ORANGE "PROTEROZOIC AND CAMBRIAN PELITlC PHYLLITEI SCHISTIGNEISS"

YELLOW "PROTEROZOIC SCHIST AND GNEISS (MODERATELY PELRIC

MAFICIFELSIC INJECTION ZO

Figure 8. Mineralogical map of the western side of the study area interpreted from the TIMS alpha residual image (Plate 2) and the TIMS alpha residual spectra (Figure 6). Some features shown here as rock are alluvium-Med washes, as discussed in the text.

Plate 3. Air photograph covering the southwest part of the study area taken simultaneously with the acquisition of the TIMS data.

Comparison of the geological map of the eastern side of the study area taken from [Karlstrom et al., 19931 with the image indicates that based on mineralogy, the alpha residual TIMS images permits excellent discrimination of the mapped geological units (Figure 4 and Plate 1). However, if two units have a similar bulk mineralogy but differ in texture or the presence of a minor constituent, it may not be possible to separate them on the images, for example, the Precambrian and Mesozoic granites or Precambrian and Paleozoic quartzites. Although there is good general agreement between the geological map and the images, there are some areas where the images led quickly to improvements of the map. For example, the area south of the East Piute pluton was mapped from traditional 1:24,000 air photographs, which show vari-

ations in brightness primarily due to iron-oxide differences that result from weathering and only indirectly relate to mineralogy, as opposed to the TIMS data that show variations in silicate mineralogy. The TIMS data show two large white areas (Plate 1, locality B) which had been mapped as Fenner Gneiss (Figure 4). However, their spectra are consistent with a more mafic mineralogy, typically amphibolite based on similar spectra from other areas. These units weather to a similar color to the Fenner Gneiss and were mapped from an air photograph; hence their confusion on the air photographs but not in the TIMS data. These areas were remapped as amphibolite; however, this has not been field checked. It is also far easier to distinguish the dike swarm on the TIMS images (Plate 1, locality F). The

15,620 HOOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNIA

Table 1. Modal Compositions for Rock Units Whose Spectra Are Shown in Figures 3a, 3b, and 6.

Rock Unit Modal Composition

Tertiary Peach Springs Tuff 10-20% sanidine; 1-3% biotite, hornblende, pyroxene, quartz, and sphene; 80-90% dass shards. devitrified glass. and bumice

Mesozoic leucocratic granite 30-35% 2535% potassium~elds'par; 5 4 0 % plagioclase feldspar; 5-10%

Mesozoic and Precambrian granites

Precambrian Fenner Gneiss

Precambrian Barrel Springs mafic

Precambrian Barrel Springs Kspporphyry

Precambrian Barrel Springs quartz syenite

Precambrian mafic dikes (amphibolite)

Precambrian biotite-rich metasediments

Cambrian (Bright Angel Schist) and Precambrian pelitic metasedihnents

muscovite 20-35% quartz; 15-50% potassium feldspar; 25-50% plagioclase feldspar; 0-15%

biotite and muscovite 15-30% quartz; 20-35% potassium feldspar; 25-35% plagioclase feldspar; 10-25%

biotite and hornblende 4 % quartz; 2WWo potassium feldspar; 26-75% biotite, hornblende, and accesso-

ries 15-20% quartz, 6&700/0 potassium feldspar; 10-15% plagioclase feldspar; 2-4%

biotite and accessories 515% quartz; 70-8070 potassium feldspar; 5-10% plagioclase feldspar; 5-15% biotite

and accessories 1-0 plagioclase feldspar; 5-0 hornblende and actinolite; 0-2Wo biotite, m

chlorite, epidote c~ino~yroxene, and accessories 20-5Wo quartz; 0-20% potassium feldspar; WO plagioclase feldspar; 2-0

biotite; 0-10% muscovite 10-5Wo quartz; 0-20% plagioclase feldspar; 1-0 biotite; 10-5Wo museovite,

staurolite, and garnet

presence of a dike swarm had been confirmed in the field. The weathering process in this area results in each unit being coated with desert varnish. The varnish also makes it difficult to distinguish units in the air photographs, therefore necessitating close inspection in the field. This is less of a problem in the thermal infrared since energy emitted by the sample is able to pass through the varnish at these wavelengths and the surface response is dominated by the underlying silicate mineralogy rather than the mineralogy of the varnish [Kahle, 19871 unless the varnish is sufficiently thick in which case the spectrum will relate to the varnish composition [Christensen and Harrison, 19931. Figure 5 shows the mineralogical map of the east part of the study area derived from the TIMS data.

The combined detailed field mapping and TIMS mineralogical map of the eastern area was then used to help produce a similar map in the less well mapped western areas.

Western Area Figure 7 and Plate 2 show the geological map and image of

the western side of the study area. There is some overlap between this image and the image from the eastern side of

C the area (Plate 1). Plate 3 shows an air photograph acquired over the study area simultaneously with the TIMS data. The location of the air photograph with respect to the study area is shown in Figure 1. While some of the mapped units can be distinguished due to differences in brightness in the air photograph, many more units can be discriminated in the TIMS image. For example, the quartzites, which appear black in the TIMS image and run N-S, are difficult to distinguish in the air photograph.

Several units in addition to those identified on the eastern side of the study area are apparent in the images. In the southwest part of the image is a large pink colored area associated with the Barrel Spring Pluton (Plate 2, locality A). The Barrel Spring Pluton has been subdivided into areas of quartz syenite, mafic syenite, K-feldspar porphyry, and porphyry dikes. It is not possible to distinguish the porphyry dikes from the quartz syenite, but it is just possible to separate the K-feldspar porphyry from the quartz syenite,

the former has a brown tint in the image (Plate 2, localities A and B). This further illustrates that TIMS is able to map subtle mineralogical differences in granitoids within a single pluton. The laboratory and image spectra from the quartz syenite are shown in Figure 6 (spectra AL, AM, AR). These are clearly different from the granitic spectra (Figures 3a and 3b, spectra D, J, and K). A mafic and felsic injection zone has been mapped in one of the northern areas of quartz syenite. In the images this zone extends farther south. It was not possible in the field to establish the difference between the southern part of this zone and the main quartz syenite. Farther south there is a dark blue unit which has been mapped as Precambrian metasedimentary rocks (Plate 2, locality C). To the east of the northern parts of the Barrel Spring pluton is a unit that is striped yellow or orange and turquoise in appearance. This unit becomes more turquoise farther east. It was originally mapped as granite augen- gneiss, but this analysis suggests the yellow and orange areas are muscovite-rich pelites (Figure 6, spectra D), whereas the turquoise areas are quartz- and biotite-rich (Figure 6, spectra E). This has been substantiated by field mapping. The broad pelitic yellow and orange colored bands versus the quartz- and biotite-rich turquoise-colored areas are readily defined at a 1:24,000 scale of the images but are difficult to map in the field due to the smaller-scale interfingering of the various units. Several areas were strongly mylonitized and mineralogically were quartz- and biotite-rich. The color of the quartz- and biotite-rich unit is similar to the Fenner Gneiss, which is also contains quartz and biotite. Farther east this unit passes into a black and orange unit that comprises the Precambrian quartzite and pelites/quartzofeldspathic schists and gneiss. The major bands of quartzite which form part of the interlayered sequence of quartz and pelite are very distinct in the images and were verified in the field. They had not been recognized in this area previously. This unit then passes into a red and purple colored unit interpreted to be the Precambrian granitic orthogneisses. In places the purple unit is clearly a river terrace, but mineralogically the surface of the terrace is not dissimilar to that of a granite (Plate 2,

I i

HOOK ET AL.: MAPPING THE PIUTE MOUNTAINS, CALIFORNIA 15,621

locality T). The terrace can easily be recognized on the air photograph.

A mineralogical map was derived from the TIMS data (Figure 8) which is clearly far more detailed than the existing geologic map for the area (Figure 7). This map was partially field-checked and showed excellent agreement with the actual geology. It is being incorporated into the latest geologic map under preparation for the western side of the study area and has proven particularly useful for mapping the Precambrian sediments, the extent of which were only

L very roughly known until this study. Analysis of the TIMS data suggests that the Proterozoic rocks in the Piute Moun- tains contain roughly equal proportions of granitoids and supracrustal gneisses. Further, the supracrustal gneisses

C contain roughly equal proportions of pelitic- and biotite-rich gneisses (-30%) with subordinate but still voluminous, clean quartzites (-10%) and amphibolites (-10%). The supracrustal package in the eastern area, in the hanging wall of the Fenner shear zone, is more amphibole-rich, whereas the western area (footwall) is more pelitic-rich. Thus the TIMS data are useful for quick estimates of the mineralogy of the Proterozoic rocks at the surface and may help distinguish tectonic blocks composed of different Proterozoic packages.

Summary and Conclusions

energy from a 12 m by 12 m pixel. For example, the images indicated the presence of large areas of Precambrian quartzite and pelitic gneiss in the western Piute Mountains and a few small areas in the eastern Piutes that had not been recognized until this study. Thus Proterozoic sedimentary sequences in this area of the Mojave were perhaps more mineralogically evolved (richer in quartz and aluminum) than many of the volcanic-rich Proterozoic supracrustal packages in central Arizona. The mineralogical map is being used in conjunction with standard field mapping techniques to produce a new geologic map and revise the geologic history for the western Piute Mountains. An additional approach to visual spectral analysis in areas where the wavelength of the minimum of the reststrahlen feature is expected to shift due to known compositional variations would be to fit a Gaussian function to the TIMS alpha residual spectrum and regress the calculated minimum against the chemical composition of field samljles. This approach has been developed by Sabine et al. [1994]; it would be a logical follow-on to this study in areas where a firm mineralogical basis now exists for separating two units, such as the Precambrian muscovite-rich versus biotite-rich sediments.

In conclusion, TIMS data are a very valuable tool for assisting in the mapping of structurally complex areas of metasedimentary and metaigneous rocks. They permit vari-

Thermal infrared multispectral scanner (TIMS) data were ations in the silicate mineralogy of the surface to be discrim- acquired over the Piute Mountains, California, in July 1990 inated and in most cases identified with fieldnaboratory to evaluate their use for lithological mapping in a structurally spectra. As these images become more readily available with complex area of metamorphosed sedimentary and igneous the launch of similar instruments on spacercraft, e.g., the rocks. The Piute Mountains were particularly well suited to advanced spaceborne thermal infrared reflectance radiometer, this study because of their extreme lithologic heterogeneity they should become a standard tool for mapping in complex and because the eastern side of the mountains was recently terrains with similar exposure to the Piute mountahs. mapped in moderate detail [Karlstrom et al., 19931 and the western side of the mountains is currently being remapped.

Ackhowldgments. The research described in this paper was The data were calibrated and corrected for atmO- camed out In part at the Jet Propulsion Laboratory, California spheric effects, and any emissivity variations were extracted Institute of Technoloev. under a contract with the National Aero- from them with the alvha residual technique. A color com- nautics and Space ~Ginistrat ion. Research sumort for Karlstrom

image was prodiced from the alpha residual data from and Miller W& provided by National science Poundation grants

TIMS channels 2, and and used to discriminate the EAR8904675 and EAR 8904320, respectively. Reference herein to any specific commefcial proauct, process, or service by trade

majority of lithological units recently mapped on the eastern names, trademark, manufacturer, m othemise does not imply side of the study area. Differences in the images were endorsement by the United States or the Jet Propulsion Laboratory, attributed to mineralogical variations and confirmed through California Institute of Technology. field checking and laboratory spectral analysis of field Sam-

I ples. Certain units were distinguished on the ground by the presence of a minor constituent or some other geologic

c property, such as relative age or deformation intensity; however, these could not be mapped in the TIMS data. This demonstrates that differences in silicate mineralogy can be

, i reliably mapped by TIMS regardless of fabric. Other additional units were identified which had been incorrectly mapped.

The mineralogical map derived from the eastern side of the study area was then used to assist in producing a mineralogical map from the TIMS data for the western side of the area. This map agreed with the existing geo16gic map and in addition permitted the discrimination of many more previously unmapped geologic units. The TIMS data were especially useful for mapping the metamorphosed Precambrian metasedimentary rocks which are particularly heterogeneous and difficult to distinguish on the field scale but appear as coherent, readily mappable packages in the TIMS data. This is due to the TIMS data providing an average of the

References Abrams, M. J., E. A. Abbott, and A. B. Kahle, Combined use of

visible, reflected infrared and thermal infrared images for mapping Hawaiian lava &ows, J. Geophys. Res., 96, 475-484, 1991.

Bender, E. E., C. F. Miller, and J. L. Wooden, The Fenner Gneiss and associated units: An Early Proterozoic composite batholith, Piute and Old Woman Mountains, CA (abstract), Geol. Soc. Am. Absfr. Programs, 22, 7, 1990.

Brown, H., Discussion: Correlation of metamorphosed Palaeozoic strata of the southeastern Mojave Desert Region, California and Arizona. Geol. Soc. Am. Bull., 95, 1482-1485, 1984.

Burchfiel, B. C., and G. A. Davis, Mojave and environs, in The Georectonic Development of California, Rubey Volume I , edited by W. G. Ernst, pp. 217-252, Prentice-Hall, Englewood Cliffs, N. J., 1981.

Christensen, P. R., and S. T. Hamson, Thermal-infrared emission spectroscopy of natural surfaces: Application to desert varnish coatings on rocks, J. Geophys. Res., 98, 19,819-19,834, 1993.

Dokka, R. K., Patterns and modes of early Miocene crustal extension, central Mojave Desert, California, in Extensional Tectonics of the Southwestern United States, edited by L. Mayer, Spec. Pap. Geol. Soc. Am., 208,75-96, 1986.


Fitzgibbon, T. T., Tectonic significance and characteristics of middle Proterozoic diabase sheets in southeastern California and Arizona (abstract), Geol. Soc. Am. Abstr. Programs, 20, 160, 1988.

Fletcher, J. M., and K. E. Karlstrom, Late Cretaceous ductile deformation, metamorphism and plutonism in the Piute Moun- tains, eastern Mojave Desert, J. Geophys. Res., 95, 487-500, 1990.

Foster, D. A., T. M. Harrison, and C. F. Miller, Age, inheritance and uplift history of the Old Woman-Piute Batholith, California and implications for K-feldspar age spectra, J. Geol., 97,232-243, 1989.

Foster, D. A., D. S. Miller, and C. F. Miller, Tertiary extension in the Old Woman Mountains area, California: Evidence from apa- tite fission track analysis, Tectonics, 10, 875-886, 1991.

Foster, D. A., C. F. Miller, T. M. Harrison, and T. D. Hoisch, Timing and character of metamorphism and tectonism in the Old Woman Mountains area, California: Evidence from 40Ar/34Ar thermochronology and thermobarometry, Geol. Soc. Am. Bull., 104, 176191, 1992.

Gerber, M. E., C. F. Miller, and J. L. Wooden, Plutonism at the interior of the Jurassic Magmatic Belt, Mojave Desert, CA, in Jurassic Magmatism and Tectonism in the Cordillera, edited by D. M. Miller and R. M. Tosdal, Spec. Pap. Geol. Soc. Am., in press, 1994.

Gillespie, A. R., A. B. Kahle, and F. D. Palluconi, Mapping alluvial fans in Death Valley, California, using multichannel thermal infrared images, Geophys. Res. Lett., 11, 1153-1156, 1984.

Gleason, J. D., C. F. Miller, and J. L. Wooden, Barrel Spring alkalic complex: 1.4 Ga anorogenic plutonism in the Old Woman-Piute Range, eastern Mojave Desert, California (abstract), Geol. Soc. Am. Abstr. Programs, 20, 164, 1988.

Hileman, G. E., C. F. Miller, and M. A. Knoll, Mid-Tertiary structural evolution of the Old Woman Mountain area: Implica-

/ tions for crustal extension across southeastern California, J. Geophys. Res., 95,581-597, 1990.

Hoisch, T. D., C. F. Miller, M. T. Hiezler, T. M. Harrison, and E. F. Stoddard, Late Cretaceous regional metamorphism in southeastern California, in Metamorphism and Crustal Evolution of the Western United States, Rubey Volume VZZ, edited by W. G. Ernst, pp. 538-571, Prentice-Hall, Englewood Cliffs, N. J., 1988.

Hook, S. J., A. R. Gabell, A. A. Green, and P. S. Kealy, A comparison of techniques for extracting emissivity information from thermal infrared data for geologic studies, Remote Sens. Environ., 42, 123-135, 1992.

Howard, K. A., and B. E. John, Crustal extension along a rooted system of imbricate low-angle faults, in Colorado River Exten- sional Corridor, California and Arizona, edited by M. P. Coward, J. F. Dewey, and P. L. Hancock, Spec. Publ. Geol. Soc. London, 28,299-31 1, 1987.

Howard, K. A., B. E. John, and C. F. Miller, Metamorphic core complexes, Mesozoic ductile thrusts, and Cenozoic detachments: Old Woman Mountains-Chemehuevi Mountains Transect, Cali- fornia and Arizona, in Geologic Diversity of Arizona and Its Margins: Excursions to Choice Areas edited by G. J. Davis and E. M. Vanderwolder, Spec. Pap. Ariz. Bur. Geol. Min. Technol., 5, 365-382, 1987.

Hunt, G. R., Electromagnetic radiation: The communication link in remote sensing, in Remote Sensing in Geology, edited by B. S. Siegal and A. R. Gillespie, pp. 545, John Wiley, New York, 1980.

Kahle, A. B., Surface emittance, temperature, and thermal inertia derived from thermal infrared multispectral scanner (TIMS) data for Death Valley, California, Geophysics, 52, 858-874, 1987.

Karlstrom, K. E., C. F. Miller, J. A. Kingsbury, and J. L. Wooden, Pluton emplacement along an active ductile thrust zone, Piute Mountains, southeastern California: Interaction between defor- mational and solidification processes, Geol. Soc. Am. Bull., 105, 213-230, 1993.

Kealy, P. S., and S. J. Hook, Separating temperature and emissivity in thermal infrared multispectral scanner data: Implications for

recovering land surface temperatures, Geosci. Remote Sens., 31, 1155-1164, 1993.

Kingsbury, J. A., C. F. Miller, J. L. Wooden, and T. M. Harrison, Monazite paragenesis and uranium-lead systematics in rocks of the eastern Mojave desert, California: Implications for thermo- chronometry, Chem. Geol., 110, 147-167, 1994.

Kneizys, F. X., E. P. Shettle, L. W. Abreu, J. H. Chetwynd Jr., G. P. Andersen, W. 0. Gallery, J. E. A. Selby, and S. A. Clough, Users guide to LOWTRAN 7, Environmental research paper 1010, Tech. Rep. AFGL-TR-88-0177, Air Force Gwphys. Lab., Bedford, Mass., 1988.

Lahren, M. M., R. A. Schweickert, and J. V. Taran&, Analysis of the northern Sierra accreted terrain, California, with airborne thermal infrared multispectral scanner data, Geology, 16, 525- 528, 1988.

Lang, H. R.. S. L. Adams, J. E. Conel, B. A. McGuffie, E. D. Paylor, and R. E. Walker, Multispectral remote sensing as strati- graphic and structural tool, Wind River Basin and Big Horn Basin areas, Wyoming, AAPG Bull., 71, 389403, 1987.

Lillesand, T. M., and R. W. Kieffer, Remote Sensing and Image Interpretation, John Wiley, New York, 1979.

Miller, C. F., K. A. Howard, and T. D. Hoisch, Mesozoic thrusting, metamorphism, and plutonism, Old-Woman-Piute Range,.southeastern California, in Mesozoic-Cenozoic Tectonic Evolution of the Colorado River, Region, California-Arizona-Nevada, edited by E. G. Frost and D. L. Martin, pp. 561-581, Cordilleran Publishers, San Diego, Calif., 1982.

Miller, C. F., J. L. Wooden, V. C. Bennett, J. E. Wright, G. C. Solomon, and R. W. Hurst, Petrogenesis of the composite pera- luminous-metaluminous Old Woman-Piute Range batholith, southeastern California: Isotopic constraints, in The Nature and Origin of Cordilleran Magmatism edited by J. L. Anderson, Mem. Geol. Soc. Am., 174,99-109, 1990.

Miller, J. S., and C. F. Miller, Tertiary extension-related volcanism, Old Woman Mountains area eastern Mojave desert, California, J. Geophys. Res., 96, 62943 , 1991.

Palluconi, F. D., and G. R. Meeks, Thermal infrared multispectral scanner (TIMS): An investigator's guide to TIMS data, JPL Publ., 85-32, 1985.

Pettijohn, F. J., P. E. Potter, and R. Siever, Sand and Sandstone, Springer-Verlag, New York, 1973.

Sabine, C., V. J. Realmuto, and J. V. Taranik, Quantitative estima- tion of granitoid composition from thermal infrared multispectral scanner (TIMS) data, Desolation Wilderness, northern Sierra Nevada, California, J. Geophys. Res., 99,42614271, 1994.

Stone, P., K. A. Howard, and W. Hamilton, Correlation of Palae- ozoic strata of the southeastern Mojave Desert region, California and Arizona, Geol. Soc. Am. Bull., 94, 1135-1147, 1983.

Streokeisen, A., To each plutonic rock its proper name, Earth Sci. Rev., 12, 1-33, 1976.

Wooden, J. L., and D. M. Miller, Chronologic and isotopic framework for Early Proterozoic crustal evolution in the eastern Mojave desert regions, SE California, J. Geophys. Res., 95, 133-146, 1990.

Wooden, J. L., D. M. Miller, and K. A. Howard, Early Proterozoic chronology of the eastern Mojave Desert (abstract), Geol. Soc. Am. Abstr. Programs, 20, 243, 1988.

S. J. Hook, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109. (e-mail: [email protected]. nasa.gov) 1

K. E. Karlstrom, Department of Geology, University of New Mexico, Albuquerque, NM 8713 1.

K. J. W. McCafTrey, Department of Geology, Trinity College, University of Dublin, Dublin, Ireland.

C. F. Miller, Department of Geology, Vanderbilt University, Nashville, TN 37235.

(Received September 27, 1993; revised February 14, 1994; accepted March 10, 1994.)

Mapping the Piute Mountains, California, with thermal ...eps different_1994_Hooketal... · JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 99, NO.B8, PAGES 15,605-15,622, AUGUST 10, 1994 Mapping

Documents