Utilizing Hyspiri Prototype Data for Geological Exploration Applications: a Southern California Case Study

geosciences

Article Utilizing HyspIRI Prototype Data for Geological Exploration Applications: A Southern California Case Study

Wendy M. Calvin * and Elizabeth L. Pace

Department of Geological Sciences and Engineering, University of Nevada, Reno, NV 89557, USA; [email protected] * Correspondence: [email protected]; Tel.: +1-775-784-1785

Academic Editors: Kevin Tansey, Stephen Grebby and Jesus Martinez-Frias Received: 4 January 2016; Accepted: 12 February 2016; Published: 24 February 2016

Abstract: The purpose of this study was to demonstrate the value of the proposed Hyperspectral Infrared Imager (HyspIRI) instrument for geological mapping applications. HyspIRI-like data were collected as part of the HyspIRI airborne campaign that covered large regions of California, USA, over multiple seasons. This work focused on a Southern California area, which encompasses Imperial Valley, the Salton Sea, the Orocopia Mountains, the Chocolate Mountains, and a variety of interesting geological phenomena including fumarole ﬁelds and sand dunes. We have mapped hydrothermal alteration, lithology and thermal anomalies, demonstrating the value of this type of data for future geologic exploration activities. We believe HyspIRI will be an important instrument for exploration geologists as data may be quickly manipulated and used for remote mapping of hydrothermal alteration minerals, lithology and temperature anomalies.

Keywords: HyspIRI; AVIRIS; MASTER; geology; remote sensing; southern California; geothermal; mineral exploration; thermal anomaly

1. Introduction Geologic exploration has taken advantage of satellite and airborne sensors to map structure, lithology and hydrothermal alteration for many decades (e.g., [1], and references therein). The launch of the Advanced Spaceborne Thermal Emission and Reflectance Radiometer (ASTER) [2] and Hyperion [3] instruments ushered in a new era of satellite remote sensing for geologic applications. ASTER has been widely used for lithologic and geologic studies (e.g., [4–8]), often in conjunction with airborne imaging spectrometer systems, such as the Advanced Visible/Infrared Imaging Specrometer (AVIRIS) or HyMap (e.g., [9,10]). Hyperion has received only limited attention for geologic applications, possibly due to the sparse global coverage [11] and the lower signal-to-noise ratio at wavelengths longer than 2.0 µm [12], that are critical for mineral mapping. An additional hurdle to using Hyperion data is that it is provided in radiance, necessitating additional calibration to obtain at-surface reflectance for geologic studies. Several efforts are currently underway to develop and launch the next generation of imaging spectrometer systems on satellite platforms for a wide range of Earth Observation goals. Many of these systems are confined to wavelengths less than approximately 1.0 µm, but several planned instruments include the full range up to 2.5 µm and will be useful for geologic exploration. Canada’s Hyperspectral Environment and Resource Observer (HERO) [13], Germany’s Environmental Mapping and Analysis Program (EnMAP) [14], Japan’s Hyperspectral Imager Suite (HISUI) [15], Italy’s (PRecusore IperSpettrale della Missione Operativa) PRISMA [16] and the USA’s Hyperspectral Infrared

Geosciences 2016, 6, 11; doi:10.3390/geosciences6010011 www.mdpi.com/journal/geosciences Geosciences 2016, 6, 11 2 of 14

Imager (HyspIRI) [17] are in varying stages of study and development. EnMAP, HISUI and PRISMA are slated for launch in 2017 or later. HyspIRI is the only mission concept that includes both shortwave and thermal infrared channels, similar to ASTER. HyspIRI was a mission recommended in the 2007 NASA Earth Science decadal survey, but has not yet been approved for a new mission start. The instrument would have 10 nm contiguous spectral channels from 0.38 to 2.5 µm (>200 channels) with a 30-m spatial resolution and a 16-day revisit rate. In addition, the HyspIRI instrument will have eight bands in the thermal infrared (TIR) from 4 to 13 µm with 60-m spatial resolution and a five-day revisit rate. Additional information on the mission architecture, planned on-board processing, and applications is provided by Lee et al. [17]. To demonstrate the effectiveness of the HyspIRI instrument for science applications NASA funded the HyspIRI airborne campaign whereby radiance data were collected via aircraft and used to generate Level 2 HyspIRI-like reflectance, emissivity, and land surface temperature data. Our funded investigation was to assess the utility of HyspIRI-like data for a variety of geologic applications including mineral and energy exploration, impacts of resource development, and natural hazards. The goal of this study was to assess the utility of several standard derived data products for geological assessments. For various environmental and ecosystem applications, the HyspIRI airborne campaign data were collected during spring, summer, and fall of 2013, 2014, and 2015. Data were collected over the same five regions (“boxes”) within the state of California, USA, each time: Southern California, Santa Barbara, San Francisco Bay, Yosemite, and Lake Tahoe. Both the AVIRIS and MODIS/ASTER Airborne Simulator (MASTER) instruments were flown at 20 km above ground level onboard NASA’s ER-2 aircraft [17]. AVIRIS collects 224 contiguous spectral channels from 0.4 to 2.5 µm [18] and MASTER collects 50 channels from 0.4 to 13 µm, though our study uses only the channels from 7.5 to 13 µm [19]. AVIRIS is considered a spectral analog for HyspIRI so no wavelength resampling is performed. AVIRIS collected Level 1 radiance data at 18 m resolution. Level 2 simulated reflectance data were generated by the AVIRIS team at 18- and 30-m resolutions, as described in Section3. Planned band centers for the seven HyspIRI TIR channels, from 7 to 13, are offset from MASTER, and MASTER has additional channels at 9.6, 10.1, and 12.8 (Table1). We did not resample MASTER data to the planned HyspIRI wavelengths, but the temperature-emissivity separation (TES) product generated by the team includes only five channels of which band centers roughly correspond to those planned for HyspIRI (Table1). MASTER collected Level 1 radiance data at 60-m resolution, which were converted to Level 2 temperature and emissivity products as described in Section3. Data were generally collected during cloud-free days at high sun-angle hours, but data were also collected over the Southern California box during two nighttime flights as per special request. The focus area for this study is shown in Figure1; it includes the southeast end of the Southern California data collection box. This region includes the Salton Sea, Imperial Valley, the Orocopia Mountains, and the Chocolate Mountains, which are just north of the US–Mexico border.

Table 1. Band centers in µm for the TIR region for HyspIRI, ASTER, MASTER.

HyspIRI [17] ASTER Channel [6] ASTER Wavelength [6] MASTER 1 MASTER-TES Processed 7.35 – – – – – – – 7.8 – 8.28 10 8.3 8.18 – 8.63 11 8.65 8.62 8.6 9.07 12 9.1 9.06 9.03 – – – 9.67 – – – – 10.10 – 10.53 13 10.6 10.59 10.63 11.33 14 11.3 11.33 11.32 12.05 – – 12.12 12.1 – – – 12.86 – 1: MASTER as ﬂown on 9/24/2013 from the on-line calibration ﬁles available on line [19]. Geosciences 2016, 6, 11 3 of 14 Geosciences 2016, 6, 11 3 of 14

Figure 1. ImperialImperial Valley, Valley, California, USA, study area. A true color image derived from from 18-m 18-m AVIRIS AVIRIS data is shown overlain on a topographic shaded reliefrelief image. Known geothermal resource areas or power plantsplants areare indicatedindicated by by initials: initials: HMS—Hot HMS—Hot Mineral Mineral Spa; Spa; SS—Salton SS—Salton Sea; Sea; B—Brawley; B—Brawley; EM—East EM— EastMesa; Mesa; H—Heber. H—Heber. The insetThe inset on upper on upper right right map map shows shows the location the location of the of study the study area. area.

2. Geologic Geologic Setting Setting The Southern California box covers land from the US–Mexico border to the Los Angeles Basin but for this study we focused only on the part from the border to the north shore of the Salton Sea (Figure1 1).). WeWe chosechose thisthis areaarea toto studystudy becausebecause ofof thethe varietyvariety inin landland surfacesurface covercover andand thethe abundanceabundance of interesting interesting geological geological phenomena, phenomena, including including dive diverserse ages ages and and lithologies lithologies of bedrock of bedrock exposures, exposures, an activean active open open pit pitgold gold mine, mine, many many geothermal geothermal ener energygy production production sites, sites, active active fumarolic fumarolic and and gas emissions from current hydrothermal activity, recentrecent volcanic eruptions and the potential for both natural and inducedinduced seismicseismic hazards.hazards. AA geologic geologic map map generalized generalized from from Jennings Jenningset et al. al.[20 [20]] is shownis shown in inFigure Figure2. 2. The northwest-trending Chocolate Chocolate Mountains Mountains mark mark the the eastern eastern boundary boundary of of Imperial Imperial Valley. Valley. The rangerange is is predominantly predominantly composed composed of Precambrian of Precambrian igneous igneous and metamorphic and metamorphic rocks, Pre-Cretaceous rocks, Pre- Cretaceousmetasedimentary metasedimentary rocks, Mesozoic rocks, granitic Mesozoic rocks, granit andic rocks, Tertiary and volcanic Tertiary rocks volcanic [21]. rocks The Chocolate [21]. The ChocolateMountains Mountains contain the contain Chocolate the Chocolate Mountain Mountain Aerial Gunnery AerialRange, Gunnery which Range, is used which by is the used Navy by andthe NavyMarines and and Marines not accessible and not to accessible the public, to butthe ispublic, an area but of is potential an area geothermal of potential energy geothermal development energy developmentby the US Navy by the [22 ].US New Navy Gold [22]. Inc.’s New MesquiteGold Inc.’s Mine Mesquite sits at Mine the southeastsits at the cornersoutheast of thecorner Southern of the SouthernCalifornia California box. The fault-controlled box. The fault-controlled epithermal goldepithermal deposit gold is hosted deposit within is hosted Mesozoic within gneisses Mesozoic where gneissesgold occurs where both gold disseminated occurs both and disseminated in veins [23 ].and The in region veins [23]. around The this region open around pit mine this was open the pit focus mine of wasa remote the focus sensing of studya remote conducted sensing by study Zhang conductedet al. [8] who by usedZhang ASTER et al.data [8] who to remotely used ASTER map alteration data to remotelymineralogy, map speciﬁcally alteration alunite,mineralogy, kaolinite, specifically montmorillonite, alunite, kaolinite, and muscovite. montmorillonite, and muscovite.

Geosciences 2016, 6, 11 4 of 14 Geosciences 2016, 6, 11 4 of 14

FigureFigure 2. 2.Orocopia Orocopia MountainsMountains and Chocolate Mountains geologic map map showing showing major major lithologic lithologic units units fromfrom Jenings Jeningset et al. al.[ 20[20].].

ToTo the the northwest northwest of of thethe ChocolateChocolate MountainsMountains are the Orocopia Mountains, Mountains, composed composed of of the the Late Late Cretaceous-EarlyCretaceous-Early Cenozoic Cenozoic OrocopiaOrocopia Schist.Schist. The OrocopiaOrocopia Schist is is particularly particularly interesting interesting because because it it waswas subducted subducted atat aa lowlow angleangle beneathbeneath thethe NorthNorth American plate during during the the Laramide Laramide Orogeny Orogeny and and exhumedexhumed during during middlemiddle CenozoicCenozoic extensionextension of the region [24]. [24]. Additionally Additionally of of interest interest in in the the region region areare the the Algodones Algodones Dunes, Dunes, a 70 a km-long70 km-long sand sand dune dune field thatfield occupies that occupies the southeastern the southeastern part of Imperialpart of Imperial Valley. Algodones dune sand is likely derived from the beaches of ancient Lake Cahuilla, a Valley. Algodones dune sand is likely derived from the beaches of ancient Lake Cahuilla, a lake that lake that formed in the geologic past whenever the Colorado River flowed into Imperial Valley formed in the geologic past whenever the Colorado River flowed into Imperial Valley instead of the instead of the Gulf of California [25]. The Salton Sea now occupies part of the region once filled by Gulf of California [25]. The Salton Sea now occupies part of the region once filled by Lake Cahuilla. Lake Cahuilla. Imperial Valley is one of the most prolific geothermal power production locations in California. Imperial Valley is one of the most prolific geothermal power production locations in California. Three electricity-producing geothermal fields are covered by the Southern California box: Salton Three electricity-producing geothermal fields are covered by the Southern California box: Salton Sea, Sea, East Mesa, and North Brawley (Figure1). Several other geothermal fields exist in the region East Mesa, and North Brawley (Figure 1). Several other geothermal fields exist in the region and are and are in various stages of use and development. Hot springs line the northwestern front of the in various stages of use and development. Hot springs line the northwestern front of the Chocolate Chocolate Mountains; some are used for spas and hot spring resorts at the Hot Mineral Spa area. Mountains; some are used for spas and hot spring resorts at the Hot Mineral Spa area. Surface Surface indicators of the geothermal activity near the south shore of the Salton Sea include the indicators of the geothermal activity near the south shore of the Salton Sea include the Davis- Davis-Schrimpf field where hot water seeps from the surface to form muddy pools and gryphons. Schrimpf field where hot water seeps from the surface to form muddy pools and gryphons. The The Imperial CO2 field is located to the north of these mudpots, where so much CO2 is emitted that Imperial CO2 field is located to the north of these mudpots, where so much CO2 is emitted that a dry aice dry production ice production facility facility was wasoperated operated from from 1933 1933to 1954. to 1954. As the As Salton the Salton Sea Seawater water level level continues continues to todrop, drop, more more mudpots, mudpots, gryphons, gryphons, and and fumaroles fumaroles are arebeing being exposed, exposed, for example for example along along the edge the edge of the of thelake lake near near Mullet Mullet Island Island where where exposure exposure began began inin 2007 2007 [26]. [26 ].Lynch Lynch and and Hudnut Hudnut [27] [27 ]surveyed surveyed other other recentlyrecently exposed exposed geothermal geothermal features features within within the the Wister Wister Unit Unit of the of Imperialthe Imperial Wildlife Wildlife Area. Area. The Mullet The IslandMullet and Island Wister and fumaroleWister fumarole fields definefields define lineaments lineaments interpreted interpreted to be to the be Calipatria the Calipatria Fault Fault [26] [26] and anand extension an extension of the of San the Andreas San Andreas Fault [ 27Fault], respectively. [27], respectively. Reath and Reath Ramsey and Ramsey [28] used [28] airborne used airborne Spatially EnhancedSpatially Enhanced Broadband Broadband Array Spectrograph Array Spectrograph System (SEBASS) System data(SEBASS) (7.6–13.5 dataµ m)(7.6–13.5 data to µm) map data geothermal to map indicatorgeothermal minerals indicator in this minerals region, in andthis Trattregion,et al.and[29 Tratt] used et al. SEBASS [29] used data SEBASS to map data ammonia to map plumes ammonia and surfaceplumes temperature. and surface Ammoniatemperature. emissions Ammonia are likelyemissi derivedons are fromlikely geothermal derived from fluids geothermal moving through fluids decomposingmoving through agricultural decomposing runoff agricultural and organic runoff material and [organic29]. material [29]. FiveFive northeast-trending northeast-trending volcanic volcanic domesdomes areare locatedlocated at the southern shore shore of of the the Salton Salton Sea. Sea. These These rhyoliticrhyolitic domes,domes, knownknown asas thethe SaltonSalton Buttes,Buttes, were extruded between between ca.ca. 59005900 and and 500 500 years years [30,31]. [30,31 ]. TheThe locations locations of of young young volcanism volcanism and and geothermal geothermal fields fiel areds aare result a result of the of regional the regional tectonics: tectonics: the Salton the SeaSalton sits Sea in a sits pull-apart in a pull-apart basin which basin opened which withinopened thewithin Brawley the Brawley Seismic Zone,Seismic a stepoverZone, a stepover between thebetween strike-slip the strike-slip Imperial faultImperial and fault San Andreas and San FaultAndreas [32]. Fault The [32]. San AndreasThe San FaultAndreas zone Fault runs zone along runs the easternalong the shore eastern of the shore Salton of Sea.the Salton The area Sea. is The host area to abundant is host to naturalabundant seismicity natural [seismicity33], as well [33], as injectionas well relatedas injection seismicity related [34 seismicity]. [34].

Geosciences 2016, 6, 11 5 of 14

3. Materials and Methods

AllGeosciences analyses 2016, 6 used, 11 data from the 2013 suite of flights as these were corrected by the5 of AVIRIS 14 and MASTER teams and were used to create the HyspIRI simulated products including at-surface reflectance3. Materials at several and Methods spatial scales, land surface temperature and emissivity [35–37]. Apparent surface reflectanceAll is analyses generated used using data from the the ATREM 2013 suite atmospheric of flights as correction these were [corrected35]. The by 30 the m AVIRIS products and were resampledMASTER using teams Gaussian and were weighted used to sampling create the with HyspIRI a 30 msimulated full-width products half-maximum including overat-surface a 90 m by 90 mreflectance area. Noise at approximatingseveral spatial scales, a HyspIRI land VSWIRsurface temperature noise function and wasemissivity added [35–37]. to radiance Apparent data [36]. Temperature-emissivitysurface reflectance is separationgenerated using takes the advantage ATREM atmospheric of concurrent correction water [35]. vapor The estimates 30m products obtained fromwere the AVIRIS resampled sensor using to Gaussian improve weighted land surface sampling temperature with a 30 [37 m]. full-width half-maximum over a 90 m by 90 m area. Noise approximating a HyspIRI VSWIR noise function was added to radiance 3.1. VSWIRdata [36]. Decorrelation Temperature-emissivity Stretch to Quickly separation Identify take Alterations advantage of concurrent water vapor estimates obtained from the AVIRIS sensor to improve land surface temperature [37]. Both modern geothermal systems and fossil hydrothermal systems associated with ore bodies commonly3.1. VSWIR display Decorrelation alteration Stretch mineralogy to Quickly atIdentify the surface. AlterationWe advocate a quick method for rapidly locating alterationBoth modern minerals geothermal is a simplesystems decorrelation and fossil hydrothermal stretch (DCS) systems of SWIRassociated bands with at 2.16,ore bodies 2.21, and 2.24 µcommonlym displayed display as red,alteration green, mineralogy and blue at (RGB), the surface. respectively. We advocate The a DCS quick technique method for removes rapidly locating correlation betweenalteration channels minerals and is createsa simple a decorrelation highly saturated stretch color(DCS) image,of SWIR andbands while at 2.16, these 2.21, band and 2.24 centers µm are similardisplayed to those as inred, ASTER green, and and blue World (RGB), View respectively. 3 sensors, The those DCS technique multichannel removes instruments correlation between have much broaderchannels band and passes. creates Using a highly the saturated 10 nm color narrow image, spectral and while channels these band of thecenters AVIRIS are similar instrument to those allowsin this techniqueASTER and to World separate View common3 sensors, alterationthose multichann mineralsel instruments based on have their much band broader centers band (Figure passes.3), so that specificUsing the mineral 10nm narrow groups spectral are highlighted channels of the in AVIR distinctIS instrument colors, as allows demonstrated this technique by Littlefieldto separate and common alteration minerals based on their band centers (Figure 3), so that specific mineral groups are Calvin [38]. This method can be performed quickly to generate an image that highlights common highlighted in distinct colors, as demonstrated by Littlefield and Calvin [38]. This method can be alteration minerals associated with argillic, phyllic, or propylitic alteration. Kaolinite and alunite are performed quickly to generate an image that highlights common alteration minerals associated with commonargillic, argillic phyllic, alteration or propylitic products, alteration. formed Kaolinite where and acidic alunite fluids are common chemically argillic weather alteration country products, rock or occurformed near fumaroles. where acidic Muscovite fluids chemically and illiteweather are country low temperature rock or occur phyllicnear fumaroles. alteration Muscovite minerals. and illite Chlorite commonlyare low occurstemperature in the phyllic propylitic alteration alteration minerals. ofChlorite mafic commonly minerals occurs and calcitein the propylitic can occur alteration in propylitic of alterationmafic butminerals may and also calcite be deposited can occur aroundin propylitic hot alteration springs asbut travertine may also be or deposited tufa. Opal around (amorphous hot springs silica) is depositedas travertine around or tufa. hot Opal springs (amorpho as sinter.us silica) Using is deposited this simple around color hot stretch, springs kaoliniteas sinter. Using and alunite this simple are blue, muscovitecolor stretch, and illite kaolinite are magenta, and alunite chlorite are blue, and muscovite calcite are and yellow, illite are epidote magenta, is yellow-green,chlorite and calcite and are opal is orangeyellow, (Figure epidote3,[ 38 is,39 yellow-green,]). The simplicity and opal of thisis orange method (Figure promotes 3, [38,39]. rapid The assessment simplicity of over this large method areas to promotes rapid assessment over large areas to help target the highest priority areas of interest. This help target the highest priority areas of interest. This decorrelation stretch technique was performed on decorrelation stretch technique was performed on the 18 m and 30 m Level 2 ortho-reflectance data the 18 m and 30 m Level 2 ortho-reflectance data collected by AVIRIS. We chose to use both the higher collected by AVIRIS. We chose to use both the higher and lower spatial resolution data in order to and lowercompare spatial results resolution and assess data what in order information to compare might results be lost and at assessthe proposed what information HyspIRI spatial might be lost atresolution. the proposed HyspIRI spatial resolution.

Figure 3. Illustration of a simple color decorrelation stretch band combination that rapidly separates Figure 3. Illustration of a simple color decorrelation stretch band combination that rapidly separates common alteration minerals into distinct color fields. Red-Green-Blue channels are chosen as common alteration minerals into distinct color ﬁelds. Red-Green-Blue channels are chosen as illustrated illustrated in the figure, and resulting mineral color associations are described in the text. Spectral in thestandards ﬁgure, and are resultingfrom the USGS mineral spectral color library associations [39]. are described in the text. Spectral standards are from the USGS spectral library [39].

Geosciences 2016, 6, 11 6 of 14

3.2. VSWIR Data for Mineral Mapping Kruse et al. [40] tested the ability of the HyspIRI instrument to discriminate various minerals at a 60-m spatial resolution over Cuprite and Steamboat Springs, Nevada, and Yellowstone, Wyoming. Recent developments in instrument design would improve the spatial resolution of the VSWIR range to 30 m [17]. The DCS color image described in the previous section quickly highlighted several regions with diverse alteration mineralogy. Detailed mineral mapping efforts were then focused on the Orocopia and Chocolate Mountains. We used 30-m AVIRIS Level 2 ortho-reﬂectance data for mineral mapping to test mapping abilities at the proposed spatial resolution. In our past remote sensing efforts focused on geothermal indicator minerals, techniques included a combination of statistical methods and experience-based approaches to map mineralogy. This methodology has worked well over a wide range of targets in the arid southwest of the USA [41]. We also used the standard spectral hourglass approach included in the ENVI software, which involves a Minimum Noise Fraction (MNF) transformation and Pixel Purity Index (PPI) algorithm. Ultimately we used carefully chosen threshold values to select pixels from Matched Filter images as mineral classes. These classes were bought into ArcGIS for interpretation with existing geologic maps.

3.3. TIR Decorrelation Stretch to Quickly Identify Lithology MASTER TIR data were provided as a Level 2 emissivity product. Two different DCS approaches to TIR data have been used in previous studies [6,42]. We examined both these decorrelation stretches to highlight different lithologies within the Orocopia and Chocolate Mountains. The ﬁrst technique is derived from Rowan and Mars [6] who used a decorrelation stretch of ASTER bands 13, 12, 10 that roughly correspond to channels 10.63, 9.03, and 8.60 µm displayed as RGB, respectively, to guide lithologic mapping near Mountain Pass, California. Another DCS combination was proposed by Vaughan et al. [42] who used 11.32, 10.66 and 9.08 µm as RGB. Both of these color products will highlight quartz-rich rocks separately from those that are more altered or intermediate in composition. These color stretches on the Salton regional area were transferred to ArcGIS where lithologies could be compared to the geologic units of Jennings et al. [20].

3.4. TIR Emissivity Data for Lithologic Mapping Several previous studies have used the emissivity spectra from ASTER to map lithologic units [6,42–45]. Mineral spectral shapes in ASTER and HyspIRI prototype data are subdued as the five channels are really insufficient to uniquely and diagnostically identify mineralogy [28,42]. While spectra of minerals and field samples convolved to ASTER band passes show diverse characteristics [44,45], actual ASTER data primarily maps units through the relative strength of emissivity near 9 µm [43–45] and lower emissivity at 8.3 and 8.6 µm (channels 10 and 11). We used the DCS methods described above along with the geologic map to identify regions of interest where unique spectral shapes were identified and mapped spectral shapes using spectral similarity tools, such as the spectral angle and matched filter.

3.5. TIR Data for Thermal Anomaly Identification Two special nighttime flights were flown to collect MASTER data over the Imperial Valley. We chose to use the data from the flight on 1 June 2013 because of its slightly better spatial coverage. It can be difficult to produce accurate land surface temperature maps from daytime data due to shading and albedo, but even nighttime remote sensing data are affected by elevation, aspect, and slope. The most accurate detection of thermal anomalies includes day/night image pairs and must account for surface thermal inertia (e.g., [46]). We wanted to explore how well the standard land surface temperature product would be able to identify the localized thermal anomalies without additional processing in a small area within Imperial Valley at the 60-m spatial resolution. This area is at the southern end of the Salton Sea, covering known mudpots and fumarolic vents and at a nearly constant elevation below 100 m. Using ArcGIS we made comparisons with previous results [26,27,29]. Geosciences 2016, 6, 11 7 of 14

4. ResultsGeosciences 2016, 6, 11 7 of 14

4.1. VSWIR4. Results Decorrelation Stretch to Quickly Identify Alteration µ A4.1. decorrelation VSWIR Decorrelation stretch Stretch of AVIRIS to Quickly bands Identify at 2.16, Alteration 2.21, and 2.24 m, displayed as RGB, respectively, was used to quickly identify areas of hydrothermal alteration (Figure4). As mentioned in Section 3.1, A decorrelation stretch of AVIRIS bands at 2.16, 2.21, and 2.24 µm, displayed as RGB, this decorrelation stretch typically highlights kaolinite, alunite, muscovite, illite, chlorite, epidote, respectively, was used to quickly identify areas of hydrothermal alteration (Figure 4). As mentioned calcite, and opal. We found that both the 18-m and 30-m images highlighted these minerals well in Section 3.1, this decorrelation stretch typically highlights kaolinite, alunite, muscovite, illite, and thatchlorite, the epidote, 30-m DCS calcite, image and isopal. very We similar found that tothe both 18-m the 18-m DCS and image 30-m in images the distribution highlighted these of colors. The 30-mminerals DCS well image and that generally the 30-m shows DCS image sharper is very color similar changes to the because 18-m DCS the image larger in the pixel distribution size does not permitof showingcolors. The subtle 30-m colorDCS image gradation generally as well. shows sharper color changes because the larger pixel size Asdoes the not Chocolate permit showing Mountains subtle color are off-limits gradation toas publicwell. access, the colors in the DCS images were validatedAs using the Chocolate the full spectral Mountains data are (see off-limits also Section to public 4.2 access,). These the DCS colors images in the highlightedDCS images were kaolinite in blue.validated Kaolinite using wasthe full distributed spectral data throughout (see also Se muchction 4.2). of These the Chocolate DCS images Mountains highlighted and kaolinite Orocopia Mountains.in blue. BlueKaolinite also was highlighted distributed the throughout alunite in themuch scene, of the a veryChocolate small Mountains area west and of theOrocopia Salton Sea Mountains. Blue also highlighted the alunite in the scene, a very small area west of the Salton Sea near the western-most corner of the black box in Figure4. Magenta was generally muscovite, and near the western-most corner of the black box in Figure 4. Magenta was generally muscovite, and yellow-green was epidote. These minerals occurred within the Chocolate Mountains and Orocopia yellow-green was epidote. These minerals occurred within the Chocolate Mountains and Orocopia Mountains.Mountains. We We identiﬁed identified calcite calcite highlighted highlighted by by yello yelloww in in the the southwest southwest corner corner of Figure of Figure 4 within4 within a a sandstone/mudstonesandstone/mudstone unit. unit. TheThe colorscolors in the the DCS DCS images images clearly clearly highlighted highlighted hydrothermal hydrothermal alteration alteration mineralmineral occurrences occurrences when when compared compared against against the fullthe spectralfull spectral proﬁles profiles for thesefor these regions. regions. From From previous experienceprevious [38 experience,41] we know [38,41] that we thisknow decorrelation that this decorrelation stretch would stretch highlightwould highlight opal as opal orange, as orange, however therehowever is no remotely there is sensedno remotely opal sensed in the opal Imperial in the Valley.Imperial Orange Valley. huesOrange on hues the easternon the eastern side of side the of scene are generallythe scene alluviumare generally or undifferentiatedalluvium or undifferentiated soils. soils.

FigureFigure 4. Decorrelation 4. Decorrelation stretch stretch of AVIRIS of AVIRIS bands bands at 2.16, at2.21, 2.16, and 2.21, 2.24 andµm 2.24 displayed µm displayed as RGB, as respectively. RGB, The samerespectively. scene is The shown same at scene two differentis shown at spatial two different resolutions: spatial (a resolutions:) 18 m, the resolution(a) 18 m, the of resolution original AVIRISof data;original and (b AVIRIS) 30 m, data; the and resolution (b) 30 m, of the the resolution proposed of the HyspIRI proposed VSWIR HyspIRI data. VSWIR The data. black The box black shows box shows the location of Figure 5. Blues, magentas, and yellows highlight hydrothermal or other the location of Figure5. Blues, magentas, and yellows highlight hydrothermal or other alteration alteration mineralogy, as described in the text. mineralogy, as described in the text. 4.2. VSWIR Data for Hydrothermal Alteration Mapping 4.2. VSWIR Data for Hydrothermal Alteration Mapping Using the 30-m VSWIR data collected by AVIRIS the matched filter classes remotely mapped epidote, Usingmuscovite, the and 30-m kaolinite VSWIR in the data Orocopia collected Mountains by AVIRIS and the the Chocolate matched Mountains ﬁlter classes (Figure remotely5). Epidote mappedwas epidote,mapped muscovite, within the and Orocopia kaolinite Schist, in thewhich Orocopia agrees with Mountains the geologic and description the Chocolate of the schist Mountains [24]. Intrusive (Figure 5). Epidoterocks was in the mapped Orocopia within Mountains the Orocopia showed epidote, Schist, muscovite, which agrees and kaolinite, with the which geologic are interpreted description to be of the schistthe [24 result]. Intrusive of hydrothermal rocks in alteration. the Orocopia Kaolinite Mountains was also mapped showed within epidote, Ceno muscovite,zoic sedimentary and units, kaolinite, including the Diligencia Formation and Maniobra Formation, both in the Orocopia Mountains. In the which are interpreted to be the result of hydrothermal alteration. Kaolinite was also mapped within Chocolate Mountains, kaolinite and muscovite were primarily mapped within granites and Tertiary Cenozoic sedimentary units, including the Diligencia Formation and Maniobra Formation, both in the Orocopia Mountains. In the Chocolate Mountains, kaolinite and muscovite were primarily mapped Geosciences 2016, 6, 11 8 of 14

Geosciences 2016, 6, 11 8 of 14 within granites and Tertiary volcanic rocks. Using tight thresholds for the standard deviations on the mappedvolcanicGeosciences spectralrocks. 2016 ,Using 6, classes,11 tight we thresholds did not identifyfor the standard opal, calcite, deviations chlorite, on the or mapped alunite tospectral the east classes, of the8 weof Salton 14 did Seanot inidentify the area opal, subset calcite, of Figurechlorite,5. Figureor alunite6 shows to the spectra east of ofthe the Salton mapped Sea in units the comparedarea subset with of Figure library 5. volcanic rocks. Using tight thresholds for the standard deviations on the mapped spectral classes, we did mineralFigure 6 standards. shows spectra of the mapped units compared with library mineral standards. not identify opal, calcite, chlorite, or alunite to the east of the Salton Sea in the area subset of Figure 5. Figure 6 shows spectra of the mapped units compared with library mineral standards.

FigureFigure 5.5.The The OrocopiaOrocopia andand ChocolateChocolate Mountains area. A A true true color color image image derived derived from from 18 18 m m AVIRIS AVIRIS datadataFigure isis shownshown 5. The overlain overlainOrocopia on onand aa Chocolate shadedshaded reliefrelief Mountains image.image. area. Remotely A true colormapped mapped image minerals minerals derived arefrom are shown shown18 m AVIRISin in colors: colors: green—epidote;green—epidote;data is shown magenta—muscovite;magenta—muscovite;overlain on a shaded blue—kaolinite.relief image. Remotely The The lo location cationmapped of of mineralsthis this figure ﬁgure are is is shownshown shown inin in colors:Figure Figure 4.4 . green—epidote; magenta—muscovite; blue—kaolinite. The location of this figure is shown in Figure 4.

FigureFigureFigure 6.6. 6.SpectraSpectra Spectra ofof of mappedmapped mapped units unitsunits (solid(solid(solid lines) compared compared with with library library library mineral mineral mineral standards standards standards from from from the the the USGSUSGSUSGS spectralspectral spectral librarylibrary library [[39]39 [39]] (dashed(dashed (dashed lines).lines).lines). Each scsc sceneeneene spectrum spectrum represents represents represents an an anaverage average average of ofmore of more more than than than severalseveralseveral hundred hundred hundred pixels. pixels. pixels.

4.3.4.3.4.3. TIR TIR TIR DecorrelationDecorrelation Decorrelation StretchStretch Stretch toto to QuicklyQuickly Quickly Identify Identify Lithology A decorrelation stretch of MASTER bands at 10.63, 9.03, and 8.60 µm displayed as RGB, AA decorrelationdecorrelation stretchstretch ofof MASTERMASTER bands at 10.63, 9.03, 9.03, and and 8.60 8.60 µmµm displayed displayed as as RGB, RGB, respectively, was used for basic lithologic mapping (Figure 7). Similar to [6], we found that red pixels respectively,respectively, waswas used for for basic basic lith lithologicologic mapping mapping (Figure (Figure 7). 7Similar). Similar to [6], to we [ 6 found], we foundthat red that pixels red in the DCS image highlight quartz-rich units, for example the Algodones dune sand. Magenta in the DCS image highlight quartz-rich units, for example the Algodones dune sand. Magenta pixelshighlighted in the DCS schist image and highlightgneiss, which quartz-rich have relatively units, for less example quartz, thefor Algodonesexample the dune Orocopia sand. Schist Magenta highlighted schist and gneiss, which have relatively less quartz, for example the Orocopia Schist highlightedwithin the schist southern and part gneiss, of the which Orocopia have relativelyMountain lesss. We quartz, found forthat example other units the (basalt, Orocopia sedimentary Schist within within the southern part of the Orocopia Mountains. We found that other units (basalt, sedimentary therocks, southern granite, part granodiorite, of the Orocopia rhyolite) Mountains. did not We nece foundssarily that correlate other units with (basalt,colors in sedimentary either of the rocks, granite,rocks,DCS stretches.granite, granodiorite, granodiorite, rhyolite) rhyolite) did not necessarily did not nece correlatessarily with correlate colors with in either colors of thein DCSeither stretches. of the DCS stretches.

Geosciences 2016,, 66,, 1111 99 ofof 1414 Geosciences 2016, 6, 11 9 of 14

Figure 7. The Orocopia Mountains and the Chocolate Mountains. Decorrelation stretch of MASTER Figurebands at 7. 10.63,The Orocopia 9.03, and MountainsMountains 8.60 µm displayed and thethe ChocolateChocolateas RGB, respectively. Mountains.Mountains. Geologic Decorrelation map units stretch from of MASTER Figure 2 bandsare outlined atat 10.63,10.63, by 9.03, 9.03,black andand lines. 8.60 8.60 Theµ µmm location displayed displayed outline as as RGB, RGB, is shown respectively. respectively. in Figure Geologic Geologic 4. map map units units from from Figure Figure2 are 2 outlinedare outlined by blackby black lines. lines. The The location location outline outline is shown is shown in Figure in Figure4. 4. 4.4. TIR Emissivity Data for Lithologic Mapping 4.4. TIR Emissivity Data for Lithologic Mapping Because colors in the DCS were not well correlated with unit boundaries, we used color transitions,Because the colorscolors geologic in in the the DCSmap, DCS were and were not other wellnot knowle correlatedwell correlateddge with based unit with approaches boundaries, unit boundaries, to we identify used colorwe regions used transitions, colorwith thecontrastingtransitions, geologic emissivitythe map, geologic and spectral other map, shapes. knowledgeand other Similar basedknowle to othe approachesdger past based studies, approaches to identify emissivity regionsto variabilityidentify with regions is contrasting primarily with emissivitycontrolledcontrasting spectralby emissivity strength shapes. spectralof the Similar feature shapes. to at other Similar 9.03 past µm, to studies, othewithr smallpast emissivity studies, changes variabilityemissivity in slope at isvariability primarily8.6 µm. Theis controlled primarily distinct byspectralcontrolled strength shapes by of thestrength identified feature ofat the are 9.03 feature shownµm, with at in 9.03 Figure small µm, changes 8. with Using small in spectral slope changes at 8.6similarity µinm. slope The tools distinctat 8.6 such µm. spectral as The a matched distinct shapes identifiedspectralfilter, the shapes mica are shown schist identified of in the Figure are Orocopio shown8. Using Mountainsin spectralFigure 8. and similarity Using a small spectral tools eastern suchsimilarity outlier as a matched toolsare mapped, such filter, as alonga thematched mica with schistseveralfilter, the of basalt the mica Orocopio units. schist However of Mountains the Orocopio the andmajority Mountains a small of the eastern and Chocolate a outlier small Mountainseastern are mapped, outlier gneiss along are and mapped, with granite several along units basalt with are units.notseveral distinguished. However basalt units. the The majority However quartz of sand the Chocolatemajoritys of the Alogodones of Mountains the Chocolate dune gneiss fieldMountains and are granite clearly gneiss units distinguished and are notgranite distinguished. using units thisare Thetechnique.not distinguished. quartz sands of The the quartz Alogodones sands of dune the fieldAlogodones are clearly dune distinguished field are clearly using distinguished this technique. using this technique.

(a) (b) (a) (b) Figure 8. Emissivity spectral shapes ( a)) and spectral locations mapped; ( b)) using a matched filterfilter Figure 8. Emissivity spectral shapes (a) and spectral locations mapped; (b) using a matched filter approach. Colors in the map correspond to the same color in the spectral plot. Blue, green and purple approach. Colors in the map correspond to the same color in the spectral plot. Blue, green and purple spectral shapes occur in the same regions, identify the schist of the Orocopio Mountains, and several spectral shapes occur in the same regions, identify the schist of the Orocopio Mountains, and several basalt outcrops. Yellow and coral spectral shapes both map the Algodones dune field field and only yellow basalt outcrops. Yellow and coral spectral shapes both map the Algodones dune field and only yellow is shown for simplicity. The area shown is identifiedidentified in Figure4 4 and and isis thethe samesame asas inin FiguresFigures5 5and and7 7 is shown for simplicity. The area shown is identified in Figure 4 and is the same as in Figures 5 and 7 for units compare withwith FiguresFigures2 2 and and7 .7. for units compare with Figures 2 and 7. 4.5. TIR Data for Thermal Anomaly IdentificationIdentification 4.5. TIR Data for Thermal Anomaly Identification We focusedfocused ourour study study of of the the land land surface surface temperature temperature data data collected collected by by MASTER MASTER on anon areaan area at the at We focused our study of the land surface temperature data collected by MASTER on an area at southernthe southern end ofend the of Salton the Salton Sea at Sea elevations at elevations below below 100 m. 100 Temperatures m. Temperatures ranged fromranged 16.85 from to 47.8516.85 ˝toC the southern end of the Salton Sea at elevations below 100 m. Temperatures ranged from 16.85 to (62.3347.85 °C to 118.13(62.33 to˝F) 118.13 and the °F) relative and the scale relative of temperature scale of temperature variation appearedvariation valid.appeared Within valid. Salton Within Sea 47.85 °C (62.33 to 118.13 °F) and the relative scale of temperature variation appeared valid. Within thereSalton are Sea areas there of are cooler areas water of cooler near water the inflow near th ofe the inflow Alamo of the and Alamo New Rivers,and New which Rivers, appear which cyan-blue appear Salton Sea there are areas of cooler water near the inflow of the Alamo and New Rivers, which appear

Geosciences 2016, 6, 11 10 of 14 Geosciences 2016, 6, 11 10 of 14 incyan-blue Figure9. in The Figure agricultural 9. The agricultural area in Imperial area Valleyin Imperial is generally Valley coolis generally due to thecool crops, due to which the crops, retain waterwhich and retain provide water shade and provide during shade the day during (blue the in Figure day (blue9). in Figure 9).

Figure 9. (a) Imperial Valley, California study area. A true color image derived from AVIRIS data is Figure 9. (a) Imperial Valley, California study area. A true color image derived from AVIRIS data shown overlain on a shaded relief image; the box shows the location of (b). (b) Land surface is shown overlain on a shaded relief image; the box shows the location of (b). (b) Land surface temperature data for the southern end of the Salton Sea below 100 m elevation; the box shows the temperature data for the southern end of the Salton Sea below 100 m elevation; the box shows the location of (c). Known geothermal areas are indicated by initials: HMS—Hot Mineral Spa; SS—Salton location of (c). Known geothermal areas are indicated by initials: HMS—Hot Mineral Spa; SS—Salton Sea. (c) Land surface temperature data for the Salton Sea geothermal area. Circles show geothermal Sea. (c) Land surface temperature data for the Salton Sea geothermal area. Circles show geothermal power plants and squares show geothermal features located by [27]. Fumarole fields are indicated by power plants and squares show geothermal features located by [27]. Fumarole ﬁelds are indicated by initials: W—Wister; MI—Mullet Island; DS—Davis Schrimpf; RI—Red Island. initials: W—Wister; MI—Mullet Island; DS—Davis Schrimpf; RI—Red Island.

Thermal anomalies occur within the Salton Sea geothermal area (red in Figure 9c). Some of the anomaliesThermal correlate anomalies with occurthe locations within of the geotherm Salton Seaal power geothermal plants. areaGeothermal (red in power Figure 9plantsc). Some utilize of thecondensers, anomalies whereby correlate large with volumes the locations of hot fluids of geothermal are run through power cooling plants. towers Geothermal to release power heat. plants Ten utilizeindividual condensers, condenser whereby units occur large in volumes this region of hotand fluids were associated are run through with thermal cooling anomalies towers to (circles release heat.in Figure Ten individual 9). Other condenserthermal anomalies units occur correlate in this regionwith the and vents were identified associated by with [26,27]. thermal Two anomalies of the (circlesstrongest in Figureanomalies9). Other highlight thermal recently anomalies exposed correlate fumaro withle fields the vents near identifiedMullet Island. by [ 26The,27 shape]. Two of of the the strongestlarger of anomaliesthese thermal highlight anomalies recently agreed exposed with the fumarole results from fields [29] near who Mullet mapped Island. it using The shape1 m spatial of the largerresolution of these SEBASS thermal data. anomalies Many of agreed the nearly with twenty the results individual from [29 geothermal] who mapped features it using noted 1 m by spatial [27] resolution(squares in SEBASS Figure 9c) data. are Manysmall or of not the very nearly active. twenty Approximately individual one-half geothermal of the features features noted previously by [27 ] (squaresidentified in do Figure not 9appearc) aresmall anomalous or not in very this active. one ni Approximatelyght-time scene. one-halfOne example of the is features the western-most previously identifiedWister area do vent not shown appear in anomalous Figure 9c, inwhich this onethe auth night-timeors describe scene. as Onea “bulldozed example ring is the in western-most pond” ([27], Wisterpage 1722). area vent This shownfeature in is Figurenot likely9c, whichvery active the authors if a bulldozer describe was as aable “bulldozed to drive through ring in pond” it, and ([ not27 ], pagesurprisingly 1722). This there feature is no significant is not likely thermal very active anomaly if a shown bulldozer in the was TIR able data to for drive this through location. it, and not surprisinglyThere are there several is no thermal significant anomalies thermal shown anomaly in the shown TIR data in that the TIRare not data correlated for this location. with geothermal powerThere plants are severalor known thermal geothermal anomalies features. shown The in anomalies the TIR data near that Mullet are not Island correlated are associated with geothermal with a powershallow plants pond orwhere known water geothermal is solar heated features. to high The temperatures. anomalies Other near Mulletanomalies Island could are be associated associated withlow aalbedo shallow as pondthere whereis great water variation is solar in field heated cover to high(different temperatures. crops at different Other anomalies stages of couldgrowth, be and associated some lowfallow albedo fields). as thereHowever is great there variation are a few in fieldremaining cover (differentthermal anomalies crops at differentthat could stages be associated of growth, with and somepreviously fallow unrecognized fields). However or shifting there are sources a few remainingof surface thermalheat flux anomalies and would that be could places be for associated future withfield previouslyvalidation. unrecognized or shifting sources of surface heat flux and would be places for future field validation. 5. Discussion

5.1. Hydrothermal Alteration Mapping

Geosciences 2016, 6, 11 11 of 14

5. Discussion

5.1. Hydrothermal Alteration Mapping The decorrelation stretch of AVIRIS bands at 2.16, 2.21, and 2.24 µm quickly identified regions where hydrothermal alteration minerals occur. In particular, alunite, calcite, epidote, muscovite and kaolinite locations were validated using the full spectral properties of the data set. The rapid identification of these minerals is useful to both the geothermal and economic mineral industries because (advanced) argillic alteration (alunite and kaolinite) is an excellent indicator of hydrothermal fluid movement. Using a combination of the DCS images and statistical methods, the 30-m HyspIRI-like VSWIR data were used to successfully map epidote, muscovite, and kaolinite within the Orocopia Mountains and Chocolate Mountains. The detailed mapping showed good correspondence with the simple DCS technique, though smaller areas were mapped by using tight tolerances on mineral spectral matches. We interpret that the epidote mapped within gneiss and schist is likely of metamorphic origin, however the kaolinite and muscovite may be hydrothermal in origin. In general, there is a good correspondence between the locations where we identified strong hydrothermal mineral abundance and the alteration zones identified in ASTER data by Zhang et al. [8] (Figure8). We did not map either alunite or montmorillonite in the HyspIRI-like data, which was identified by [8]. The areas mapped by [8] as alunite were identified as kaolinite. We also did not identify the extensive montmorillonite mapped by Zhang et al. [8] and they did not map epidote. These discrepancies in mineral mapping are likely because [8] used multi-channel ASTER data and we used the much higher spectral resolution data of the HyspIRI-like reflectance product, with tight tolerances on similarity to mineral spectral features.

5.2. Lithologic Mapping We used two decorrelation stretches of MASTER bands to quickly differentiate lithologies in the Imperial Valley, but ultimately found that this method was only able to reliably highlight quartz-rich areas. The various combinations showed the Algodones dune sand and the Orocopia Schist but did not highlight other igneous and sedimentary units. Using a knowledge-based approach, five different emissivity shapes were identified and mapped using spectral similarity tools. The spectra also matched the dune field, and the Orocopoia schist. The spectra also mapped several basalt outcrops but did not distinguish the units of the Chocolate Mountains. In order to achieve the best lithologic mapping, the complementary nature of the VSWIR and TIR channels will need to be employed, similar to past successful studies with ASTER [43–45,47,48].

5.3. Thermal Anomaly Mapping We found that the Level 2 land surface temperature nighttime image derived from MASTER TIR data identified numerous thermal anomalies. Most of the strongest thermal anomalies highlighted geothermal power plants, with other anomalies highlighting fumarole fields. Of the many naturally occurring small anomalies previously identified, many features were not evident at the proposed 60-m HyspIRI TIR resolution. However, all power plant condenser units were correlated with thermal anomalies. The largest and recently exposed fumarole fields near Mullet Island would be identified in the 60-m HyspIRI data. The high temporal resolution of HyspIRI is well suited to study a place like the Salton Sea where new geothermal features are being exposed on the order of months to years.

6. Conclusions In this study we demonstrated the geological applications of HyspIRI-like data in the Imperial Valley region of Southern California. We used a decorrelation stretch of SWIR bands to quickly highlight hydrothermal alteration, and a decorrelation stretch of TIR bands to quickly highlight quartz-rich units. We used DCS images to guide mineral mapping efforts. Using the VSWIR data we were able to explicitly map the hydrothermal alteration minerals present in the scene: epidote, Geosciences 2016, 6, 11 12 of 14 muscovite, and kaolinite. TIR mapping highlighted quartz rich sands, schist and some basalt outcrops. We also used land surface temperature data derived from the TIR data to identify thermal anomalies, which were associated with geothermal power plants, fumarole fields, and shallow ponds. We advocate that HyspIRI will be a useful instrument to the geology community, especially to those interested in mineral distribution, for example in geothermal and mineral exploration. The high resolution of the VSWIR data allows for identification of very specific mineralogy with narrow absorption features. The 30 m spatial resolution of the VSWIR data is high enough to map relatively small mineral occurrences and rock outcrops, for example, the Salton Buttes. The 60-m resolution of the TIR data is high enough to distinguish some geologic units and identify small thermal anomalies such as the Wister fumarole fields. The 5- and 16-day revisit time for the TIR and VSWIR instruments, respectively, will provide high temporal resolution for the monitoring of rapidly changing geologic conditions, for example the continued exposure of new fumarole fields as the Salton Sea continues to recede.

Acknowledgments: This work was supported by NASA Grant #NNX12AQ17G to UNR. Derived data product support from NASA Grant #NNX12AP08G. Author Contributions: Elizabeth Pace performed the VSWIR analysis and drafted the initial version of the manuscript. Wendy Calvin performed the TIR analysis and mapping and wrote the final version of the text. Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations The following abbreviations are used in this manuscript:

ASTER Advanced Spaceborne Thermal Emission and Reﬂectance Radiometer AVIRIS Advanced Visible/Infrared Imaging Specrometer DCS decorrelation stretch HyspIRI Hyperspectral Infrared Imager MASTER MODIS/ASTER Airborne Simulator TIR Thermal Infrared VSWIR Visible and Short-Wave Infrared

References

1. Sabins, F.F. Remote sensing for mineral exploration. Ore Geol. Rev. 1999, 14, 157–183. [CrossRef] 2. Yamaguchi, Y.; Kahle, A.B.; Tsu, H.; Kawakami, T.; Pniel, M. Overview of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). IEEE Trans. Geosci. Remote Sens. 1998, 36, 1062–1071. [CrossRef] 3. Pearlman, J.S.; Barry, P.S.; Segal, C.C.; Shepanski, J.; Beiso, D.; Carman, S.L. Hyperion, a space-based imaging spectrometer. IEEE Trans. Geosci. Remote Sens. 2003, 41, 1160–1173. [CrossRef] 4. Van der Meer, F.D.; van der Werff, H.M.A.; van Ruitenbeek, F.J.A.; Hecker, C.A.; Bakker, W.H.; Noomen, M.F.; van der Meijde, M.; Carranza, E.J.M.; de Smeth, J.B.; Woldai, T. Multi- and hyperspectral geologic remote sensing: A review. Int. J. Appl. Earth Obs. Geoinf. 2012, 14, 112–128. [CrossRef] 5. Hewson, R.D.; Cudahy, T.J.; Huntington, J.F. Geologic and alteration mapping at Mt Fitton, South Australia, using ASTER satellite-borne data. In Proceedings of the 2001 IEEE Geoscience and Remote Sensing Symposium, Sydney, Australia, 9–13 July 2001; pp. 724–726. 6. Rowan, L.C.; Mars, J.C. Lithologic mapping in the Mountain Pass, California area using Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data. Remote Sens. Environ. 2003, 84, 350–366. [CrossRef] 7. Mars, J.C.; Rowan, L.C. Spectral assessment of new ASTER SWIR surface reflectance data products for spectroscopic mapping of rocks and minerals. Remote Sens. Environ. 2010, 114, 2011–2025. [CrossRef] 8. Zhang, X.; Pamer, M.; Duke, N. Lithologic and mineral information extraction for gold exploration using ASTER data in the south Chocolate Mountains (California). ISPRS J. Photogramm. Remote Sens. 2007, 62, 271–282. [CrossRef] Geosciences 2016, 6, 11 13 of 14

9. Rowan, L.C.; Hook, S.J.; Abrams, M.J.; Mars, J.C. Mapping hydrothermally altered rocks at Cuprite, Nevada, using the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), a new satellite-imaging system. Econ. Geol. 2003, 98, 1019–1027. [CrossRef] 10. Bedini, E. Mineral mapping in the Kap Simpson complex, central east Greenland, using Hymap and ASTER remote sensing data. Adv. Space Res. 2011, 47, 60–73. [CrossRef] 11. Middleton, E.M.; Ungar, S.G.; Mandl, D.J.; Ong, L.; Frye, S.W.; Campbell, P.E.; Landis, D.R.; Young, J.P.; Pollack, N.H. The Earth Observing One (EO-1) satellite mission: Over a Decade in Space. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2013, 6, 243–256. [CrossRef] 12. Green, R.O.; Pavri, B.E.; Chrien, T.G. On-orbit radiometric and spectral calibration characteristics of EO-1 Hyperion derived with an underflight of AVIRIS and in situ measurements at Salar de Arizaro, Argentina. IEEE Trans. Geosci. Remote Sens. 2003, 41, 1194–1203. [CrossRef] 13. Bergeron, M.; Hollinger, A.; Staenz, K.; Maszkiewicz, M.; Neville, R.A.; Qian, S.E.; Goodenough, D.G. Hyperspectral environment and resource observer (HERO) mission. Can. J. Remote Sens. 2008, 34, S1–S11. [CrossRef] 14. Guanter, L.; Kaufmann, H.; Segl, K.; Foerster, S.; Rogass, C.; Chabrillat, S.; Kuester, T.; Hollstein, A.; Rossner, G.; Chlebek, C.; et al. The EnMAP spaceborne imaging spectroscopy mission for earth observation. Remote Sens. 2015, 7, 8830–8857. [CrossRef] 15. Matsunaga, T.; Iwasaki, A.; Tsuchida, S.; Tanii, J.; Kashimura, O.; Nakamura, R.; Yamamoto, H.; Tachikawa, T.; Rokugawa, S. Current status of Hyperspectral Imager Suite (HISUI). In Proceedings of the 2014 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Quebec, QC, Canada, 13–18 July 2014. 16. Lopinto, E.; Ananasso, C. The Prisma Hyperspectral Mission. Available online: http://www.earsel.org/ symposia/2013-symposium-Matera/proceedings.php (accessed on 4 January 2016). 17. Lee, C.M.; Cable, M.L.; Hook, S.J.; Green, R.O.; Ustin, S.L.; Mandl, D.J.; Middleton, E.M. An introduction to the NASA Hyperspectral Infrared Imager (HyspIRI) mission and preparatory activities. Remote Sens. Environ. 2015, 167, 6–19. [CrossRef] 18. AVIRIS Web Site. Available online: http://aviris.jpl.nasa.gov (accessed on 16 February 2016). 19. MASTER Web Site. Available online: http://masterweb.jpl.nasa.gov (accessed on 16 February 2016). 20. Jennings, C.W.; Strand, R.G.; Rogers, T.H. Geologic Map of California, Scale 1:750,000; California Division of Mines and Geology: Piscataway, NJ, USA, 1977. 21. Jennings, O.P. Geologic Map of California: Salton Sea Sheet, Scale 1:250,000; California Division of Mines and Geology: Sacramento, CA, USA, 1967. 22. Bjornstad, S.; Alm, S.; Huang, W.; Tiedeman, A.; Frazier, L.; Page, C.; Sabin, A.; Veazey, D. An Update on Geothermal Energy Resource Investigations, Chocolate Mountains Aerial Gunnery Range, Imperial Valley, California. Trans. Geotherm. Resour. Counc. 2011, 35, 713–720. 23. New Gold Inc. Technical Report on the Mesquite Mine, Imperial County, California, USA; NI 43–101 Report; Roscoe Postle Associates Inc.: Toronto, ON, Canada, 2014. 24. Jacobsen, C.E.; Grove, M.; Vućić,A.; Pedrick, J.N.; Ebert, K.A. Exhumation of the Orocopia Schist and associated rocks of southeastern California: Relative Roles of Erosion, Synsubduction Tectonic Denudation, and Middle Cenozoic Extension. Geol. Soc. Am. Spec. Pap. 2007, 419, 1–37. 25. Norris, R.M.; Norris, K.S. Algodones Dunes of southeastern California. Geol. Soc. Am. Bull. 1961, 72, 605–620. [CrossRef] 26. Lynch, D.K.; Hudnut, K.W.; Adams, P.M. Development and growth of recently-exposed fumarole fields near Mullet Island, Imperial County, California. Geomorphology 2013, 195, 27–44. [CrossRef] 27. Lynch, D.K.; Hudnut, K.W. The Wister mud pot lineament: Southeastward Extension or Abandoned Strand of the San Andreas Fault? Bull. Seismol. Soc. Am. 2008, 98, 1720–1729. [CrossRef] 28. Reath, K.A.; Ramsey, M.S. Exploration of geothermal systems using hyperspectral thermal infrared remote sensing. J. Volcanol. Geotherm. Res. 2013, 265, 27–38. [CrossRef] 29. Tratt, D.M.; Young, S.J.; Lynch, D.K.; Buckland, K.N.; Johnson, P.D.; Hall, J.L.; Westberg, K.R.; Polak, M.L.; Kasper, B.P.; Qian, J. Remotely sensed ammonia emission from fumarolic vents associated with a hydrothermally active fault in the Salton Sea Geothermal Field, California. J. Geophys. Res. 2011, 116, D21308. [CrossRef] 30. Schmitt, A.K.; Martin, A.; Stockli, D.F.; Farley, K.A.; Lovera, O.M. (U-Th)/He Zircon and archaeological ages for a late prehistoric eruption in the Salton Trough (California, USA). Geology 2013, 41, 7–10. [CrossRef] Geosciences 2016, 6, 11 14 of 14

31. Wright, H.M.; Vazquez, J.A.; Champion, D.E.; Calvert, A.T.; Mangan, M.T.; Stelten, M.; Cooper, K.M.; Herzig, C.; Schriener, A. Episodic Holocene eruption of the Salton Buttes rhyolites, California, from paleomagnetic, U-Th, and Ar/Ar dating. Geochem. Geophys. Geosyst. 2015, 16, 1198–1210. [CrossRef] 32. Brothers, D.S.; Driscoll, N.W.; Kent, G.M.; Harding, A.J.; Babcock, J.M.; Baskin, R.L. Tectonic evolution of the Salton Sea inferred from seismic reflection data. Nat. Geosci. 2009, 2, 581–584. [CrossRef] 33. McGuire, J.J.; Lohman, R.B.; Catchings, R.D.; Rymer, M.J.; Goldman, M.R. Relationships among seismic velocity, metamorphism, and seismic and aseismic fault slip in the Salton Sea Geothermal Field region. J. Geophys. Res. Solid Earth 2015, 120, 2600–2615. [CrossRef] 34. Wei, S.J.; Avouac, J.P.; Hudnut, K.W.; Donnellan, A.; Parker, J.W.; Graves, R.W.; Helmberger, D.; Fielding, E.; Liu, Z.; Cappa, F.; et al. The 2012 Brawley swarm triggered by injection-induced aseismic slip. Earth Planet. Sci. Lett. 2015, 422, 115–125. [CrossRef] 35. Thompson, D.R.; Gao, B.C.; Green, R.O.; Roberts, D.A.; Dennison, P.E.; Lundeen, S.R. Atmospheric correction for global mapping spectroscopy: ATREM Advances for the HyspIRI Preparatory Campaign. Remote Sens. Environ. 2015, 167, 64–77. [CrossRef] 36. Orthorectified HyspIRI Preparatory Campaign Products. Available online: ftp://popo.jpl.nasa.gov/2013_ HyspIRI_Prep_Data/AVIRIS_HyspIRI_Simulated_Data_Products.readme (accessed on 21 December 2015). 37. Grigsby, S.P.; Hulley, G.C.; Roberts, D.A.; Scheele, C.; Ustin, S.L.; Alsina, M.M. Improved surface temperature estimates with MASTER/AVIRIS sensor fusion. Remote Sens. Environ. 2015, 167, 53–63. [CrossRef] 38. Littlefield, E.F.; Calvin, W.M. Geothermal exploration using imaging spectrometer data over Fish Lake Valley, Nevada. Remote Sens. Environ. 2014, 140, 509–518. [CrossRef] 39. Clark, R.N.; Swayze, G.A.; Wise, R.; Livo, E.; Hoefen, T.; Kokaly, R.; Sutley, S.J. USGS Digital Spectral Library Splib06a; U.S. Geological Survey: Reston, VA, USA, 2007. 40. Kruse, F.A.; Taranik, J.V.; Coolbaugh, M.; Michaels, J.; Littlefield, E.F.; Calvin, W.M.; Martini, B.A. Effect of reduced spatial resolution on mineral mapping using imaging spectrometry-examples using Hyperspectral Infrared Imager (HyspIRI)-simulated data. Remote Sens. 2011, 3, 1584–1602. [CrossRef] 41. Calvin, W.M.; Littlefield, E.F.; Kratt, C. Remote sensing of geothermal-related minerals for resource exploration in Nevada. Geothermics 2015, 53, 517–526. [CrossRef] 42. Vaughan, R.G.; Hook, S.J.; Calvin, W.M.; Taranik, J.V. Surface mineral mapping at Steamboat Springs, Nevada, USA, with multi-wavelength thermal infrared images. Remote Sens. Environ. 2005, 99, 140–158. [CrossRef] 43. Hewson, R.D.; Cudahy, T.J.; Mizuhiko, S.; Ueda, K.; Mauger, A.J. Seamless geological map generation using ASTER in the broken Hill-Curnamona province of Australia. Remote Sens. Environ. 2005, 99, 159–172. [CrossRef] 44. Rowan, L.C.; Mars, J.C.; Simpson, C.J. Lithologic mapping of the Mordor, NT, Australia ultramafic complex by using the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER). Remote Sens. Environ. 2005, 99, 105–126. [CrossRef] 45. Rowan, L.C.; Schmidt, R.G.; Mars, J.C. Distribution of hydrothermally altered rocks in the Reko Diq, Pakistan mineralized area based on spectral analysis of ASTER data. Remote Sens. Environ. 2006, 104, 74–87. [CrossRef] 46. Coolbaugh, M.F.; Kratt, C.; Fallacaro, A.; Calvin, W.M.; Taranik, J.V. Detection of geothermal anomalies using Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) thermal infrared images at Bradys Hot Springs, Nevada, USA. Remote Sens. Environ. 2007, 106, 350–359. [CrossRef] 47. Rockwell, B.W.; Hofstra, A.H. Identification of quartz and carbonate minerals across northern Nevada using ASTER thermal infrared emissivity data—Implications for geologic mapping and mineral resource investigations in well-studied and frontier areas. Geosphere 2008, 4, 218–246. [CrossRef] 48. Hook, S.J.; Dmochowski, J.E.; Howard, K.A.; Rowan, L.C.; Karlstrom, K.E.; Stock, J.M. Mapping variations in weight percent silica measured from multispectral thermal infrared imagery—Examples from the Hiller Mountains, Nevada, USA and Tres Virgenes-La Reforma, Baja California Sur, Mexico. Remote Sens. Environ. 2005, 95, 273–289. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).