SERIES

Google Earth® Models withCOLLADA and WxAzygy®Transparent Interface:An example from GrottoCreek, Front Ranges,Canadian Cordillera

Katherine J.E. BoggsDepartment of Earth Sciences, Mount RoyalUniversity, 4825 Mount Royal Gate SW,Calgary, Alberta, Canada, T3E 6K6Corresponding author: kboggs@mtroyal.ca

Mladen M. DordevicDepartment of Physics, Room 306,OCNPS Building, Old DominionUniversity, 4600 Elkhorn Ave. Norfolk,Virginia, USA, 23529mdordevi@odu.edu

Scott T. ShipleyWxAnalyst, Ltd., 4458 OakdaleCrescent Court, Suite 1226, Fairfax,Virginia, USA, 22030sshipley@wxanalyst.com

SUMMARYVirtual globes represent a paradigmshift for geoscience education. It isnow possible to explore real worldexperiences across the entire Earth, theMoon, and Mars, and also to combinemultiple 2-D images into one 3-D

image with topography. Modelsviewed in Google Earth® are moreintuitive for visualizing 3-D geologicalstructures than traditional paper mapsand cross-sections. Here a student-constructed geological map and cross-sections from an introductory fieldschool are used to illustrate the cre-ation of a draped geological map overtopography. A custom vertical sliderelevates the cross-sections abovetopography and a horizontal sliderrestores thrust faulting. Models locatedin situ in topography are madequeryable via a ‘cut-away’ using theWxAzygy® transparent interface.

SOMMAIRELa notion de « globes virtuels » con-stitue un changement de paradigmedans le domaine de l’éducation en géo-science. Il est maintenant possible detraiter de la réalité de tous les recoinsde la Terre, de la lune et de Mars, etaussi de combiner de multiples images2-D en une image 3-D affichant latopographie. Les modèles de GoogleEarth® permettent une visualisation 3-D plus intuitive des structuresgéologiques que ne le permettent lescartes papiers usuelles et les coupes.Dans la présente, une carte géologiqueet une coupe réalisées par un étudiantd’un cours d’introduction au travail deterrain sont utilisées pour illustrer laconfection d’une carte géologiqueappliquée sur la topographie corre-spondante. Un curseur vertical per-sonnalisé dessine les coupes au-dessusde la topographie, et un curseur hori-zontal permet de restaurer les failles dechevauchement. Ces modèles ancrésau droit de la topographie peuvent êtreexploiter au moyen d’un écorché pro-duit par l’interface transparent WxAzygy®.

INTRODUCTION TO Google Earth®MODELS AND VIRTUAL FIELD TRIPSGoogle Earth® and other virtualglobes have only been available for adecade, yet they represent a significantparadigm shift for geoscience educa-tion. Now it is possible for students(agile and disabled) to explore theentire Earth, Moon, and Mars. InGoogle Earth®, these students canalso be exposed to real-world typeexperiences, rather than a typical‘canned laboratory’ exercise. Here aGoogle Earth® model is presentedthat provides some interactive compo-nents critical for creating excellent vir-tual field trips (VFTs, Hurst 1997;Woerner 1999; Tuthill and Klemm2002; Arrowsmith et al. 2005).

As outlined by Mogk (2011),the novice geology student is impactedby multiple factors such as distractionby irrelevant features (Goodwin 1994;Reynolds et al. 2006), spatial and scalerelations (Wilson 2001, 2002; Monteloet al. 2004), internal versus externalviews (Bryant et al. 1992), and theaffective domain (values, motivations,emotions, attitudes, etc; Krathwohl etal. 1973) all of which cause the firstfield experience to be overwhelming.A VFT is best employed as a supple-mental introduction to real field experi-ences (Bellan and Scheurman 2001;Tuthill and Klemm 2002; Arrowsmithet al. 2005; Smedley and Higgins 2005)and could potentially lessen the over-whelming nature of first field experi-ences.

Virtual globes emerged at theclose of the last millennium (Bailey2000). Gore (1998) strongly endorsedthis technology and predicted that vir-tual globes would revolutionize spatialvisualization and data analysis for geo-science research and education. Earth-

56

Co-author Scott Shipley has posted supplemental files to YouTube, collectively at:http://www.youtube.com/user/WxAzygy?ob=0&feature=results_mainSpecific files of note for this article are located on the GAC open access site:Canada GEESE Series: http://www.gac.ca/wp/?page_id=6991

GEOSCIENCE CANADA Volume 39 2012 57

browser was the first working digitalglobe application to appear in 1998,followed by other ‘first generation’ vir-tual globes until 2004. Developmentof the XML-based markup language‘Keyhole Markup Language’ (KML) byKeyhole Corporation in the early 2000smade virtual globes easily extensible.Google Earth® was released in 2005as the newly named KML-based geo-browser (after Google Inc. purchasedKeyhole Inc.). Google ensured thesuccess of Google Earth® by releasingcontrol of KML to the Open Geospa-tial Consortium in 2008, thereby estab-lishing KML as the standard for inter-active digital globes. KML is now thescripting language for all major virtualglobes such as ESRI’s Arc Explorer,Bing Maps 3-D (formerly MSN VirtualEarth), Google Earth®, and NASA’sWorld Wind (Wilson et al. 2007, 2008;Wernecke 2009; Whitmeyer et al. 2010;De Paor et al. 2010; De Paor andWhitmeyer 2011).

Google Earth® capabilitieswere greatly expanded with the adop-tion of COLLADA (COLLAborativeDesign Activity) for 3-D content mod-elling in 2006 (with version 4), andAnimation in 2007 (with version 5).COLLADA was originally developedby Sony Entertainment Corporationfor exchange of 3-D graphics content(Arnaud and Barnes 2006). It is nowan open standard maintained byKhronos Group (Khronos Group2012). Google developed SketchUp(which it recently sold to Trimble Nav-igation Inc.) as a user-friendly applica-tion for creation of 3-D content, suchas buildings. SketchUp version 8 wasused in this paper to create emergentCOLLADA models of subsurfacestructures and to animate COLLADAmodels with the Google Earth®Application Program Interface (API)and JavaScript (De Paor and Whitmey-er 2011). Emergent and animatedCOLLADA models are presented hereusing the Grotto Creek mapping proj-ect from an introductory field schooltaught at Mount Royal University, Cal-gary, Alberta, Canada (Fig. 1).

The Grotto Creek exercise(Fig. 2) was initially created for educa-tion, to guide novice geology studentswhile developing the necessary spatialcognitive abilities during their voyagetoward becoming professional geolo-

gists. Novice students and non-professionals typically experience diffi-culties when attempting to visualize 3-D geological structures with paper geo-logic maps and cross-sections (Piburnet al. 2002; Kastens and Ishikawa2006). As emphasized by Whitmeyeret al. (2010), interactive geologicalmaps in Google Earth® communicate3-D geological data more intuitivelythan the traditional format of separate

paper maps and cross-sections. Drap-ing geological maps over topographycombined with creating emergentcross-sections permits students toremove much of the cognitive barrierinvolved in trying to piece together 2-D paper maps and paper cross-sections.

Using the method of Dorde-vic (in press), multiple slider controlscan be created for the Google Earth®

Figure 1. Location of Grotto Creek map area in Google Earth®. This map area(coloured portion draped over topography) is in the McConnell Thrust of theCanadian Cordillera Front Ranges, 10 km east of Canmore, Alberta, near theTrans-Canada Highway (yellow line). The body of water in the south-central portion of this figure is the Spray Lakes Reservoir (©2012 Google, GeoEye, Terra-Metrics, and Cnes/Spot Image).

Figure 2. The Grotto Creek geological map with Fell’s cross-sections (Fell 2008)in Google Earth®. Lac Des Arcs is the body of water in the middle of the maparea. This illustrates one of the strengths of Google Earth® models, the ability tocombine different 2-D figures into one 3-D image that includes topography. Asemphasized by Whitmeyer et al. (2010), these interactive maps should be moreintuitive for inexperienced people to visualize 3-D geological structures than tradi-tional 2-D paper maps and cross-sections (©2012 Google, GeoEye, TerraMetrics,and Cnes/Spot Image).

58

API that are controlled usingJavaScript code. Here a vertical sliderwas used to create emergent cross-sec-tions, while a horizontal slider recreatesthe formation of the faults and foldsobserved in the field area. In a furtherinnovation, WxAnalyst’s WxAzygy®Transparent Interface (Shipley et al.2009, 2010) was used to create aqueryable ‘cut-away’ cross-section intothe topography. The main purpose ofthis paper is to highlight the practicaland accessible nature of GoogleEarth® when combined withSketchUp, COLLADA, and WxAzy-gy®, and the ability to create engaging3-D interactive models from geologicalmaps and cross-sections constructedby students or professionals.

THE GROTTO CREEK MAPPINGPROJECTThe Grotto Creek map area is locatedin the McConnell Thrust Sheet in theFront Ranges of the CanadianCordillera, near the Trans-CanadaHighway between Calgary and Banff,Alberta (Fig. 1). Devonian–Mississip-pian carbonate–clastic strata weredeposited on the western margin ofancestral North America and thenrepeated by folding and faulting alongthe Lac des Arcs Thrust Fault. The>440 m thick Devonian to Carbonifer-ous stratigraphic sequence in the Grot-to Creek map area progresses from theshelf margin wedge systems thatonlapped bathymetric highs in Albertaduring a sea-level rise (Southesk/Alexo

and Palliser formations, Fig. 3A, B), tothe transgressive systems tract of theExshaw Formation (Caplan and Bustin1996, 1998). This was accompanied byeutrophication of surface waters andreduced carbonate accumulation rateswith the destruction of the carbonateplatform due to changes in the paleo-ceangraphic circulation marked by theblack organic-rich shales of the lowermember of the Exshaw Formation.The bioturbated shelf siltstones andcross-stratified nearshore marine sand-stones of the upper member of theExshaw Formation record an abruptrelative sea-level fall (Caplan andBustin 1996, 1998) that continuedthrough to the development of a sec-ond carbonate platform represented bythe Banff and Livingstone formations.

This mapping project waschosen because the five mapped for-mations are important to the localpetroleum industry in the WesternCanadian Sedimentary Basin. Forexample, the Exshaw Formation (Allanand Creaney 1991; Barson et al. 2000;Manzano-Kareah et al. 2004 and multi-ple references therein) is an importantsource rock for many petroleum reser-voirs in Western Canada. The cleanand pure limestone of the Palliser For-mation is quarried in the map area forcement. Karst topography within theLivingstone Formation formed theRat’s Nest Cave system (Yonge 2001)in Grotto Mountain. The cliff-forminglimestone of the Palliser and Living-stone formations form the spectacular

landscapes that draw millions oftourists to our geological backyard.Access into this area is excellent withroad-cuts along Highway 1A and pop-ular hiking trails up the two Grottocreeks. West Grotto Creek follows oneof the syncline hinge lines, while EastGrotto Creek cuts up the trace of asmall thrust fault.

Jonathan Fell (Fell 2008) con-structed the geological map and thetwo shorter cross-sections used here(Fig. 2), during his introductory fieldschool at Mount Royal University (Cal-gary, Alberta). As per many CanadianB. Sc. geology major programs, theintroductory field school is taught fortwo weeks during the end of summerbefore the second year geology cours-es. The Grotto Creek field area repre-sents the final four-day mapping proj-ect of the field school.

Four days were spent mappingthe toe of Grotto Mountain, west andeast Grotto creeks and the road-cutsalong Highway 1A in the western por-tion of the field area (see Fig. 4 forgeographic features mentionedthroughout this paper). Two stopswere made at outcrops along thesouthern side of the Trans-CanadaHighway to include those on the mapand to examine the structural featuresof the mapping area from a distance.A stratigraphic column was measuredalong Highway 1A road-cuts in theeastern portion of the map area earlierin the field school. Two days werespent in Jura Creek (2.7 km northwest

Figure 3. Palliser Formation limestone cliffs in Google Earth®, southern portion of map area. (A) Cliff-forming limestonesof the Palliser Formation (green) occur at the steepest portion of this range, the slope shallows as the contact with theSouthesk/ Alexo Formation dolostones (purple) is crossed toward the bottom of this range. (B) Frequently, the steeper Pallis-er Formation outcrops are not vegetated, whereas the less steep outcrops of Southesk/Alexo Formation are vegetated. Unfor-tunately the image quality in Google Earth® at this location is blurry (©2012 Google, GeoEye, TerraMetrics, and Cnes/SpotImage).

A B

GEOSCIENCE CANADA Volume 39 2012 59

of the mapping area) mapping the topand bottom contacts of the Palliserand Exshaw formations. Jura Creek isthe type locality for the Exshaw For-mation (MacQueen and Sandberg1970; Barson et al. 2000) which isrecessive and eroded away within theGrotto Creek area. To complete themap and to practice structural contour-ing, the contacts were structurally contoured north and south of the out-crops.

Due to time constraints, theregion between the East Grotto Creekthrust fault and the Lac des Arcs thrust

fault was not visited (Fig. 4). To com-plete the geological map, the studentswere asked to extrapolate the geologyfrom the final mapping project area(western portion) to the eastern road-cuts. So, while this geological mapdoes include the anticline–syncline pair(West Grotto Creek) and the smallthrust fault (East Grotto Creek) addedby Simony (in Spratt et al. 1995) to theMcMechan (1995) and Price (1970)geological maps for this region, theportion between the two thrust faultsdoes not correspond to the work ofPrice (1970), McMechan (1995) and

Simony (in Spratt et al. 1995) becausethe students were not aware of thiswork. These differences are not thefocus of this paper, rather we aim tohighlight how student mapping can beenhanced and presented in an engaging3-D interactive setting using GoogleEarth®, COLLADA and WxAzygy®.

Robyn Thompson (Instruc-tional assistant at Mount Royal Univer-sity), digitized Fell’s map and cross-sections (Fell 2008), and used these tocreate an emergent model and terrainoverlay using SketchUp and KMLcode. Thompson used an early versionof the Grotto Creek model to create aYoutube video (Thompson 2011) illus-trating the difference between the >3km apparent map thickness of thegreen Palliser Formation limestone andthe ~250 m true thickness observed inthe cross-sections (Fig. 5A, B). Thisillustrates one powerful capacity forthese Google Earth® models for edu-cation purposes.

This Grotto Creek GoogleEarth® model will be used in intro-ductory geology courses to illustratethe ‘rule of Vs’ and why inclined con-tacts are not always straight lines. Theterrain overlay combined with thetopography in Google Earth® greatlyenhances what may be difficult forsome students to visualize with onlythe topographic contours on topo-graphic maps. Thompson’s Youtubevideo (Thompson 2011) will be used inthe introductory field school, andstructural geology courses to graphical-ly illustrate the difference between true

Figure 4. Features mentioned in paper. The red lines mark West and East Grottocreeks, which are part of the Grotto Creek hiking trails. PQ is the quarry wherePalliser Formation limestone is currently being mined. AQ is the now abandonedquarry where the Southesk/Alexo Formation dolostones were exposed. BQ is thenow abandoned quarry where the Banff Formation was once quarried. JC is JuraCreek, the type locality for the Exshaw Formation. Under RNC is the Rat’s NestCave system in Grotto Mountain. ERC is the eastern road-cuts, WRC the westernroad-cuts (©2012 Google, GeoEye, TerraMetrics, and Cnes/Spot Image).

Figure 5. This figure is useful for educational purposes when explaining the difference between true thickness (visible in cross-section) and apparent thickness (visible in map view; as per Thompson’s Youtube video: (Thompson 2011)). (A) The thickorange arrow represents the >3 km of apparent thickness of the green Palliser Formation limestone (©2012 Google, GeoEye,TerraMetrics, and Cnes/Spot Image). (B) In cross-section, the black arrow represents the true thickness of 250 m. Also in thisview, one can see the folding and faulting that explains the much larger apparent thickness in map view.

A B

60

and apparent thickness. The verticaland horizontal sliders will be used inthese courses to illustrate how foldsand faults can develop and how toconstruct proper cross-sections. Bothconcepts can be troublesome fornovice students.

Perhaps the most powerfulapplication of Google Earth® modelsfor the geosciences is in the creation ofinteractive components to create effec-tive virtual field trips (VFTs; Hurst1997; Woerner 1999; Tuthill andKlemm 2002; Arrowsmith et al. 2005).Most geology students are completelyoverwhelmed during the first day ofintroductory field schools due to fac-tors listed above – distraction by irrele-vant features, spatial and scale rela-tions, internal versus external views,and the affective domain. Studentresponses in Arrowsmith et al. (2005)support the conclusion that VFTsassist students in preparing for field-work and that most students felt morecomfortable during the real field expe-rience after being exposed to a VFT.These same students also mentionedthat VFTs should not replace real fieldexperiences. The Grotto CreekGoogle Earth® model presented hererepresents the first stage of creating amuch larger VFT through the Canadi-an Cordillera.

EMERGENT COLLADA MODELS ANDSTRUCTURAL RECONSTRUCTIONS

Stage Design for the Grotto CreekModelThe senior author created the longercross-section (Fig. 6) and stages for thestructural emergent model (Fig. 7)from Fell’s field school geological map(Fell 2008). As emphasized by Enkin etal. (1997, 2000), extensive mapping andstructural analysis has outlined thearchitecture of the Front Ranges(Clark 1949; Bally et al. 1966;Dahlstrom 1970; Price 1970; Price andMountjoy 1970; Monger and Price1979; Price 1994; McMechan 1995;Simony in Spratt et al. 1995), howeverthe timing, mechanism and environ-ment of Cordilleran deformationremain incompletely understood.Wheeler et al. (1974) proposed thatapproximately 140 km of shortening inthe Front Ranges occurred at a rate of~2 mm/yr between Early Cretaceous

(136–100 Ma) and Early Eocene(54–49 Ma). A rate of ~3 mm/yr forthe 30 km of movement was also esti-mated for along the McConnell Thrustbetween early Campanian to pre-Palaeocene. The main component ofmovement along the thrust faults inthe field area is proposed to haveoccurred between stages D and E (Fig.7).

Dahlstrom (1970) proposedthe ‘piggyback’ model for the ForelandBelt of the Canadian Cordillera, wheremain thrust faults develop ‘insequence’ from the hinterland towardthe foreland and from top to bottomof the accretionary wedge (Price 2001).In the Front Ranges of the ForelandBelt, there are four main SW-dipping,NW-striking thrust sheets that are,from west (oldest) to east (youngest):Bourgeau, Sulphur Mountain, Rundle,and McConnell thrust sheets.

Typically, each thrust sheet hasresistant Paleozoic carbonates overlainby recessive Triassic to Middle Jurassicmarine shales and siltstones, cappedlocally by Early Jurassic to Cretaceouscontinental, siliciclastic deposits (Price2001). The current field area is locatedin the easternmost McConnell ThrustSheet. The Late Devonian to EarlyCarboniferous formations are repeatedby the Lac des Arcs Thrust Fault in theeastern portion of the map area.

The Foreland Belt of theCanadian Cordillera is a typical thrustand fold belt (Stockmal et al. 2007)

with ‘thin-skinned’ geometry where theundeformed Paleoproterozoic crys-talline basement extends southwest-ward, from the Western Canada Sedi-mentary Basin beneath the CanadianRockies beyond the Rocky MountainTrench (Shaw 1963; Bally et al. 1966;Keating 1966; Price and Mountjoy1970; Price 1981; Cook et al. 1988).The SW-dipping, listric faults, whichflatten with depth and merge into aregional décollement above the base-ment, are the dominant structures inthe Foreland Belt (Bally et al. 1966).For the purpose of the demonstrationpresented here, only the strataobserved and measured during theMount Royal University introductoryfield school were included in the cross-sections, however the faults are impliedto be listric as suggested by Bally et al.(1966).

Dahlstrom (1970) defined a‘Foothills family’ as four structural fea-tures that include: i) late normal faults(frequently listric), ii) tear faults, iii)low-angle thrust faults, and iv) concen-tric folds underlain by a décollement.The décollement and the associatedroom problems with concentric foldswill not be observed in the cross-sec-tions because the cross-sections do notextend below the Southesk and Alexoundivided formations. Tear faults werenot observed in the map area, but low-angle thrust faults were observed.

Figure 6. Grotto Creek geological map with long cross-section line used for thestructural reconstruction (yellow line) in Figure 7 (©2012 Google, GeoEye, Terra-Metrics, and Cnes/Spot Image).

GEOSCIENCE CANADA Volume 39 2012 61

Creation of COLLADA ModelsGIMP (The GNU Image ManipulationProgram; freely distributed software at[www.gimp.org]) was used to retouchthe jpeg cross-section files (this is a

free alternative to the commercial Pho-toshop® application). An alpha layerwas added to enable transparency ofthe cross-section background. Thefiles were imported into SketchUp,

where they were scaled and aligned toa reference point. This was chosenwhere the cross-section line crossedthe Lac Des Arcs Thrust Fault. Thisreference point provides a commonlatitude, longitude, and altitude for allcross-sections used in the restoration.The cross-sections were then exportedas 3-D models in COLLADA format.Sliders were constructed to elevate andanimate the motion of the restoredcross-sections, using interactive screenoverlays, JavaScript, and the GoogleEarth® API (described in Dordevic, inpress). Moving the vertical slider willelevate the model from an anchorpoint, which is the point of origin inSketchUp at the moment of export(here at 1000 m elevation). Animationis created through the use of the hori-zontal slider, which loads in differentstages of the Grotto mapping areastructural reconstruction as texturesapplied to the side of the model. Justas with cartoon animation, a highernumber of stages would smooth outthis animation.

Adding Contours to Google Earth®Contours are important for map inter-pretation and are built into GoogleMaps but strangely not into GoogleEarth®. We devised four methods foradding contours to the Google Earth®terrain.

In the first method (Fig. 8A) ascreen shot of the region of interest istaken in Google Maps with its terrainfeature switched on. This is importedinto Google Earth® as a ground over-lay and draped over the matchingtopography. This works well becauseGoogle Maps and Google Earth®share the same DEM (if a map from adifferent source were draped onto theGoogle Earth® terrain, the resultantcontour lines might be far from hori-zontal). A disadvantage of this methodis that paths or polygons drawn withGoogle Earth® tools will be coveredby the ground overlay image (pathsand polygons cannot be given a higherKML draw order). However, if thetransparency of the image is adjusted,map features beneath become visible.This approach is the fastest and it pro-duces acceptable visual results if theregion of interest is not too large – onthe order of 10 km2 – otherwise, thecontours become too dense and the

Figure 7. Five stages of structural reconstruction. Stage (A) represents the origi-nal horizontal sediments. By Stage (B), the beds were warped due to the change toa convergent setting off the west coast of North America. With further foldingand the beginning of faulting, movement along the East Grotto Creek Thrust Faultcut through the Southesk/Alexo and Palliser formations – Stage (C). By Stage (D),most of the movement along the thrust faults was progressing at rates between 2and 3 mm/yr. What is seen today is represented by Stage (E).

A

B

C

D

E

62

image file becomes very large.The second method (Fig. 8B)

involves manually tracing the contoursusing the Google Earth® ‘Add Path’tool. Here, the first step is to create arectangular polygon similar in size tothe region of interest. When the alti-tude mode of this rectangle is set to‘absolute’ (the height is 1800 m in Fig.8B) the contact with topography canbe seen (indicated with the white boldline) and traced out by hand. Thisprocess is then repeated for all thedesired altitudes.

For the third method, theDEM from Google Earth® is import-ed into SketchUp using its ‘Get Cur-rent View’ tool (Fig. 8C). Once theDEM is imported, by intersecting itwith a set of equally spaced horizontalplanes, contours can be produced andexported back into Google Earth® asa 3-D model. Two disadvantages of

this approach include the limit of ~4km2 for the area that can be importedand the large file size.

The fourth method (Fig. 8D)is a derivative of the second method.Instead of tracing the intersection ofthe rectangular polygons with theDEM, the rectangles are made 90%transparent (10% opaque) and stackedby copying them while increasing theiraltitudes. Then a solid colour base rec-tangle is clamped to the ground underthis stack. Here the base colour is blueand the altitude ascends from bluetoward white because the base layer iscovered with more rectangular planesat each elevation. This method can becompleted fairly quickly and canextend over large regions (subject tothe effect of Earth’s curvature).

Opportunities Provided by COLLADA ModellingThe above discussion demonstrates thepotential for Google Earth® to beused for visualizing complex geologicstructures and processes. Additionalsteps can be taken to show under-ground geologic structures in situ(meaning in their original locations) asshown in Figure 9. These KML objectswere created using image processingtools to 1) capture a surface image tile,and 2) remove sections of that tilewhich would obstruct the user’s viewof the cross-sections. In version 5 ofthe Google Earth® desktop applica-tion, surface tiles are made transparentby highlighting ‘Primary Database’ inthe Layers sidebar and then movingthe transparency slider all the way tothe left (see ‘A’ in Figure 10). InGoogle Earth® version 6.2, the pri-mary database is always opaque;

Figure 8. (A) Contours imported as ground overlay from Google Maps terrain snapshot into Google Earth®. Transparency isset at 50%. (B) Contours (thick white line) constructed by manually tracing out the intersection of an elevated rectangle (blueplane) with the DEM. (C) Contours drawn in SketchUp by intersecting an imported DEM with parallel horizontal planes. (D) Astack of transparent rectangular planes over a solid blue basal plane which is clamped to the ground (©2012 Google, GeoEye,TerraMetrics, and Cnes/Spot Image).

A B

C D

GEOSCIENCE CANADA Volume 39 2012 63

instead the Radar item in the WeatherLayer must be made transparent. Thecut-away tile is reinserted as a groundoverlay and is draped on the terrain.An obvious next step would use theline drawn along the surface (the yel-low line in Fig. 9) to define the cross-section(s) and perform the surfaceimage tile cut-away. The image tile cutis also a split, where the two resultingsections of the ground overlay can beused to view the cross-sections in situfrom opposing directions.

Now that the cross-sectionshave been established in situ, we canapply WxAnalyst’s WxAzygy® Tools tointerrogate the cross-sections as shownin Figure 10. This add-on is available(WxAnalyst 2012; Shipley 2012a) forGoogle Earth® versions running onMicrosoft Windows which supportsthe GE COM API. WxAnalyst’sWxAzygy® Transparent Interface(Shipley et al. 2009, 2010) interpretsuser mouse clicks on Google Earth®to calculate the corresponding positionof that click on a COLLADA surface.The user selects the ‘focus’ COLLA-DA layer that is to be interrogated thenclicks on that COLLADA surface toobtain the coordinates of the pointselected on that surface. This methoduses the GE COM API to establish theuser’s Point Of View (POV) in GoogleEarth®, so that the resulting coordi-nates are independent of GoogleEarth® pan, zoom, rotation, or tilt.The tool also provides a visualizationof the COLLADA models using iconsand lines which assists the user in iden-tifying the layer that has been selected.The example in Figure 10 shows layerKB_B in situ selected as the FocusLayer with the geoCursor tool in COL-LADA triangle [p0,p1,p2]. WxAna-lyst’s WxAzygy® interface shows theresults of the click in theSouthesk/Alexo stratum at a height of1000 m above the Earth Geoid, whichcombined with model displacement of200 m at that location corresponds toa depth of approximately 200 m belowthe surface. The interface also providesthe coordinates of the point on thedraped image [i=2056, j=467] and itscolour at that point in hexidecimal[A=80h, R=ABh, G=39h, B=93h]. An‘External’ command is also visible inthe cursor for access to additionalfunctions defined by the COLLADA

Figure 9. Cross-sections in situ, meaning that the subterranean structures areshown in their original locations. In this case, the Google Earth® primary databaseis removed from view and a ‘cut-away’ surface tile is inserted in its place as a<GroundOverlay> with terrain exaggeration set to one. The ‘cut-away’ can beachieved automatically using the line (yellow) which indicates the location of thesecross-sections along the surface (©2012 Google, GeoEye, TerraMetrics, andCnes/Spot Image).

Figure 10. WxAzygy® Tools operating in 3-D on the selected COLLADA cross-section. The WxAzygy® interface provides the ability to ‘point and click’ on aCOLLADA surface without leaving Google Earth®. This add-on uses the GECOM API to interpret a user click event on the GE surface as an intersection witha COLLADA model along the user’s line of sight. Note that the coordinates of theselected point provide the real world location {long, lat} and height {metres}above the geoid, which, in this example, is below ground level. The ability toremove the Primary Database (A) is currently supported only in Google Earth®5.x and 6.0. In version 6.2, the Radar item in the Weather folder must be madetransparent. User highlights the COLLADA surface (B), then selects that surface asthe Focus Layer (C) and uses the geoCursor (D) to click on that surface (©2012Google, GeoEye, TerraMetrics, and Cnes/Spot Image).

64

model, as discussed in the WxAzygy®User Manual (WxAnalyst 2010; Shipley2012b, c, d, e, and f).

The WxAzygy® TransparentInterface was developed by WxAnalystunder NASA SBIR Phase 1 funding bythe NASA Langley Research Center.Phase 3 funding was subsequently pro-vided by NOAA’s National ClimaticData Center (NCDC) to provide afreeware version of the tools online(WxAnalyst 2012; Shipley 2012a). Asof this publication, the GE COM APIis available in GE versions 5.x through6.2 for Microsoft Windows. Googlehas announced that the GE COM APIhas been deprecated and may nolonger be supported. Additional func-tions developed beyond the freewareversion of the tools include the abilityto edit COLLADA models, to drawfreehand on COLLADA model sur-faces, and to change colours of theimages draped on the COLLADAmodels, with all operations performedin situ without leaving Google Earth®(Shipley 2012a, b, c, d, and e for moredetailed user instructions).

DISCUSSIONThe ultimate goal for future GoogleEarth® models expanding upon theGrotto Creek model presented here isto create a series of 3-D interactiveVFTs through the Canadian Cordillera.As mentioned by Whitmeyer et al.(2010), these interactive GoogleEarth® geological maps are more intu-itive for non-professionals than tradi-tional paper maps and cross-sections.It is also hoped that VFTs employed asa supplemental introduction to realfield experiences will greatly decreasethe overwhelming nature of first fieldexperiences (as summarized in Mogk,2011). In the future, this pilot GoogleEarth® model will be expanded toinclude the extensive mapping andstructural analysis outlined by Clark(1949), Balley et al. (1966), Price(1970), Dahlstrom (1970), Price andMountjoy (1970), Monger and Price(1979), Price (1994), McMechan(1995), and Simony in Spratt et al.(1995).

Google Earth® and SketchUpwere used because they are well inte-grated with each other, supported bymultiple platforms, relatively simple touse, widely available, and free. For

most geosciences tasks, SketchUp issufficiently powerful; however it can-not wrap a texture around a sphere ordraw a line on a curved surface. Thereare free Ruby plug-ins available forSketchUp that enable these tasks.Other software is available that couldcreate similar models to the one pre-sented here, such as Gplates fromEarthbyte, Layerscape from Microsoft,Move software from Midland Valley,NASA World Wind, GEON, OpenTopography, etc. SketchUp was usedbecause the others are either not free,not cross platform, or have a steeplearning curve. However, with the saleof SketchUp to Trimble NavigationInc., the future of free SketchUp isunclear.

Custom sliders were createdusing COLLADA, JavaScript, andGoogle Earth® APIs. Once again,these were used because they are wellintegrated, supported by multiple plat-forms, relatively simple to use, widelyavailable, and free. While there aremany other possibilities for creating 3-D models such as LightWave 3-D,Maya, 3ds Max, many of these areheavy weight modelling applicationswith steep learning curves that wouldbe underused for geoscience model-ling.

Finally, Google Earth® Layerswere adjusted and manipulated usingWxAzygy® to create a queryable ’cut-away’ transparent view of in situ geolo-gy. One powerful application ofWxAzygy® is that it provides a con-venient platform for creating COLLA-DA and KML directly in GoogleEarth®. Unfortunately, GoogleEarth® recently announced that it willno longer support the GE COM API,which makes the creation of transpar-ent tiles much more difficult.

CONCLUSIONSGoogle Earth® and other virtualglobes present excellent media foranalysis of surface data such as geolog-ical maps draped over topography.The addition of Google Earth® API,KML, and COLLADA has greatlyenhanced the 3-D visualization capaci-ty for Google Earth® models. COL-LADA models retain azimuthal orien-tations (unlike placemark icons), whileKML permits users to add custom datain a variety of formats, and the Google

Earth® API combined with JavaScriptenables users to manipulate COLLA-DA model behaviour (De Paor andWhitmeyer 2011).

Another significant applicationfor Google Earth® models is the abili-ty to immediately ‘fly and explore’ fea-tures embedded within the text ofpapers such as this. Regardless of thepotential for assisting students inacquiring the necessary spatial cogni-tive skills during their voyage fromnovice student to professional geolo-gist, most people are engaged by thisincredible technology.

Further Google Earth® appli-cations introduced here include the useof vertical sliders to create emergentcross-sections capable of slidingthrough topography, horizontal slidersthat reconstruct a structural recon-struction model, and queryable ‘cut-away’ cross-sections embedded intotopography.

ACKNOWLEDGEMENTSThis paper is dedicated to the memoryof Dr. Cynthia Riediger. The JuraCreek Exshaw Formation type localityoutcrop was informally named ‘Cindy’soutcrop’ during the MRU field school2010. Amanda Taylor’s digitizing ofthe cross-sections for the structuralreconstruction model was greatlyappreciated as is her input on potentialeducational applications of theseGoogle Earth® models. In addition toJonathan Fell, the numerous fieldschool students at Mount Royal Uni-versity and the University of Calgaryhave been instrumental in illustratingthe need for Google Earth® modelssuch as presented here to augmenttheir educational experience.

The custom slider construc-tion was partially supported by NSFTUES 1022755, NSF GEO 1034643,and a 2010 Google Faculty ResearchAward supporting Dordevic. Anyopinions, findings, and conclusions orrecommendations expressed in thispaper are those of the authors and donot necessarily reflect the views of theNational Science Foundation orGoogle Inc.

WxAnalyst’s WxAzygy®Transparent Interface was developed inpart under NASA SBIR Phase I Con-tract NNX09CF28P and NOAANational Climatic Data Center SBIR

GEOSCIENCE CANADA Volume 39 2012 65

Phase III funding under ContractNEEF4100-10-18847. WxAzygy® is aregistered trademark of WxAnalyst,Ltd.

The manuscript benefittedgreatly from reviews by Glen Stockmaland Fraser Keppie. This paper wouldnot have been possible without thegreatly appreciated coordination andeditorial support of Declan De Paor.

REFERENCESAllan, J., and Creaney, S. 1991, Oil families

of the Western Canada Basin: Bulletinof Canadian Petroleum Geology, v. 39,p. 107-122.

Arnaud, R., and Barnes, M.C., 2006, COL-LADA® - Sailing the Gulf of 3DDigital Content Creation, KhronosGroup Open Standard: A.K. Peters,Ltd., Wellesley, Massachusetts, 240 p.

Arrowsmith, C., Counihan, A., andMcGreevy, D., 2005, Development ofa multi-scaled virtual field trip for theteaching and learning of geospatialscience: Development using Informa-tion and Communivation Technology,v. 1, p. 42-56.

Bailey, J., 2000, Virtual globes, in Waif, B.,ed., Encyclopedia of Geography, SagePublications Inc., Los Angeles, Cali-fornia, 3528 p.

Bally, A.W., Gordy, P.L., Stewart, G.A.,1966, Structure, seismic data, and oro-genic evolution of the southern Cana-dian Rockies: Bulletin of CanadianPetroleum Geology, v. 14, p. 337-381.

Barson, D., Bartlett, R., Hein, F., Fowler,M., Grasby, S., and Riediger, C., eds.,2000, Hydrogeology of Heavy Oil andTar Sand Deposits: Water Flow andSupply, Migration and Degradation, v.14, 162 p.

Bellan, J.M., and Scheurman, G., 2001,Actual and virtual field trips; makingthe most of field trips, Chapter 14, inStevens, R.L., ed., Homespun: Teach-ing Local History in Grades 6-12,Heinemann, Portsmouth, New Hamp-shire.

Bryant, P.E., Tversky, B., and Franklin, N.,1992, Internal and external spatialframeworks for representing describedscenes: Journal of Memory and Lan-guage, v. 31, p. 74-98.

Caplan, M.L., and Bustin, R.M., 1996, Fac-tors governing organic matter accu-mulation and preservation in a marinepetroleum source rock from theUpper Devonian to Lower Carbonif-erous Exshaw Formation, Alberta:Bulletin of Canadian Petroleum Geol-ogy, v. 44, p. 474-494.

Caplan, M.L., and Bustin, R.M., 1998, Sedi-mentology and sequence stratigraphyof Devonian-Carboniferous strata,

southern Alberta: Bulletin of Canadi-an Petroleum Geology, v. 46, p. 487-314.

Clark, L.M., 1949, Geology of RockyMountain Front Ranges near BowRiver, Alberta: Bulletin of the Ameri-can Association of Petroleum Geolo-gists, v. 33, p. 614-633.

Cook, F.A., Green, A.G., Simony, P.S.,Price, R.A., Parrish, R.R., Milkereit, B.,Gordy, P.L., Brown, R.L., Coflin, K.C.,and Patenaude, C., 1988, LITHO-PROBE seismic reflection structure ofthe southeastern Canadian Cordillera:initial results: Tectonics, v. 7, p. 157-180.

Dahlstrom, C.D.A., 1970, Structural geolo-gy in the eastern margin of the Cana-dian Rocky Mountains: Bulletin ofCanadian Petroleum Geology, v. 18, p.332-406.

De Paor, D.G., Whitmeyer, S., and Gobert,J., 2010, The next phase of theGoogle Earth® revolution: Geologi-cal Society of America Annual Con-ference Program and Abstracts, Paper172-1.

De Paor, D.G., and Whitmeyer, S.J., 2011,Geological and geophysical modelingon virtual globes using KML, COL-LADA, and JavaScript: Computersand Geosciences, v. 37, p. 100-110.

Dordevic, M.M., in press, Designing inter-active screen overlays to enhanceeffectiveness of Google Earth® Geo-science resources, Geological Societyof America Special Paper 492.

Enkin, R.J., Osadetz, K.G., Wheadon, P.M.,and Baker, J., 1997, Paleomagneticconstraints on the tectonic history ofthe Foreland Belt, southern CanadianCordillera: preliminary results: Cana-dian Journal of Earth Sciences, v. 34,p. 260-270.

Enkin, R.J., Osadetz, K.G., Baker, J., andKisilevsky, D., 2000, Orogenic remag-netizations in the Front Ranges andInner foothills of the southern Cana-dian Cordillera: Chemical harbingerand thermal handmaiden of Cordiller-an deformation: Geological Society ofAmerica Bulletin, v. 112, p. 929-942.

Fell, J., 2008, Grotto Creek Geological Mapand Cross-sections; Grotto Creek andArea Stratigraphy and StructuralStudy: Introductory Field School Map-ping Project, Mount Royal University,Calgary, AB.

Goodwin, C., 1994, Professional Vision:American Anthropologist, v. 96, p.606-633.

Gore, A., 1998, The Digital Earth: Under-standing our Planet in the 21st centu-ry, California Science Center, LosAngeles, California;[http://www.isde5.org/al_gore_speech.htm#], accessed 6 July, 2012.

Hurst, S.D., 1997. Use of “Virtual” FieldTrips in teaching introductory geolo-gy: Computers and Geosciences, v. 24,p. 653-658.

Kastens, K.A., and Ishikawa, T., 2006, Spa-tial thinking in the geosciences andcognitive sciences: A cross-disciplinarylook at the intersection of the twofields, in Manduca, C.A., and Mogk,D.W., eds., Earth and Mind: HowGeologists Think and Learn about theEarth: Geological Society of AmericaSpecial Paper, 413, p. 53-76.

Keating, J.F., 1966, Exploration in theCanadian Rockies and Foothills: Cana-dian Journal of Earth Sciences, v. 3, p.713-723.

Khronos Group, 2012, COLLADA - 3DAsset Exchange Schema,[http://www.khronos.org/collada/],accessed 6 July, 2012.

Krathwohl, D.R., Bloom, B.S., and Masia,B.B.,1973, Taxonomy of EducationalObjectives, the Classification of Edu-cational Goals. Handbook II: Affec-tive Domain, David McKay Co., Inc.,New York, New York.

MacQueen, R.W., and Sandberg, C.A.,1970, Stratigraphy, age and inter-regional correlation of the ExshawFormation, Alberta Rocky Mountains:Canadian Society of Petroleum Geol-ogists Bulletin, v. 18, p. 32-66.

Manzano-Kareah, B.K., Stevenson, P.M.,and Riediger, C.L., 2004, Geochem-istry of medium gravity crude oils insouthern Alberta and northern Mon-tana: Bulletin of Canadian PetroleumGeology, v. 52, p. 105-120.

McMechan, M.E., 1995, Geology, RockyMountain Foothills and Front Rangesin Kananaskis country, Alberta: Geo-logical Survey of Canada, Map 1865A,scale 1:100,000.

Mogk, D., 2011, Get ready, get set, go…ona field trip, in Wirth, K.R., Goodge, J.,Perkins, D., and Stokes, A., TeachingMineralogy, Petrology and Geochem-istry in the Field: GSA Field Guides, v.24, p. 477-482. doi:10.1130/2011.0024(23)

Monger, J.W.H., and Price, R.A., 1979,Geodynamic evolution of the Canadi-an Cordillera—progress and prob-lems: Canadian Journal of Earth Sci-ences, v. 16, p. 771-791.

Montelo, D.R., Waller, D., Hegarty, M.,Richardson, A.E., 2004, Spatial memo-ry of real environments, virtual envi-ronments, and maps, in Allen, G.L.,and Haun, D., eds., Human SpatialMemory: Remembering Where,Lawrence Erlbaum Associates, Mah-wah, New Jersey, p. 251-285.

Piburn, M.D., Reynolds, S.J., Leedy, D.E.,McAuliffe, C.M., Burk, J.P., and John-son, J.K., 2002, The Hidden Earth:

66

Visualization of Geologic Featuresand their Subsurface Geometry:National Association for Research inScience Teaching, p. 1-48.

Price, R.A., 1970, Geology, Canmore, easthalf, Alberta: Geological Survey ofCanda, Map 1265A, 1:50,000 scale.

Price, R.A., 1981, The Cordilleran forelandthrust and fold belt in the southernCanadian Rocky Mountains, inMcClay, K.R., Price, N.J., eds., Thrustand Nappe Tectonics, The GeologicalSociety of London, Special Publica-tion 9, p. 427-448.

Price, R.A., 1994, Cordilleran tectonicsand the evolution of the WesternCanada Sedimentary Basin, inMossop, G., and Shetsen, I., comp.,Geological atlas of the Western Cana-da Sedimentary Basin, Canadian Socie-ty of Petroleum Geologists and Alber-ta Research Council, Calgary andEdmonton, p. 13-24.

Price, R.A., 2001, An evaluation of modelsfor the kinematic evolution of thrustand fold belts: structural analysis of atransverse fault zone in the FrontRanges of the Canadian Rockies northof Banff, Alberta: Journal of Structur-al Geology, v. 23, p. 1079-1088.

Price, R.A., and Mountjoy, E.W., 1970,Geologic structure of the CanadianRocky Mountains between Bow andAthabasca Rivers—a progress report,in Wheeler, J.O., ed., Structure of theSouthern Canadian Cordillera, Geo-logical Association of Canada, SpecialPaper 6, p. 7-25.

Reynolds, S.J., Piburn, M.D., Leedy, D.E.,McAuliffe, C.M., Birk, J.P., and John-son, J.K., 2006, The HiddenEarth–Interactive computer-basedmodules for geosciences learning, inManduca, C., and Mogk, D., eds., Earthand Mind: How Geoscientists Thinkand Learn about the Earth, GeologicalSociety of America Special Publica-tion, v. 413, p. 171-186.

Shaw, E.W., 1963, Canadian Rockies–orien-tation in space and time, in Childes,O.E., ed., Backbone of the Americas,American Association of PetroleumGeologists, Tulsa, p. 231-242.

Shipley, S.T., Cantrell, S., and Peterlin, A.,2009, GIScience Operations with Vir-tual Globes: American GeophysicalUnion Fall Meeting, Paper IN22A-05.Slides and full video presentationavailable at[http://wxanalyst.com/WxAzygy/],accessed 6 July, 2012.

Shipley, S.T., Cantrell, S., and Peterlin, A.,2010, WxAzygy® Transparent Inter-face for Google Earth®, AmericanMeteorological Society Annual Meet-ing, IIPS Paper J6.1, Atlanta, Georgia,[http://ams.confex.com/ams/90annu-

al/techprogram/paper_157157.htm],accessed 6 July, 2012.

Shipley, S.T., 2012a, Grotto Creek demofor Google Earth® - How to down-load and install WxAzgyg® “FairUse” software;[http://www.youtube.com/watch?v=MLzblTU3zy4], accessed 6 July, 2012.

Shipley, S.T., 2012b, 20120329 GrottoCreek demo - Using WxAzygy® onSample Data,[http://www.youtube.com/watch?v=eXnVZ5qiejA], accessed 6 July, 2012.

Shipley, S.T., 2012c, 20120328 GrottoCreek demo - shows how to removethe Primary Database in GoogleEarth® 6.2 for work underground,[http://www.youtube.com/watch?v=psiUYgeJ9NE], accessed 6 July, 2012.

Shipley, S.T., 2012d, How to create Figure10 in Boggs et al. (2012) GeoscienceCanada (Google Earth® 5.2),[http://www.youtube.com/watch?v=ulER7AxkEGs], accessed 6 July, 2012.

Shipley, S.T., 2012e, Grotto Creek -WxAzygy® Sample Data; instructionsand links to the KML used to createFigures 9 and 10,[http://wxanalyst.com/WxAzygy/Grotto_Creek/], accessed 6 July, 2012.

Shipley, S.T., 2012f, How to work under-ground: Supplement to Boggs et al.(2012) in Geoscience Canada, Vol 39,[http://wxanalyst.com/WxAzygy/Grotto_Creek/WorkingUnderground.pdf],accessed 6 July, 2012.

Smedley, T.M., and Higgins, K., 2005, Vir-tual technology: Bringing the worldinto the special education classroom,Intervention in School and Clinic, v.41, p. 114-119.

Spratt, D.A., Simony, P.S., Price, R.A., andMacKay, P.A., 1995, Fault-relatedFolding in the Foothills and FrontRanges of Southern Alberta: GSAPenrose Conference, Field TripGuidebook, 96 p.

Stockmal, G.S., Beaumont, C., Nguyen, M.,and Lee, B., 2007, Mechanics of thin-skinned fold an thrust belts; insightsfrom numerical models, in Sears, J.W.,Harms, T.A., and Evenchick, C.A.,eds., Whence the Mountains? Inquiriesinto the Evolution of Orogenic Sys-tems: A Volume in Honor of Ray-mond A. Price. Geological Society ofAmerica Special Paper, v. 433, p. 63-98.

Thompson, R., 2011, Grotto Creek Learn-ing Object using Comtasio Studio, AYoutube video created for MDDE610,a course at Athabasca University,[http://www.youtube.com/watch?v=3RzF4y4A7-w], accessed 6 July, 2012.

Tuthill, G., and Klemm, E.B., 2002, Virtualfield trips: Alternatives to actual fieldtrips: International Journal of Instruc-

tional Media, v. 29, p. 453-468.Wernecke, J., 2009, The KML Handbook:

Geographic Visualization for the Web:Addison-Wesley, Upper Saddle River,New Jersey, 368 p.

Wheeler, J.O., Charlesworth, H.A.K., Mon-ger, J.W.H., Muller, J.E., Price, R.A.,Reesor, J.E., Roddick, J.A., and Simo-ny, P.S., 1974, Western Canada, inSpencer, A.M., ed., Mesozoic-CenozoicOrogenic belts–Data for OrogenicStudies: Special Publication of theGeological Society of London, 4, p.809.

Whitmeyer, S.J., Nicoletti, J., and De Paor,D.G., 2010, The digital revolution ingeologic mapping: GSA Today, v. 20,p. 4-10.

Wilson, M., 2001, The case for sensorimo-tor coding in working memory: Psy-chometric Bulletin and Review, v. 8, p.44-57.

Wilson, M., 2002, Six views of embodiedcognition: Psychometric Bulletin andReview, v. 9, p. 625-636.

Wilson, T., Burggraf, D., Lake, R., Patch, S.,McClendon, B., Jones, M., Ashbridge,M., Hagemark, B., Wernecke, J., andReed, C., 2007, KML 2.2–an OGCbest practice Open Geospatial Con-sortium, Document OGC 07-113rl,153 p.[http://portal.opengeospatial.org/files/?artifact_id=23689]; accessed 6 July,2012.

Wilson, T., Burggraf, D., Lake, R., Patch, S.,Martell, R., McClendon, B., Jones, M.,Ashbridge, M., Hagemark, B., Wer-necke, and J., Reed, C., 2008. OGC®KML Open Geospatial Consortium,Document OGC 07-147r2, 251 p.[http://www.opengeospatial.org/stan-dards/kml]; accessed 6 July,, 2012.

Woerner, J.J., 1999, Virtual field trips in theEarth Science classroom, Preparationand Classroom Applications of Virtu-al Field Trips Workshop,[http://www.eric.ed.gov/PDFS/ED446901.pdf], accessed 6 July, 2012.

WxAnalyst, 2010, WxAzygy® v21r User’sManual, available online at[http://wxanalyst.com/WxAzygy/WxAzygy_Users_Guide_v21r.pdf],accessed 6 July, 2012.

WxAnalyst, 2012, freeware version of theWxAzygy® Transparent Interface,made available under NOAA/NCDCSBIR Phase III Contract, availablethrough portal at[http://wxanalyst.com/WxAzygy/],accessed 6 July, 2012.

Yonge, C.J., 2001, Under Grotto Mountain;Rat’s Nest Cave: Rocky MountainBooks, Surrey, BC, 144 p.

Received February 2012Accepted as revised August 2012