1: \documentclass[preprint2]{aastex}
2:
3: \shorttitle{Introducing Sky in Google Earth}
4: \shortauthors{Scranton et al.}
5: \begin{document}
6:
7: \title{Sky in Google Earth:\\The Next Frontier in Astronomical\\Data
8: Discovery and Visualization\\{\tt http://earth.google.com/sky/}}
9:
10: \author{Ryan Scranton\altaffilmark{1}, Andrew Connolly\altaffilmark{2,1},
11: Simon Krughoff\altaffilmark{2}, Jeremy Brewer\altaffilmark{3,1},
12: Alberto Conti\altaffilmark{4}, Carol Christian\altaffilmark{4},
13: Craig Sosin\altaffilmark{1}, Greg Coombe\altaffilmark{1},
14: Paul Heckbert\altaffilmark{1}}
15: \email{scranton@google.com,ajc@astro.washington.edu}
16:
17: \altaffiltext{1}{Google, Inc.}
18: \altaffiltext{2}{University of Washington, Seattle, WA 98195}
19: \altaffiltext{3}{University of Pittsburgh, Pittsburgh, PA 15260}
20: \altaffiltext{4}{Space Telescope Sciences Institute, Baltimore, MD 21218}
21:
22: \begin{abstract}
23: Astronomy began as a visual science, first through careful
24: observations of the sky using either an eyepiece or the naked
25: eye, then on to the preservation of those images with photographic
26: media and finally the digital encoding of that information via
27: CCDs. This last step has enabled astronomy to move into a fully
28: automated era -- where data is recorded, analyzed and interpreted
29: often without any direct visual inspection. {\it Sky} in Google Earth
30: completes that circle by providing an intuitive visual interface to
31: some of the largest astronomical imaging surveys covering the full
32: sky. By streaming imagery, catalogs, time domain data, and ancillary
33: information directly to a user, {\it Sky} can provide the general public as
34: well as professional and amateur astronomers alike with a wealth of
35: information for use in education and research. We provide here a brief
36: introduction to {\it Sky} in Google Earth, focusing on its extensible
37: environment, how it may be integrated into the research process and
38: how it can bring astronomical research to a broader community. With an
39: open interface available on Linux, Mac OS X and Windows, applications
40: developed within {\it Sky} are accessible not just within the Google
41: framework but through any visual browser that supports the Keyhole
42: Markup Language. We present {\it Sky} as the embodiment of a virtual
43: telescope.
44: \end{abstract}
45:
46: \keywords{Astrophysical Data, Data Analysis and Techniques, Tutorials}
47:
48: \section{Introduction}
49:
50: The purpose of this paper is to provide an introduction to {\it Sky} in
51: Google Earth\footnote{\tt http://earth.google.com} -- describing the data
52: used in its creation, why certain features appear the way they do and, most
53: importantly, how astronomers can use {\it Sky} to create, explore, and share
54: their data with a broad community. We will generally avoid technical details
55: about the underlying mechanisms for serving the imagery and instead focus on
56: the operations required to place catalogs, images and educational tools within
57: {\it Sky} and how to serve these data sets efficiently across the web. This
58: paper should be viewed as a starting point for the reader's interaction with
59: {\it Sky} and the beginning of a discussion about what can be accomplished
60: with this new tool.
61:
62: Astronomical data is deceptively complex. At first glance, it is
63: merely the result of pointing a telescope at the sky and recording
64: what you see. In reality, however, a scientifically useful
65: description of a single observation requires knowledge of when it was
66: taken, from where, over what wavelengths, in what conditions, covering
67: what area, and so on. Combined, the underlying images and associated
68: meta-data provide a data set that is not just esthetically pleasing
69: but is also scientifically meaningful. The goal of {\it Sky} is to create a
70: general framework that will enable users to access images, catalogs
71: and any associated meta-data across the full sky in a seamless manner
72: \citep{2001Sci...293.2037S}.
73: It will serve optical, infrared, x-rays, ultraviolet and radio images,
74: enable overlays of these images (with transparency), catalogs and
75: ancillary data associated with the underlying images, static,
76: time-varying and transient data. In short, it can deliver a view of the
77: night sky across the full electromagnetic spectrum to a user in an
78: efficient and scalable manner.
79:
80: \section{Design}
81:
82: Before discussing how to add your own data to {\it Sky}, we begin with a
83: description of the two principal components of {\it Sky}: the {\it basemap}
84: and the overlays. The first is simply the set of images laid
85: out on the celestial sphere to match their positions on the sky. The
86: second is the collection of lines, placemarks, and images that can be overlaid
87: on the base imagery and the annotations associated with the basic visual data.
88: Below we will describe the contents of both pieces, along
89: with some discussion of the design decisions that went into their
90: development.
91:
92: \subsection{Imagery}
93:
94: The core of Google Earth is a huge RGB color image pyramid projected on a
95: sphere, served via the internet by Google, with client software that runs on
96: each user's computer. This client takes advantage of graphics hardware in
97: their computer to display a window into the huge image database with real time,
98: continuous panning and zooming. Data is stored and served in such a way that
99: redisplay is as smooth as possible, giving the user the illusion that the full
100: detail of all the images is resident locally on their computer when, in fact,
101: only a small portion of it is cached there. {\it Earth} was created for
102: imagery of a sphere from the outside, but for {\it Sky} we reversed that
103: perspective, using the same infrastructure to serve images of space viewed
104: from inside the celestial sphere.
105:
106: As with Google Earth, the basemap in {\it Sky} provides a full-sphere view
107: of the sky by taking images with different resolutions and depths from
108: a range of sources, registering them to a common coordinate system and
109: rendering them as the user's vantage point sweeps across the sky. For
110: {\it Sky} there are three primary imagery sources: the Digitized Sky Survey
111: (DSS; \cite{DSS}), the Sloan Digital Sky Survey (SDSS; \cite{SDSS}) and images
112: from the Hubble Space Telescope (HST)\footnote{\tt http://www.stsci.edu}. The
113: DSS is derived from plate scans of the photographic plates from the Second
114: Palomar Sky Survey (POSSII; \cite{POSSII}) and the UK Schmidt Southern Sky
115: Survey in the $J$ and $F$ passbands. The digitization of the photographic
116: plates was undertaken by the Space Telescope Science Institute \citep{DPOSS}
117: and contains a total of 1788 plates in the Northern and Southern hemispheres.
118: In the north Galactic cap, the DSS images are replaced by data from the Sixth
119: Data Release of the Sloan Digital Sky Survey (SDSS), which covers $\sim$8000
120: square degrees and contains false color images generated from the $g$, $r$,
121: and $i$ passbands. Finally, there are 130 high resolution images drawn from
122: the Hubble Space Telescope which cover some of the most
123: interesting regions on the sky.
124:
125: The union of these data sets covers the full sky. The underlying
126: imagery used in {\it Sky} resides in a lat/long projection \citep{HANDBOOK}.
127: This results in substantial distortion at the poles even when re-projecting
128: onto the sphere. Thus, for regions within five degrees of the pole we
129: replace the original images with a lower resolution view of the sky
130: derived from the Tycho II catalog \citep{Tycho2}. These derived images
131: contain stars to a depth of $B \sim 12$ and each star is represented
132: by a Gaussian, scaled and colored based on the star's magnitude and $B-V$
133: color.
134:
135: \subsection{Data Registration and False Color Images}
136:
137: The underlying projection and registration of images in {\it Sky} is based
138: on the technology used in Google Earth. This provides a mature
139: visualization platform on which to develop {\it Sky}, a very large user
140: base, a simple but extensible interface and a well-defined and on-going
141: support and development mechanism. It does, however, lead to a small
142: number of trade-offs -- related to the way geospatial data is served -- that
143: were made in the course of adapting the system to serve astronomical data.
144:
145: The first of these concerns the bounds used for geodetic coordinates
146: ($GEO$), which are different from those used in Equatorial coordinates
147: ($EQ$). The geodetic coordinates range in longitude from -180 degrees
148: West to +180 degrees East. Hence, there is a simple translation that
149: must be made from the latter to the former:
150: \begin{eqnarray}
151: RA_{\rm GEO} &=& RA_{\rm EQ} - 180^\circ \nonumber \\
152: DEC_{\rm GEO} &=& DEC_{\rm EQ}
153: \end{eqnarray}
154: A second issue arises due to the fact that the Earth is an oblate
155: spheroid which will be discussed in \S\ref{sec:image_overlays}.
156:
157: All images to be served through {\it Sky} were re-projected from their
158: native tangent plane (or gnomic) projections onto the aforementioned
159: lat/long projection. SDSS images were taken from jpeg images
160: in the SDSS
161: archive\footnote{\tt http://www.sdss.org/dr6/data/das\_users\_guide\_dr2}
162: created as part of the standard SDSS reduction pipeline. The color of the SDSS
163: images was derived from the $g$, $r$ and $i$ passbands using the color
164: transformation proposed by \cite{LuptonRGB}. HST press release images were
165: taken from high resolution TIFF images created by the STScI OPO
166: group\footnote{\tt http://hubblesite.org}. For the DSS data, no color images
167: were readily available that matched the color range of the SDSS data. Color
168: images were, therefore, generated from the original FITS format images
169: (allowing for the effects of differential chromatic abberation) in the $J$
170: and $F$ bands and calibrating the photographic plates such that their stellar
171: locus matched that of the SDSS data. As with Google Earth, the basemap is
172: not static and we anticipate continued improvements as we refine our
173: treatment of the current imagery and new large-scale datasets become available.
174:
175: \subsection{Overlays and Annotation}
176:
177: As mentioned above, the built-in overlays provide a means to annotate
178: the basemap imagery. In its initial release {\it Sky} utilizes this
179: functionality to provide a simple and intuitive introduction to
180: objects that are visible to the general public and amateur
181: astronomers. For example, we outline the 88 IAU constellations
182: (Figure~\ref{fig:large_scale}) provide a basic introduction to the
183: morphologies of galaxies visible on the sky and a description of the stages
184: of stellar evolution using examples in the basemap to illustrate these tours.
185: Likewise, each of the images from the HST is accompanied by a pop-up balloon
186: containing a snippet of the press release for that image and links for further
187: information. Finally, we have placed icons on objects from the Messier, New
188: General and Yale Bright Star catalogs (Figure~\ref{fig:small_scale}). These
189: placemarks identify the object, provide basic observational data about the
190: source (names, positions, distance, colors and brightnesses along with links
191: to NED and SIMBAD where appropriate; Figure~\ref{fig:small_scale_info}) and,
192: in some cases, include a snippet from Wikipedia about what is known about
193: that star, star cluster, nebula or galaxy.
194:
195:
196: In addition to these static layers, we also provide a time-based layer
197: which shows the position of the Moon and Solar System planets over the
198: course of three months' time. The ephemeris data was generated using the
199: JPL Horizons interface\footnote{\tt http://ssd.jpl.nasa.gov/?horizons}. The
200: associated placemarks indicate the distance, magnitude and angular extent
201: of each body as a function of time, controlled by a slider bar that appears
202: in the upper right corner of the window when these layers are activated.
203: It is worth noting that planets' and the Moon's icons are not scaled according
204: to their actual appearance on the sky so that they can be more easily
205: distinguished from the background stars. The duration of the time interval is
206: sufficient to show both prograde and retrograde motion in the planetary orbits.
207:
208: \section{Adding Data to {\it Sky}}
209:
210: While the basemap provides a common reference frame and a view of the
211: optical sky, the strength of {\it Sky} comes from its ability to
212: incorporate and display user generated data (images, catalogs, time
213: variable data, tours of the sky) on top of the basemap. This wide array of
214: functionality is one the advantages of a mature platform, as {\it Earth}
215: has been performing similar tasks for the GIS community for a number of years.
216: For example, the x-ray flux from a galaxy cluster can be displayed on top of
217: the optical galaxies seen in the basemap, which can themselves be tagged with
218: placemarks containing the survey data for each of the member galaxies.
219: Likewise, infrared, ultra-violet and radio imaging of a single galaxy can be
220: overlaid simultaneously, with the user in complete control of the transparency
221: of each layer. In this section, we will discuss the basic tools available for
222: incorporating data within {\it Sky}, as well as point to some examples
223: currently available for download to supplement the basic {\it Sky} package.
224:
225:
226: \subsection{Keyhole Markup Language}
227:
228: Most of the methods for putting data into {\it Sky} involve some use of
229: Keyhole Markup Language (KML). An introduction to the KML schema
230: can be found at the Google API documentation
231: site\footnote{\tt http://code.google.com/apis/kml/documentation/}.
232: Like HyperText Markup Language (HTML) and XML, KML files are composed
233: of tags and attributes. Each tag describes an entity (such
234: as the coordinates for a point on the sky or the bounding box for an
235: image) and each attribute defines parameters associated with that
236: entity. KML files can be used to add catalog data with placemarks,
237: imagery with overlays, and vector data in the form of lines and
238: polygons. As with HTML, the basic building blocks can be combined in
239: a variety of ways to create data sets as utilitarian or richly
240: formatted as the user desires. The KML standard also evolves over time,
241: adding features and interfaces based on user feedback and proposed usage
242: cases.
243:
244: \subsection{Placemarks}
245:
246: The most basic element for displaying data is the {\it placemark}.
247: This acts as a single push-pin located at a discrete point on the sky,
248: possibly with some associated data; indeed, the default icon for a
249: placemark is an image of a push-pin. KML allows users to customize
250: the icon associated with a placemark, as well as the information that
251: appears in the pop-up balloon when a user clicks on the placemark.
252: More advanced features include the ability to control when the
253: placemark appears and how the camera moves when the user clicks on the
254: placemark. For example, the appearance of the default catalog data in
255: {\it Sky} is tuned to the objects' magnitudes. This keeps fainter
256: objects from appearing until the user is at a higher zoom level, which
257: avoids having the viewer overwhelmed by a unmanageable number of placemarks
258: on the sky at any given time. Likewise, visibility of placemarks can be
259: governed by the region of the sky shown in the viewport at any given time,
260: preventing multiple catalogs across the full sky from being rendered at the
261: same time and improving the efficiency of serving the data.
262:
263:
264: \subsection{Vector Data}
265:
266: The next stage of sophistication is vector data: lines and polygons.
267: These are essentially collections of placemarks, with line segments
268: connecting them and (in the latter case) a solid color filling the
269: enclosed area. In both cases, users can control both the color and
270: transparency of the lines and polygons. This makes vector data ideal for a
271: quick description of a observational footprint or uncertainty region
272: associated with a survey or observation, as one might have with the
273: observation of a gamma ray burst, for example.
274:
275: \subsection{Image Overlays} \label{sec:image_overlays}
276:
277: The KML tag name {\it GroundOverlay} is inherited from {\it Earth}, but
278: these are simply images projected against the basemap in {\it Sky}.
279: GroundOverlays are defined by associating an image with some bounding
280: region on the sky, the latter being given by the bounds in the four
281: cardinal directions and a rotation angle. For small-area images, the
282: difference between the tangent plane and lat/long projection is slight enough
283: that images can be aligned to the basemap using only a linear transformation.
284: Google Earth has a graphical interface for aligning images this
285:
286:
287: Larger images (or those nearer the poles) can be more difficult. As
288: mentioned previously, since {\it Sky} shares a rendering engine with
289: {\it Earth}, the geometry of the sky is, in fact, a slightly oblate spheroid
290: (technically, the WGS84 projection). The GIS community has developed a
291: number of tools for handling re-projections of images using this geometry,
292: most notably the Geospatial Data Abstraction Library (GDAL). This software
293: can warp images from one projection (including a tangent plane) to another
294: and encodes the geometric information necessary for registering the
295: image on the sphere in either an image header (analogous to a FITS
296: header) or an associated ``world file'', depending on the image
297: format. While users may want to become more familiar with the GDAL
298: software themselves, we provide a simple, open source tool
299: {\it wcs2kml}\footnote{\tt http://code.google.com/p/wcs2kml/} which will read
300: in an image in a variety of formats and WCS information from a FITS header and
301: generate a properly warped image and overlay KML for you. This tool is
302: available in both Python and C++ versions.
303:
304: \subsection{Network Links and Regionation}\label{sec:regionation}
305:
306: While the basic elements of KML are static, network links allow users to
307: load new KML dynamically. This may be triggered by a change in the view of
308: the camera in {\it Sky} or after a specified interval has elapsed. This
309: flexibility allows for a limitless number of applications: updating the
310: location of satellites in orbit or their current observing target, displaying
311: feeds from gamma ray burst trackers, providing real-time updating of
312: observations to collaborators back home, and so on. Likewise, the ability
313: to have network links activated based on the viewport of the sky gives
314: KML authors the ability to {\it regionate} their data.
315:
316: {\it Regionation} is simply splitting a given data set into a hierarchical
317: structure, similar to what happens with the imagery and placemarks that are
318: part of the basic {\it Sky} data set. When the camera is far from the
319: surface of the celestial sphere, the imagery served is at a very low resolution
320: (finer details would be lost anyway, so displaying them is unnecessary) and
321: only the brightest stars and galaxies are annotated to avoid overwhelming the
322: user. As the camera zooms in, higher resolution imagery is fetched
323: automatically and fainter objects are tagged. The same can be accomplished
324: with network links to other files that are tied to progressively smaller
325: regions on the sky. This allows vast amounts of catalog data and high
326: resolution imagery to be displayed progressively on demand, instead of
327: clogging the user's internet connection all at once.
328: {\it Regionator}\footnote{\tt http://code.google.com/p/regionator/},
329: an open source Python program, provides a simple interface for regionating
330: KML files that have already been created (like those produced by the Python
331: version of {\it wcs2kml}, for example). The C++ version of {\it wcs2kml}
332: includes a feature which will regionate its outputs as well, producing a
333: hierarchical set of network linked KML files and sub-sampled versions of
334: astronomical images with associated WCS information. For large mosaics (or
335: even images taken with large format CCDs), this can vastly improve the
336: performance of the resulting image overlay. The {\it wcs2kml} package
337: also includes utilities for converting FITS catalog data to KML placemarks
338: and regionating them as well.
339:
340:
341: \subsection{Sharing \& Publishing Data}
342:
343: Since KML is completely platform independent, data sharing is easy.
344: Simply construct a KML file describing your data and email it to
345: your collaborator. Even better, place it on a webpage where it can be
346: discovered and shared with the world. Larger KML files (which are
347: fundamentally just simple ASCII text files) can be compressed into a
348: {\it zip} archive (generally tagged with a ``.kmz'' suffix to differentiate
349: them from uncompressed files) along with imagery to form a self-contained file
350: that can be opened natively by {\it Sky}. Alternatively, imagery can be
351: loaded with URL pointers, just like in standard HTML.
352:
353:
354: \subsection{Working with KML in {\it Sky} mode}
355:
356: KML was created to annotate Google Earth with images, placemarks and ancillary
357: information. As the KML schema is designed to support this Earth-centric
358: annotation some of the features and tags present within KML do not translate
359: naturally to annotation of the sky. We provide here a brief overview of these
360: tags and features and provide workarounds for their use in {\it Sky}.
361:
362: To differentiate between KML for viewing on the {\it Sky} as opposed to on
363: {\it Earth} a hint attribute is available within the initial specification
364: of the KML schema. Adding the {\it Sky}
365: hint\footnote{\tt <kml xmlns="http://earth.google.com/kml/2.2" hint="target=sky">} to your {\tt <kml>} tag will cause the client to prompt a user to open the
366: KML under sky mode (if the view is currently {\it Earth}-based). All KML saved
367: from within the Google Earth client while it is in {\it Sky} mode will have
368: this attribute set by default.
369:
370: Placemarks, image overlays, lines, polygons and styles are all supported under
371: {\it Sky}. Tilt and roll angle for the camera position are ignored
372: when viewing images and placemarks under {\it Sky} as the current astronomical
373: imagery is a simple projection onto a sphere without altitudinal information.
374: Range for the LookAt and other positional tags (i.e.~the altitude above the
375: Earth that the camera zooms to when a placemark is double-clicked) are
376: measured in meters. To convert to an angular system that is more appropriate
377: for viewing the sky we suggest the following relation,
378: \begin{equation}
379: R \sin(\alpha/2 - \beta/2) = (R - r) \sin(\alpha/2)
380: \end{equation}
381: where $r$ is the range to use in the LookAt tag, $R$ is the radius of the
382: Earth (6371 km), $\alpha$ is the angle subtended by the viewer when you are
383: viewing from the maximum distance from the sky (i.e.~you are pulled all the
384: way back to the center of the Earth; for the current client
385: $\alpha = 80^\circ$) and $\beta$ is the angular diameter of your region.
386: For $r \ll R$, this reduces to,
387: \begin{equation}
388: \beta = 2r\tan(\alpha/2)/R
389: \end{equation}
390: where $\beta$ is in radians.
391:
392: Images that are incorporated into {\it Sky} are correctly oriented (i.e.~East
393: to the left) and do not require any transformations other than rotation.
394: Rotation in {\it Sky} is in the clockwise direction (opposite from that in
395: {\it Earth} mode). 3D sketchup models for objects within {\it Sky} are not
396: officially supported but may work.
397:
398:
399:
400: \section{Future}
401:
402: The research applications of {\it Sky} are limitless. Data can be
403: visualized from an all-sky view all the way to native resolution of
404: the underlying imagery with minimal pre-processing. Catalog data can
405: be immediately placed on the sky imagery with KML placemarks enabling
406: new discoveries to be made available in almost real time. Data can be
407: shared within a research group or to the broader community securely by
408: simply sending KML files or posting them on a webserver. Finding charts are
409: as simple to create as producing a screenshot or zooming to a location just
410: prior to observing. Survey progress can be visualized in real time by
411: updating KML available over the internet (via network links) as images come
412: off the CCDs. Finally, when the observations are ready to be made public,
413: {\it Sky} is an ideal platform to reach parts of the general public interested
414: in astronomy and put your results in the context of the broader astrophysical
415: data universe.
416:
417: As a tool with a simple, extensible and intuitive interface we see {\it Sky}
418: in Google Earth as the embodiment of a Virtual Telescope, providing
419: open access to data through open standards. It leverages the power of
420: the Google infrastructure for serving data to a wide range of people
421: while also enabling any user to integrate and share their own images
422: and catalogs. We hope that this step towards the democratization of
423: science will enable a new era of visual discovery and science
424: communication.
425:
426: \acknowledgments
427:
428: The authors wish to thank Chikai Ohazama, Lior Ron, Effie Seiberg, Steve
429: Zelinka, Andrew Moore, Robert Pike, Brian McClendon, Linne Ha, Mohammad Khan,
430: Brian McClean, Dylan Myers, Wei Luo, Peter Birch, Andrew Kirmse, Michael
431: Ashbridge, Bent Hagemark, Chris DiBona, Sam Roweis, and Keir Mierle for their
432: tireless efforts to make {\it Sky} what it is and what it will be.
433:
434: Please see the {\it Sky} partners
435: homepage\footnote{\tt http://earth.google.com/sky/partners.html} for a list
436: of the institutions contributing imagery to {\it Sky}.
437:
438: \begin{thebibliography}{}
439: \bibitem[Hambly et al.(2001)]{superCOSMOS} Hambly, N.~C.,
440: MacGillivray, H.~T., Read, M.~A., Tritton, S.~B., Thomson, E.~B.,
441: Kelly, B.~D., Morgan, D.~H., Smith, R.~E., Driver, S.~P., Williamson,
442: J., Parker, Q.~A., Hawkins, M.~R.~S., Williams, P.~M., Lawrence, A.,
443: 2001, \mnras, 326, 1279
444: \bibitem[H{\o}g et al.(2000)]{Tycho2} H{\o}g, E., Fabricius, C.,
445: Makarov, V.~V., Urban, S., Corbin, T., Wycoff, G., Bastian, U.,
446: Schwekendiek, P., Wicenec, A., 2000, \aa, 355, L27
447: \bibitem[Lasker et al.(1996)]{DPOSS} Lasker, B.~M., Doggett, J., McLean, B.,
448: Sturch, C., Djorgovski, S., de Carvalho, R.~R., Reid, I.~N., 1996,
449: Astronomical Society of the Pacific Conference Series, 101, 88
450: \bibitem[Lupton et al.(2004)]{LuptonRGB} Lupton, R., Blanton, M.~R.,
451: Fekete, G., Hogg, D.~W., O'Mullane, W., Szalay, A., Wherry, N., 2004,
452: \pasp, 116, 133
453: \bibitem[Madore et al.(1992)]{NED} Madore, B.~F., Helou, G.,
454: Corwin, Jr., H.~G., Schmitz, M., Wu, X., Bennett, J., 1992,
455: Astronomical Society of the Pacific Conference Series, 25, 47
456: \bibitem[McLean et al.(2000)]{DSS} McLean, B.~J., Greene, G.~R.,
457: Lattanzi, M.~G., Pirenne, B., 2000,
458: Astronomical Society of the Pacific Conference Series, 216, 145
459: \bibitem[Reid et al.(1991)]{POSSII} Reid, I.~N., Brewer, C., Brucato, R.~J.,
460: McKinley, W.~R., Maury, A., Mendenhall, D., Mould, J.~R., Mueller, J.,
461: Neugebauer, G., Phinney, J., Sargent, W.~L.~W., Schombert, J.,
462: Thicksten, R., 1991, \pasp, 103, 661
463: \bibitem[Snyder(1926)]{HANDBOOK} Snyder, J.~P., 1926,
464: {\it Map Projections -- A Working Manual}, 90-91
465: \bibitem[Szalay \& Gray(2001)]{2001Sci...293.2037S} Szalay, A. \& Gray, J.,
466: 2001, Science, 293, 2037
467: \bibitem[Wenger et al.(2000)]{SIMBAD} Wenger, M., Ochsenbein, F.,
468: Egret, D., Dubois, P., Bonnarel, F., Borde, S., Genova, F.,
469: Jasniewicz, G., Lalo{\"e}, S., Lesteven, S., Monier, R., 2000,
470: Astronomy and Astrophysics Supp., 143, 9
471: \bibitem[York et al.(2000)]{SDSS} York, D.~G. et al. 2000, \aj, 120, 1579
472: \end{thebibliography}
473:
474:
475: \begin{figure*}
476: \begin{center}
477: \includegraphics[width=450pt]{SkyImageLargeScale.jpg}
478: \end{center}
479: \caption{\label{fig:large_scale} Large scale image of the sky showing
480: constellations and Messier (red) and NGC (blue) object tags. The basemap
481: imagery is largely dark thanks to the re-sampling of the DSS and SDSS imagery
482: to lower resolution, although a number of stars are visible at the junctions
483: of the constellation lines. Icons are also available for objects in the
484: Yale Bright Star catalog, although they are deactivated in the current image.}
485: \end{figure*}
486:
487: \begin{figure*}
488: \begin{center}
489: \includegraphics[width=450pt]{SkyImageSmallScale.jpg}
490: \end{center}
491: \caption{\label{fig:small_scale} Closer view of the imagery for the area of
492: sky around M101. As the credits at the lower middle of the image indicate,
493: this imagery is a combination of data from the SDSS and the HST, with the
494: latter providing a higher resolution mosaic of the central galaxy region. The
495: icon at the top of the M101 HST imagery contains information related to the
496: press release issued by the STScI for this image, as well as links to
497: associated papers and observational data.}
498: \end{figure*}
499:
500: \begin{figure*}
501: \begin{center}
502: \includegraphics[width=450pt]{SkyImageSmallScaleInfo.jpg}
503: \end{center}
504: \caption{\label{fig:small_scale_info} Same as Figure~\ref{fig:small_scale},
505: but with the information balloon for one of the NGC objects opened. These
506: balloons contain basic observational data as well as links to the NED and
507: Simbad databases for further information and a snippet of Wikipedia text for
508: each object, if available. This particular balloon, chosen so as not to
509: obscure the imagery, shows the bare minimum of information available for a
510: given object.}
511: \end{figure*}
512:
513:
514: \end{document}
515: