THE Ĝ INFRARED SEARCH FOR EXTRATERRESTRIAL CIVILIZATIONS WITH LARGE ENERGY SUPPLIES. III. THE REDDEST EXTENDED SOURCES IN WISE

Hart (1975) argued that the failure of SETI to date was because humanity is alone in the Milky Way based on a comparison of likely colonization timescales for the Milky Way and its age. Hart's argument also implies that any galaxy with a spacefaring species will become thoroughly colonized in a short time compared to the galaxy's age, suggesting that most galaxies should either contain no spacefaring species or be filled with them.

Kardashev (1964) parameterized potential alien civilizations by their energy supply compared to the starlight available to them, with a Type i civilization (K1 in our notation) commanding its planet's entire stellar insolation, a Type ii civilization commanding an entire star's luminosity (i.e., a Dyson sphere, K2), and a Type iii civilization (K3) commanding most of the stellar luminosity in a galaxy. Expressed in these terms, Hart's argument is that the timescale for the appearance of the first K2 to its growth into a K3 is very short, implying that we should expect many K3 civilizations in the universe if spacefaring life is common.

Indeed, the technological sophistication required to construct a Dyson sphere seems far greater than that required for achieving interstellar travel: while humanity's solar panels currently fall short of complete coverage of the Sun by a factor of $\sim {{10}^{17}}$, our deepest space probes today fall short of the distance to the nearest star by a factor of only a few thousand.

If Hart's reasoning is sound, then we should expect that, unless intelligent, spacefaring life is unique to Earth in the local universe, other galaxies should have galaxy-spanning supercivilizations, and a search for K3s may be fruitful. If there is a flaw in it, then intelligent, spacefaring life may be endemic to the Milky Way in the form of many K2s, in which case a search within the Milky Way would be more likely to succeed. It is prudent, therefore, to pursue both routes.

Dyson (1960) and Slysh (1985) demonstrated that waste heat would be an inevitable signature of extraterrestrial civilization, and that such signatures might be detectable to mid-infrared (MIR) instrumentation for civilizations with energy supplies comparable to the luminosity of their host star. The first effective all-sky search sensitive to such namesake "Dyson spheres" was performed by IRAS, but the infrared cirrus and poor angular resolution of IRAS limited its sensitivity to only the brightest sources.

Jugaku & Nishimura (1991) performed a series of studies of IRAS sources with anomalously red $(K-12\;\mu )$ colors (Jugaku & Nishimura 1991; Jugaku et al. 1995; Jugaku & Nishimura 1997, 2000), and found no highly complete Dyson spheres around any of the 365 solar-type stars within the 25 pc studied or around another 180 stars within the same distance (Jugaku & Nishimura 2004).

Carrigan (2009a, 2009b) used the IRAS low-resolution spectrometer to determine whether candidate Dyson spheres' SEDs were consistent with blackbodies with T = 100–600 K. Carrigan (2009a) concluded from the infrared colors and low resolution spectra that the best of these Dyson sphere candidates were typically reddened and dusty objects such as heavily extinguished stars, protostars, Mira variables, AGB stars, and planetary nebulae (PNe). Nonetheless, of the 11,000 sources he studied, Carrigan (2009a) identified a few weak Dyson sphere candidates with spectra consistent with carbon stars. One candidate, IRAS 20369+5131, showed a nearly featureless blackbody spectrum with T = 376 K, but Carrigan (2009a) concluded it is likely a distant red giant with no detectable SiC emission.

To date, the only search for the waste heat of a K3 civilization in the peer-reviewed literature has been that of Annis (1999) who searched for outliers to the Tully–Fischer relation to identify K3s intercepting a significant fraction of their starlight. Carrigan (2012) also suggested searching for the morphological signatures of K3s, especially in elliptical galaxies.

The advent of large solid angle, sensitive MIR surveys makes a waste-heat-based K3 search more feasible today. The Wide-field Infrared Survey Explorer (WISE; Wright et al. 2010) performed an all-sky MIR survey at 3.4, 4.62, 12, and 22 μ (the $W1$, $W2$, $W3$, and $W4$ bands) with superior angular resolution (by a factor of 5) and sensitivity (by a factor of 1000) compared to IRAS. WISE is thus the first sensitive survey for both K2s in the Milky Way and K3s among the approximately 1×10⁵ galaxies it resolved. The Spitzer Space Telescope is another powerful tool for waste heat searches, having superior sensitivity and angular resolution. Its survey of the Galactic Plane will be a powerful tool in the search for K2s in the Milky Way. Since its large area surveys are generally restricted to star-forming regions and the Galactic Plane, where sensitivity to K3s is more limited, we have restricted our efforts in this paper to WISE.

In Wright et al. (2014a), we developed the AGENT formalism for quantifying the expected MIR spectra from galaxies hosting K3s in terms of the energy supply of an alien civilization, and outlined the methodology of the Glimpsing Heat from Alien Technology (Ĝ) search for such civilizations in the local universe using the results of the WISE All-sky MIR survey. In particular, we argued in that work that extended sources have the lowest false positive rate because many of the confounding sources, primarily dusty and extinguished stars and cosmological sources, would not be present in that sample.

The AGENT formalism parameterizes the power used by an alien civilization in terms of starlight absorbed (represented by the parameter α), energy generated by other means (), thermal waste heat emitted (γ), and other energy disposal (ν).

Most relevant to the present paper are the parameters γ (waste heat luminosity, expressed as a fraction of the starlight available to the civilization) and ${{T}_{{\rm waste}}}$, the characteristic temperature of the waste heat (which dictates its infrared colors). For values of ${{T}_{{\rm waste}}}$ in the 100–600 K range, values of γ near 1 would imply that most of the luminosity of a galaxy is in the MIR (in the form of the waste heat from alien factories), while values near 0 would imply that the alien waste heat was very small compared to the output of the stars in the galaxy. For dust-free elliptical galaxies with little of their luminosity in the MIR, values of γ of a few percent would be detectable as an anomalous MIR excess.

As an essential step in our waste heat search, we have produced a clean catalog of the reddest sources resolved by WISE. The purpose of our focus, in this work, on resolved sources is twofold: resolved sources present their own challenges of interpretation and photometry, necessitating this separate effort, and we wish to first deal with a relatively small and clean sample of nearby galaxies, consistent with a search for nearby K3s. WISE  resolves approximately 1 × 10⁵ galaxies (see Section 7.2).

Sources unresolved by WISE include a wide variety of potential false positives, such as false detections, data artifacts, cosmologically redshifted objects, dusty objects at cosmological distances, dusty stars, and heavily extinguished stars. The study of these sources requires a different sort of analysis from those needed for nearby galaxies, which we will describe in later papers. By contrast, there are many fewer, and more easily excluded, sources of false positives among the extended sources in the all-sky catalog.

Our primary objective in this paper is to "map the landscape" among galaxies resolved by WISE by identifying the nature of the very reddest of these extended sources in the WISE All-sky survey, using several metrics for "redness," including the AGENT parameter γ. A byproduct of this effort is a clean catalog of the reddest extended sources in WISE, which we present here.

Our secondary objective is to identify and explain the most extreme objects in this catalog, which by their superlative nature are inherently scientifically interesting, regardless of the origin of their MIR luminosity. In most cases, these are well-known objects; in many of the remaining cases, their nature seems clear. A few cases, however, are new to the scientific literature and their nature is uncertain. These sources of uncertain nature are natural candidates for follow-up, and those that appear consistent with Kardashev civilizations warrant follow-up by SETI communication efforts, in particular.

As a tertiary objective, we place a zeroth-order upper limit on the energy supplies of nearby K3s by identifying the most MIR-luminous galaxies in our sample. This upper limit can be pushed down to the degree that these galaxies' MIR emission can be shown to have purely natural origins. A rigorous upper limit will require a more detailed analysis of these galaxies' SEDs, and a more precise calculation of the number of galaxies considered in our sample. We save this exercise for another paper in this series.

Finally, we also illustrate the waste heat approach by performing a quick check of two classes of anomalous galaxies to confirm that they are not hosts to MIR-bright K3s.

In Section 2, we describe how we have analyzed the WISE All-sky catalog, and how we performed a series of cuts to select a sample of only real, red, extended sources (our "Extended Gold Sample" of 30,808 sources). Section 2.6 describes how we calibrated the WISE photometry and our estimates of our photometric precision.

Section 3 describes our efforts to classify the sources in the Extended Gold Sample, mostly via SIMBAD object types, so that we could identify previously unstudied sources and reject many Galactic sources.

Section 4 describes our more detailed efforts to understand the reddest sources in the Extended Gold Sample using careful visual inspection and literature searches.

Section 5 describes how we performed a second round of vetting on the Extended Gold Sample, using our calibrated photometry from Section 2.6, our classifications from Section 3, and our visual inspections and literature searches from Section 4 to confidently and carefully identify the reddest sources and determine the best photometry for them on a case-by-case basis. The result is the 563 source "Platinum Sample," which we present in our catalog, and whose fields are described in Table 4.

Section 6 describes the reddest objects in the Platinum Sample for several definitions of "red" including all six combinations of the WISE bandpasses and the AGENT parameter γ. Section 6.3 describes five sources that are effectively new to science, having little or no literature presence (beyond having been detected with IRAS).

We present our conclusions in Section 8, and in the appendices we use the WISE imagery to examine two categories of anomalous galaxies, H i dark galaxies and so-called "passive spirals," and show that they do not exhibit sufficient MIR emission to have their anomalous natures explained by the presence of a K3.

The following methodology was used to select the full sample of 12 μm selected sources with extended photometric profiles from the WISE All-sky catalog. The instrumental profile-fit reduced $\chi _{\nu }^{2}$ (W3RCHI2)⁶ was used to distinguish sources with extended profiles (W3RCHI2 ≥3) from sources with point source profiles (W3RCHI2 < 3). Given that the majority of sources in the WISE All-sky catalog are unresolved in W3, i.e., angular sizes $\lt 6\buildrel{\prime\prime}\over{.} 5$, we adopt a conservative $\chi _{\nu }^{2}$ threshold of 3, which excludes only a small number of marginally resolved sources while admitting a large but manageable number of point sources with anomalously poor point-spread function (PSF) fits. We also required that the uncertainty in the W3 "standard" aperture magnitude be measured (i.e., not null). A null result means that the W3 "standard" aperture magnitude is a limit, or that no aperture measurement was possible. Finally, we applied a cut in the Galactic latitude ($|b|\geqslant 10$) to remove contamination from nebular emission in the Galactic Plane:

$$\begin{eqnarray*}&&{\rm W}3{\rm RCHI}2\geqslant 3\ {\rm and}\ |b|\geqslant 10\ {\rm and}\ {\rm W}3{\rm SIGM}\;{\rm is}\;{\rm not}\;{\rm null}.\end{eqnarray*}$$

These search criteria yield a total of 202,851 sources, which comprises our parent sample.

The WISE All-sky database provides photometry measurements using a variety of methods: PSF profile fitting, variable aperture photometry (eight circular apertures), curve of growth (COG) aperture photometry, and elliptical aperture photometry (for sources matched to the 2MASS Extended Source Catalog (XSC)).

The profile fitting photometry is referred to as WXMPRO and is defined as the magnitude measured with profile-fitting photometry, where X is 1,2,3, or 4 corresponding to the four WISE bands. In addition to magnitudes, this procedure also derives the signal-to-noise ratio and WXRCHI2 (i.e., the goodness of fit to the PSF model of the source).

The COG or "standard" aperture photometry is referred to as WXMAG. According to the WISE explanatory supplement, "This is the COG corrected source brightness measured within an 825 radius circular aperture centered on the source position. The background sky reference level is measured in an annular region with inner radius of 50'' and outer radius of 70''.⁷

The WISE pipeline performed nested circular aperture photometry using eight apertures from 55 to 2475 in W1–W3 and 11'' to 495 in W4 (see footnote 7). The WISE team provides these in the form of parameters named WXMAG_1 for the first aperture of 55 in W1 through W3 and 11'' in W4, WXMAG_2 for the second aperture, WXMAG_3 for the third aperture, and so on.

The elliptical aperture measurements, referred to as WXGMAG were based on matched sources between WISE and the 2MASS XSC. The shape of the elliptical aperture was determined by utilizing the shape information of the source provided by the 2MASS XSC. We discuss the uncertainties in our best photometry from calibrating aperture magnitudes to careful extended source photometry in Section 2.6.

Since our analysis is primarily interested in identifying galaxy-scale extraterrestrial civilizations, "K3s," we required that photometric measurements of extended sources in WISE be as reliable as possible. The PSF profile fitting photometry is best used when extracting photometry from point-like sources in the WISE survey. For the majority of extended sources in WISE, the COG photometry generally provides more reliable measurements, though the strict 825 radius circular aperture fails for large and extended galaxies. Measuring reliable photometry for extended galaxies has historically proven to be a non-trivial and difficult task, the reason being that galaxies come in all shapes and sizes, and a single "standard" aperture fails to encompass the full range of galaxy sizes and structures. 

Recently, Jarrett et al. (2012) have attempted to extract reliable galaxy photometry for extended sources in the WISE catalog. They have developed a complex algorithm in order to construct the WISE High Resolution Galaxy Atlas and present their initial results in Jarrett et al. (2013). Even more recently, Tom Jarrett has measured reliable photometry for ∼67,000 extended sources in the WISE catalog contained within the south Galactic Cap (SGC: $b\lt -60$). We use this analysis to calibrate the aperture magnitudes as presented by the WISE All-sky database, as we describe in Section 2.6.

In order to remove spurious sources, we use the contamination and confusion flags (CC_FLAG) given for W3. In Figure 1, we describe the definitions and provide the number counts of objects having the various CC_FLAGs. We retain sources with the following CC_FLAGs: "0," "d," "h," and "o." We consider sources with CC_FLAGs "P," "p," "D," "H," and "O" to be contaminated and we removed them from further analysis. Our decision for choosing these particular flags is motivated by empirical examination of a significant fraction of the flagged entries (i.e., we found that the rejected flags have very good reliability in flagging false sources, but the ones we retain often appear for real sources). Using these quality flags reduces the 12 μm sample to 132,651 sources.

Zoom In Zoom Out Reset image size

In addition to removing the Galactic Plane, we also applied three additional coordinate cuts to remove the most obvious high-density regions of foreground contamination from objects in star-forming regions. Region 1 is composed of the Galactic Bulge, (i.e., −12 < l < 10 and 20 < b < 10); the total area covered by this region is 212.23 deg² and contained 9323 sources. We removed all sources in Region 1 from our sample. Region 2 comprises sources associated with the Large Magellanic Cloud (LMC); this region contains a total of 13,104 sources. Region 3 is composed of sources associated with the Orion Nebula and contains a total of 12,733 sources. Instead of imposing a blanket coordinate cut for these regions of patchy contamination, we found that we could rather reliably identify foreground sources by rejecting highly clustered sources and retaining relatively isolated sources using a surface density algorithm. This reduces our sensitivity to K3s in regions 2 and 3 without rendering us completely blind to them. After we removed the sources in these three regions, the 12 μm sample was reduced to 97,491 sources.

We use two simple color cuts to eliminate the most obvious stellar contaminants (it should be noted that these cuts also eliminate a large number of elliptical galaxies since these generally tend to be blue in the infrared). We identify the stellar locus in the lower left (blue) corner of Figure 2 (since stellar photospheres have MIR colors near zero) and use the following criteria

$$\begin{eqnarray*}\begin{array}{rcl} {\rm W}2{\rm MPRO}-{\rm W}3{\rm MPRO} & \lt & 2\ {\rm and}\ {\rm W}3{\rm MPRO} \\ {} & {} & -{\rm W}4{\rm MPRO}\leqslant 1 \\ \end{array}\end{eqnarray*}$$

to remove them from further analysis. Use of the profile fit photometry is appropriate because these sources are typically stars, and thus should have reliable profile fitting photometry.

Zoom In Zoom Out Reset image size

We visually examined a representative sample of sources in this region and concluded that they are indeed stellar-like objects with anomalously high W3RCHI2 values and since we are primarily concerned with red extended objects, this cut is compatible with our overall search. These criteria identified a total of 21,645 stellar-like sources, leaving 75,846 sources in the 12 μm extended sample for further inspection. We present a color–color map of these sources in Figure 3.

Zoom In Zoom Out Reset image size

We constructed $2^{\prime} \times 2^{\prime}$ WISE color images ((W1+W2)/2 = blue, W3 = green, and W4 = red) for the remaining 75,846 sources and have visually classified them into five primary groups. The five groups are: stellar artifacts, low coverage artifacts, nebular, needs closer inspection, and high quality. Representative examples are presented in Figure 4.

Zoom In Zoom Out Reset image size

The stellar artifacts comprise 11,033 sources which were caused by bright saturated stars (i.e., halos, streaks, and latents). Most of these did not have any CC_FLAGs indicating a problem in the All-sky release.

Low coverage artifacts are sources we identified as artifacts because they were observed fewer than six times by WISE under nominally good conditions (i.e., W3M < 6, where W3M gives the number of individual W3 exposures on which a profile-fit measurement of the source was possible). To put this number in perspective, the median W3M for all 202,851 sources was 13. There were a total of 14,595 low coverage sources. We visually examined this sample and recovered 215 sources which, though having low coverage, appear to be real astrophysical sources and are considered for further analysis.

The nebular sources total 12,989 sources and we determined them to be locally bright regions of large, nebular networks of Galactic dust and so not discrete objects. These sources required images with a much larger FOV $(20^{\prime} \times 20^{\prime} )$ to be classifiable. Since these are highly extended in nature, we first identified sources within a $10^{\prime}$ radius and constructed $20^{\prime} \times 20^{\prime}$ color images. The brightest W3 source within a cluster was used as the center of a single color image for the field. We recovered 192 sources that appeared to be "nebular" in the $2^{\prime} \times 2^{\prime}$ image, but which we reclassified as being discrete objects after inspecting the $20^{\prime} \times 20^{\prime}$ images.

There were 4727 sources lacking a 2MASS association that we labeled for closer inspection, since they appear to be legitimate sources with good coverage. The majority of these sources were duplicates of sources already included in the high quality sample. The sources that were unique were rematched to the 2MASS catalog using a two-fold process. The first matching used a 30'' radius and recovered photometry for 509 sources. The second matching used a 60'' search radius and recovered photometry for 44 sources.

Figure 5 illustrates the typical colors and magnitudes of these categories of sources, and of the W3 Extended Gold Sample (i.e., our sample of real sources extended in W3 with red MIR colors).

Zoom In Zoom Out Reset image size

We identified 32,502 sources which had 2MASS associations and appear to be real astrophysical objects based on their color images as high quality sources. Some extended objects appear in the catalog multiple times. We removed the duplicate entires (those within 60'') of the brightest entry in the catalog. We identified a total of 1694 duplicated sources which we removed from the sample.

The number of galaxies we have in our final sample here makes sense. Most of the high latitude extended sources in WISE are galaxies, and as we show in Section 7.2, there are approximately 100,000 galaxies that would be extended in the W3 WISE band if they had significant W3 emission. We have thus selected approximately the reddest 1/3 of the galaxies on the sky larger than the angular resolution of the WISE W3 band.

We have thus constructed a sample of mostly real, extended, and discrete sources in the WISE All-sky release which we refer to as the W3 Extended Gold Sample.

We summarize the vetting procedure that produced the W3 Extended Gold Sample in Figure 6 and present the sky distribution for this sample in Figure 7.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The W3 Extended Gold Sample comprises 30,808 apparently extended sources from the high quality sample, 597 sources from the for closer inspection sample, and 192 sources from the nebular sample, bringing the final sample to 31,597 sources (shown in the lower right plot of Figure 5). As discussed in Section 2.5, the WISE COG photometry is generally reliable for most sources in this sample, but some care must be taken before using these measurements.

Tom Jarrett has kindly provided us with a preliminary version of his extended source catalog for the south Galactic Cap ($b\lt -60$) in advance of this catalog's publication. To further calibrate the extended source photometry and determine the degree of systematic errors in the WISE pipeline extraction for extended sources, we have cross-matched 1907 WISE All-sky sources with sources in Jarrett's preliminary catalog (210 sources did not match, usually because the source was only marginally extended in one of the catalogs). We found that for our W3 Extended Gold Sample, the distribution of differences between the All-sky COG magnitudes and Jarrett's magnitudes were significantly offset from zero (by 0.4, 0.4, 0.1, and −0.15 magnitudes in W1, W2, W3, and W4) and skewed with long tails past 1 magnitude, with the Jarrett magnitudes typically brighter than the COG magnitudes. In Figure 8, we present the comparison between the COG magnitudes presented in the All-sky release to those measurements presented in Jarrett's preliminary catalog.

Zoom In Zoom Out Reset image size

We estimated the size of the sources using the difference between the profile, 4th, and 8th aperture magnitudes, W_profile, WX_4, and WX_8. For unblended point sources these magnitudes are typically identical, but for extended sources the larger aperture magnitudes are more faithful because they measure flux further up the curve of growth. We formed "size" parameters from the quantities (WX_8-WX_4) and (WX_8-W_profile), and found that by using the WX_8 magnitudes, with quadratic corrections in the size parameters and constant offsets, we could reproduce the Jarrett magnitudes to within 0.06 mag in most cases and colors to within 0.08 mag.

The transformation we used to generate accurate extended source magnitudes WX from the WISE pipeline photometry is

$$\begin{eqnarray}&&{\rm S}1={\rm WX}\_8-{\rm WX}\_4\end{eqnarray} \tag{ 1 }$$

$$\begin{eqnarray}&&{\rm S}2={\rm WX}\_8-{{{\rm W}}_{{\rm profile}}}\end{eqnarray} \tag{ 2 }$$

$$\begin{eqnarray}\begin{array}{rcl} {\rm WX} & = & {{a}_{0}}+{{a}_{8}}{\rm WX}\_8+{{a}_{s1,1}}{\rm S}1+{{a}_{S2,1}}{\rm S}2 \\ {} & {} & +{{a}_{S1,2}}{\rm S}{{1}^{2}}+{{a}_{S2,2}}{\rm S}{{2}^{2}}. \\ \end{array}\end{eqnarray} \tag{ 3 }$$

We performed a singular value decomposition to determine the best values for the coefficients and report our coefficients in Table 1.

Table 1.  Correction Coefficients to Produce Corrected Magnitudes for Extended Sources

Band	a0	a8	${{a}_{s1,1}}$	${{a}_{S2,1}}$	${{a}_{S1,2}}$	${{a}_{S2,2}}$
W1	−1.083	1.093	0.107	−0.591	0.021	−1.114
W2	−1.211	1.103	0.246	−1.104	0.062	−1.540
W3	−0.419	1.050	0.030	−0.449	0.049	−1.130
W4	−0.588	1.106	0.006	−1.021	−0.228	0.676

Download table as:  ASCIITypeset image

We present a comparison of our corrected photometry to that of the preliminary Jarrett photometry in Figure 10. Because these distributions are roughly Gaussian but with extended wings, we report the precision of our photometry in three ways. In our figures, we show the best-fit Gaussians to each of the residual distributions and report the width of these Gaussians as σ; this represents the typical systematic error due to limitations of the WISE pipeline for most of our extended sources. The long wings of the residual distribution typically represent blended and highly structured sources for which our simple calibration scheme failed to match Jarrett's more careful analysis, and a small number of extreme outliers inflate the standard deviation of these distributions beyond utility. To quantify the systematic photometric errors including the bulk of the non-Gaussian wings in a more robust manner, we calculated a "robust sigma" including outlier rejection⁸ and also the value of the 68th percentile absolute deviation from the median for each magnitude and color combination. We present the results of all three error estimates in Table 2.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

While a detailed study of individual targets would benefit from the more precise photometry of Jarrett's final catalog, this precision is sufficient for the initial exploration of the WISE data set and identification of superlative objects.

We have grouped the SIMBAD classifications for extragalactic sources into five broad categories:

• 1.  
  normal galaxy: includes the following SIMBAD types, Galaxy (G), Galaxy in Pair (GiP), Galaxy in Group (GiG), Galaxy in Cluster (GiC), Cluster of Galaxies (GlC), Group of Galaxies (GrG), Brightest galaxy in a Cluster (BiC), and Compact Group of Galaxies (CGG);
• 2.  
  active galaxy: LINERS (LIN), QSOs, Radio Galaxies (rG), Active galactic Nuclei (AGNs), Seyfert Galaxies (Sy1 + Sy2), BL Lac-type objects (BLL), and Blazars (Bla);
• 3.  
  star-forming galaxy: Emission Galaxy (EmG), H ii Galaxy (H2G), Starburst Galaxy (SBG), and Blue Compact Galaxy (bcG);
• 4.  
  interacting galaxy: Interacting Galaxy (IG) and Pair of Galaxies (PaG);
• 5.  
  low surface brightness galaxy: low surface brightness galaxy (LSB).

To characterize the galaxy population, we construct a simple color–magnitude diagram (W2–W3 versus W3). Figure 11 presents W2–W3 versus W3 for a variety of (mostly extragalactic) sources. The two most striking features in this figure are the almost complete absence of galaxies with W3 $\gt \;10.0$ and W2–W3 $\geqslant \;4.5$.

Zoom In Zoom Out Reset image size

The faint limit is significantly above the WISE detection limit in W3 of ∼11.2 (for point sources) and so requires explanation. This limit originates in our requirement that a galaxy be both detected and extended in WISE in the W3 band. We measure extendedness with W3RCHI2, and it appears that this parameter rarely takes values above 3 for sources with W3 ≳ 9, no matter how extended the object is. This lower limit is thus an artifact of the WISE data pipeline and our Extended Gold Sample selection criteria.

The upper left plot shows where the normal galaxies and the SIMBAD classification Part of a Galaxy reside in this color–magnitude space. These sources compose ∼82% of the total galaxy population. In the upper right, we show the color–magnitude diagram for star-forming and LSB galaxies, which compose ∼12% of the population. The lower left shows the active and interacting galaxies, composing ∼6% of the population. The lower right plot shows the generic SIMBAD classifications of (IR) and X-ray (X) sources. These last two are somewhat ambiguous classifications, and we include them only for completeness.

These results are quite consistent with our expectations validating the SIMBAD types. The normal galaxies appear to occupy the overall parameter space seen in the other populations, suggesting this to be a mixture of a variety of galaxy types. The star-forming galaxies appear to congregate at redder W2–W3 colors (W2–W3 $\geqslant \;3$), as expected. Interestingly, the LSB galaxies appear to tightly congregate on the fainter end of the star-forming population. The interacting galaxies appear to populate the same parameter space as seen by the star-forming galaxies, which is an expected result as a majority of interacting galaxies are experiencing high star formation rates. The active galaxies populate similar parameter space as seen in the normal galaxy population. The IR sources appear to be a mixture of both extragalactic sources and Galactic sources, though we do see a tight clustering occupying a similar parameter space as the LSB galaxies.

Though Galactic sources (i.e., sources within the Milky Way Galaxy) compose only ∼7% of our sample, it is important to understand how they might produce contamination in our investigation. SIMBAD provides classifications for over 70 different types of these Galactic sources, which we have reduced to six primary types: normal stars, Young Stellar Objects (YSO), Variables, Evolved, Evolved + IR, and stars in clusters. Our grouping methodology is as follows:

• 1.  
  normal star: Includes the following types: Star (*), Emission-line Star (Em*), Peculiar Star (Pe*), High proper-motion Star (PM*), Star in Double System (*i*), Be Star (Be*), and Eclipsing Binary (EB*);
• 2.  
  YSO: T Tau-type Star (TT*), Variable Star of FU Ori-type (FU*), Herbig–Haro Object (HH), Pre-main-sequence Star (pr*), and Young Stellar Object (Y*O);
• 3.  
  variable star: Variable Star (V*), Variable Star of beta Cep-type (bC*), Variable Star of Orion-type (Or*), Semi-regular Pulsating Star (sr*), Variable Star of W Vir-type (WV*), Long-period Variable Star (LP*), CV DQ Her-type (intermediate polar) (DQ*), Variable Star of delta Sct-type (dS*), Variable Star of alpha2 CVn-type (a2*), Cepheid Variable Star (Ce*), Variable Star with Rapid Variations (RI*), Variable Star of RV Tau-type (RV*), Pulsating Variable Star (Pu*), Variable Star of Mira Cet-type (Mi*), Variable Star of R CrB-type (RC*), Eruptive Variable Star (Er*), Variable Star of Irregular Type (Ir*), and Cataclysmic Variable Star (CV*);
• 4.  
  evolved star: Asymptotic Giant Branch Star (He-burning) (AB*), Red Giant Branch Star (RG*), Wolf–Rayet Star (WR*), and S Star (S*);
• 5.  
  evolved+IR star: Post-AGB Star (proto-PN) (pA*), Star with envelope of OH/IR-type (OH*), and Carbon Star (C*);
• 6.  
  star in cluster: Star in Cluster (*iC), Cluster of Stars (Cl*), and Star in Nebula (*iN).

For completeness we must also consider two other types of Galactic sources, PNe and Solar System Objects (SSOs) such as comets, asteroids, and planets. The SSOs are identified by utilizing the WISE All-sky Known SSO Possible Association List. We also identify two IR-bright planets, Neptune and Uranus, which are included in the sample.

To characterize the WISE high latitude Galactic population, we present Figure 12 which shows the color–magnitude diagrams for the various Galactic classifications. As before we find that the majority of the Galactic population is relatively rare at W3 $\gt \;10$ and W2–W3 $\geqslant \;4.5$, though we do find sources with very extreme W2–W3 colors, i.e., W2–W3 $\geqslant \;6$. We also find, as was expected, that the two types of sources that confuse and contaminate our search the most are the YSOs and PNe. While we do find contaminating sources within the Galactic sample, the scarcity of astrophysical sources with W2–W3 $\geqslant \;4.5$ should make our search relatively straightforward for high γ K3s.

Zoom In Zoom Out Reset image size

For every source we apply the methodology described in Wright et al. (2014a) to estimate the AGENT parameters γ (the fraction of starlight reradiated as alien waste heat) and ${{T}_{{\rm waste}}}$ (the waste heat's characteristic temperature) for each galaxy, assuming $\nu =0$ (no nonthermal alien power disposal/emission) and $\alpha =\gamma$ (that is, all of the waste heat originated as stellar power). We fit all four WISE bands to the three parameters $(L/4\pi {{d}^{2}})$ (the bolometric flux of the source), γ, and ${{T}_{{\rm waste}}}$.

For consistency, we modeled every galaxy as having an intrinsic SED represented by the old elliptical SED model Ell13 of Silva et al. (1998), of which a fraction α of the starlight is absorbed and re-emitted as a blackbody at temperature ${{T}_{{\rm waste}}}$. Of course, many galaxies have a significant blue stellar population and dust. Because the Ell13 template has the least MIR emission of all of the Silva et al. (1998) galaxy templates, using it will allow us to measure the maximum amount of MIR emission from each galaxy that could be attributed to ETIs from WISE broadband photometry alone. Its is thus consistent with our desire to set upper limits on K3 waste heat emission.

Using γ as the primary parameter for prioritizing sources, we inspected, reviewed, and graded every source in the W3 Extended Gold Sample with $\gamma \geqslant 0.25$, $\sim \;4000$ in all.

To facilitate this visual review, we constructed an interactive GUI system. This system provided an "at-a-glance chart" for each source (Figure 13) consisting of $2^{\prime} \times 2^{\prime}$ multiwavelength imaging cutouts (DSS, SDSS, 2MASS, and WISE) centered on the WISE source position, a variety of hyperlinks to supplementary imaging data and database queries centered on the source position (e.g., WISE (L1B + Atlas images), SIMBAD, NED), and printed information including source coordinates, WISE catalog colors and magnitudes, number of publications listed in SIMBAD, number of times WISE observed the source position in each filter (WXM), and number of times WISE detected the source in each filter (WXNM).

Zoom In Zoom Out Reset image size

The sample of red, extended objects in WISE is a combination of sources, galactic and extragalactic, real and instrumental, well-studied and new-to-science. We developed a sorting scheme to help us understand it better and prioritize sources for further investigation.

We defined a simple alphabetical grading scheme (A–F).

• 1.  
  A—Astrophysical red source with no previous publications in SIMBAD, little or no ancillary survey data. Highest priority candidate for observational follow-up.
• 2.  
  B—Astrophysical red source with some ancillary data or few publications, which do not provide convincing evidence that its nature is understood. Good candidate for follow-up.
• 3.  
  C—Visual review of source and/or publication list provides convincing evidence that its nature is understood. (Although in some cases, the SIMBAD classification may be incorrect.)
• 4.  
  D—Astrophysical source that creates artifacts detected as red extended sources by WISE. Most frequently associated with a bright star or large region of bright nebulosity.
• 5.  
  F—Fake/false source or artifacts, such as a latent image. Not a real astrophysical red source. Does not belong in any astronomical catalog.

We next provide some details of the decision process leading to the assignment of the various grades above, working backward from F to A.

Grade F. Persistence artifacts, or "latents," in the W3 band constitute the most common class of false, extended red source meriting an F grade. Most latents are culled by the initial visual classification (see Section 2.5), but some did make it into the Extended Gold Sample. The most pernicious cases involve W3 latents that happen to fall on the positions of legitimate W1+W2 sources; these closely resemble our expectations for a high-γ KIII galaxy in the WISE images. The visual review process readily discards such sources. They have very low W3NM/W3M, failing to appear in the majority of the individual WISE Level 1b frames (planets, comets, and asteroids also exhibit this behavior, but do so in all bands and appear in the Known Solar System Object Possible Association List; see Section 3.2). The wider-field WISE atlas images typically reveal the bright star responsible for producing a W3 latent.

Grade D. Unlike F-grade sources, sources assigned D grades are real astrophysical objects, but these are not valid, red extended WISE sources. There are two main classes: (1) saturated stars (W3 $\leqslant \;3.8\;{\rm mag}$) which would otherwise be point sources, and (2) bright "knots" within larger regions of mid-IR nebulosity (for example, associated with Galactic foreground emission). Our visual review of the WISE images readily identifies such cases.

Grade C. For sources that pass the quality control checks associated with F and D grades above, we next evaluate the citations listed for the associated SIMBAD source(s).

Sources that receive C grades typically have $\gtrsim 4$ citations, which, taken together, convince us that its astrophysical nature is understood.

If our review of the literature for a given source points toward the original object type listed in SIMBAD, we accept that object type. In many cases, we found either that (1) the preponderance of literature pointed to a different object type or (2) the nearest SIMBAD source returned by our automated matching was not the appropriate counterpart for the WISE source. If necessary, we manually overrode the SIMBAD object type and/or matching source for our catalog.

An example of a grade C source is shown in Figure 13.

Grades B and A. Our final two grades, B and A, represent real astrophysical extended red sources with scant ($\lesssim 4$) or zero existing literature citations, respectively; these sources should be given high priority for further observational follow-up to determine their nature.

Grade B sources are cited only in large survey catalogs containing minimal interpretation of individual sources. Commonly encountered catalog papers containing Grade B (and also C) extended red WISE sources are listed in Table 3.

Table 4.  Platinum Sample Published Catalog

Parameter	Type	Entry	Description
DESIGNATION	STRING	J190101.24–181215.0	WISE Designation
NAME	STRING	PNA6651	SIMBAD Identifier
GRADE	STRING	C	Grade (see Section 4.3)
SIMBAD_TYPE	STRING	PN	Object Type
R.A.	DOUBLE	285.25520	R.A. (degrees)
Decl.	DOUBLE	−18.204188	Decl. (degrees)
W1BEST	FLOAT	15.4980	W1 optimal photometry (mag)
W2BEST	FLOAT	13.889	W2 optimal photometry (mag)
W3BEST	FLOAT	9.0770	W3 optimal photometry (mag)
W4BEST	FLOAT	3.0690	W4 optimal photometry (mag)
PHOT_SOURCE	INT	1	Photometric System, see Section 5.1
TWASTE	FLOAT	72.923	Waste Heat Temperature (K)
GAMMA	FLOAT	0.99294	AGENT parameter γ
BOL_FLUX	FLOAT	255.83	Bolometric flux at Earth (${{L}_{\odot }}$/pc2)

Download table as:  ASCIITypeset image

The most promising candidates for observational follow-up are grade A sources that are isolated, meaning they are neither part of a cluster of red objects (for example, a young embedded star cluster or a galaxy cluster hosting many mergers) nor associated with diffuse mid-IR nebulosity (a hallmark of embedded star clusters in H ii regions or an indication that they may be associated with a large, extended galaxy). An example of an isolated "A" source and a cluster of red sources is presented in Figure 14.

Zoom In Zoom Out Reset image size

Note that this grading scheme only loosely tracks "interest" from a SETI perspective: a well-studied galaxy might have an anomalously high MIR luminosity and thus be an outstanding SETI target, while a new-to-science protostar might be manifestly ordinary and so very low priority. The "follow-up" of class A sources is thus primarily to determine their true nature and determine if they warrant further study from either a natural astrophysics or SETI perspective.

Figure 15 illustrates the colors and magnitudes of the sources we have graded, and shows where these sources reside in this parameter space.

Zoom In Zoom Out Reset image size

The Extended Gold Sample was constructed to be relatively free from non-extended or non-astrophysical sources, but our curation was deliberately liberal so that borderline cases could be handled more carefully on a case-by-case basis. Having constructed the Gold Sample, we performed this more careful analysis on those objects whose calibrated photometry indicate that they are especially red.

As discussed in Section 2.6, we have performed magnitude corrections using the photometry provided by the WISE pipeline. For every source in the Extended Gold Sample we now have photometry using four different photometric systems (i.e., MPRO (profile), MAG (aperture corrected magnitude), CMAG (corrected magnitude, see Section 2.6), and GMAG (elliptical aperture)). While the corrected magnitudes are reliable for most sources, there are special cases where a different photometric system may yield more reliable results. Given this, we have measured γ, ${{T}_{{\rm waste}}}$, and $L(/4\pi {{d}^{2}})$, (see Section 4.0, and Paper II) in each of the four different photometric systems.

To ensure that our choice of photometry does not cause us to miss any red sources, we selected from the Extended Gold Sample all sources with $\gamma \geqslant 0.25$ in any of the photometric systems. This identified a total of 3145 sources. We used our collected imaging, grading, and literature search results (Section 4.3) to carefully and visually vet this sample to identify and remove the most obvious contaminants, mostly nebulosity and saturated stars that survived our very liberal curation of the Extended Gold Sample. We identified and removed 366 obvious contaminants, reducing our list to 2779 objects.

We found that in a small percentage of cases, especially with blended sources, an extended source is broken up into two or more components in the All-sky catalog. In these cases we choose the entry coordinates closest to the extended source center and the photometric system that best captures the true magnitudes of the extended source. In some cases, the W3RCHI2 value of the catalog entry centered on an extended source is below three because of the details of how the All-sky catalog breaks up composite sources.

We segregated this sample into four different categories and determined which photometric system best fits the source in question.

• •  
  extended source: source is truly extended in the WISE images.
• •  
  point-source galaxy: source appears to be a galaxy from O/IR imaging or other sources, but appears unresolved in the WISE imagery, despite having a high W3RCHI2 value.
• •  
  point-source star: source appears to be stellar in nature.
• •  
  junk: source is a contaminant, and should not be in any catalog of real astrophysical sources.

In addition to classifying these sources as described above, we also determine which photometric system best estimates the true photometry of the given source. We use the following convention to indicate the best possible WISE photometry in the source tables:

• •  
  0: MPRO (Profile fit photometry)
• •  
  1: MAG (Aperture-corrected magnitude)
• •  
  2: CMAG (Corrected magnitude)
• •  
  3: GMAG (2MASS XSC elliptical aperture).

Using these conventions we identify 1296(46.7%) classified as extended, 974(35%) classified as unresolved galaxies, 263(9.4%) classified as stellar, and 246(8.9%) classified as junk. Of the extended sources, 563 have $\gamma \gt 0.25$ in the photometric system most appropriate for that source. These compose our W3 Extended Platinum Sample of real, extended, red sources in WISE.

We present a catalog of 563 sources deemed to be extended and real in the WISE 12 μm filter and identified as the Platinum Sample as a FITS file associated with this manuscript. This list of sources has also been selected as those having $\gamma \geqslant 0.25$ in the preferred photometric system. Table 4 gives the detailed parameter description for this electronic catalog. For brevity, we present a minimal number of parameters representing the most important measurements for these sources, and the WISE identifier so that users may cross-reference our catalog to the WISE All-sky or ALLWISE catalogs. Presented in this parameter list are the coordinates, optimal WISE photometry and the photometric system used (see Section 5.1), the AGENT parameters, and pertinent information derived from the SIMBAD database. Tables 6– 9 have been derived using this catalog.

Table 5.  Five Spirals Red in Optical, ${\rm NUV}-r$, and MIR Colors

WISE designation	$({\rm NUV}-r)$	${{T}_{{\rm waste}}}$	γ	SIMBAD name
J085428.94+084444.4	4.62	179	0.258	2MASX J08542894+08444430
J113325.93+141618.9	4.94	180	0.198	2MASX J11332590+1416186
J142619.17+473357.8	4.51	180	0.202	SDSS J142619.17+473357.7
J142859.55+605000.5	4.56	163	0.212	2MASX J14285959+6050005
J230616.43+135856.4	4.78	239	0.172	2MASXJ23061640+1358560

Download table as:  ASCIITypeset image

Table 6.  Extreme WISE Colors

R.A.	Decl.	SIMBAD	Grade	W1	W2	W3	W4	Color	$\gamma$	${{T}_{{\rm waste}}}$	Photometry	Note
hhmmss.ss	ddmmss.s	type		(mag)	(mag)	(mag)	(mag)	(mag)		K
W1–W2 Most Extreme Sources
072533.95	+292919.9	PN	C	11.20	8.89	4.33	0.44	2.31	0.83	123	2	NGC 2371
213652.95	+124719.2	PN	C	11.68	9.49	5.25	0.53	2.19	0.92	98	2	NGC 7094
180600.64	+002235.5	PN	C	15.12	13.01	8.52	3.56	2.11	0.90	248	1	PN-G028.0+10.2
054206.16	+090511.6	PN	C	10.06	7.98	4.07	−0.33	2.09	0.83	106	2	NGC 2022
060253.97	−710310.0	LIN	C	11.59	9.52	6.14	3.25	2.07	0.66	349	0	2MASX J06025406-7103104
122430.65	−184700.7	PN	C	9.61	7.66	4.33	0.00	1.95	0.71	108	2	NGC 4361
214035.92	+663525.8	Y*O	B	10.05	8.25	5.58	3.05	1.80	0.45	403	1	2MASS J21403852+6635017
194357.75	−140913.5	PN	C	8.85	7.09	3.14	−0.88	1.75	0.72	118	2	NGC 6818
193121.41	−723921.3	Sy2	C	10.29	8.54	5.55	2.06	1.75	0.36	137	2	Antennae
003443.48	−000226.6	Sy2	C	11.72	9.98	6.57	3.84	1.74	0.54	347	2	2MFGC-403
W2–W3 Most Extreme Sources
091742.98	+131214.8	Comet	C	12.76	11.50	5.02	0.99	6.49	0.93	204	2	29 P/Schwassmann-Wachmann 1 1927 V1
155812.25	−141806.2	IR	B	16.48	15.39	9.13	4.03	6.26	0.99	89	0	IRAS-15553-1409
235821.58	−320748.6	Comet	C	16.56	15.71	9.92	4.97	5.79	0.97	93	1	Garradd 2009 P1
045608.93	−674947.5		C	14.39	14.51	8.79	6.52	5.71	0.51	211	1	Object in LMC
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	5.34	0.85	120	2	NGC 4355
052703.47	−664959.2		C	15.10	14.63	9.32	7.90	5.31	0.45	282	0	Object in LMC
100918.85	−805128.1	PN	C	10.81	9.80	4.61	2.88	5.19	0.59	290	2	NGC 3195
055542.61	+032331.8	H2G	C	10.99	10.12	4.94	1.52	5.18	0.72	142	2	UGCA 116
061452.97	−062244.4	Y*O	B	8.80	8.31	3.15	1.78	5.16	0.42	289	2	IRAS 06124-0621
004036.06	+410117.9	HII	C	14.43	13.76	8.62	6.45	5.13	0.52	263	0	BA1-441
W1–W3 Most Extreme Sources
091742.98	+131214.8	Comet	C	12.76	11.50	5.02	0.99	7.74	0.93	204	2	29 P/Schwassmann-Wachmann 1 1927 V1
155812.25	−141806.2	IR	B	16.48	15.39	9.13	4.03	7.34	0.99	89	0	IRAS-15553-1409
072533.95	+292919.9	PN	C	11.20	8.89	4.33	0.44	6.87	0.83	123	2	NGC 2371
235821.58	−320748.6	Comet	C	16.56	15.71	9.92	4.97	6.64	0.97	93	1	Garradd 2009 P1
180600.64	+002235.5	PN	C	15.12	13.01	8.52	3.56	6.60	0.90	248	1	PN-G028.0+10.2
213652.95	+124719.2	PN	C	11.68	9.49	5.25	0.53	6.43	0.92	98	2	NGC 7094
190101.24	−181215.0	PN	C	15.50	13.89	9.08	3.07	6.42	0.99	73	1	PN A6651
185334.98	+330152.2	PN	C	11.79	10.33	5.48	0.85	6.30	0.93	100	1	Ring Nebula
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	6.23	0.85	120	2	NGC 4355
041415.79	−124420.7	PN	C	9.12	7.98	2.91	−0.05	6.21	0.69	258	2	NGC 1535
W3–W4 Most Extreme Sources
190101.24	−181215.0	PN	C	15.50	13.89	9.08	3.07	6.01	0.99	73	1	PNA6651
095551.65	+694046.6	Em*	C	6.38	5.40	3.73	−1.79	5.53	0.75	77	1	[DOB2000]-19
032910.31	+312158.0	Y*O	C	6.76	5.85	3.62	−1.63	5.25	0.73	83	1	[LAL96]-213
155812.25	−141806.2	IR	B	16.48	15.39	9.13	4.03	5.10	0.99	89	0	IRAS-15553-1409
215935.03	−392308.7	PN	C	16.42	15.54	10.75	5.70	5.05	0.95	90	1	IC 5148
180600.64	+002235.5	PN	C	15.12	13.01	8.52	3.56	4.96	0.90	248	1	PN-G028.0+10.2
235821.58	−320748.6	Comet	C	16.56	15.71	9.92	4.97	4.94	0.97	93	1	Garradd 2009 P1
071448.80	−465733.6	PN	C	11.68	10.09	6.39	1.59	4.79	0.86	95	2	ESO 256-19
213652.95	+124719.2	PN	C	11.68	9.49	5.25	0.53	4.72	0.92	98	2	NGC 7094
160426.50	+404059.1	PN	C	12.54	12.04	7.48	2.78	4.70	0.87	98	2	NGC 6058
W1–W4 Most Extreme Sources
155812.25	−141806.2	IR	B	16.48	15.39	9.13	4.03	12.45	0.99	89	0	IRAS-15553-1409
190101.24	−181215.0	PN	C	15.50	13.89	9.08	3.07	12.43	0.99	73	1	PN A6651
091742.98	+131214.8	Comet	C	12.76	11.50	5.02	0.99	11.77	0.93	204	2	29 P/Schwassmann-Wachmann 1 1927 V1
235821.58	−320748.6	Comet	C	16.56	15.71	9.92	4.97	11.58	0.97	93	1	Garradd 2009 P1
180600.64	+002235.5	PN	C	15.12	13.01	8.52	3.56	11.56	0.90	248	1	PN-G028.0+10.2
213652.95	+124719.2	PN	C	11.68	9.49	5.25	0.53	11.15	0.92	98	2	NGC 7094
185334.98	+330152.2	PN	C	11.79	10.33	5.48	0.85	10.94	0.93	100	1	Ring Nebula
072533.95	+292919.9	PN	C	11.20	8.89	4.33	0.44	10.76	0.83	123	2	NGC 2371
215935.03	−392308.7	PN	C	16.42	15.54	10.75	5.70	10.71	0.95	90	1	IC 5148
160628.19	−354520.9	PN	C	10.61	9.25	4.52	0.08	10.53	0.89	105	2	PN-G341.5+12.1
W2–W4 Most Extreme Sources
155812.25	−141806.2	IR	B	16.48	15.39	9.13	4.03	11.36	0.99	89	0	IRAS-15553-1409
190101.24	−181215.0	PN	C	15.50	13.89	9.08	3.07	10.82	0.99	73	1	PN A6651
235821.58	−320748.6	Comet	C	16.56	15.71	9.92	4.97	10.73	0.97	93	1	Garradd 2009 P1
091742.98	+131214.8	Comet	C	12.76	11.50	5.02	0.99	10.52	0.93	204	2	29 P/Schwassmann-Wachmann 1 1927 V1
215935.03	−392308.7	PN	C	16.42	15.54	10.75	5.70	9.83	0.95	90	1	IC 5148
185334.98	+330152.2	PN	C	11.79	10.33	5.48	0.85	9.48	0.93	100	1	Ring Nebula
180600.64	+002235.5	PN	C	15.12	13.01	8.52	3.56	9.45	0.90	248	1	PN-G028.0+10.2
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	9.30	0.85	120	2	NGC 4355
160426.50	+404059.1	PN	C	12.54	12.04	7.48	2.78	9.25	0.87	98	2	NGC 6058
160628.19	−354520.9	PN	C	10.61	9.25	4.52	0.08	9.17	0.89	105	2	PN-G341.5+12.1

Download table as:  ASCIITypeset images: 1 2

Table 7.  Galaxies with Extreme WISE Colors

R.A.	Decl.	SIMBAD	Gradea	W1	W2	W3	W4	Color	$\gamma$	${{T}_{{\rm waste}}}$	Photometryb	Note
hhmmss.ss	ddmmss.s	type		(mag)	(mag)	(mag)	(mag)	(mag)		K
W1–W2 Most Extreme Galaxies
060253.97	−710310.0	LIN	C	11.59	9.52	6.14	3.25	2.07	0.66	349	0	2MASX J06025406-7103104
193121.41	−723921.3	Sy2	C	10.29	8.54	5.55	2.06	1.75	0.36	137	2	SUPER ANTENNAE
003443.48	−000226.6	Sy2	C	11.72	9.98	6.57	3.84	1.74	0.54	347	2	2MFGC-403
080106.63	−660914.9	G	C	10.20	8.51	5.24	2.17	1.69	0.52	338	2	6dFGSg J080106.6-660915
002153.60	−791007.8	Sy2	C	10.06	8.39	5.22	2.14	1.67	0.49	341	2	2MASX J00215355-7910077
100125.94	+154612.2	G	C	7.99	6.33	4.06	1.14	1.65	0.34	401	2	NGC 3094
033639.05	−205406.8	GiG	C	9.58	7.98	4.69	1.50	1.60	0.50	328	2	NGC 1377
183820.32	−652539.0	Sy2	C	9.37	7.80	4.32	1.21	1.57	0.52	322	2	ESO 103-35
173801.51	+561325.9	Sy2	C	11.08	9.57	6.19	3.51	1.51	0.45	343	2	2MASX J17380143+5613257
112402.72	−282315.4	Sy2	C	10.81	9.31	5.98	3.60	1.50	0.43	359	2	IRAS11215-2806
W2–W3 Most Extreme Galaxies
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	5.34	0.85	120	2	NGC 4355
055542.61	+032331.8	H2G	C	10.99	10.12	4.94	1.52	5.18	0.72	142	2	UGCA 116
042826.03	−043349.2	LIN	C	11.62	10.99	5.92	2.00	5.06	0.79	121	2	IRAS 04259-0440
103833.62	−071014.4	G	C	9.49	9.10	4.14	0.80	4.97	0.61	146	1	IC 630
001850.88	−102236.6	EmG	C	10.76	10.29	5.41	2.37	4.89	0.52	162	0	MCG 02-01-051
181338.76	−574356.9	G	C	11.26	10.84	5.99	2.57	4.85	0.60	142	0	IC 4686
225234.71	+244349.4	AGN	C	11.44	10.87	6.06	2.67	4.81	0.60	143	2	Mrk 309
010426.95	−640712.2	H2G	C	11.23	10.81	6.03	2.78	4.78	0.54	150	0	ESO 79-16
195405.20	+495647.0	G	B	12.57	12.16	7.39	4.81	4.76	0.39	195	0	LEDA 200363
043400.03	−083444.9	AGN	C	8.74	8.21	3.54	−0.07	4.67	0.63	133	2	NGC 1614
W1–W3 Most Extreme Galaxies
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	6.23	0.85	120	2	NGC 4355
055542.61	+032331.8	H2G	C	10.99	10.12	4.94	1.52	6.06	0.72	142	2	UGCA 116
205826.80	−423900.3	LIN	C	11.08	9.69	5.33	1.75	5.75	0.65	135	2	ESO 286-19
042826.03	−043349.2	LIN	C	11.62	10.99	5.92	2.00	5.69	0.79	121	2	IRAS 04259-0440
003652.44	−333316.8	AGN	C	10.52	9.09	4.84	1.17	5.69	0.66	131	2	ESO 350-38
124930.16	−112403.4	Sy2	C	11.46	10.10	5.90	2.79	5.55	0.59	286	1	IRAS 12468-1107
060253.97	−710310.0	LIN	C	11.59	9.52	6.14	3.25	5.45	0.66	349	0	2MASX J06025406-7103104
012002.63	+142142.5	LIN	C	10.84	9.67	5.41	2.24	5.42	0.49	154	0	MCG+02-04-025
011607.20	+330521.7	Sy2	C	11.09	9.78	5.70	2.57	5.38	0.54	286	0	NGC 449
225234.71	+244349.4	AGN	C	11.44	10.87	6.06	2.67	5.38	0.60	143	2	Mrk 309
W3–W4 Most Extreme Galaxies
131503.51	+243707.8	Q?	C	10.44	10.32	7.05	2.45	4.60	0.61	99	2	IC 860
153457.25	+233011.5	SyG	C	9.52	8.96	4.55	0.19	4.35	0.79	107	2	Arp 220
125145.54	+254628.5	IG	C	9.74	9.58	6.24	1.92	4.32	0.53	107	2	NGC 4747
000820.57	+403755.9	Sy2	C	12.41	12.27	8.58	4.44	4.14	0.54	113	3	2MASX J00082041+4037560
042759.96	−475445.8	LIN	C	9.52	9.44	6.79	2.81	3.98	0.27	115	2	CARAFE NEBULA
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	3.96	0.85	120	2	NGC 4355
065558.96	−404912.2	G	B	11.12	10.72	7.34	3.41	3.93	0.43	119	2	6dFGSg J065559.0-404912
042826.03	−043349.2	LIN	C	11.62	10.99	5.92	2.00	3.92	0.79	121	2	IRAS 04259-0440
102508.18	+170914.1	IG	C	12.56	12.15	8.06	4.15	3.92	0.58	121	0	NGC 3239
202825.49	−330420.5	G	B	10.33	10.32	7.15	3.29	3.86	0.32	121	2	ESO 400-28
W1–W4 Most Extreme Galaxies
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	10.19	0.85	120	2	NGC 4355
042826.03	−043349.2	LIN	C	11.62	10.99	5.92	2.00	9.61	0.79	121	2	IRAS 04259-0440
055542.61	+032331.8	H2G	C	10.99	10.12	4.94	1.52	9.47	0.72	142	2	UGCA 116
003652.44	−333316.8	AGN	C	10.52	9.09	4.84	1.17	9.35	0.66	131	2	ESO 350-38
153457.25	+233011.5	SyG	C	9.52	8.96	4.55	0.19	9.33	0.79	107	2	Arp 220
205826.80	−423900.3	LIN	C	11.08	9.69	5.33	1.75	9.33	0.65	135	2	ESO 286-19
205724.32	+170738.5	G	C	10.57	9.88	5.27	1.43	9.15	0.70	124	2	IRAS F20550+1655-SE
133955.96	−313824.4	AGN	C	8.27	7.00	2.95	−0.73	9.01	0.60	130	2	NGC 5253
150029.00	−262649.2	H2G	C	11.96	11.36	6.82	3.08	8.88	0.65	128	2	2MASX J15002897-2626487
231546.75	−590314.5	Sy2	C	10.69	9.67	5.34	1.87	8.83	0.57	139	2	ESO 148-2
W2–W4 Most Extreme Galaxies
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	9.30	0.85	120	2	NGC 4355
042826.03	−043349.2	LIN	C	11.62	10.99	5.92	2.00	8.98	0.79	121	2	IRAS 04259-0440
153457.25	+233011.5	SyG	C	9.52	8.96	4.55	0.19	8.77	0.79	107	2	Arp 220
055542.61	+032331.8	H2G	C	10.99	10.12	4.94	1.52	8.60	0.72	142	2	UGCA 116
205724.32	+170738.5	G	C	10.57	9.88	5.27	1.43	8.45	0.70	124	2	IRAS F20550+1655-SE
234709.20	+153548.3	Sy2	C	11.02	10.56	5.91	2.20	8.36	0.65	129	2	MCG+02-60-017
103833.62	−071014.4	G	C	9.49	9.10	4.14	0.80	8.31	0.61	146	1	IC 630
150029.00	−262649.2	H2G	C	11.96	11.36	6.82	3.08	8.29	0.65	128	2	2MASX J15002897-2626487
043400.03	−083444.9	AGN	C	8.74	8.21	3.54	−0.07	8.28	0.63	133	2	NGC 1614
181338.76	−574356.9	G	C	11.26	10.84	5.99	2.57	8.27	0.60	142	0	IC 4686

^a Grade refers to how well understood an object is (see Section 4.3). All of these objects have grade C indicating that they are understood, having been discussed in the refereed literature. ^b Photometry refers to the source of the photometry used in the table: 0: profile fit; 1: aperture-corrected; 2: calibrated, 3: elliptical aperture. See Section 5.1.

Download table as:  ASCIITypeset images: 1 2

Table 8.  Galaxies with Extreme WISE $\gamma$ Values

R.A.	Decl.	SIMBAD	Gradea	W1	W2	W3	W4	$\gamma$	${{T}_{{\rm waste}}}$	Photometryb	Note
hhmmss.ss	ddmmss.s	type		(mag)	(mag)	(mag)	(mag)		K
122654.61	−005239.1	Sy2	C	10.04	9.15	3.81	−0.15	0.85	120	2	NGC 4355
042826.03	−043349.2	LIN	C	11.62	10.99	5.92	2.00	0.79	121	2	IRAS 04259-0440
153457.25	+233011.5	SyG	C	9.52	8.96	4.55	0.19	0.79	107	2	Arp 220
055542.61	+032331.8	H2G	C	10.99	10.12	4.94	1.52	0.72	142	2	UGCA 116
205724.32	+170738.5	G	C	10.57	9.88	5.27	1.43	0.70	124	2	IRAS F20550+1655-SE
060253.97	−710310.0	LIN	C	11.59	9.52	6.14	3.25	0.66	349	0	2MASX J06025406-7103104
003652.44	−333316.8	AGN	C	10.52	9.09	4.84	1.17	0.66	131	2	ESO 350-38
150029.00	−262649.2	H2G	C	11.96	11.36	6.82	3.08	0.65	128	2	2MASX J15002897-2626487
234709.20	+153548.3	Sy2	C	11.02	10.56	5.91	2.20	0.65	129	2	MCG +02-60-017
205826.80	−423900.3	LIN	C	11.08	9.69	5.33	1.75	0.65	135	2	ESO 286-19
043400.03	−083444.9	AGN	C	8.74	8.21	3.54	−0.07	0.63	133	2	NGC 1614
131503.51	+243707.8	Q?	C	10.44	10.32	7.05	2.45	0.61	99	2	IC 860
103833.62	−071014.4	G	C	9.49	9.10	4.14	0.80	0.61	146	1	IC 630
181338.76	−574356.9	G	C	11.26	10.84	5.99	2.57	0.60	142	0	IC 4686
225234.71	+244349.4	AGN	C	11.44	10.87	6.06	2.67	0.60	143	2	Mrk 309
151806.13	+424444.8	LIN	C	10.69	10.05	5.60	1.96	0.60	132	2	IRAS F15163+4255-NW
133955.96	−313824.4	AGN	C	8.27	7.00	2.95	−0.73	0.60	130	2	NGC 5253
121539.36	+361935.1	SBG	C	11.13	10.70	6.31	2.59	0.59	128	1	NGC 4228
124930.16	−112403.4	Sy2	C	11.46	10.10	5.90	2.79	0.59	286	1	IRAS 12468-1107
102508.18	+170914.1	IG	C	12.56	12.15	8.06	4.15	0.58	121	0	NGC 3239
231546.75	−590314.5	Sy2	C	10.69	9.67	5.34	1.87	0.57	139	2	ESO 148-2
134442.10	+555313.3	Sy2	C	10.22	9.00	5.36	1.51	0.55	123	2	Mrk 273
121346.00	+024840.3	LIN	C	12.11	11.39	7.02	3.54	0.55	139	2	LEDA 39024
062722.52	−471046.7	IG	C	11.52	11.06	6.66	3.11	0.55	135	3	ESO 255-7
025941.29	+251415.0	G	C	11.89	11.46	6.99	3.49	0.54	138	1	NGC 1156
011607.20	+330521.7	Sy2	C	11.09	9.78	5.70	2.57	0.54	286	0	NGC 449
010426.95	−640712.2	H2G	C	11.23	10.81	6.03	2.78	0.54	150	0	ESO 79-16
003443.48	−000226.6	Sy2	C	11.72	9.98	6.57	3.84	0.54	347	2	2MFGC-403
000820.57	+403755.9	Sy2	C	12.41	12.27	8.58	4.44	0.54	113	3	2MASX J00082041+4037560
021037.63	−154624.2	G	C	10.95	10.49	6.01	2.54	0.54	140	2	NGC 814
005404.02	+730505.7	G	C	9.31	8.75	4.23	0.84	0.54	143	2	MCG+12-02-001
130220.39	−154559.0	GiG	C	9.86	9.43	5.14	1.55	0.53	134	2	MCG-02-33-099
125145.54	+254628.5	IG	C	9.74	9.58	6.24	1.92	0.53	107	2	NGC 4747
134818.91	−505838.8	G	B	11.08	10.65	6.00	2.73	0.52	149	0	2MASX J13481892-5058391
001850.88	−102236.6	EmG	C	10.76	10.29	5.41	2.37	0.52	162	0	MCG-02-01-051
080106.63	−660914.9	G	C	10.20	8.51	5.24	2.17	0.52	338	2	6dFGSg J080106.6-660915
183820.32	−652539.0	Sy2	C	9.37	7.80	4.32	1.21	0.52	322	2	ESO 103-35
043548.45	+021529.6	G	C	11.17	10.03	6.03	2.54	0.52	138	2	UGC 3097
002131.65	−483728.7	IG	C	8.71	8.24	3.89	0.44	0.51	140	2	NGC 92
102751.28	−435414.5	GiG	C	7.48	6.83	2.58	−0.88	0.51	139	2	6dFGSg J102751.3-435414
054323.63	+540044.2	G	B	11.54	11.07	6.52	3.24	0.51	149	2	2MASX J05432362+5400439
064651.13	−645727.7	IG	C	12.10	11.78	7.30	3.90	0.50	142	0	ESO 87-41
105918.14	+243234.6	LIN	C	10.90	10.12	5.63	2.44	0.50	153	0	2XMM J105918.1+243234
130842.02	−242257.8	Sy2	C	11.29	10.22	5.91	2.72	0.50	153	2	PKS 1306-241
033639.05	−205406.8	GiG	C	9.58	7.98	4.69	1.50	0.50	328	2	NGC 1377
002153.60	−791007.8	Sy2	C	10.06	8.39	5.22	2.14	0.49	341	2	2MASX J00215355-7910077
012002.63	+142142.5	LIN	C	10.84	9.67	5.41	2.24	0.49	154	0	MCG+02-04-025
034648.35	+680546.5	GiG	C	6.95	6.61	2.54	−1.12	0.49	131	2	IC 342
022557.05	−244240.8	IG	C	11.80	11.57	7.55	3.82	0.49	127	1	AM 0223-245
233614.11	+020917.9	AGN	C	9.31	9.04	4.73	1.24	0.49	138	2	NGC 7714

^a Grade refers to how well understood an object is (see Section 4.3.) Grade B is reserved for objects with very little presence in the refereed literature, such that we are not convinced the object's true nature has been carefully verified. Grade A is given to objects that are effectively new to science having, at most, been detected in surveys but never examined. ^b Photometry refers to the source of the photometry used in the table: 0: profile fit; 1: aperture-corrected; 2: calibrated, 3: elliptical aperture. See Section 5.1.

Download table as:  ASCIITypeset image

Table 9.  Extreme WISE $\gamma$ sources (Grades A+B)

NLS #	R.A.	Decl.	SIMBAD	Gradea	W1	W2	W3	W4	$\gamma$	${{T}_{{\rm waste}}}$ (K)	Photometryb	Note
	hhmmss.ss	ddmmss.s	type		(mag)	(mag)	(mag)	(mag)		K
1	224436.12	+372533.6		A	11.27	9.79	6.92	4.44	0.35	381	2
2	073504.83	−594612.4		A	12.86	12.59	8.03	5.80	0.30	229	0
3	162721.02	+824538.7	IR	A	13.39	13.18	9.41	6.31	0.27	156	0	IRAS 16329+8252
4	155812.25	−141806.2	IR	B	16.48	15.39	9.13	4.03	0.99	89	0	IRAS 15553-1409
5	031805.43	−663024.0	Cl*	B	12.12	11.69	6.92	3.42	0.61	138	1	[L2004]-n1313-341
6	005342.56	−723920.9	EB*	B	13.17	12.78	8.43	4.77	0.56	131	0	OGLE J005342.56-723921.4
7	125552.98	−765608.7	HH	B	12.38	10.67	8.14	6.92	0.53	516	3	HH-54H2
8	134818.91	−505838.8	G	B	11.08	10.65	6.00	2.73	0.52	149	0	2MASX J13481892-5058391
9	054323.63	+540044.2	G	B	11.54	11.07	6.52	3.24	0.51	149	2	2MASX J05432362+5400439
10	214140.73	−454223.0	IR	B	11.57	10.54	7.33	3.28	0.50	115	2	IRAS F21384-4556
11	012933.33	−733344.3	Y*?	B	10.57	9.73	6.57	2.57	0.45	117	2	[GQH2007]-50
12	214035.92	+663525.8	Y*O	B	10.05	8.25	5.58	3.05	0.45	403	1	2MASS J21403852+6635017
13	031803.99	−663234.1	WR*	B	13.30	12.93	8.61	5.32	0.45	147	1	[HC2007]-15
14	065558.96	−404912.2	G	B	11.12	10.72	7.34	3.41	0.43	119	2	6dFGSg J065559.0-404912
15	163600.57	+101340.4	G	B	13.09	12.56	8.22	5.09	0.43	156	1	2MASX J16360060+1013396
16	061452.97	−062244.4	Y*O	B	8.80	8.31	3.15	1.78	0.42	289	2	IRAS 06124-0621
17	093016.68	+212150.5	G	B	11.75	11.31	6.94	3.86	0.41	159	2	SDSSCGB-51158.1
18	020805.42	−291432.7	G	B	11.64	10.16	6.98	4.19	0.41	346	2	2dFGRSTGS306Z043
19	224458.08	−014600.4	G	B	11.10	10.55	6.18	3.16	0.41	162	2	2MASX J22445816-0145589
20	195405.20	+495647.0	G	B	12.57	12.16	7.39	4.81	0.39	195	0	LEDA 200363
21	044738.41	−172601.8	G	B	11.74	11.39	7.38	4.02	0.39	144	2	ESO 552-5
22	113231.06	+742242.0	G	B	11.59	11.30	7.06	3.89	0.39	153	2	2MASX J11323098+7422421
23	154813.36	−245309.6	EmG	B	11.29	11.00	6.74	3.59	0.39	155	1	ESO 515-7
24	181603.19	+473705.4	G	B	11.46	11.13	6.79	3.74	0.39	160	2	2MASX J18160312+4737056
25	024143.24	+454626.6	G	B	11.71	11.35	7.22	3.99	0.38	150	2	2MASX J02414325+4546272
26	045510.80	+053512.6	GiC	B	12.52	12.12	8.00	4.81	0.38	152	2	2MASX J04551070+0535126
27	181041.37	+250723.2	G	B	11.88	11.64	7.19	4.23	0.38	166	0	2MASX J18104135+2507238
28	061647.61	−090133.6	IR	B	8.64	8.18	3.34	1.60	0.37	280	2	IRAS 06144-0900
29	185222.44	−293620.7	EmG	B	10.77	10.49	6.52	3.18	0.37	144	2	6dFGSg J185222.4-293621
30	134547.40	+700445.9	G	B	11.24	10.87	6.59	3.58	0.37	162	2	2MASX J13454733+7004455
31	025559.96	+474819.3	G	B	10.74	10.49	6.39	3.16	0.37	150	2	MCG+08-06-022
32	052134.60	−660547.1	Y*O	B	13.02	12.39	7.88	5.30	0.36	197	1	[GC2009]J052134.67m660548.3
33	083538.40	−011407.1	G	B	11.68	10.65	6.82	4.38	0.36	310	2	2MASX J08353838-0114072
34	173429.01	−040541.7	G	B	13.11	12.58	8.20	5.44	0.35	180	0	6dFGSg J173428.9-040542
35	045542.30	−712034.9	Y*?	B	13.52	13.18	8.30	6.55	0.35	273	1	GMP 19
36	195751.88	−322128.2	EmG	B	10.70	10.40	6.28	3.15	0.35	155	2	6dFGSg J195751.9-322128
37	191227.31	−290235.7	EmG	B	11.66	11.33	6.97	4.10	0.35	171	0	ESO 459-7
38	132021.98	−233225.9	EmG	B	11.51	10.67	6.71	3.71	0.35	164	2	2MASX J13202200-2332256
39	152309.66	−393448.2	G	B	11.74	11.28	7.13	4.17	0.34	165	0	2MASX J15230967-3934481
40	035715.22	−261859.3	G	B	13.96	13.51	9.37	6.39	0.34	164	0	APMBGC 483+082+074
41	054746.71	−444953.8	IR	B	11.95	11.63	7.49	4.47	0.33	162	2	IRAS 05463-4450
42	064233.41	−645925.4	G	B	13.41	13.12	9.01	5.96	0.33	160	0	AM 0642-645
43	025609.76	−153943.5	G	B	12.60	12.01	8.33	5.04	0.33	147	3	6dFGSg J025609.8-153946
44	042851.45	+693447.1	G	B	11.72	11.32	6.89	4.26	0.33	190	0	2MASX J04285125+6934469
45	150253.22	+165508.4	GiG	B	10.85	10.51	6.25	3.40	0.33	172	0	MCG+03m38m076
46	025039.58	+414012.5	IR	B	10.91	10.73	7.24	3.65	0.32	133	1	IRAS 02474+4127
47	102050.93	−171859.4	EmG	B	11.95	11.63	7.38	4.52	0.32	172	0	MCG-03-27-005
48	182552.75	+375241.6	IR	B	12.10	11.91	7.77	4.74	0.32	161	0	IRAS 18241+3750
49	202825.49	−330420.5	G	B	10.33	10.32	7.15	3.29	0.32	121	2	ESO 400-28
50	140330.96	−504643.1	HI	B	13.45	13.38	9.81	6.26	0.32	134	0	HIPASS J1403-50
51	092338.22	−251634.9	EmG	B	11.11	10.86	6.69	3.74	0.32	166	2	6dFGSg J092338.2-251635
52	013506.96	−412611.9	G	B	9.64	9.55	5.76	2.40	0.32	143	2	6dFGSg J013506.9-412612
53	093548.86	−291955.6	EmG	B	11.84	10.62	7.34	5.08	0.32	355	0	ESO 434-13
54	194112.82	+630542.9	G	B	11.76	11.42	7.30	4.37	0.31	167	2	2MASX J19411289+6305430
55	105416.74	−394019.3	G	B	11.28	11.07	7.12	3.98	0.31	155	2	ESO 318-23
56	125324.16	−234545.6	EmG	B	12.69	12.43	8.40	5.37	0.31	161	0	6dFGSg J125324.2-234546
57	044016.10	−444525.7	G	B	11.77	11.46	7.38	4.43	0.31	166	2	6dFGSg J044016.1-444526
58	054421.56	−135311.9	G	B	12.02	11.65	7.29	4.70	0.31	192	0	2MASX J05442151-1353116
59	174709.21	−643817.6	WR?	B	9.54	9.18	6.11	2.37	0.30	125	2	[CB2009]-A1-C1
60	004250.05	−365243.2	G	B	11.46	11.24	6.87	4.22	0.30	187	0	6dFGSg J004250.1-365241
61	111859.15	−400013.2	G	B	11.45	11.08	6.55	4.42	0.30	248	0	2MASX J11185912-4000135
62	051521.42	−262817.1	EmG	B	11.52	11.25	7.34	4.25	0.30	157	2	ESO 486-39
63	040819.05	−611605.7	IG	B	11.11	10.89	6.91	3.87	0.30	160	2	ESO 118-4
64	111932.31	−471015.7	G	B	12.05	11.83	7.94	4.82	0.30	155	2	2MASX J11193186-4710218
65	050147.35	−181000.8	IG	B	12.20	11.99	7.57	5.02	0.30	195	0	NGC 1739
66	063226.08	−243208.3	G	B	12.50	12.32	8.13	5.29	0.30	173	0	ESO 490-11
67	070912.66	−440731.7	G	B	12.58	12.40	8.17	5.38	0.29	177	0	ESO 256-16
68	045605.74	+015932.1		B	12.17	11.96	7.75	4.97	0.29	177	0
69	173431.80	+471301.7	G	B	11.75	11.49	7.29	4.54	0.29	179	2	2MASX J17343177+4713010
70	222149.97	+395024.0	GiG	B	12.43	12.08	7.63	5.31	0.29	219	1	IRAS 22196+3935
71	105052.15	+010944.2	GiG	B	12.09	11.75	7.36	4.94	0.29	208	0	IC 649 S
72	114543.59	−114712.6	EmG	B	11.30	11.00	6.55	4.21	0.29	216	0	6dFGSg J114543.6-114712
73	024506.40	−020727.7	G	B	12.01	11.83	7.93	4.86	0.29	158	3	2MFGC 2186
74	142837.03	−394844.1	IG	B	11.93	11.79	7.60	4.81	0.28	176	2	ESO 326-24
75	232610.59	−303106.1	EmG	B	11.45	11.14	6.91	4.29	0.28	189	2	6dFGSg J232610.6-303106
76	161833.99	+132425.9	G	B	13.08	12.63	9.02	5.85	0.28	152	3	2MASX J16183392+1324253
77	075803.01	−642929.2	G	B	11.89	11.54	7.11	4.93	0.28	238	0	6dFGSg J075803.0-642930
78	225451.03	+374220.8	IG	B	12.75	12.12	7.96	5.49	0.28	206	1	2MASX J22545096+3742205
79	064540.95	+433407.5	G	B	11.81	11.61	7.46	4.72	0.27	180	2	2MASX J06454097+4334069
80	064339.31	−271217.6	GiG	B	10.83	10.64	6.79	3.74	0.27	159	2	6dFGSg J064339.3-271218
81	202933.28	−441815.2	G	B	12.20	11.97	7.60	5.18	0.27	206	0	2MASX J20293311-4418140
82	174949.84	−484826.9	PN	B	12.20	12.19	9.16	5.40	0.27	124	0	PN-G343.2-10.8
83	030445.12	+074739.2	G	B	12.47	12.30	8.34	5.42	0.27	167	0	2MASX J03044511+0747394
84	055652.46	−052303.8	LSB	B	11.58	11.40	7.78	4.55	0.26	149	2	6dFGSg J055652.3-052309
85	051646.24	−122059.4	G	B	10.53	10.33	6.44	3.47	0.26	164	1	6dFGSg J051646.2-122100
86	165033.09	−672039.6	G	B	11.89	11.53	7.28	4.88	0.26	210	0	LEDA 96592
87	194605.40	+640850.1	G	B	10.96	10.55	6.97	3.84	0.26	154	2	2MASX J19460544+6408494
88	120913.87	+265237.4	G	B	12.00	11.82	8.05	5.00	0.26	159	2	LEDA 38612
89	235521.97	−563457.1	G	B	13.02	12.71	8.42	6.07	0.26	214	0	6dFGSg J235522.0-563456
90	054738.65	−103552.8	G	B	12.79	12.50	8.88	5.76	0.25	155	3	6dFGSg J054738.7-103552
91	140736.99	+160121.6	G	B	12.03	11.68	7.70	4.98	0.25	181	2	2MASX J14073693+1601212
92	093930.15	+062613.0	G	B	11.98	11.72	7.49	5.04	0.25	203	1	2MASX J09393017+0626133
93	095814.56	−381356.5		B	12.82	12.68	8.80	5.87	0.25	166	0

^a Grade refers to how well understood an object is (see Section 4.3.) Grade B is reserved for objects with very little presence in the refereed literature, such that we are not convinced the object's true nature has been carefully verified. Grade A is given to objects that are effectively new to science having, at most, been detected in surveys but never examined. ^b Photometry refers to the source of the photometry used in the table: 0: profile fit; 1: aperture-corrected; 2: calibrated, 3: elliptical aperture. See Section 5.1.

Download table as:  ASCIITypeset images: 1 2

The WISE filter system allows for the measurement of six different photometric colors, i.e., W1–W2, W2–W3, etc. In Table 6, we present a list of the 10 reddest (per color) objects contained within the Extended Platinum Sample. These objects have been ordered by decreasing "redness," so that the 10th object in the list represents the 10th reddest object within the color explored.

The most extreme MIR objects in the high latitude sky ($|b|\geqslant 10$) appear to be dominated by PNe. Other objects with extreme colors include a pair of comets, three YSOs, a handful of Type ii Seyfert galaxies, two uncataloged sources in the LMC that escaped our mask of that region, and some objects we discuss in Section 6.2.

Since this study is primarily interested in finding ETIs of extra-galactic origin, we present Table 7, which is a list of the 10 reddest (per color) high latitude extragalactic objects seen in the Extended Platinum Sample. We find that the colors of galaxies are in general not as extreme as Galactic sources. Most of the galaxies on this list are Grade C, meaning that their nature has been identified in the literature.

Table 7 is dominated by galaxies classified in SIMBAD as AGNs of various stripes, which is unsurprising since those galaxies are often characterized by their extreme MIR emission. The host galaxies of AGNs are thus, as expected, our primary confounders in our waste heat search for K3 civilizations.

In the AGENT formalism (Section 4.1, Wright et al. 2014a), the parameter γ represents the fraction of starlight reemitted in the MIR, at temperature ${{T}_{{\rm waste}}}$. We have measured maximal values for this parameter assuming that there is no dust in any of our sources, and that their underlying stellar population is that of an old elliptical galaxy (so, virtually dust-free). Since this is the parameter of interest in searches for alien waste heat, we have sorted the Extended Platinum Sample by this parameter. We present the top 50 such galaxies in Table 8.

The galaxy with the highest measured $\gamma (=0.85)$ is NGC 4355 (=NGC 4418), which also has the most extreme colors in four of the six WISE color combinations. It is a well-studied Type ii Seyfert galaxy in the Virgo cluster with a (W1–W4) color of 10.19, with the extreme MIR emission being due to the AGN.

The second source on our list is IRAS 04259–0440, a marginally resolved galaxy with a modest presence in the literature. It has been studied in the context of being a Seyfert galaxy or LINER (Wu et al. 2011), so we are convinced that the MIR emission is understood. Nonetheless, given the extreme nature of this galaxy's infrared emission, this galaxy would appear to warrant more attention than it has received to date.

The third galaxy in our list is the well-studied Arp 220, the quintessential local starburst galaxy and a known active galaxy. Fourth and fifth are UGCA 116 and IRAS F20550+1655-SE, both pairs of interacting galaxies. In all three cases, the extreme MIR colors are clearly due to star formation triggered by galaxy interactions.

One of the primary objectives of this investigation is to search for and identify the rarest and most extreme sources in the high latitude infrared sky. The majority of the sources in our tables have already been discovered and discussed in various articles, but there are still a small number of objects not previously discovered or discussed in the literature.

In Table 9, we present 3 objects ($\gamma \geqslant 0.25$) classified as As, meaning that they are essentially new to science. In this section, we also discuss an extreme and apparently anomalous object that we gave a B grade, IRAS 15553–1409, and a particularly interesting lower-γ grade A source.

6.3.1.  IRAS 04287+6444: an Unusual Cluster of MIR Sources with No Optical Counterparts

Our most unusual objects are a cluster of MIR point sources with no optical counterparts associated with IRAS 04287+6444. These appear consistent with being either a cluster of completely enshrouded stars, either by dust or, potentially, Dyson spheres, or else a cluster of extremely red cosmological sources. Closer examination leads us to conclude they are likely the first of these, a previously unstudied cluster of protostars or YSOs.

Our search was designed to identify extended objects somewhat redder than these objects, but two effects conspired to allow them to slip into our catalog. The first is that the brightest of these sources is slightly blended, which complicated our magnitude corrections, giving our source an erroneously high γ value before our quality checks corrected the error. Second, this blending gave the source a high W3RCHI2 value, which suggested this was an extended source. Rather than reject this source, as we did in similar cases where blends made our color and angular size estimates of point sources unreliable, we chose to retain it due to its extreme MIR-optical colors, unusual nature, and virtual absence in the literature.

There appear to be at least seven very red point sources clustered in this region in all. We identify and number four fainter sources in Figure 16 (bottom right); the other three or more all contribute to the brighter blend in the NE.

Zoom In Zoom Out Reset image size

NED reports five entries within 2$^{\prime}$ of these sources' positions. The two nearest detections are from the 2MASS Extended Source Catalog (see below), and the third nearest entry is the IRAS counterpart to this source. The other two NED entries are a ROSAT detection (1WGA J0433.4+6451 Voges et al. 1999) centered $\sim 33^{\prime\prime}$ away (outside the $20^{\prime\prime}$ astrometric precision of ROSAT), and the 1.4 GHz radio source NVSS J043322+645120 (Condon et al. 1998) centered $\sim 0\buildrel{\,\prime}\over{.} 8$ away (consistent with being associated with our source).

Strauss et al. (1992) included the IRAS center in their optical spectroscopic survey, and identified it as a "cirrus or dark cloud," although they note that a significant number of such entries may in fact be galaxies. This non-detection is not surprising given the lack of optical counterparts to these objects.

We have found a serendipitous archival Spitzer MIPS image of these sources, taken because they are within 15$^{\prime}$ of HD 28495, a target observed with IRAC as part of the FEPS program (Meyer et al. 2006). Figure 17 shows how the superior resolution of this 22 μ imagery reveals substructure to the SE component of the bright blend and many sources undetected by WISE (without color information, it is unclear whether these sources are associated with IRAS 04287+6444).

Zoom In Zoom Out Reset image size

Objects #1 and #3 are barely detected in the WISE images (not appearing in the W4 band at all) and thus provide little clues as to their true nature, but appear to be fainter versions of the other red objects, with similar colors. Objects #2 and #4 appear cleanly detected in all four bands.

The angular proximity of these four objects to HD 28495 is intriguing, but these are likely unassociated since a common distance at 25 pc (van Leeuwen 2009) would imply a projected separation of $\sim 2\times {{10}^{4}}\;{\rm AU}$. Nonetheless, the lack of an optical counterpart complicates efforts to rule out this scenario from proper motion.

One possibility is that this is a previously uncataloged moderate-latitude ($b=11\buildrel{\circ}\over{.} 5$) dark cloud, and that these are an embedded cluster of YSOs or protostars. Figure 18 shows the 2MASS imagery for this region, which has significantly better angular resolution. This NIR imagery reveals that the brightest source in the WISE imagery comprises at least three sources, only one of which is evident in the J band. Supporting this interpretation, Yang et al. (2002) detected CO (J = 1 $\to$ 0) emission in the direction of the IRAS source⁹ with an LSR radial velocity of −13.27 km s⁻¹ and an FWHM of 3.4 km s⁻¹, implying a molecular cloud exists in this direction at a kinematic distance of 840 pc (and thus a height of ∼170 pc above the Sun's position in the plane).

Zoom In Zoom Out Reset image size

If the sources are extragalactic, the most natural explanation is that they are members of a galaxy cluster. Edelson & Malkan (2012) identified the IRAS source as having a modest chance ($\sim 50\%$) of being an AGN of some flavor based on the WISE, 2MASS, and X-ray fluxes, however, the X-ray detection may be unassociated, and it appears this probability does not incorporate the fact that the source has no optical counterpart or that it is not isolated.

The lack of an optical counterpart could be due to redshift and internal extinction. A significant population of high redshift ($z\ \gt 2$) and more luminous (${{L}_{{\rm IR}}}\gt {{10}^{13}}$ ${{L}_{\odot }}$) Dust Obscured galaxies (called Hot Dust Obscured Galaxies, or "Hot DOGs") have recently been identified by the WISE survey, (see Eisenhardt et al. 2012; Wu et al. 2012; Bridge et al. 2013; Stern et al. 2014, for a detailed discussion of these types of objects). Indeed, the strong 22 μm emission for these objects is reminiscent of Dust Obscured Galaxies (DOGs), either local ($z\sim 0$; Hwang & Geller 2013) or at high redshift ($z\geqslant 2$; Dey et al. 2008).

However, our objects are inconsistent with Hot DOGs since Hot DOGs tend to have very little or no emission in the shorter WISE bands. Such extreme examples of dusty galaxies are not typically as highly clustered as our sources are.

We tentatively favor the interpretation that this is a cluster of YSOs embedded in and heavily extinguished by their parent molecular cloud. We are intrigued by these objects, and we hope that spectroscopic observations can and will reveal their true nature in the future.

6.3.2. WISE J224436.12+372533.6: a New MIR-bright Galaxy

We gave the object WISE J224436.12+372533.6 (shown in Figure 19) an A grade because it has no presence in the astronomical literature beyond having been noted in the 2MASS Extended Source Catalog. It is MIR-bright and red, and DSS and 2MASS imaging shows what appears to be a galaxy. It is just barely detected by IRAS (it appears in only two bands in the IRAS Faint Source catalog; Moshir et al. 1992), and so may have evaded prior notice for that reason. It also appears as a 1.4 GHz source in the NRAO VLA radio survey (Condon et al. 1998). This source deserves further study to understand its superlative nature.

Zoom In Zoom Out Reset image size

6.3.3.  IRAS 16329+8252: an MIR-bright Galaxy at z = 0.04?

We give this source an A grade because it has virtually no presence in the literature beyond a single low resolution spectrum by Chen et al. (2011). If their identification of the emission lines in this spectrum is correct, then it is a galaxy at z = 0.039. Its strong MIR emission suggests large amounts of star formation, possibly triggered by the disturbance of a nearby neighbor (see Figure 20).

Zoom In Zoom Out Reset image size

6.3.4. WISE J073504.83–594612.4: a Pair of Merging Galaxies?

We graded the source WISE J073504.83–594612.4 (Entry 2 in Table 9) A because it has virtually no presence in the literature, appearing only in a handful of photometric catalogs, including the 2MASS extended source catalog. It has no IRAS counterpart. It appears to be extragalactic.

The WISE color image (Figure 21) shows little structure; it is barely extended in W3 and W4. The H-band and W2 imagery, which presumably trace the stellar populations of the bulges of the galaxies in this imagery, reveals two sources, one at the position of the WISE source and one $\sim 20^{\prime\prime}$ to the north. The DSS B-band image shows significant structure between these positions, suggesting that this is a disturbed or merging galaxy pair. The bright W4 emission suggests that this disruption is generating significant star formation in the southern galaxy.

Zoom In Zoom Out Reset image size

6.3.5.  IRAS 15553–1409: a Large, MIR Nebula around a Classical Be Shell Star

The infrared source IRAS 15553–1409 is the reddest or second reddest source in four of the six WISE colors. We give this source a B grade because it has virtually no presence in the literature beyond Carballo et al. (1992), who identify it as a potentially "evolved Galactic object." WISE imagery reveals it to be an MIR-bright nebula associated with the classical Be star 48 Librae, and so not a good K3 candidate, but interesting nonetheless. Figure 22 shows that the MIR morphology and colors are similar to those of reflection nebulae, but in this case there is no apparent emission in the optical.

Zoom In Zoom Out Reset image size

48 Librae is known to be a classical Be shell star (e.g., Jaschek & Jaschek 1992) and a member of the Sco–Cen Association. It has a distance of ∼140 pc (van Leeuwen 2009), expansion velocities of 25 km s⁻¹ (Hoffleit & Warren 1995), and an age $\lt 20\;{\rm Myr}$ (E. Mamajek 2015, private communication¹⁰ and references therein). It is sometimes listed in the literature as a giant star, but this is likely because its rapid rotation gives it an anomalously low surface gravity, complicating its spectrally derived luminosity classification.

This source is superlative because of its extreme MIR colors and extent, although this is a byproduct of its proximity; similar Be shell stars at greater distance would not be resolved.

Adams et al. (2013) note that the infrared excesses of main-sequence stars typically have one of four origins: circumstellar debris disks, protoplanetary and protostellar disks around very young stars, "cirrus hot spots" caused by the illumination of ambient interstellar dust, and the excretion disks of classical Be stars.

Cirrus hot spots (van Buren & McCray 1988; Adams et al. 2013) can be simple reflection nebulae (the "Pleiades phenomenon," Kalas et al. 2002) or the result of bow-shocking by the star's wind or radiation as it moves through the ISM (e.g., van Buren & McCray 1988; Povich et al. 2008; Everett & Churchwell 2010). Inspection of the morphology of typical reflection nebulae (including that of the Pleiades themselves and those discovered by Kalas et al. 2002) shows that 48 Librae does not appear to be a typical example. The 48 Librae nebula has no optical counterpart, suggesting that the 22 micron emission is thermal. It also appears to have symmetries about the position of the star, suggesting that it has some connection to the star beyond being illuminated by it (see Figure 23).

Zoom In Zoom Out Reset image size

The most likely explanation of this nebula is that it is therefore a ring or bow shock. The bow shock and ring interpretations are complicated by the fact that despite being a Sco–Cen Association star, 48 Librae does not appear to be embedded in a region of high density gas and dust, and its space motion is not large.

The proper motion of 48 Librae is SW ($({{\mu }_{\alpha }},{{\mu }_{\delta }})=(-12.44,-16.73)$ mas yr⁻¹ van Leeuwen 2009), which is not in the direction of the brightest sources of emission, but it appears that most of this apparent motion is actually due to motion of the Sun with respect to the local standard of rest. Correcting for the solar motion (using radial velocity $7.5\;\pm$ 1.8 km s⁻¹ (Gontcharov 2006) and the LSR of Mihalas & Binney 1981), the proper motion is still roughly SW ($({{\mu }_{\alpha }},{{\mu }_{\delta }})=(-2.4,-0.9)$ mas yr⁻¹), corresponding to an LSR space velocity of 4 km s⁻¹, with an error of a few km s⁻¹ from our LSR correction.

Further complicating the bow shock interpretation is that both sides of the 48 Librae nebula have similar arcs consistent with being portions of rings inclined $\sim 20{}^\circ$ from edge-on and centered on the star.

Given these difficulties with the typical bow shock and ring models, we suggest that the nebula originated with the star itself. Classical and shell Be stars are not typically considered significant sites of dust formation or sources of MIR excess (Rivinius et al. 2013), although NIR excesses, presumably due to circumstellar dust, are common, and many unclassified Be stars are known to have MIR excesses (e.g., Miroshnichenko & Bjorkman 2000; Miroshnichenko et al. 2001; Lee et al. 2011). 48 Librae has one of the larger and stronger disks known among classical Be stars (T. Rivinius 2015, private communication).

Since 48 Librae is known to have shells, a time variable disk (Štefl et al. 2012), and significant mass loss, it is reasonable that some of this shell material would condense into dust near the star before being lost (Lee et al. 2011). The nebula is not too large for this: the nebula has an angular size of the order of a few arcminutes, which at the distance of 48 Librae (∼150 pc) corresponds to ∼15,000 AU. The expansion velocity of the shells is 25 km s⁻¹, yielding a characteristic timescale of ∼30,000 yr, significantly shorter than the Be phase of a star.

So we may be seeing shocked dust where the shells (really rings) of excreted material collide, either with the ambient ISM or with previously ejected material. If so, then we expect the rings we observe in the nebula to have a rough correspondence to the geometry of the excretion disk. The rings (which we fit by eye) have an inclination of 70° (i.e., 20° from edge-on), and a position angle of 72${}^\circ$. The actual inclination of the excretion disk must be $\gtrsim 65{}^\circ$ (because it shows absorption lines from the disk; T. Rivinius 2015, private communication), and the actual position angle is known from interferometry to be $50{}^\circ \pm 9{}^\circ$ (Štefl et al. 2012), consistent with our rings at 1–2σ, within the rough precision with which we can define them.

Alternatively, we may be seeing excreted dust being illuminated by UV radiation from 48 Librae itself, although this interpretation is complicated by the lack of apparent optical scattered light and by the patchiness of the emission.

It is unclear if the nebula contains more dust than could be plausibly explained by the currently observed mass loss rate, but of course 48 Librae could have had episodes of higher mass loss rates in the past.

This object may be revealing to us that Be shell stars are, in fact, common sites of dust generation. This phenomenon warrants further study.

We can use those sources in our study with the largest amounts of thermal emission to set an upper limit for waste heat emission among the galaxies we have surveyed. We have found no sources with $\gamma \gt 0.85$, and the 50 galaxies we have found with $\gamma \gt 0.5$ appear to have natural origins for most of their MIR emission, although we have not rigorously verified this.

These values for γ were calculated under the assumption that our target galaxies are composed of only two components: an old stellar population and alien waste heat originating entirely from reprocessed starlight. Since real galaxies have other sources of MIR emission, these numbers are upper limits on the integrated alien waste heat emitted by these galaxies.

Our assumption that the origin of the alien waste heat is intercepted starlight (i.e., $\gamma =\alpha$, where α is the fraction of starlight intercepted; see Section 4.1) means that we have assumed that the flux in the W1 and W2 bands, which in our model constrain the total stellar luminosity, does not include the starlight lost to alien factories. If, instead, we assume that all alien waste heat is generated by other means and that only a negligible fraction of starlight is occulted ($\alpha \sim 0$), then we would infer a higher value for the fraction of starlight emitted as waste heat. Specifically, our limiting values of $\gamma =0.85$ and 0.5 correspond to ${{\gamma }_{\alpha =0}}=5.7$ and 1 (see Section 3.2; Wright et al. 2014a).

In other words, we shave shown that there are no galaxies resolved by WISE with MIR luminosities consistent with alien energy supplies in excess of 5.7 times the starlight in their galaxies (i.e., we have ruled out $\gamma =\epsilon \gt 5.7$). If all 50 galaxies in our list turn out to have purely natural origins to their emission, then this upper limit drops to $\gamma =\epsilon \lt 1$.

If we assume that any large alien energy supply will be based on starlight (that is, $\gamma \sim \alpha$ and $\epsilon \sim 0$), then our upper limit is much tighter: no resolved galaxies exist in our search area with more than 85% of their starlight reprocessed by alien factories, a limit which will drop to 50% when our 50 high-γ galaxies are more carefully vetted.

Translating these numbers into physical units (erg s⁻¹) will require a more detailed modeling of the stellar and nonstellar components of the galaxies we have surveyed, a project which is beyond the scope of this paper. We hope to pursue this in a future paper.

Translating our upper limits into an upper limit on the frequency of K3s requires knowledge of the number of galaxies we have effectively surveyed. We cannot use our Gold or Platinum Samples to estimate this number because they included color cuts to remove stars that also removed elliptical and other dust-free galaxies. Even if we had imposed no such cuts, there are many galaxies that would be resolved in the W3 band if they were MIR bright, but are unresolved—or in some cases undetected—in WISE because their MIR surface brightness is below the WISE detection limits.

To estimate the number of galaxies that would have been included in our sample if they had $\gamma \gt 0.5$, we can use the number of sources in the W1 and W2 bands. The W1 band, in particular, has better angular resolution than W3, and is primarily sensitive to stellar photospheres, so is in many ways a clean band for estimating the angular extent of galaxies around which alien factories might reside.

One concern with using WISE data for this purpose is that, as we have seen, the source counts include many non-galaxies (including artifacts), and many point sources have erroneously large values of WXRCHI2. To mitigate this, we have used the 2MASS Extended Source Catalog (XSC), which is relatively clean of point sources and is composed almost entirely of galaxies.

We first cross-matched WISE to the 2MASS XSC, and selected only those sources with $|b|\geqslant 10$. We then examined the NIR properties of MIR-red galaxies in the 2MASS XSC by examining the relationship between the WXRCHI2 values for the W1 and W2 bands and the W3RCHI2 parameter, which we used for the Platinum Sample. We restricted our analysis to 9589 matched sources with (W1–W3 ≥3.8) for the W1RCHI2 analysis, and 14,927 sources with (W1–W3 ≥3.5) for the W2RCHI2 analysis.

The left hand panels of Figure 24, show the relationship between the W1RCHI2 and W2RCHI2 parameters (which describe the degree to which galaxies are resolved in those bands, see Section 2.1) and the W3RCHI2 parameter we used to define a source as "extended" in our survey.

Zoom In Zoom Out Reset image size

These panels show that for the red sources we used to construct this figure, we can use W1RCHI2 $\gt \;12$ and W2RCHI2 $\gt \;3$ as proxies for our actual criterion W3RCHI2 $\gt \;3$. Using the full WISE-2MASS XSC cross-matching (that is, not imposing any color cuts), we find 1,589,099 sources common to both catalogs. Of these, 1,463,781 have $|b|\geqslant 10$. Of these, 111,617 have W1RCHI2 $\geqslant \;12$ and 104,039 have W2RCHI2 $\geqslant \;3$. These figures are consistent, suggesting that we have surveyed 1 × 10⁵ galaxies. The only previous search for K3s in the refereed literature, that of Annis (1999), surveyed 163 galaxies.

As a check, we also used the R_EXT parameter in the XSC, which corresponds to a measure of the NIR angular size of these galaxies. The right-hand panels in Figure 24 show that R_EXT corresponds to the angular sizes we are interested in with good sensitivity (i.e., the test R_EXT $\gt \;18^{\prime\prime}$ has a low false negative rate), although it is not very specific (i.e., it has roughly a 50% false positive rate) for these red sources.

A query of the entire XSC with $|b|\geqslant 10$ and R_EXT $\geqslant \;18$ yields 229,813 sources. A random sampling of 100 of these sources reveals that all are present in the WISE All-sky catalog, 45 have W1RCHI2 $\gt \;12$, and 43 have W2RCHI2 $\gt \;3$. These numbers are consistent with the specificity we estimated among the MIR-red sources in Figure 24. We also tested 100 random WISE sources satisfying our extended source criteria in W1 and W2, and find that 86 and 87 of them, respectively, have R_EXT $\gt \;\;18^{\prime\prime}$, also consistent with the sensitivity suggested in Figure 24.

We conclude that there are $\sim 1\times {{10}^{5}}$ galaxies with sufficient angular size that they would have been included in our Platinum Sample if they had had significant W3 emission. In our survey of these $\sim 1\times {{10}^{5}}$ galaxies, we have found that there are no alien, non-stellar energy supplies in excess of 5.7 times the stellar luminosity of their host galaxy, and no alien supercivilizations reprocessing as much as 85% of their starlight into the MIR. We have found 50 galaxies consistent with 50% reprocessing, all of which are presumably extraordinary, but entirely natural, star-forming galaxies. Verification of the natural origin of the MIR flux in these 50 galaxies will thus lower our upper limit to 50%.