THE COLORS OF EXTREME OUTER SOLAR SYSTEM OBJECTS

The dynamical and physical properties of small bodies in our solar system offer one of the few constraints on the formation, evolution, and migration of the planets. The Kuiper Belt has been found to be dynamically structured with several observed dynamical classes (Trujillo et al. 2001; Kavelaars et al. 2008, 2009); see Figure 1. Classical Kuiper Belt objects (KBOs) have semimajor axes 42 ≲ a ≲ 48 AU with moderate eccentricities (e ∼ 0.1) and inclinations. These objects may be regarded as the population originally predicted for the Kuiper Belt, but their relatively large eccentricities and inclinations were unexpected. The dynamics of the classical KBOs have shown that the outer solar system has been highly modified through the evolution of the solar system (Duncan & Levison 1997; Petit et al. 1999; Ida et al. 2000; Morbidelli & Levison 2004; Gomes et al. 2005). Resonant KBOs are in mean motion resonances with Neptune and generally have higher eccentricities and inclinations than classical KBOs. These objects, which include Pluto and the Plutinos in the 3:2 resonance, were likely captured into these resonances from the outward migration of Neptune (Malhotra 1995; Hahn & Malhotra 2005; Levison et al. 2008). Scattered disk objects have large eccentricities with perihelia near the orbit of Neptune (q ∼ 25–35 AU). The scattered disk objects are likely to have been scattered into their current orbits through interactions with Neptune (Duncan & Levison 1997; Duncan 2008a; Gomes et al. 2008).

Zoom In Zoom Out Reset image size

A new class of outer solar system object, called the extended or detached disk (Figure 1), has only recently been recognized (Gladman et al. 2002; Emel'yanenko et al. 2003; Morbidelli & Levison 2004; Allen et al. 2006). To date only a few detached disk objects are known. The detached disk objects have large eccentricities, but unlike the scattered disk objects the detached disk objects have perihelia q ≳ 38 AU, which do not appear to be directly caused by Neptune interactions alone (Gladman & Collin 2006; Levison et al. 2008). Though unexpected, the discovery of these detached disk objects have given us a new understanding of our solar system's chaotic history.

A few objects have been found that have very large semimajor axes and eccentricities (Sedna, 2006 SQ₃₇₂, and 2000 OO₆₇). Through dynamical simulations these objects are best described as coming from the inner Oort Cloud (Brown et al. 2004; Kenyon & Bromley 2004; Morbidelli & Levison 2004; Kaib et al. 2009). Two objects have also been found to have retrograde orbits in the outer solar system (2008 KV₄₂ and 2008 YB₃). These two retrograde objects along with the very high inclination object 2002 XU₉₃ (i ∼ 78°) could be from the outer Oort Cloud or a possibly yet to be discovered high inclination source region (Gladman et al. 2009).

Some trans-Neptunian objects (TNOs) have likely not experienced significant thermal evolution since their formation. The amount of thermal evolution depends on how close the object formed to the Sun and how close it has approached the Sun during its lifetime (Meech et al. 2009). The objects in the Kuiper Belt dynamical classes had varied histories with some experiencing very little thermal evolution, making them some of the most primitive bodies in the solar system. Optical observations of TNOs and Centaurs have shown some of these objects have the reddest material known in the solar system (Figure 2; Jewitt & Luu 2001; Peixinho et al. 2004; Doressoundiram et al. 2008; Tegler et al. 2008). This ultra-red material is currently thought to be rich in organic material (Gradie & Veverka 1980; Vilas & Smith 1985; Cruikshank et al. 2005; de Bergh et al. 2008). The ultra-red color may be from Triton tholins and ice tholins, which can be produced by bombarding simple organic ice mixtures with ionizing radiation (Doressoundiram et al. 2003; Barucci et al. 2005a; Emery et al. 2007; Barucci et al. 2008).

Zoom In Zoom Out Reset image size

Interestingly, short-period comets that are believed to have originated from the Kuiper Belt do not show this ultra-red material (Figure 2; Jewitt 2002). The reason is because comet surfaces have been highly processed from their relatively close passages to the Sun (Jewitt 2002; Grundy 2009). This demonstrates that the surfaces of comets are not reliable for understanding the original compositions of the comets. Some Centaurs, which are the precursors to the short-period comets, do show these ultra-red colors probably because they have not yet been near the Sun for a long enough time to have their surfaces highly modified from thermal, sublimation, or evaporation processes. No long-period comets from the Oort Cloud have been sufficiently observed before any significant heating would have taken place on their surfaces. Thus we do not have a good knowledge of what color an Oort Cloud comet may have been before it started to thermally evolve (Meech et al. 2009).

There have been one or possibly two subsets of TNOs that appear to be dominated by the ultra-red material (Figure 2). First are the low inclination "cold" classical KBOs that also have large perihelions (Tegler & Romanishin 2000; Trujillo & Brown 2002; Doressoundiram et al. 2005; Gulbis et al. 2006; Fulchignoni et al. 2008; Peixinho et al. 2008). These objects likely formed in the more distant solar system unlike the higher inclination KBOs, which may have formed closer to the Sun and were transported to and captured in the Kuiper Belt during the planet migration process (Levison & Morbidelli 2003; Gomes 2003; Levison et al. 2008). Sedna, an object well beyond the Kuiper Belt edge at 50 AU (Jewitt et al. 1998; Allen et al. 2001), also has an ultra-red color and could be a new class of object, possibly from the inner Oort Cloud (Brown et al. 2004; Morbidelli & Levison 2004; Kenyon & Bromley 2004; Brasser et al. 2006; Barucci et al. 2005b). Some previous works (Tegler & Romanishin 2000; Trujillo & Brown 2002; Doressoundiram et al. 2005) have noted that objects with larger perihelion distances tend to have redder surfaces, but most of the ultra-red objects observed were in the main classical Kuiper Belt. No systematic survey of the colors of the large perihelion detached disk population has been performed to date.

In this work, the optical colors were observed for most of the known detached disk objects, possible inner Oort Cloud objects, and other outer solar system objects that exhibit extreme orbits in terms of their inclination, semimajor axis, or perihelion distance. Understanding any color trends or correlations, in particular the ultra-red material, will constrain where these extreme objects may have formed in the solar system and thus how they may have ended up on their current orbits. This in turn will allow us to determine how the planets may have migrated and what amount of this ultra-red organic-rich material may have been incorporated into the planets.

Observations of the outer solar system objects presented in this work were obtained with the twin Magellan Baade and Clay 6.5 m telescopes at Las Campanas, Chile and the 8.2 m Subaru telescope atop Mauna Kea in Hawaii. Table 1 shows the various observational circumstances for the 33 objects observed. The LDSS3 camera on the Clay telescope was used on the nights of 2005 November 2 and 3, 2008 May 7 and 8, 2009 January 28, 2009 May 23 and 24, and 2009 August 25 and 26. LDSS3 is a CCD imager with one STA0500A 4064 × 4064 CCD and 15 μm pixels. The field of view is about 8.3 arcmin in diameter with a scale of 0.189 arcsec per pixel. The IMACS camera on the Baade telescope was used on the nights of 2008 October 19 and 2008 December 3. IMACS is a wide-field CCD imager that has eight 2048 × 4096 pixel CCDs with a pixel scale of 0.20 arcsec per pixel. The eight CCDs are arranged in a box pattern with four above and four below and about 12 arcsec gaps between chips. Only chip 2, which is just north and west of the camera center, was used in the IMACS color analysis. The Suprime-Cam imager on the Subaru telescope was used on the night of 2009 October 15. Suprime-Cam is a wide-field CCD imager that has ten 2048 × 4096 pixel CCDs with a pixel scale of 0.20 arcsec per pixel (Miyazaki et al. 2002). The 10 CCDs are arranged in a 5 × 2 box pattern similar to the IMACS imager. Only chip 5, which is just west of the camera center, was used in the Suprime-Cam color analysis.

Dithered twilight flat fields and biases were used to reduce each image. Images were acquired through either the Sloan g', r', or i' filter while the telescope was auto-guiding at sidereal rates using nearby bright stars. Exposure times were between 300 s and 450 s. Southern Sloan standard stars were used to photometrically calibrate the data (Smith et al. 2005). In order to more directly compare our results with previous works, the Sloan colors were converted to the Johnson–Morgan–Cousins BVRI color system using transfer equations from Smith et al. (2002). To verify the color transformation, the known ultra-red (44594) 1999 OX₃ and gray (19308) 1996 TO₆₆ TNOs were observed (Tegler & Romanishin 1998, 2000; Jewitt & Luu 2001; Barucci et al. 2005a). The BVRI photometric results are shown in Table 2 (Figures 2 and 3) and the Sloan results in Table 3 (Figure 4).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Table 3. Sloan g', r', i' Optical Photometry

Name	$m_{r^{\prime }}$	Sa	g' − r'	r' − i'	g' − i'
	(mag)		(mag)	(mag)	(mag)
(90377) Sedna	20.75 ± 0.03	26.3 ± 3	0.85 ± 0.03	0.45 ± 0.03	1.31 ± 0.04
(48639) 1995 TL8	21.31 ± 0.01b	29.4 ± 2	0.96 ± 0.02	0.43 ± 0.02	1.39 ± 0.02
(19308) 1996 TO66	21.31 ± 0.01	-0.15 ± 1	0.46 ± 0.02	0.12 ± 0.02	0.58 ± 0.02
(181874) 1999 HW11	23.15 ± 0.03	12.2 ± 3	0.63 ± 0.03	0.31 ± 0.03	0.94 ± 0.04
(44594) 1999 OX3	21.25 ± 0.01	31.5 ± 2	0.94 ± 0.02	0.48 ± 0.02	1.42 ± 0.02
(148209) 2000 CR105	24.06 ± 0.04	13.7 ± 6	0.59 ± 0.09	0.38 ± 0.09	0.94 ± 0.11
(118702) 2000 OM67	23.52 ± 0.03	14.4 ± 3	0.61 ± 0.04	0.38 ± 0.04	0.97 ± 0.04
(87269) 2000 OO67	22.37 ± 0.02	24.1 ± 3	0.85 ± 0.03	0.40 ± 0.03	1.26 ± 0.03
2000 PE30	21.83 ± 0.02	4.4 ± 3	0.55 ± 0.04	0.17 ± 0.03	0.72 ± 0.04
(82075) 2000 YW134	21.10 ± 0.03	17.1 ± 3	0.74 ± 0.04	0.34 ± 0.03	1.05 ± 0.04
(182397) 2001 QW297	23.71 ± 0.05	25.0 ± 5	0.80 ± 0.07	0.46 ± 0.06	1.26 ± 0.07
2002 GB32	23.35 ± 0.01	17.4 ± 3	0.67 ± 0.02	0.40 ± 0.02	1.07 ± 0.03
(84522) 2002 TC302	20.65 ± 0.02	28.2 ± 3	0.90 ± 0.02	0.45 ± 0.02	1.34 ± 0.03
(127546) 2002 XU93	21.37 ± 0.01	4.4 ± 2	0.55 ± 0.02	0.17 ± 0.02	0.72 ± 0.02
(120132) 2003 FY128	20.54 ± 0.01	20.4 ± 3	0.83 ± 0.02	0.34 ± 0.02	1.17 ± 0.02
2003 FZ129	22.98 ± 0.02	9.9 ± 3	0.62 ± 0.04	0.25 ± 0.03	0.88 ± 0.04
2003 HB57	23.38 ± 0.02	12.5 ± 3	0.62 ± 0.03	0.33 ± 0.03	0.95 ± 0.04
2003 QK91	23.18 ± 0.03	11.2 ± 3	0.66 ± 0.04	0.26 ± 0.03	0.92 ± 0.04
2003 UY291	23.59 ± 0.05b	19.9 ± 7	0.67 ± 0.07	0.46 ± 0.07	1.14 ± 0.08
2004 OJ14	23.76 ± 0.02	15.1 ± 3	0.69 ± 0.03	0.33 ± 0.03	1.02 ± 0.04
2004 VN112	23.06 ± 0.04	11.3 ± 4	0.69 ± 0.06	0.24 ± 0.06	1.01 ± 0.06
2004 XR190	21.76 ± 0.03	10.3 ± 3	0.58 ± 0.04	0.31 ± 0.04	0.89 ± 0.04
2005 EO297	23.64 ± 0.03b	14.0 ± 6	0.63 ± 0.05	0.36 ± 0.08	0.98 ± 0.08
2005 PU21	22.63 ± 0.02	30.0 ± 3	0.92 ± 0.02	0.47 ± 0.02	1.38 ± 0.02
2005 SD278	22.36 ± 0.02	17.1 ± 3	0.76 ± 0.02	0.32 ± 0.02	1.09 ± 0.03
(145480) 2005 TB190	21.11 ± 0.02	18.1 ± 3	0.76 ± 0.03	0.34 ± 0.03	1.10 ± 0.04
2006 SQ372	21.74 ± 0.02	24.6 ± 3	0.82 ± 0.03	0.44 ± 0.04	1.26 ± 0.04
2007 JJ43	20.46 ± 0.01	17.7 ± 3	0.81 ± 0.02	0.29 ± 0.02	1.11 ± 0.02
2007 TG422	21.89 ± 0.01	13.3 ± 1	0.67 ± 0.04	0.30 ± 0.04	0.98 ± 0.02
2007 VJ305	22.39 ± 0.02	14.6 ± 3	0.70 ± 0.03	0.31 ± 0.03	1.01 ± 0.03
2008 KV42	23.70 ± 0.04b	7.7 ± 4	0.61 ± 0.06	0.21 ± 0.06	0.88 ± 0.06
2008 OG19	20.68 ± 0.01b	18.3 ± 2	0.72 ± 0.01	0.38 ± 0.01	1.10 ± 0.01
2008 YB3	18.46 ± 0.01	9.6 ± 0.5	0.59 ± 0.01	0.28 ± 0.01	0.87 ± 0.01

Notes. ^aThe normalized spectral gradient for the optical colors of the observed objects (see the text for details). ^bThese few objects showed large light variations during the observations indicating possible significant rotational light curves (> 0.1 mag). Their colors were consistent throughout the observations since the variations caused by possible light curves were similar in all filters. Filters were rotated during the observations in order to prevent any rotational light curve from influencing the color results. The apparent magnitude is based on the average of the photometry.

Download table as:  ASCIITypeset image

Photometry was performed by optimizing the signal-to-noise ratio of the faint small outer solar system objects. Aperture correction photometry was done by using a small aperture on the TNOs (057–095 in radius) and both the same small aperture and a large aperture (246–340 in radius) on several nearby unsaturated bright field stars with similar point-spread functions (PSFs). The magnitude within the small aperture used for the TNOs was corrected by determining the correction from the small to the large aperture using the PSF of the field stars (cf. Tegler & Romanishin 2000; Jewitt & Luu 2001). For a few of the brighter objects (Sedna, 2003 FY₁₂₈, 2007 JJ₄₃, 2008 YB₃), both small apertures and the full large apertures were used on the TNOs to confirm both techniques obtained similar results.

The orbital parameters of the 33 outer solar system objects observed in this work are shown in Table 4. There were three main classes of objects in the observation sample: (1) objects dynamically linked to the inner Oort Cloud, (2) outer solar system retrograde and high inclination objects, and (3) extended or detached disk and outer classical belt objects. Each class is discussed in the subsections below. In addition, the well-measured gray object (19308) 1996 TO₆₆ that is part of the Haumea KBO collisional family and ultra-red object (44594) 1999 OX₃ were observed to confirm the photometry is consistent with previous works.

As seen in Figure 2, all objects observed appear to have correlated broadband optical colors. In other words, the objects appear to follow a nearly linear red slope in the optical. This has also been confirmed through spectroscopy and correlation analysis on other TNOs (Doressoundiram et al. 2008). Because of the near linearity in the optical colors, we can obtain the spectral gradient, S, of the objects using two unique optical broadband filters. The spectral gradient is basically a very low resolution spectrum of an object and is usually expressed in percent of reddening per 100 nm in wavelength. We follow Doressoundiram et al. (2008) and express the spectral gradient as S(λ₂ > λ₁) = (F_2,V − F_1,V)/(λ₂ − λ₁), where λ₁ and λ₂ are the central wavelengths of the two filters used for the calculation and F_1,V and F_2,V are the fluxes of the object in the two filters normalized to the V-band filter. S is the measure of the reddening of an object's surface determined between two wavelength measurements (two different filters). We determined the spectral gradient of the observed objects using the g' and i' filters, which have well-separated central wavelengths of 481.3 and 773.2 nm, respectively. The spectral gradient results for the observed objects are shown in Table 3, and the spectral gradient determined for known small body populations in the solar system are shown in Table 5. Ultra-red color is here defined as including the reddest 90% of the measured low inclination classical KBOs (ultra-red: S ≳ 25, B − R ≳1.6, V − I ≳ 1.2, B − I ≳ 2.2, V − R ≳ 0.6, R − I ≳ 0.6, and using Sloan colors g' − i' ≳ 1.2, g' − r' ≳ 0.8, and r' − i' ≳ 0.4 mag).

3.1. Inner Oort Cloud Objects

3.2. Retrograde and High Inclination Objects

3.3. Extended/Detached Disk and Outer Classical Belt Objects

4.1. Detached and Scattered Disk

The scattered disk is probably made up of two main source populations. Some scattered disk objects are likely the surviving members of a relic population of objects that were scattered during Neptune's migration in the very early solar system (Gomes et al. 2008). A second source for the scattered disk is from recently dislodged objects from the Kuiper Belt through various slow dynamical processes (resonances) or collisions (Duncan et al. 1995; Levison & Duncan 1997; Duncan & Levison 1997; Nesvorny & Roig 2001; Gomes et al. 2008).

How the detached disk may have formed is still an open question (Gomes et al. 2008; Morbidelli et al. 2008; Kenyon et al. 2008; Duncan et al. 2008b; Gladman et al. 2008). For high inclination objects (i > 50°) the Kozai resonance can allow scattered objects to obtain large perihelion distances (Thomas & Morbidelli 1996; Gallardo 2006). For objects with moderate inclinations, the Kozai mechanism only works in increasing the perihelion distance of a scattered object if the object is also in a mean motion resonance with Neptune (Gomes 2003). Using Neptune mean motion resonances and the Kozai mechanism, Gomes et al. (2008) believe the high perihelia and relatively large semimajor axes of some moderate inclination detached objects can be explained through the above mechanism, specifically 2000 YW₁₃₄, 2005 EO₂₉₇, and 2005 TB₁₉₀ as well as the high inclination object 2004 XR₁₉₀, since they are all in or near Neptune mean motion resonances. These objects were likely at some point scattered disk objects that simply had their perihelia raised through Neptune mean motion resonances and the Kozai effect.

Based on the similar average spectral gradients of the two populations, the origin of the objects in the detached disk could be similar as the scattered disk (Table 5). The scattered disk spectral gradient (S = 10.1 ± 5) shown in Table 5 uses the strict definition similar to Gladman et al. (2008) which eliminates objects thought to be in any resonance with Neptune from being called a scattered disk object (called here the strict scattered disk: objects not in an obvious high-order resonance with Neptune, perihelia less than 35 AU, and semimajor axis between 30 and 100 AU). If objects in high-order resonances with Neptune are allowed in the definition used for what is a scattered disk object, the spectral gradient increases slightly and is almost the same as the detached disk average spectral gradient (14.5 ± 5). It is interesting to note that very red (S ≳ 20) objects are absent in the strict definition of the scattered disk but are not when including the higher order resonance objects. This may hint that many high-order resonance scattered disk objects are coming from the ultra-red low inclination classical belt or outer classical belt objects. It may be that the only efficient way to dislodge these fairly dynamically stable ultra-red objects is through some resonance interactions.

To further compare the scattered disk to the detached disk population, the Student's t-test and the Kolmogorov–Smirnov (K–S) test were performed on the spectral gradients of the two populations (Figure 5). The differences in the two population distributions were not statistically significant (< 3σ) in either test and thus are consistent with both populations coming from the same parent population (Table 6). This is true no matter if the high-order outer resonance objects are considered scattered disk objects or not (Figure 6). The similarity of spectral gradients may hint that Neptune mean motion and Kozai resonances allowed scattered disk objects to become detached overtime from significant Neptune influence and that the detached disk is a simple extension of the scattered disk (Gallardo 2006; Lykawka & Mukai 2007b; Emel'yanenko & Kiseleva 2008; Gladman et al. 2008; Gomes et al. 2008). Based on the spectral gradients and dynamics of the objects in the detached and scattered disk, it appears that they likely contain many objects from the same source region.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Table 6. t-test and Kolmogorov–Smirnov Test Results

Type 1a	Type 2a	Nb	t-statc	t-testd	D-state	K–Sd
Detached disk	Cold classical	41	8.88	99.99%	0.88	99.99%
Scattered strict	Cold classical	35	10.49	99.98%	0.86	99.98%
3σ		...	...	...	...	99.73%
High-order res.	Cold classical	38	2.48	83%	0.56	98.97%
Outer belt	Scattered strict	12	−3.76	99.88%	0.86	98.96%
Detached disk	Outer belt	18	3.87	97.7%	0.77	98.79%
Detached disk	Scattered strict	20	−1.68	98.1%	0.64	97.3%
2σ		...	...	...	...	95%
Outer belt	Cold classical	33	3.20	92%	0.56	92%
High-order res.	Scattered strict	17	−2.08	99.09%	0.56	90%
Detached disk	High-order res.	23	2.70	88%	0.42	80%
Outer belt	High-order res.	15	1.13	10%	0.50	75%

Notes. ^aThe dynamical groups being tested. ^bThe number of objects used in the test. ^cThe t-statistic from the t-test. ^dThe level of confidence that the two groups are not drawn from the same parent population using the t-test or the K–S test. ^eThe D-statistic from the K–S test.

Download table as:  ASCIITypeset image

4.2. Ultra-red Colors and the Outer Classical Belt

4.3. Spectral Gradients Versus Orbital Dynamics

To further explore the origins of the detached disk and outer classical belt objects, their eccentricities versus spectral gradient were plotted (Figure 7). There is an apparent trend that the lower the eccentricity the redder the object. The Pearson correlation coefficient is −0.49 using the 18 known spectral gradients of the detached disk and outer classical belt objects. The correlation with eccentricity is only significant at about the 97% level and additional low eccentricity outer classical belt objects need to be found to confirm or reject this possible trend (Table 7). Including the strict scattered disk objects increases the significance of the correlation with eccentricity to 99.1%. If the low inclination "cold" main classical KBOs are also included, the trend is even stronger with a Pearson correlation coefficient of −0.80 and a significance at the 99.99% confidence limit (Figure 8). There is no trend of spectral gradient with the inclination or perihelion distances of the detached disk and outer classical belt objects (Table 7). Including the strict scattered disk also finds no trend with spectral gradient and inclination or perihelion distance.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Table 7. Pearson Correlations for Spectral Gradients and Dynamics

Parameter 1a	Parameter 2a	Nb	rcorrc	Sigd
S(Det,Out,ScatStr,LowIncl)	e	53	−0.80	99.99%
S(Det,Out,ScatStr)	e	25	−0.53	99.1%
S(Det,Out,ScatStr,HighRes)	e	35	−0.41	98.5%
S(Det,Out)	e	18	−0.49	97%
S(Det,Out,ScatStr)	q	25	0.39	85%
S(Det,Out)	q	18	−0.18	60%
S(Det,Out,ScatStr)	i	25	0.06	20%
S(Det,Out)	i	18	−0.03	0%

Notes. ^aThe two parameters being compared through the Pearson correlation coefficient. S is the spectral gradient with the dynamical groups used in parentheses where Det is the detached disk, Out is the outer main belt, ScatStr is the scattered strict, LowIncl is the low inclination "cold" classical belt, and HighRes is the high-order Neptune resonance objects. ^bThe number of objects used in the correlation. ^cr_corr is the Pearson correlation coefficient. ^dSig is the confidence of significance of the correlation.

Download table as:  ASCIITypeset image

Thirty-three extreme outer solar system objects were observed to determine their optical colors.

• 1.  
  The three possible inner Oort Cloud objects (Sedna, 2006 SQ₃₇₂, and 2000 OO₆₇) all have ultra-red surfaces (spectral gradient S ∼ 25). These ultra-red surfaces are abundant in the low inclination "cold" classical KBO population and is believed to be associated with organic-rich material. Because the ultra-red material is only seen in the very outer parts of the observable solar system, it is likely that this material has not been significantly thermally altered. The red lobe of the Centaur distribution could thus either be from the low inclination classical KBO population or from the inner Oort Cloud population.
• 2.  
  For the first time, a systematic color determination of extended or detached disk objects was obtained. Most detached disk objects have only moderately red surfaces (10 ≲ S ≲ 18). Though slightly redder on average than the scattered disk, the detached disk colors are consistent with being from the same source region as the scattered disk objects. The only ultra-red objects observed with scattered disk-like orbits appear to be objects in high-order resonances with Neptune.
• 3.  
  The outer classical KBOs, which have semimajor axes beyond the 2:1 resonance with Neptune and low eccentricities, were found to be very red (S ≳ 20) and are on average redder than the detached disk objects. Unlike the scattered disk and detached disk, the outer classical belt objects have spectral gradients similar to the ultra-red low inclination "cold" classical KBOs though they appear to be less red on average.
• 4.  
  The two retrograde objects with perihelia in the outer solar system (2008 KV₄₂ and 2008 YB₃) and the extremely high inclination object (127546) 2002 XU₉₃ show only moderately red colors (S ∼ 9). These colors are very similar to the known comets, dead comets, damocloids, Jupiter Trojans, Neptune Trojans, irregular satellites, D-type main belt asteroids, scattered disk objects, and the neutral lobe of the Centaurs. 2008 YB₃ perihelion is near Jupiter; thus this object has had its surface thermally altered over the age of the solar system as is probably true for all the above moderately red populations. 2008 KV₄₂ has a rather large perihelion at 21 AU and it is unknown if it has ever approached closer to the Sun. The moderately red surface color suggests its surface has likely been thermally altered.
• 5.  
  The detached disk and outer classical KBOs show a trend that the lower the eccentricity the redder the object. This trend is currently not statistically significant since only a few of these objects are known. The trend is strengthened when adding the strict scattered disk and low inclination "cold" classical KBOs. The trend must be confirmed through discovering and measuring the colors of more outer classical belt objects.

I thank Chadwick A. Trujillo and Henry H. Hsieh for helpful comments that improved this manuscript. This paper includes data gathered with the 6.5 m Magellan telescopes located at Las Campanas Observatory, Chile. This work is based in part on data collected with the Subaru telescope, which is operated by the National Astronomical Observatory of Japan. This work was partially supported by the National Aeronautics and Space Administration through the NASA Astrobiology Institute (NAI) under Cooperative Agreement No. NNA04CC09A issued to the Carnegie Institution of Washington.