A Deep and Wide Twilight Survey for Asteroids Interior to Earth and Venus

The small body population in our solar system interior to Earth's and Venus' orbits has not been extensively explored to date because of the difficulty in observing near the glare of the Sun. Twilight observations near Venus' orbit from ground-based Earth telescopes occur at a very high airmass, with a bright sky background, poorer than average seeing, and with the objects of interest being at high phase angles during a short observational window. These factors create well over a magnitude of signal-to-noise loss when compared to observations overhead at night. This region of space is important for understanding the distribution of Near Earth Objects (NEOs) and objects that may have stable orbits in resonance with Venus.

One way to estimate the true number of small NEOs better is to include more objects found interior to Earth's orbit in population calculations, increasing the completion of orbital NEO types (Granvik et al. 2018; Harris & Chodas 2021). Currently the population models are biased toward NEOs found exterior to Earth's orbit as they are the easiest to find observationally. Only about 25 asteroids are known that have orbits completely interior to Earth's orbit and have well-determined orbits (called Atira or Apohele asteroids). This is compared to the thousands of known NEOs with orbits that cross Earth's orbit such as Aten and Apollo NEOs with semimajor axes interior and exterior to Earth, respectively (Mainzer et al. 2014; Schunova-Lilly et al. 2017; Morbidelli et al. 2020).

Understanding the relative populations of objects with orbits interior to Earth versus exterior to Earth will help us understand how objects are transported throughout the inner solar system (Strom et al. 2015). In addition, disruption processes of NEOs, such as fragmenting from thermal stresses, breaking up from rotational excitation, or tidal disruption from passing near the Sun or planets can be analyzed by comparing the predicted number of NEOs found interior to Earth versus exterior to Earth and correlating this with the composition or type of objects found (Granvik et al. 2016). Such correlations and analysis will further yield insights into the structure and strength of the different types of NEO classes. Various internal stresses on NEOs such as tidal deformation and solar heating will cause more fragile asteroid compositions and structures to erode or break-up as they approach the planets and Sun (Li & Jewitt 2013; Ye & Granvik 2019).

In addition to NEOs, which are strongly influenced by the Earth, some objects found interior to Earth's orbit may exhibit strong influences from Venus. There are a few known well observed asteroids that have orbital periods similar to Venus': (322756) 2001 CK32, (524522) 2002 VE68, 2012 XE133, 2013 ND15, and 2015 WZ12 as well as some more recently discovered NEOs with semimajor axes near Venus' between 0.722 and 0.725 au: 2020 BT2, 2020 CL1, 2020 QU5, 2021 XA1, 2021 XO3, 2022 BL5, and 2022 CD (see Figures 1 and 2). All of these near-Venus co-orbital asteroids are dynamically unstable on million year timescales since they have high-eccentricity orbits that cross the orbit of the Earth (Mikkola et al. 2004; de la Fuente Marcos & de la Fuente Marcos 2012, 2013, 2014, 2017). The same situation has been found for the known co-orbitals of Earth (Brasser et al. 2004; Connors et al. 2011; Hui et al. 2021; Santana-Ros et al. 2022). The population of low-eccentricity Venus and Earth co-orbitals that are stable resonant objects may be small due to perturbations from terrestrial planets or nongravitational effects (Morais & Morbidelli 2006; Malhotra 2019; Pokorny & Kuchner 2021). There are no known satellites of Venus, likely because the Hill sphere of Venus is mostly dynamically unstable to long term satellites (Sheppard & Trujillo 2009). To date, the population of 1 km and smaller objects near Venus' orbit is relatively unconstrained observationally as most surveys in this region are only sensitive to larger objects.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Data from the two space missions HELIOS and STEREO indicate the presence of a narrow ring of dust in Venus' orbit (Leinert & Moster 2007; Jones et al. 2013). This Venus co-orbital dust ring could come from a population of low-eccentricity (e < 0.3) stable Venus resonant co-orbital objects (Pokorny & Kuchner 2019). These authors also found that about 8% of an initial population of Venus resonant objects remain stable for the age of the solar system. Thus, these putative low-eccentricity Venus resonant co-orbitals could be the leftover remnants of the planetesimals that formed near Venus, unlike most NEOs and high-eccentricity Venus co-orbitals that are believed to have recently escaped from the much further out main asteroid belt (Granvik et al. 2017). This theorized population of low-eccentricity Venus resonant objects has yet to be observationally ruled out since most surveys to date have not covered the large areas of sky to the faint depths needed to find such a population of low-eccentricity objects interior to Earth. This population could be too faint for small class telescopes to detect efficiently in the glare of the Sun as objects less than about 1 km in size would mostly be fainter than about 20th magnitude (Morbidelli et al. 2020; Masiero et al. 2020).

Surveys that have covered a significant portion of the sky looking for asteroids near Venus have used small class telescopes (Myhrvold 2016). Recently the 48 inch telescope Zwicky Transient Facility (ZTF; Ye et al. 2020) survey has done the most extensive search for asteroids near Venus. ZTF found the first ever asteroid that has an orbit entirely interior to Venus, (594913) 'Ayló'chaxnim (2020 AV2; Bolin et al. 2020), though the orbit is only stable for a few million years (Greenstreet 2020; de la Fuente Marcos & de la Fuente Marcos 2020; Bolin et al. 2022). This survey has covered several thousand square degrees of sky interior to Earth's orbit, but is limited in its discovery ability of smaller objects as it uses a 1.2 m telescope (Ye et al. 2020). Based on 'Ayló'chaxnim's absolute magnitude, it is likely a relatively large asteroid of ∼1.5 km in size.

We performed a pilot survey in 2019 September to determine if a search for objects near Venus' orbit is feasible with the medium class Blanco 4 m telescope using the large field-of-view Dark Energy Camera (DECam), which is the largest sky area camera on a 4 m or larger telescope. We searched about 35 square degrees of sky some 40°–50° in elongation away from the Sun near the end of nautical and start of astronomical twilight time in the evening. The images went to over 21st mag in the r band and we were able to put some moderate limits on the size of any stable Venus resonant co-orbital population (Pokorny et al. 2020). Here we discuss an additional 589 square degrees of twilight sky near to and interior to Venus' orbit, yielding a total of 624 square degrees searched in the survey to date.

The twilight survey for objects near to and interior to Venus' orbit uses the DECam on the Cerro-Tololo Inter-American Observatory (CTIO) 4 meter Blanco telescope in Chile. DECam has 61 working science CCDs arranged in a circular type pattern at the prime focus of the telescope. Each CCD has 2048 × 4096 pixels with a pixel scale of about 0.264 arcseconds, yielding a field of view of about 2.7 square degrees per image (Flaugher et al. 2015). Images were calibrated by subtracting a median bias image and divided by a median dome flat.

Observations were obtained just after nautical twilight ends in the evening and just before nautical twilight begins in the morning. The Sun is usually between about −15 and −12 degrees below the horizon when the images are taken, giving about 10 minutes of observation time each twilight. Images were generally taken at airmasses between about 2.3 and 2.5, near the 23° elevation limit of the telescope. Most fields were observed near to or interior to Venus' orbit when projected onto the sky (Figure 3). The r-band filter with an exposure time of 28 s was used. With simultaneous readout and offsetting the telescope to the next position a few degrees away, it takes just under 1 minute between the start of two successive images. Two images of each field are taken separated by about 2–4 minutes. The images usually only had a background of a few thousand counts, while the detector does not saturate until several tens of thousands of counts.

Zoom In Zoom Out Reset image size

The depth of each image mostly depends on the seeing and the sky brightness. The signal-to-noise ratio of asteroids in the twilight images is generally worse than when observed exterior to Earth's orbit because of the high phase angles of the asteroids, bright background sky, and high airmass. In the twilight images taken near the Sun, the main belt asteriods are much farther away in our fields than when imaged in darker skies exterior to Earth's orbit. Thus, because of the relatively large distances of the main belt asteroids, our survey will not usually find unknown main belt asteroids.

An object that is interior to Earth's orbit will generally have an apparent motion of over 100 arcseconds per hour, with objects near Venus' orbit expected to show motions near 150 arcseconds per hour (Figure 4). The exposure time is limited by the fast motion of the objects, as an object with an apparent motion of 150 arcseconds per hour will start to show significant trailing after about 30 s. The survey used 28 s for the exposure time to limit any signal-to-noise trailing losses of the asteroids. The moving object search algorithm was the same as that used to find outer solar system objects (Trujillo et al. 2001; Sheppard et al. 2019) and it was set to flag any objects that were found to have an apparent motion faster than 75 arcseconds per hour. This allows us to identify easily the objects of interest that could be NEOs or Venus resonant co-orbitals while rejecting the many objects within the main asteroid belt and beyond, as their apparent motion is less than about 80 arcseconds per hour. Since some main belt asteroids have apparent motions faster than our 75 arcsecond per hour lower limit, we used the known main belt asteroids that we detected moving faster than 75 arcseconds per hour to help characterize our survey's detection efficiency.

Zoom In Zoom Out Reset image size

We determined the limiting magnitude or depth to find moving objects in our survey fields based on two techniques. First, we determined to what magnitude we detected known main belt asteroids as well as known NEOs. Second, we implanted artificial moving objects into some fields to determine our moving object detection efficiency. The techniques are complementary and agreed to within a tenth of a magnitude of each other. The known main belt asteroids had their V-band Minor Planet Center magnitudes converted to r-band magnitudes using the simple conversion V − r = 0.2 mag (Smith et al. 2002; Sheppard 2012; Sergeyev & Carry 2021). Artificial objects were placed in some of the fields ranging from 19th to 23rd mag in the r band with apparent motions as expected for NEOs between 80 and 170 arcseconds per hour. Because of the short time-base of only ∼4 minutes between images and the short integration time of 28 s, we found no significant differences in our detection of objects with the fastest or slowest apparent motions.

We find that for the average field with a seeing of about 14, we detected about 50% of artificially implanted objects for m_r ∼ 21.3 mag, which is what we take as the overall limiting magnitude of the survey, though different nights had different limiting magnitudes as shown in Table 1 and Figure 5, with some of the best seeing fields reaching nearly 22nd mag in depth. The detection efficiency curve has a similar shape each night as shown in Figure 6, just shifted in m_r to the value as shown in Table 1, which mostly depended on the seeing. Figure 5 shows the limiting magnitude of each field versus the elongation of the field from the Sun. The survey was able to detect asteroids near Venus' orbit of only a few hundred meters in size assuming moderate albedos like S-type asteroids (Pokorny et al. 2020).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

From Jedicke et al. (2016) we use Equation (15) to fit the detection efficiency curve shown in Figure 6: $\epsilon ({m}_{r})\,={\epsilon }_{o}{\left[1+\exp ({m}_{r}-{r}_{50 \% }/{r}_{\mathrm{width}})\right]}^{-1}$, where r_50% is the magnitude in the r band where the efficiency is 50%, _o is the maximum survey efficiency (i.e., the brightest object efficiency), and r_width is the magnitude range in m_r that the efficiency decreases from near 75% to 25%. We find _o = 0.95 ± 0.01, r_50% = 21.3, and r_width = 0.19.

As the survey was active every few nights, most of the asteroid recovery of new discoveries was done using DECam. D. Tholen also recovered new discoveries with the University of Hawaii 88 inch telescope and the Canada-France-Hawaii 3.6 m telescope (CFHT). The 6.5 m Baade–Magellan telescope at Las Campanas in Chile was also used in recovery observations as was the 1 meter global network of telescopes operated by Las Cumbres Observatory, mainly through the European Space Agency's (ESA) Planetary Defense Office (PDO).

Using DECam, we covered 624 square degrees of sky near to and interior to Venus' orbit. Three new NEOs were discovered, with two likely being about 1 km or larger in size (Table 2). In addition, we serendipitously detected several known NEOs moving well over 90 arcseconds per hour, including Atira-, Aten-, Amor-, and Apollo-type NEOs, with several being Potentially Hazardous Asteroids (PHAs).

3.1. 2021 PH27: Atira Type

Asteroid 2021 PH27 was discovered by S. Sheppard on 2021 August 13 in twilight images taken near the Sun by the Local Volume Complete Cluster Survey (LoVoCCS; Dell'Antonio 2020; Fu et al. 2022) in collaboration with the DECam twilight asteroid survey (Sheppard et al. 2021). The asteroid was recovered the next night on August 14 at both the Baade–Magellan telescope and again using DECam on the Blanco 4 m telescope. As the object was relatively bright at 19.2 mag, it could be tracked by 1 meter class telescopes. On August 15 the asteroid was observed by DECam and Magellan as well as the Las Cumbres 1 m telescopes in South Africa and Chile. It was tracked with 1 m class telescopes for almost a month. In addition, pre-discovery images of the asteroid were found from DECam archival images on 2017 July 16, allowing the orbit to be well determined. 2021 PH27 was also recovered again in March 2022.

2021 PH27 has the smallest semimajor axis of any known asteroid and thus the shortest orbital period for an asteroid of about 113 days (Table 2). Only the planet Mercury has a smaller known semimajor axis for any object in our solar system. 2021 PH27 has an Atira-type asteroid orbit that always keeps it inside of Earth's orbit (Figure 7). Because 2021 PH27 has a fairly eccentric orbit, it crosses both the orbits of Mercury and Venus and approaches the Sun to within about 0.133 au.

Zoom In Zoom Out Reset image size

To understand the orbital behavior of 2021 PH27, we ran a numerical orbital simulation using the SWIFT RMVS4 integrator described by Levison & Duncan (1994) and included the planets Mercury, Venus, Earth+Moon (barycenter), Mars (barycenter), Jupiter (barycenter), Saturn (barycenter), Uranus (barycenter), and Neptune (barycenter). The time step of the simulations was 0.1 days. We generated 1000 clones using the JPL covariance matrix of 2021 PH27 from 2022 April 18. The randomly generated points follow a six-dimensional normal distribution defined by the covariance matrix and the orbital elements as described by Namouni & Morais (2018).

The orbit of 2021 PH27 is dynamically unstable within a few million year timescale (Figures 8 and 9). There is a 0.7% chance that it will collide with Venus in the next million years, and it will pass within Venus' Hill sphere in the next 950 to 1050 yr, as all of our clones travel through Venus' Hill sphere in that time. 2021 PH27's very close perihelion to the Sun means 2021 PH27 experiences the largest general relativistic effects on any known object in our solar system, including Mercury. We used the ReboundX code to determine the general relativistic effects on 2021 PH27 from the Sun (Tamayo et al. 2020). The orbital precession rate of 2021 PH27 is about 53 arcseconds per century, which is faster than Mercury's orbital precession rate of about 43 arcseconds per century (Will 2018). The general relativity effects are not important in 2021 PH27's orbital evolution as planetary perturbations are much more significant. The asteroid interacts strongly with Venus and is likely in some sort of Kozai–Lidov oscillation from Jupiter and/or the inner planets as seen by the coupling of the eccentricity and inclination of the evolution of 2021 PH27's orbit in Figure 9 (Kozai 1962; Lidov 1962; de la Fuente Marcos & de la Fuente Marcos 2021). 2021 PH27's orbit and Venus overlap enough that it is a PHA to Venus with a Venus Minimum Orbit Intersection Distance (MOID) of only 0.015. If 2021 PH27 is or becomes an active asteroid, it could create meteor showers in Venus' atmosphere (Albino et al. 2022).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

2021 PH27 can reach surface temperatures of around 500°C during its closest approach to the Sun. This means its surface has been strongly thermally processed over time and its internal structure likely has been heavily stressed from the intense and changing thermal environment (Li & Jewitt 2013; Lisse & Steckloff 2022).

2021 PH27 was relatively bright at discovery, being some 19.2 mag in the r band. As the aphelion distance is only 0.79 au, which is just beyond Venus' orbit, this object, though relatively bright and around 1 km in size, would be very hard for most NEO surveys to find as they generally do not observe near Venus' orbit or interior to it. The large inclination of about 32° also means 2021 PH27 spends most of its time well away from the ecliptic.

Though 2021 PH27 most likely came from the main asteroid belt, it is possible that it could have originated much closer to the Sun from a possible stable reservoir of small objects in resonance with Venus or even the hypothetical Vulcanoid population, which is a theoretical stable area of small bodies interior to Mercury (Greenstreet et al. 2012).

3.2. 2021 LJ4: Atira Type

3.3. 2022 AP7: Apollo Type

The main design of the survey was to find or put constraints on the population and orbital properties of objects near to or interior to Venus' orbit. We did not find any relatively stable Venus co-orbital or resonant objects, allowing us to put strong upper limits on the size of such a population if it exists. Though not the main goal of the survey, four Atira-type asteroids were detected, which is ∼15% of the known Atira population. This allows us to put significant constraints on the size of the Atira population, which have orbits completely interior to Earth's. In addition, this survey is one of the few that can put modest constraints on the Vatira population, which has orbits completely interior to Venus' orbit.

To determine what percentage of each interior population's orbits we may have surveyed, we performed several Monte Carlo simulations. These simulations consisted of generating 100,000 clones of orbits of a particular population and then determining how many of these objects our survey would have been expected to detect based on the fields observed, limiting magnitude of each field, and the efficiency of the survey. This gives a rough percentage of the entire population of orbits. We then use Poisson statistics to put limits or upper limits, in the case of no detections, of each population simulated.

For each simulation we assumed asteroids of six spectral types (S-type with 0.22 albedo, M with 0.17, E with 0.45, C with 0.06, P with 0.04, and D with 0.05; Mainzer et al. 2011, 2012; Shevchenko et al. 2016; Morbidelli et al. 2020) and three diameters (D = 0.5, 1.0, and 1.5 km) using the same methods as detailed by Pokorny et al. (2020). See Pokorny et al. (2020) for more details on our simulations, including the phase function, apparent visual magnitude, and diameter conversions we used in our simulations. The asteroid type and hence the albedo of the asteroid only matters for detecting objects less than 1 km as all objects, no matter their type, would have similar detection efficiencies near or above 1 km as our survey was deep enough to make almost all 1 km or bigger inner asteroids efficiently detected.

4.1. Stable Venus Resonant Objects

Pokorny & Kuchner (2019) found that some low-eccentricity, low-inclination Venus co-orbital objects could be in resonance and stable for the age of the solar system. If this population exists, it would be a very interesting population to find and study as these objects could be the remnant of the inner solar system's formation disk. Pokorny et al. (2020) performed a limited survey with DECam of the space near Venus' orbit to put modest constraints on this possible population of stable Venus resonant co-orbitals. Here we extend the Pokorny et al. (2020) analysis using the new, much larger survey area covered near Venus' orbit with DECam. For this population we followed 100,000 clones of low-eccentricity, low-inclination Venus co-orbital asteroids that were shown to be stable for the age of the solar system by Pokorny & Kuchner (2019).

Based on our simulations, this DECam twilight survey should have observed some 21%–23% of the entire population of stable Venus co-orbitals larger than 1 km, which depends on the spectral type and thus albedo of the objects (Table 3). Since we found no relatively stable Venus resonant objects and there are none known, this population is not likely to be very large. We calculate the upper limit on the population size of 1 km or larger stable Venus co-orbitals to be only about ${4}_{-3}^{+6}$, depending on the asteroid type (Table 3). Similar constraints are found for even smaller 0.5 km sized higher albedo asteroids types like the S, M, and E types, as they would be relatively easier to find because of their high reflectance. For the darker asteroid types, like the C, P, and D types, smaller sized asteroids of 0.5 km are basically unconstrained as most would be too faint for our survey to detect efficiently (Table 3).

Table 3. Stable Venus Co-orbital Completeness Percentage and Population

Type	D = 0.5	D = 1.0	D = 1.5
	(km)	(km)	(km)
S-type	22%(${4}_{-3}^{+6}$)	23%(${4}_{-3}^{+6}$)	23%(${4}_{-3}^{+6}$)
M-Type	21%(${4}_{-3}^{+7}$)	23%(${4}_{-3}^{+6}$)	23%(${4}_{-3}^{+6}$)
E-Type	23%(${4}_{-3}^{+6}$)	23%(${4}_{-3}^{+6}$)	23%(${4}_{-3}^{+6}$)
C-Type	0.8%(${100}_{-80}^{+170}$)	21%(${4}_{-3}^{+7}$)	23%(${4}_{-3}^{+6}$)
P-Type	0.2%(${540}_{-400}^{+900}$)	20%(${4}_{-3}^{+8}$)	23%(${4}_{-3}^{+6}$)
D-Type	0.4%(${210}_{-150}^{+350}$)	21%(${4}_{-3}^{+7}$)	23%(${4}_{-3}^{+6}$)

Note. Survey completeness based on a Monte Carlo simulation of our survey for each population in percent assuming the shown diameter (D) or larger and asteroid type. In "()" is the computed upper limit on the population size based on our survey coverage and null result.

Download table as:  ASCIITypeset image

4.2. Unstable Venus Near Co-orbitals

There are five well observed asteroids that currently have orbital periods similar to Venus: (322756) 2001 CK32, (524522) 2002 VE68, 2012 XE133, 2013 ND15, and 2015 WZ12. All of these, as well as the newer objects found since 2020 are much smaller than 1 km and are on dynamically unstable orbits on the order of a few million years, as discussed in Section 1. We generated 100,000 clones using the above five object's semimajor axes, eccentricities, and inclinations by randomly selecting the remaining orbital elements. Though we did not observe any of these known objects, through this simulation, we find our survey would have detected about 5%–10% of the objects similar to these known dynamically unstable Venus near co-orbitals and larger than 1 km in size, depending on their albedos and spectral types (Table 4). From this simulation there could be a few of these near unstable Venus co-orbitals to discover around the 1 km size regime.

Table 4. Unstable Venus Co-orbital Completeness Percentage and Population

Type	D = 0.5	D = 1.0	D = 1.5
	(km)	(km)	(km)
S-type	8%(${10}_{-7}^{+16}$)	10%(${8}_{-6}^{+13}$)	10%(${8}_{-6}^{+13}$)
M-Type	6%(${12}_{-9}^{+20}$)	10%(${8}_{-6}^{+13}$)	10%(${8}_{-6}^{+13}$)
E-Type	10%(${8}_{-6}^{+13}$)	10%(${8}_{-6}^{+13}$)	10%(${8}_{-6}^{+13}$)
C-Type	0.4%(${220}_{-170}^{+360}$)	7%(${11}_{-7}^{+17}$)	10%(${8}_{-6}^{+13}$)
P-Type	0.2%(${440}_{-330}^{+720}$)	5%(${14}_{-10}^{+23}$)	9%(${8}_{-6}^{+13}$)
D-Type	0.3%(${310}_{-230}^{+510}$)	7%(${11}_{-8}^{+19}$)	10%(${8}_{-6}^{+13}$)

Note. Survey completeness based on a Monte Carlo simulation of our survey for each population in percent assuming the shown diameter or larger and asteroid type. In "()" is the computed upper limit on the population size based on our survey coverage and null result.

Download table as:  ASCIITypeset image

4.3. Atira/Apohele Asteroid Population

4.4. Vatira Asteroid Population

Vatira asteroids have orbits completely interior to Venus' orbit. Since the Vatira population's orbital distribution is still highly uncertain, we simulated two different Vatira populations to understand how sensitive our survey would be to finding Vatira asteroids. First, we used the semimajor axis, eccentricity, and inclination of the only known Vatira asteroid (594913) 'Ayló'chaxnim (2020 AV2) and generated 100,000 clones by randomly selecting the remaining orbital elements. In this simulation, we would have found about 5% of the ∼1 km or larger Vatira asteroids with similar orbits as 2020 AV2 (Table 6). This puts only minimal constraints on a Vatira population with orbits like 2020 AV2, as there could still be tens of objects like 2020 AV2 to be found since they spend most of their time at smaller elongations than many of our survey fields.

Table 6. Vatira Asteroids like 2020 AV2 Population Completeness Percentage

Type	D = 0.5	D = 1.0	D = 1.5
	(km)	(km)	(km)
S-type	5%(${17}_{-13}^{+30}$)	5%(${16}_{-12}^{+30}$)	5%(${16}_{-12}^{+30}$)
M-Type	5%(${19}_{-15}^{+30}$)	5%(${16}_{-12}^{+30}$)	5%(${16}_{-12}^{+30}$)
E-Type	5%(${16}_{-12}^{+30}$)	5%(${16}_{-12}^{+30}$)	5%(${16}_{-12}^{+30}$)
C-Type	0.7%(${130}_{-100}^{+220}$)	5%(${18}_{-14}^{+30}$)	5%(${16}_{-12}^{+30}$)
P-Type	0.5%(${170}_{-130}^{+270}$)	5%(${19}_{-15}^{+30}$)	5%(${16}_{-12}^{+30}$)
D-Type	0.5%(${160}_{-120}^{+270}$)	5%(${19}_{-15}^{+30}$)	5%(${16}_{-12}^{+30}$)

Note. Survey completeness based on a Monte Carlo simulation of our survey for each population in percent assuming the shown diameter (D) or larger and asteroid type. In "()" is the computed upper limit on the population size based on our survey coverage and null result.

Download table as:  ASCIITypeset image

In a second simulation, we made a 100,000 asteroid hypothetical Vatira population by creating a somewhat circular and moderately inclined population through randomly selecting orbital elements with 0.50 < a < 0.60 au, 0.0 < e < 0.1, and 0 < i < 15° and randomly selecting the other remaining orbital elements between 0 and 360°. As many of the objects in this hypothetical Vatira population would not have elongations as great as 2020 AV2, it would be much harder to detect these objects in our fields. Thus our survey would only find at most about 1% of objects in this hypothetical Vatira population (Table 7).

Table 7. Circular Vatira Asteroid Population Completeness Percentage

Type	D = 0.5	D = 1.0	D = 1.5
	(km)	(km)	(km)
S-type	1%	1%	1%
M-Type	1%	1%	1%
E-Type	1%	1%	1%
C-Type	0.3%	0.9%	1%
P-Type	0.3%	0.9%	1%
D-Type	0.3%	0.9%	1%

Note. Survey completeness based on a Monte Carlo simulation of our survey for each population in percent assuming the shown diameter (D) or larger and asteroid type.

Download table as:  ASCIITypeset image

The twilight survey using DECam on the CTIO Blanco 4 m telescope is one of the largest area and sensitive searches ever performed for objects interior to Earth's and near Venus' orbit. Though no objects with orbits similar to or interior to Venus' orbit were found, the survey did find three relatively large NEOs, including two Atira and one Apollo orbital type. The new discovery 2021 PH27 is about 1 km in size and has the smallest semimajor axis and thus shortest orbital period around the Sun of any known asteroid. 2021 PH27 has strong interactions with Venus, with it likely passing through Venus' Hill sphere in 950–1050 yr from now, making it a PHA to Venus. The eccentric orbit of 2021 PH27 means it crosses both Mercury's and Venus' orbit and its very low perihelion of only 0.13 au creates about a 53 arcsecond per century precession in its orbit from general relativistic effects, which are the largest known in our solar system. The newly discovered 2022 AP7 is an Apollo-type NEO that is probably the largest potentially hazardous NEO to Earth found in several years, being some 1.5 km in size. There are likely several more 1 km sized Atira-type asteroids left to find, which probably have low semimajor axes and high inclinations, like 2021 PH27, making them hard to find for most asteroid surveys. The DECam twilight survey is covering sky geometries and areas that most other surveys do not cover to depths not usually obtained, filling an important niche in the survey for the last few remaining relatively large unknown NEOs. Interestingly, the twilight survey has discovered more larger asteroids (≳1 km) than smaller ones even though the survey is sensitive to smaller asteroids. This might suggest the smaller asteroids are dynamically less stable and/or more susceptible to break-up from the extreme thermal and gravitational environment near the Sun, though additonal discoveries of asteriods with orbits near the Sun must be made to determine statistically if the smaller asteroids are under-abundant since in general they are also harder to detect.

Observations were obtained at Cerro-Tololo Inter-American Observatory, National Optical Astronomy Observatory, which are operated by the Association of Universities for Research in Astronomy, under contract with the National Science Foundation. This project used data obtained with the Dark Energy Camera (DECam), which was constructed by the Dark Energy Survey (DES) collaborating institutions: Argonne National Lab, University of California Santa Cruz, University of Cambridge, Centro de Investigaciones Energeticas, Medioambientales y Tecnologicas-Madrid, University of Chicago, University College London, DES-Brazil consortium, University of Edinburgh, ETH-Zurich, University of Illinois at Urbana-Champaign, Institut de Ciencies de l'Espai, Institut de Fisica d'Altes Energies, Lawrence Berkeley National Lab, Ludwig-Maximilians Universitat, University of Michigan, National Optical Astronomy Observatory, University of Nottingham, Ohio State University, University of Pennsylvania, University of Portsmouth, SLAC National Lab, Stanford University, University of Sussex, and Texas A&M University. Funding for DES, including DECam, has been provided by the U.S. Department of Energy, National Science Foundation, Ministry of Education and Science (Spain), Science and Technology Facilities Council (UK), Higher Education Funding Council (England), National Center for Supercomputing Applications, Kavli Institute for Cosmological Physics, Financiadora de Estudos e Projetos, Fundação Carlos Chagas Filho de Amparo a Pesquisa, Conselho Nacional de Desenvolvimento Científico e Tecnológico and the Ministério da Ciência e Tecnologia (Brazil), the German Research Foundation-sponsored cluster of excellence "Origin and Structure of the Universe" and the DES collaborating institutions. D.T. was supported by NASA grant 80NSSC21K0807. P.P. was supported by NASA ISFM EIMM award, the NASA Cooperative Agreement 80GSFC21M0002 and NASA solar system Workings award 80NSSC21K0153. T.S.R. acknowledges funding from the NEO-MAPP project (H2020-EU-2-1-6/870377). C.M. was partially supported by the international Gemini Observatory, a program of NSF's NOIRLab, which is managed by the Association of Universities for Research in Astronomy (AURA) under a cooperative agreement with the National Science Foundation, on behalf of the Gemini partnership of Argentina, Brazil, Canada, Chile, the Republic of Korea, and the United States of America. This work was (partially) funded by the Spanish MICIN/AEI/10.13039/501100011033 and by "ERDF A way of making Europe" by the "European Union" through grant RTI2018-095076-B-C21, and the Institute of Cosmos Sciences University of Barcelona (ICCUB, Unidad de Excelencia María de Maeztu') through grant CEX2019-000918-M. This work makes use of observations from the Las Cumbres Observatory global telescope network. This paper includes data gathered with the 6.5 m Magellan Telescopes located at Las Campanas Observatory, Chile.