Physical properties of five near-Earth asteroids within the NEOROCKS project

ABSTRACT

We present photometric data of five near-Earth asteroids (NEAs) obtained in the framework of the NEO Rapid Observation, Characterization, and Key Simulations (NEOROCKS) project. The selected asteroids are (351545) 2005 TE15, (438908) 2009 XO, (501647) 2014 SD224, 2015 FC35, and 2016 CO247. Light curves were obtained for all the asteroids using the 0.46-m TAR2 telescope at the Teide Observatory (OT), and spectrophotometric data in g, r, i, z|$_s$| filters were obtained for asteroids (438908) 2009 XO and (501647) 2014 SD224 using the MuSCAT-2 instrument attached to the 1.52-m Carlos Sanchez Telescope also at the OT. We derived the rotational period of the five NEAs using Fourier series analysis of the rotational light curves. We found |$P = 10.6035 \pm 0.0010$| h for (351545) 2005 TE15, |$7.9140 \pm 0.006$| h for (501647) 2014 SD224, |$3.4211 \pm 0.0075$| h for 2015 FC35, and |$4.800 \pm 0.005$| h for 2016 CO247. We observed two periodic variations for (438908) 2009 XO, a long one with two possible periods of |$P_1 = 157.2 \pm 0.4$| h and |$P_2 = 304.0 \pm 0.4$| h and a large amplitude of 0.881 mag, and a short one with a period of |$P = 13.448 \pm 0.008$| h. Radar data suggest that (438908) 2009 XO could be a contact binary and its double period points towards a non-principal axis rotator. Finally, we used the spectrophotometric data to obtain colours and taxonomic classification for (438908) 2009 XO and (501657) 2014 SD224, concluding that both NEAs are S-type objects.

1 INTRODUCTION

Near-Earth objects (NEOs) are asteroids and comets that have orbits close to that of the Earth (perihelion distance less than 1.3 au). The majority of the population of NEOs are asteroids, named as near-Earth asteroids (NEAs). Due to their proximity to Earth, NEAs are the most accessible vestiges from the building blocks that formed the Solar system, are accessible to spacecrafts, and allow ground-based facilities to do detailed observations of very small (a few meter sized) objects, otherwise being too faint to be observed. They can also be studied using ground-based observational techniques like radar that cannot be used with other small bodies of the Solar system. In addition, and also due to their proximity to Earth, NEAs deserve special attention since they can collide with our planet, and so, they are a real hazard for life on Earth. Those NEAs that have a minimum orbital intersection distance with the Earth smaller than 0.05 au and diameters larger than 140 m are classified as potentially hazardous asteroids (PHAs).¹

The discovery rate of NEAs is large and only a small fraction of the total are physically characterized (the number of NEAs with known physical properties depends on their size), mainly due to the long required time to perform a complete physical characterization and the small number of telescopes dedicated to the NEO observations. According to Rondón et al. (2019), of the 19 000 known NEAs at the time of their publication, and despite the existence of several dedicated works, photometric properties had been partially determined only for less than 10 per cent of the known NEAs, about 1300 objects had a rotational period with good reliability, i.e. a quality code greater than 2 – according to the light-curve (LC) data base (Warner, Harris & Pravec 2009), and there were slightly more than 1 500 NEAs with taxonomic classification (Binzel et al. 2019). As of 2023 July 29, the number of known NEAs has increased significantly, with a total of 32 392 objects.

In 2018, in the framework of several European funded projects, the Solar System Group of the Instituto de Astrofísica de Canarias (IAC), and in collaboration with the Astronomical Institute of the Romanian Academy (AIRA), started a joint Visible NEAs Observations Survey (ViNOS). The main goal of the programme is to characterize the NEA population using spectroscopic, spectrophotometric, and LC observations. The programme focuses on recently discovered NEAs, small NEAs, those classified as potentially hazardous (PHAs), possible targets for space missions, and those observed with other techniques, including the radar, to provide complementary data at optical wavelengths. ViNOS is based on observations done with the facilities located at the Observatorios de Canarias (OCAN), in Spain, and managed by the IAC, namely the Teide Observatory (OT), located in the island of Tenerife, and the El Roque de los Muchachos Observatory (ORM), in the island of La Palma.

The observations presented in this paper are part of our contribution to the NEO Rapid Observation, Characterization and Key Simulations (NEOROCKS) project funded through the H2020 European Commission programme to improve the knowledge on near-Earth objects by connecting expertise in performing small body astronomical observations and the related modelling needed to derive their dynamical and physical properties. The members of Solar System Group of the IAC led a specific task on the NEOROCKs project, consisting on doing observations of NEAs in support to the Arecibo Observatory radar programme, using the facilities of the OCAN. These observations included times-series photometry (LCs), spectroscopy, and colour photometry, mainly in the visible. After the collapse of Arecibo occurred on 2020 December 1, we focused in those targets observed or planned to be observed in the frame of the Arecibo Observatory radar programme. After 2021 January, and to increase our sample, we started to observe also NEAs with radar observations from the National Aeronautics and Space Administration (NASA)’s Goldstone Solar System radar.

Radar observations are a powerful technique for characterizing the properties of NEAs: radars can spatially resolve these objects, making it possible to identify satellites, contact binaries, and very irregular shapes. Radar observations allow also to estimate the target’s rotation period and albedo (Benner et al. 2015). By combining observations obtained with radar and optical telescopes, we can have a better determination of the physical properties of NEAs.

We obtained photometric data (rotational LCs and colours) of five NEAs that were also observed or planned to be observed by Arecibo or Goldstone radar facilities: (351545) 2005 TE15, (438908) 2009 XO, (501647) 2014 SD224, 2015 FC35, and 2016 CO247. Rotational periods for the five targets were retrieved from the rotational LCs, while colours provided a taxonomical classification for two of the NEAs. Colour taxonomy can be used to infer gross compositional information on the targets, while rotational periods can tell us about the structure of the object, i.e. a strengthless rubble-pile will not be able to rotate faster than about 2.2 h without breaking up. Thus, slow rotations can be the result of an intense collisional evolution, while fast rotations are indicative of a monolithic body (Pravec, Harris & Michalowski 2002).

This paper is organized as follows. In Section 2, we describe the target selection, the observations, and the data reduction for all asteroids. In Section 3, we present the methods to obtain the taxonomical classification using the colours of (438908) 2009 XO and (501647) 2014 SD224. Section 4 describes the obtained results and we present their implications and conclusions in Section 5.

2 OBSERVATIONS AND DATA REDUCTION

We selected our five targets by considering the following criteria: (1) most of them has reliable photometric parameters determined on any photometric public data base; (2) they were observed or planned to be observed with either Arecibo or Goldstone radar facilities. These NEAs are (351545) 2005 TE15, (438908) 2009 XO, (501647) 2014 SD224, 2015 FC35, and 2016 CO247. Their orbital elements are shown in Table 1. The information includes semimajor axis (a), eccentricity (e), sine of the inclination (i), and absolute magnitude (H). The orbital elements were extracted from the JPL Small-Body Database Browser.²

Time-series photometry for the five NEAs were obtained using the 0.46-m TAR2 telescope, one of the three TAR (Telescopio Automático Remoto, or Remote Automatic Telescope in English) telescopes, located at the OT, as part of a dedicated large programme to obtain LCs of NEAs that uses TAR2 almost every clear night. The instrumental set-up and the image acquisition process are described with more detail in Licandro et al. (2023). The images were acquired with no filter.

Aperture photometry of the final images was done by using the photometry pipeline (PP) software (Mommert 2017).³PP employs the widely used Source Extractor software for source identification and aperture photometry and the scamp⁴ software for image registration. Both image registration and photometric calibration are based on matching field stars with star catalogues [e.g. Sloan Digital Sky Survey (SDSS), Gaia, URAT-1]. Circular aperture photometry is performed using Source Extractor; an optimum aperture radius is identified using a curve-of-growth analysis. Field stars are used to flux calibrate the results. The unfiltered images are calibrated to the |$r^{\prime }$| Sloan filter using Pan-STARRS catalogue. Calibrated magnitudes obtained by PP are typically accurate within 0.05 mag and astrometric accuracies are of the order of 0.3|$^{\prime \prime }$|relative to the catalogues used in the registration. Final calibrated photometry for each field source is written into a queryable data base; target photometry is extracted from this data base. Moving targets are identified using JPL Horizons ephemerides. LC analysis (including rotation period determination) was carried out using the tycho package.⁵ Magnitudes were (H-G) and LC corrected and the rotation period was obtained using a Fourier analysis algorithm, like in Harris & Lupishko (1989).

Spectrophotometric observations were done with the 1.52-m Carlos Sanchez Telescope (TCS) located at OT as part of a large programme to obtain colours of NEAs, based on which we scheduled observations every two nights per month (Popescu et al., 2021). As for the case of the time-series photometry, the instrumental set-up and the acquisition process for the TCS colour images are described in Licandro et al. (2023). We also used Photometry Pipeline to obtain the calibrated photometry.

A summary of the observational circumstances for each asteroid studied in this work is presented in Table 2. The information includes the telescope, date and the starting time (Date-UT) of the observations, the total observed time in hours, the phase angle (⁠|$\alpha$|⁠), the longitude and latitude of the phase angle bisector (⁠|$L_{PAB}$|⁠, |$B_{PAB}$|⁠), the distance to the Sun (r) and to the Earth (⁠|$\Delta$|⁠), and the apparent visual magnitude (⁠|$m_V$|⁠) of the asteroid at the time of observation.

3 COLOUR TAXONOMY

To perform the taxonomic classification, we used both the K-Nearest Neighbour (KNN) and the Random Forest (RF) algorithms. We implemented each algorithm using the scikit-learn package from python. The KNN algorithm implies classifying an object based of the label values or taxonomy of its neighbours, while the RF assigns the final label of an object using decision-tree structures, which are also drawn using objects with known taxonomy. Thus, the first step was to build a training set, namely a set of objects for which we know both the photometric and spectral data. We searched the available spectral information from the SMASS-MIT-Hawaii Near-Earth Object Spectroscopic Survey (MITHNEOS) programme (Binzel et al. 2019) and from the Modelling for Asteroids (M4AST) data base (Popescu, Birlan & Nedelcu 2012; Popescu et al. 2019), and we found spectral data for a total of 84 objects that had colour photometry obtained with the TCS.

In addition to these 84 NEAs, we computed so-called synthetic colours from the visible spectra of another 70 NEAs. These synthetic colours were obtained using the spectral observations published by Popescu et al. (2019) and Perna et al. (2018). Thus, the final training set consisted of 154 asteroids, and included 5 A-types, 8 V-types, 34 Q-types, 48 objects belonging to the S-complex, 7 B-types, 15 objects belonging to the C-complex, 9 D-types, and 28 objects belonging to the X-complex. Note that we only classified the targets into the major complexes (S-, C-, and X-complex) and end-member classes, because colours do not have enough spectral resolution to separate into different subclasses.

Since the spectral classification has been made based on a small number of features, we chose to implement the KNN algorithm so that the taxonomy is given by calculating the euclidean distance to the first three nearest neighbours. Also, because the assigned spectral type is sensitive to the position of each object relative to the training set, we needed to account for the magnitude errors. To do that, we started from the colour value and its error and generated three normal distributions (one for each colour) of 10 000 fictitious colour values. Then, in each of these cases we classified the object. Finally, the assigned taxonomy was the one with the highest frequency.

For the classification based on the RF algorithm, we used nine different decision-trees, each of them with a maximum of 50 leaf nodes. To account for the magnitude errors, a procedure similar to the one for the KNN method was used. We applied the RF algorithm for each of the 10 000 colour sets and chose the predicted taxonomy with the highest frequency. In the end, having the output of each algorithm, we needed to pick a final taxonomy. For that, we compared the prediction probability of each method and chose the taxonomy corresponding to the algorithm having a higher probability. We emphasize that the goal of these classification methods was to constrain the composition group these objects are part of, namely to distinguish between carbonaceous, silicate, or basaltic composition. In order to distinguish between two spectral types within the same composition group (i.e. Sa and Sq types), we need either more training data or more features used for classification.

4 RESULTS

We obtained time-series photometry for our five NEAs, and colour photometry for two of them, asteroids (438908) 2009 XO and (501647) 2014 SD224. In an attempt to enhance the visualization and classification of the results, we have also incorporated quality codes to LCs according to the LC data base (Warner et al. 2009). In this section, we present the results for each asteroid.

4.1 (351545) 2005 TE15

Asteroid (351545) 2005 TE15 is an NEA from the Apollo group. It has an absolute magnitude of |$H = 19.81$|⁠, which gives a diameter of |$D \sim$| 370 m, assuming an albedo value of |$p_V$| = 0.15. It was on the list of asteroids planned to be observed with the Goldstone radar facility in 2021 March. Unfortunately, due to maintenance works at the Goldstone Observatory, no radar data was obtained.

We observed (351545) 2005 TE15 on 2021 March 9 and 15 with TAR2. The LC shown in Fig. 1 is the phased curve using the data from both nights and fitted using Fourier series analysis. Using a third-order fitting, we derived a rotation period of |$P = 10.6035 \pm 0.0099$| h and a LC amplitude of |$\text{Amp.} = 0.838$| mag. The periodogram in Fig. 1 shows another minimum at |$P \sim$| 11.29 h, similar to the one at 10.60 h, so we cannot completely discard it. In fact, by slightly increasing the fitting order, the |$P = 11.29 \pm 0.01$| h solution becomes the best one. This makes the period determination a bit uncertain. Because of the possibility of the other rotational period we attribute to our data the quality code U|$=2+$|⁠. However, Skiff et al. (2023) found similar results, with a value of period |$P = 10.610$| using data from 2009 September 21, 24, 25, 26, but with more sparse data and rather noisy in comparison with our data. Other remarkable feature is the high value of the LC amplitude with |$\text{Amp.} = 0.838$| mag from our data, compared to found by Skiff et al. (2023) with |$\text{Amp.} = 0.52$| mag. This amplitude variation can be attributed to a change in the geometry of the observations, variations of either its aspect or phase angle.

4.2 (438908) 2009 XO

Asteroid (438908) 2009 XO is another NEA of the Apollo group, classified as a PHA, which means that this asteroid has the potential to make threatening close approaches to the Earth. It was on the list of asteroids planned to and actually observed by the Arecibo radar facility in May 2020. Delay-Doppler images were obtained on several days in 2020 May,⁶ including some bistatic observations with Arecibo transmitting and the Green Bank Telescope receiving (Fig. 2). The data show a very narrow Doppler bandwidth (less than 0.1 Hz). This means that either it is a slow rotator or that it was observed nearly pole-on. Since its angular position changed by about 60|$^{\circ }$| over the course of the radar observations, it could not have been pole-on for every radar observation; therefore, the narrow bandwidth indicates slow rotation. Radar observations from both Arecibo and Green Bank radar facilities (private communication) are used to determine that (438908) 2009 XO is possible an elongated contact binary, with a maximum length of about 180 m (⁠|$\pm$| 25 per cent). Fig. 2 shows a sum of the best Arecibo delay-Doppler images of 2009 XO from 2020 May 3.

On the other hand, we obtained time-series photometry of this asteroid on 2020 May 15, 17, 18, 19, and 21, using the TAR2 telescope, while the Arecibo radar observations were performed May 3–7. We first noticed that there was a rather large brightness variation (⁠|$\sim 0.9$| mag) that could be periodic (see Fig. 3) above the |$\lt $|0.1 mag calibration accuracy obtained using the solar-type field stars (usually |$\sim$| 100 stars because the large field of view of 1 deg|$^2$| from TAR2). To confirm this variation we downloaded images of (438908) 2009 XO obtained by the Asteroid Terrestrial-impact Last Alert System⁷ survey along 24 d starting 2020 May 9, and measured its brightness in the o and c ATLAS filters. By applying a |$o-c= 0.3$| mag colour correction (Denneau, personal communication), we plotted the ATLAS magnitudes together with ours in Fig. 3 and looked for a periodicity using the Fourier analysis. In this way, we obtain the periodogram shown in the Fig. 4, lower panel, founding two possible periods, |$P_1 = 157.2 \pm 0.4$| h and |$P_2 = 304.0 \pm 0.4$| h, with rotational LC shown in Fig. 4, upper panel and lower panel, respectively. These two values are in agreement with the asteroid being a slow rotator, as suggested from radar data. In both cases, the amplitude is |$\text{Amp.} = 0.881$| mag. We also see shorter period variations in our TAR2 LCs. After applying a magnitude offset to each TAR2 LC to remove the long-period variation, we determined that there is another periodic variation with a period |$P = 13.448 \pm 0.008$| h (see Fig. 5) and an amplitude of |$\text{Amp.} = 0.157$| mag. Since the data lack high quality, it leaves a gap to support the hypothesis that the object is closer to the characteristics of a conventional binary system as opposed to a contact binary. In this context, the large period is indicative of the orbital period of the system, while the short period aligns with the rotational period of the primary body. Due to the poor data coverage of the rotational LC in the Fig. 4, we assigned a quality code U = |$2-$|⁠; however, for the rotational LC in Fig. 5 we assigned a quality code U = |$2+$|⁠. No phase angle correction was applied as the intrinsic large-amplitude variation of the asteroid magnitude, and the short-phase angle range of the observations (only 2.6|$^{\circ }$|⁠) makes it impossible to derive any reasonable fit. Anyhow, as the range of phase angles is small also the effect is small, likely in the range of the absolute calibration uncertainty.

We also obtained simultaneous observations of (438908) 2009 XO using broad-band filters g, r, i, |$z_s$| in order to obtain its taxonomical classification. With three colour features and the KNN algorithm, we predicted that (438908) 2009 XO is an S-type with 93 per cent of probability. On the other hand, using the RF algorithm we predicted that it is an S-type with 52 per cent probability. Using both algorithms the result is consistent with a silicate-like composition. We can see this classification in Fig. 6, where the pink circle, representing the asteroid colours, is well inside the region of the S-types.

Regarding radar observations, Benner et al. (2008) find a very strong correlation between circular polarization ratio (SC/OC) and visible-infrared taxonomical class. In an attempt to confirm the taxonomical classification obtained from the colours, we reproduce here fig. 1 from Benner et al. (2008) with the regions corresponding to the S-, C-, E-, V-, and X-classes, as shown in Fig. 7. The obtained SC/OC values for asteroid (438908) 2009 XO using the Arecibo radar data from 2020 May 3, 4, and 7 are 0.28, 0.32, 0.20, giving an average of SC/OC = 0.266, with this value and knowing that the absolute magnitude is H = 20.65, we observe that the NEA (438908) 2009 XO is in the S-type region (Fig. 7). This taxonomy is characterized by having an average value of SC/OC |$\sim$| 0.275 (Benner et al. 2008; Aponte-Hernandez, Rivera-Valentín & Taylor 2020).

Using the median albedo value from WISE for asteroids belonging to the S-complex (Mainzer et al. 2011), |$p_V = 0.223 \pm 0.073$|⁠, and the asteroid’s absolute magnitude, we derive a diameter of |$D = 209 \pm 36.828$| m, in agreement with the value suggested from radar observations.

4.3 (501647) 2014 SD224

Asteroid (501647) 2014 SD224 is a NEA that belongs to the Aten group and was discovered on 2014 September 22 by Pan-STARRS 1. It was in the list of asteroids planned to be observed with both the Arecibo and Goldstone radar facilities in 2020 December, and so, it was included in our target list. Delay-Doppler radar images of (501647) 2014 SD224 were obtained with Goldstone radar facility on 2020 December 26, between 16:44 and 21:02 ut.⁸ Images, with a resolution of 7.5 m × 0.08 Hz, confirm that the asteroid has an irregular and elongated shape, and a size of somewhat larger than 60–70 m (Fig. 8, Lance Benner’s private communication).

We obtained colour photometry with the TCS on 2020 December 10 and time-series photometry on 2020 December 17, using the TAR2 telescope. The phased LC is shown in Fig. 9. The best derived rotation period is |$P = 7.914 \pm 0.006$| h. It was obtained by fitting a third-order Fourier series. The periodogram in Fig. 9 shows that there are several other possible rotation periods with similar likelihood, for this reason, we associate the quality code U|$=3-$|⁠. Unfortunately, there is no more data available to do a better determination. The derived amplitude is quite high, |$\text{Amp.} = 0.705$| mag, suggesting that this is an elongated object. Colour photometry on broad-band filters g, r, i, |$z_s$| provided colours that, according to the two algorithms (KNN and RF), give a 100 per cent of probability for the asteroid being an S-type (orange circle in Fig. 6). As for the case of (438908) 2009 XO, we use the albedo value from WISE for asteroids belonging to the S-complex and the asteroid’s absolute magnitude to infer a diameter of |$D = 95 \pm 17.156$| m, in agreement with the size inferred from the radar images.

4.4 2015 FC35

Asteroid 2015 FC35 is also a NEA from the Apollo group. It has an absolute magnitude of |$H = 22.5$|⁠, which assuming a visual albedo of |$p_V = 0.15$| provides a diameter of |$D \sim 130$| m. It was on the list of asteroids planned to be observed with the Arecibo radar facility around 2020 March 28, before the telescope collapsed. However, observations were not attempted because the expected signal-to-noise ratio (SNR) was too low. Regarding Goldstone, they were doing maintenance works at that time.

We obtained time-series photometry on 2020 March 28 with the TAR2 telescope (Fig. 10). The data were phased using the best-fitting rotation period |$P = 3.4211 \pm 0.0074$| h obtained using a third-order Fourier series. Increasing the fitting order produces very similar results. Currently, there is no other published data of this asteroid. The LC clearly shows the typical two maxima and two minima with an amplitude |$\text{Amp.} = 0.209$| mag; however, the data shown a large dispersion, probable caused by the short observed time. We assigned the quality code U|$=2-$|⁠.

4.5 2016 CO247

Asteroid 2016 CO247 is an NEA of the Apollo group classified as a PHA, and it was on the list of asteroids planned to be observed with Goldstone radar facility in 2021 January.

We obtained photometric data of 2016 CO247 on 2021 January 19 and 20 from TAR2 telescope. The rotational period was obtained by fitting a second-order Fourier series, with period |$P = 4.800 \pm 0.005$| h and amplitude |$\text{Amp} = 0.316$| mag, the data shown a large dispersion of the fit, for this reason, we attribute to this rotational LC a quality code U|$=1+$|⁠. Other solutions are possible to see in the periodogram but with greater Root Mean Square Error (RMSE) (see the lower panel of Fig. 11). More data are necessary to improve the determination of the rotational period of this object. Regarding radar observations, only continuous-wave (CW) echo-power spectra were obtained for little less than 3 h. In a private communication, the Goldstone team informed us that they did not notice any variation in the bandwidth over these 3 h. Assuming a diameter of about 200 m and an equatorial view, it would imply a rotation period of about 5 h. As we obtained a rotational period for this target of about 4.8 h, very close to the value inferred from radar data, we can assume, with a certain level of confidence, that the diameter of the asteroid is |$\sim$| 200 m.

A near-infrared spectrum of asteroid 2016 CO247 was available in the MITHNEOS⁹ data base (Binzel et al. 2019) and is shown in Fig. 12. The taxonomical classification is not as precise as with the visible wavelengths, and visual inspection is needed to decide on the best fit. With this spectrum, we cannot confidently determine a unique taxonomy, but we can say that the asteroid is a primitive-like object, belonging to either the C- or the B-complexes. Nevertheless, and from Fig. 12, it seems that it is closer to be a B-type than a C-type. Using the median albedo value for B-types from Mainzer et al. (2011), |$p_V = 0.120 \pm 0.022$|⁠, and the asteroid’s absolute magnitude (H = 20.62), we derive a diameter of |$D \sim 280$| m, in good agreement with the size inferred from radar data. Another spectrum of this asteroid was published in Hromakina et al. (2021), where they determined this asteroid spectra fit mean C-type classification using Johnson–Cousins filters.

After converting the Johnson filter values (from Hromakina et al. 2021) for asteroid 2016 CO247 to the SDSS filter system, we derived |$(r-i) = 0.118 \pm 0.08$| and |$(i-z) = 0.059 \pm 0.11$|⁠. These values are illustrated as the dark blue point in Fig. 6. Consistent with (Hromakina et al. 2021) classification, this object remains within the C-complex region. Performing the same calculations described previously and using the median albedo value for C-types from Mainzer et al. (2011), |$p_V = 0.050 \pm 0.006$|⁠, we derive a diameter of |$D \sim 447$| m, double of the size inferred from radar data.

5 DISCUSSION AND CONCLUSIONS

In this paper, we presented newly obtained photometric data for the NEAs (351545) 2005 TE15, (438908) 2009 XO, (501647) 2014 SD224, 2015 FC35, and 2016 CO247 that include the determination of the rotational LC for all these objects and the taxonomy classification for two of them, using the telescopes 0.46-m TAR2 and 1.5-m TCS. The results are summarized in Table 3. The information includes the total observed time (h), rotational periods and LC amplitudes, spectral types, visual albedos, and derived diameters. Whenever available, we also include information derived from radar observations. For those objects with no spectral type, we use the typical visual albedo value for asteroids, |$p_V = 0.15$|⁠, to derive their diameter. For asteroid (438908) 2009 XO, we include the two long periods (⁠|${\rm long1}$| and |${\rm long2}$|⁠), as well as the short period (⁠|${\rm short}$|⁠) and their associated amplitudes.

For the NEA (351545) 2005 TE15, we found a rotational period of |$P = 10.6035 \pm 0.0099$| h. Similar value was obtained by Skiff et al. (2023), with a lower coverture that obtained in this work. We note that the asteroid showed a high value of LC amplitude and an irregular shape of the secondary minima. Such irregularity could indicate the object’s irregular shape. No radar data was obtained for this asteroid due to maintenance works at the Goldstone Observatory.

We found rotational periods |$P = 13.448 \pm 0.008$| h and |$P = 7.914 \pm 0.006$| h for NEAs (438908) 2009 XO and (501647) 2014 SD224, respectively. In the case of (438908) 2009 XO, we also found a large rotational period of |$P_2 = 304.0 \pm 0.4$| h, indicating it could be in a non-principal axis rotation state. Delay-Doppler radar images of this NEA suggest it could be a contact binary.

Using the broad-band filters g, r, i, and |$z_s$|⁠, we estimated the taxonomical classification for asteroids (438908) 2009 XO and (501647) 2014 SD224. The KNN algorithm gave 93 per cent and 100 per cent probabilities of (438908) 2009 XO and (501647) 2014 SD224 being S-types, respectively. Using the RF algorithm, this probability is maintained for asteroid 2014 SD224, while it decreases to a 52 per cent for asteroid 2009 XO. This results is consistent with the average of the circular polarization ratio (SC/OC|$=0.27$|⁠) associated with S-type taxonomical class.

In the case of the asteroids 2015 FC35 and 2016 CO247, we have only one and two observation nights, respectively. For 2015 FC35, it was not possible to obtain radar data, and so, its rotational period can only be considered as an estimation. For 2016 CO247, despite the scarcity of photometric data, the additional information inferred from both radar and near-infrared spectra points to a rotation period about 5 h and a diameter of about 200 m. Finally, it is interesting to note that all the studied asteroids in this work have diameters below 500 m, i.e. they can all be considered small NEAs.

With this, we have given support to radar observations with photometric data, in accordance to the assigned responsibilities in the frame of the EU NEOROCKS project, and helped to increase our knowledge of the physical properties of NEAs. According to Durech et al. (2015), results based on individual well-studied asteroids can be generalized to other members of the population. Nevertheless, a statistically large sample of asteroids with known properties is crucial to derive general effects that may play important roles for the whole population.

ACKNOWLEDGEMENTS

We thank the reviewer E. Rondón for their helpful comments and suggestions. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 870403 (NEOROCKS). HM, JL, JdL, MRA, and MP acknowledge support from the ACIISI, Consejería de Economía, Conocimiento y Empleo del Gobierno de Canarias, and the European Regional Development Fund (ERDF) under grant with reference ProID2021010134. JL, and JdL acknowledge support from the Agencia Estatal de Investigacion del Ministerio de Ciencia e Ińnovacion (AEI-MCINN) under grant ‘Hydrated Minerals and Organic Compounds in Primitive Asteroids’ with reference PID2020-120464GB-100. The work of MP was supported by a grant of the Romanian National Authority for Scientific Research − UEFISCDI, project number PN-III-P1-1.1-TE-2019-1504. NP-A acknowledges support from the Center for Lunar and Asteroid Surface Science (CLASS), a NASA’s SSERVI team funded in CAN3, and from the Arecibo Planetary Radar Programme that is fully supported by NASA’s Near-Earth Object Observations Programme in NASA’s Planetary Defense Coordination Office through grant no. 80NSSC19K0523 awarded to University of Central Florida (UCF). The Arecibo Observatory is a facility of National Science Foundation operated under cooperative agreement by UCF, Yang Enterprises, Inc., and Universidad Ana G. Mendez. Photometry Pipeline is supported by NASA grants NNX15AE90G and NNX14AN82G and has been developed in the framework of the Mission Accessible Near-Earth Objects Survey (MANOS). This article is based on observations made with the Telescopio Carlos Sánchez, and TAR2 telescopes operated on the island of Tenerife by the Instituto de Astrofísica de Canarias in the Spanish Observatorio del Teide. This work has made use of data from the Asteroid Terrestrial-impact Last Alert System (ATLAS) project. ATLAS is primarily funded to search for NEAs through NASA grants NN12AR55G, 80NSSC18K0284, and 80NSSC18K1575; by-products of the NEO search include images and catalogs from the survey area. The ATLAS science products have been made possible through the contributions of the University of Hawaii Institute for Astronomy, the Queen’s University Belfast, the Space Telescope Science Institute, the South African Astronomical Observatory (SAAO), and the Millennium Institute of Astrophysics (MAS), Chile. Part of the data utilized in this publication were obtained and made available by the MITHNEOS MIT-Hawaii Near-Earth Object Spectroscopic Survey. The IRTF is operated by the University of Hawaii under contract 80HQTR19D0030 with the NASA. The MIT component of this work is supported by NASA grant 80NSSC18K0849.

DATA AVAILABILITY

The data underlying this article will be shared on reasonable request to the corresponding author.

Footnotes

References