Open Access
Issue
A&A
Volume 690, October 2024
Article Number A193
Number of page(s) 12
Section Planets and planetary systems
DOI https://doi.org/10.1051/0004-6361/202451225
Published online 08 October 2024

© The Authors 2024

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1 Introduction

Traditionally, small Solar System bodies (SSSBs) in the inner region have been thought to be distinctively classified as either comets or asteroids. In that case, comets are icy bodies that formed beyond the snow line and display comae and tails near perihelia due to the volatile sublimation. Meanwhile, asteroids are rocky objects that originate from the inner part of the Solar System and appear as point sources in the observed images. However, discoveries have expanded this traditional classification to a broader continuum, with comets and asteroids as the end members (Jewitt & Hsieh 2022). Active asteroids, whose orbits and spectra are similar to those of asteroids but display cometlike activity, are a kind of object between comets and asteroids (Jewitt 2012).

107P/(4015) Wilson-Harrington (hereafter, WH) is an intriguing object that lies between comets and active asteroids. In November 1949, when it was discovered, WH displayed a tail in photographic plate images (Cunningham 1950), but never again (Cunningham 1950; Bowell et al. 1992; Ishiguro et al. 2011a). Accordingly, it was classified as a dormant comet whose activity was extinct (e.g., Chamberlin et al. 1996; Coradini et al. 1997; Lupishko & Lupishko 2001). Fernández et al. (1997) conducted Finson–Probstein modeling on the tail of WH and found that the dust size would be tens to hundreds of micrometers, which contradicts the blue color (B − R ~ −1) of the tail reported by Bowell et al. (1992). Thus, they suggested that the observed tail is dust-less and due to CO+ and H2O+ ion fluorescence, which are commonly found in comets. To date, it remains the only explanation for the activity of WH.

As Fernández et al. (1997) pointed out, however, the existence of an ion tail without the presence of dust is questionable if this object was almost a dormant comet. This is because the surface of the object would be covered with a dust mantle, and the drag force of sublimating gas from beneath the mantle would naturally carry out dust particles from the object. In addition, the spectrum indicated an asteroidal rather than cometary (i.e., C-complex asteroids or carbonaceous chondrites, Tholen 1984; Kareta & Reddy 2023). Moreover, Bottke et al. (2002) suggested from their dynamical simulation that its origin is more likely from the outer main belt (65%) rather than the Jupiter family comet (4%). This implies that the object could be an active asteroid that originated from the main asteroid belt. If this is indeed an active asteroid, several dust ejection mechanisms, such as impacts or rotational breakups, found in several active asteroids, could also be possible triggers for the activity (Jewitt 2012).

In this study, we aim to determine whether WH is a dormant comet or an active asteroid and to constrain its activity mechanism. First, we conducted the first polarimetric study on the object and compared its surface scattering properties with those of comets and asteroids. Polarimetry is a unique tool for discriminating between asteroid-like and comet-like objects, as demonstrated by Geem et al. (2022a). Moreover, polarization degrees obtained at high phase angles (i.e., Sun-object-observer’s angle α > 40°) of the active asteroid (3200) Phaethon are significantly higher than those of a comet nucleus, further suggesting the usability of polarimetry at large phase angles (Ito et al. 2018; Kuroda et al. 2015). Thus, it is safe to say that polarimetry provides critical information for distinguishing between comets and asteroids. Next, we reanalyzed the photographic plates of WH in 1949 to inspect the color of its tail. Besides, We conducted a dust ejection simulation to determine physical quantities related to its activity, following the method in Ishiguro et al. (2007). Considering all this information (i.e., nuclear polarimetry, dust color, and dust ejection mechanism), we provide a comprehensive discussion about a possible mechanism for the 1949’s activation.

2 Methods

2.1 Polarimetry

Table 1 provides a summary of our polarimetric observations of WH. We conducted polarimetric observations from May 25, 2022 to January 25, 2023 using three telescopes in Japan: the 1.6-m Pirka telescope at the Nayoro Observatory (NO) of the Faculty of Science, Hokkaido University (Minor Planet Center observatory code Q33), the 2-m Nayuta telescope at the Nishi-Harima Astronomical Observatory (NHAO), operated by the University of Hyogo1, and the 1.5-m Kanata telescope at the Higashi-Hiroshima Observatory (HHO)2. The observations covered a wide phase angle range, from 2.7° to 80.4°. At NO, we utilized the polarimetry mode of the Multi-Spectral Imager (MSI), which has a pixel scale of 0.39″ pixel−1 (Watanabe et al. 2012). Meanwhile, at the NHAO, we employed the polarimetry mode of the Wide Field Grism Spectrograph 2 (WFGS2), which has a pixel scale of 0.198″ pixel−1 (Uehara et al. 2004; Kawakami et al. 2021), and at the HHO, we used the Hiroshima Optical and Near-InfraRed camera (HONIR), which has a pixel scale of 0.29″ pixel−1 (Akitaya et al. 2014). We used a standard RC-band filter and set an exposure time range of 60 to 180 seconds to obtain polarimetric accuracy of less than 1%p in non-sidereal tracking mode. The observations were conducted for 16 nights at NO, 2 nights at NHAO, and one night at HHO.

We conducted our data analysis using the same methods as those described in detail in Ishiguro et al. (2022) for MSI data and in Geem et al. (2022b) for WFGS2 and HONIR data. The analysis process can be outlined as follows:

  1. preprocessing of the data,

  2. masking of the field stars near the target and bad pixels,

  3. aperture photometry to derive the source flux of ordinary and extraordinary components,

  4. derivation of the Stokes parameters (Q/I, and U/I),

  5. correction of the polarization efficiency, instrumental polarization, and position angle offset, and

  6. derivation of the polarization degree (P), the position angle (θP), the polarization degree with respect to the scattering plane normal (Pr), and the position angle with respect to the scattering plane normal (θr).

2.2 Photometry

We obtained two digitized photographic plate images of WH in 1949 from the Mikulski Archive for Space Telescopes (MAST) Digitized Sky Survey (DSS)3. One was taken with a photographic plate with an emulsion 103aO, sensitive to blue wavelengths (hereafter, blue plate), and the other was with an emulsion 103aE, sensitive to red wavelengths (red plate). These plates were taken on November 19, 1949, during the first Palomar Observatory Sky Survey (POSS). The coordinates of the image centers are taken from Table 1 of Fernández et al. (1997), and the size of each image is 30′ × 30′. The pixel sizes of those images are 15.0 μm (1.0″) and 25.284 μm (1.67″) for blue and red plates, respectively.

Files downloaded from the MAST contain the photographic emulsion density multiplied by a constant factor (6553.4) to adjust the 16-bit representation (Laidler et al. 1994). We linearized these emulsion densities to intensities following the technique in Cutri et al. (1993). Detailed procedures are given in Appendix A. Then, we subtracted the background using Source-Extractor (Bertin & Arnouts 1996). We set the mesh size for the background estimation to be 512 pixels, large enough not to be significantly affected by the diffuse tail of WH. Moreover, we estimated the error of these intensities using the method described in Appendix B. The error estimated in Appendix B does not include the inherent root mean square (RMS) error of diffuse source photometry from the POSS plate. Cutri et al. (1993) suggested the amount of the RMS error to be 0.1–0.3. Therefore, we added the RMS error of 0.2 to our original results.

Because of the sidereal tracking, the nucleus appeared elongated in the images. We determined regions where the WH nucleus and tail were located from the images (see Fig. 1). We visually determined the rectangular regions and hexagonal regions that enclose the nucleus and the tails in each band for the photometry. The leftmost and rightmost sides of the tail area have relative position angles of 157° and 144° with respect to the horizontal axis of each image (i.e., the WH’s velocity vector projected on the sky plane). We obtained these angles from Fig. 9 of Fernández et al. (1997). The distance between the lower side of the nucleus area and the tail area corresponds to 42″ on both plates, although the horizontal lengths are different due to the different exposure times. As a result, the areas of the tail regions are 3834.8 arcsec2 and 8292.2 arcsec2 for blue and red plates, respectively.

We calculated the intensities and uncertainties of the nucleus and tail within these regions for blue and red plates. For the tail, which is an extended source, we derived the surface brightness by dividing the intensity by the photometric areas. Once the intensities and surface brightness were obtained, we estimated the mass of the dust tail (Mtail). It can be given as Mtail =4πN0ρ3(4+q)[amax4+qamin4+q]$\[M_{\text {tail }}=\frac{4 \pi N_0 \rho}{3(4+q)}\left[a_{\max }^{4+q}-a_{\min }^{4+q}\right] \text {, }\]$(1)

where ρ is the dust mass density. We assumed ρ = 1300 kg m−3 based on the measurement of Ryugu samples due to the similarity of the spectral type (i.e., C-complex). We supposed a simple power law size distribution of the dust particles in the radius range from amin to amax. q denotes the power index of the differential size distribution, and N0 is the reference number of dust particles with the radius a0 = 1 m. Assuming the scattering property (i.e., the albedo and the scattering phase functions) of dust particles are the same as that of the nucleus, we derived N0 as below: N0=RWH2(3+q)amax3+qamin3+q(Itail IWH),$\[N_0=\frac{R_{\mathrm{WH}}^2(3+q)}{a_{\max }^{3+q}-a_{\min }^{3+q}}\left(\frac{I_{\text {tail }}}{I_{\mathrm{WH}}}\right),\]$(2)

where IWH and Itail are intensities in the blue plate within the enclosed areas after sky subtractions. We cited the radius of WH nucleus (RWH) to be 2200 m (i.e., the best-fit value obtained in Bach et al. 2017).

Table 1

Summary of polarimetric observation.

2.3 Dust ejection simulation

Assuming that the tail is associated with scattered light by ejected dust, we analyzed the tail properties. We conducted a simple simulation of the dust ejection to estimate the ejection epoch and particle size. The model descriptions are given in, for example, Ishiguro et al. (2007).

First, we produced the synchrone–syndyne network, where the zero ejection speed was considered. The synchrone curves are the lines indicating the distribution of dust particles ejected at given epochs. On the other hand, the syndyne curves represent the location of dust particles with the same β value, a ratio between solar radiation pressure (Fr) and gravity (Fg). It is given as βFrFg=3 LQpr16πGMcρa=KQprρa,$\[\beta \equiv \frac{F_{\mathrm{r}}}{F_{\mathrm{g}}}=\frac{3 ~L_{\odot} Q_{\mathrm{pr}}}{16 \pi G M_{\odot} c \rho a}=\frac{K Q_{\mathrm{pr}}}{\rho a},\]$(3)

where L, G, M, and c are the solar luminosity, gravitational constant, solar mass, and light speed, respectively. Substituting these constant values, K is equal to 5.7 × 10−4 kg m−2. Qpr is the radiation pressure coefficient averaged over the solar spectrum. We considered large absorbing particles and assumed Qpr = 1, following Ishiguro et al. (2007). The synchrone–syndyne network was drawn on the XY-plane of the digitized blue photographic plate image taken on November 19, 1949. Furthermore, we conducted a three-dimensional analysis of the dust tail, allowing consideration of the terminal ejection velocity of the particles, described in Ishiguro et al. (2007). This model assumes a terminal ejection velocity (υej) of particles as vej=V0βu1(rh1 au)u2,$\[v_{\mathrm{ej}}=V_0 \beta^{u_1}\left(\frac{r_{\mathrm{h}}}{1 ~\mathrm{au}}\right)^{-u_2},\]$(4)

where V0 is the reference ejection velocity (m s−1) of the particles with β = 1 at the heliocentric distance of rh = 1 au. We set u1 and u2 as 0.5. With the given υej and β, we solved Kepler’s equation to determine the spatial distribution of dust particles. Moreover, we assumed the differential size distribution of dust particles in the size range between amin and amax and the power index q.

thumbnail Fig. 1

Digitized photographic plate images of WH in 1949 used for the analysis. (a) and (c) are the blue and red plate images taken with exposure times of 720 seconds and 2700 seconds, respectively. In these images, the nucleus and tails are stretched out due to the sidereal tracking mode of the telescope mount. The tail in the red plate image is not as clear as the one in the blue plate because of the lower sensitivity of the red plate. (b) and (d) are zoomed images within the white squares of images (a) and (c). These images are rotated to align the WH’s velocity vectors in a horizontal direction. The areas surrounded by red rectangles indicate the region for the nucleus photometry, while those surrounded by the lower left parts of the red rectangles and yellow lines indicate the region for the tail photometry.

3 Results

In this section, we report our polarimetric results in Sect. 3.1, photometric results from the 1949 observation images in Sect. 3.2, and the tail analysis in Sect. 3.3, as described below.

3.1 Polarimetric results

Throughout our observations, we did not detect any signatures of comet-like activity (i.e., a tail and a coma) in our polarimetric images in 2022–2023. Therefore, we characterize the polarimetric properties of the nuclear surface.

We summarized nightly-averaged linear polarization degrees in Fig. 2 and Table 2. Similar to other SSSBs, it indicates the negative branch at α ≲ 20° and the positive branch at α ≳ 20° (Ishiguro et al. 2022). Although we cannot determine the maximum polarization degree because of insufficient phase angle coverage, we find Pr(α) > 30% at α ~ 80°.

We compared the polarization phase curve (Pr(α), PPC) of WH with other asteroids (Fig. 3). At a glance, the PPC’s slope near the inversion angle is predominantly steeper than that of S-complex asteroids and consistent with those of C-complex asteroids. A careful look reveals that the WH’s PPC is more moderate than that of Ryugu and steeper than that of Phaethon around the low phase angles. This fact suggests that the WH’s geometric albedo value would be between these two asteroids. We also compared WH’s PPC with those of a comet nucleus (209P/LINEAR) and cometary dust (Fig. 4). Although there is only one comet whose nuclear PPC is available, we find that WH’s PPC indicates a higher polarization degree than 209P. Considering most SSSBs have polarization maxima around α > 90° while WH was observed at α≪90° but indicated the large polarization degree (Pr(80°) >30%), this object likely shows a polarization maximum larger than that of comet nuclei.

To determine the polarimetric parameters quantitatively, we employed PyMC3 (Salvatier et al. 2016) to fit the PPC using the empirical trigonometric function (TF) shown in Ishiguro et al. (2022), modified from Lumme & Muinonen (1993) as shown below: Pr(α)=h(sinαsinα0)c1(cosα2cosα02)c2sin(αα0),$\[P_r(\alpha)=h\left(\frac{\sin \alpha}{\sin \alpha_0}\right)^{c_1}\left(\frac{\cos \frac{\alpha}{2}}{\cos \frac{\alpha_0}{2}}\right)^{c_2} \sin \left(\alpha-\alpha_0\right),\]$(5)

where α0 is an inversion angle where Pr(α0) = 0% is satisfied, and h is a polarimetric slope at α = α0. c1 and c2 are constant values obtained from fitting. Meanwhile, We also fitted PPC using a modified linear-exponential function (LE) from Bach et al. (2024) modified from the original forms in Muinonen et al. (2002); Kaasalainen et al. (2003), Pr (α)=h(1eα0/k)α(1eα0/k)α01(1+α0/k)eα0/k,$\[P_r~(\alpha)=h \frac{\left(1-e^{-\alpha_0 / k}\right) \alpha-\left(1-e^{-\alpha_0 / k}\right) \alpha_0}{1-\left(1+\alpha_0 / k\right) e^{-\alpha_0 / k}},\]$(6)

such that αmin=kln{kα0(1eα0/k)},$\[\alpha_{\min }=-k \ln \left\{\frac{k}{\alpha_0}\left(1-e^{-\alpha_0 / k}\right)\right\},\]$(7)

where α0 and h are the same parameter to Eq. (5), k is a scaling constant, and αmin is a phase angle where minimum polarization degree (Pmin) locates. The best-fit parameters and their 1-sigma uncertainties derived from the fitting using various phase angle ranges are summarized in Table C.1. From the fitting using all observation data, we also derive the minimum polarization degree of Pmin=0.960.34+0.32%$\[P_{\min }=-0.96_{-0.34}^{+0.32} \%\]$ at the phase angle of αmin =10.72.8+2.2$\[\alpha_{\text {min }}={10.7^{\circ}}_{-2.8^{\circ}}^{+2.2^{\circ}}\]$ from TF and Pmin = 1.46 ± 0.23% and αmin =8.90.5+0.4$\[\alpha_{\text {min }}={8.9^{\circ}}_{-0.5^{\circ}}^{+0.4^{\circ}}\]$ from LE function.

The polarimetric slope, h, is known to have a strong correlation to the geometric albedo in the V band (pV) (Widorn 1967). The relation is given as log10(pV)=C1log10(h)+C2,$\[\log _{10}\left(p_V\right)=C_1 \log _{10}(h)+C_2,\]$(8)

where C1 and C2 are constant values: C1 = −1.111 ± 0.031 and C2 = −1.781 ± 0.025 in Cellino et al. (2015), and C1 = −1.016 ± 0.010 and C2 = −1.719 ± 0.012 in Lupishko (2018). We assumed this relation and constant parameters are applicable to RC-band data and derive the geometric albedo as shown in Table 3. We converted the RC-band geometric albedo (pRC$\[p_{R_{\mathrm{C}}}\]$) into the standard V-band geometric albedo (pV) using the color index: pV=pRC×100.4[(VRC)(VRC)WH],$\[p_V=p_{R_{\mathrm{C}}} \times 10^{0.4\left[\left(V-R_{\mathrm{C}}\right)_{\odot}-\left(V-R_{\mathrm{C}}\right)_{\mathrm{WH}}\right]},\]$(9)

where (V − RC) = 0.356 ± 0.003 (Ramírez et al. 2012) and (V − RC)WH = 0.378 ± 0.025 (converted from a r′ − i′ color using relations from Jester et al. 2005) are V − RC colors of the Sun and WH, respectively (Urakawa et al. 2011). The estimated albedo values using h values from the TF fitting are 0.094 ± 0.018 with coefficients in Cellino et al. (2015) and 0.093 ± 0.015 with those in Lupishko (2018), while those from the LE relation fitting are 0.076 ± 0.010 and 0.077 ± 0.008 with coefficients from Cellino et al. (2015) and Lupishko (2018), respectively. Therefore, it is safe to say that the possible range of pV of WH is 0.066–0.112. These results are also summarized in Table 3, together with results from data with different α ranges used for the fitting.

thumbnail Fig. 2

Polarization phase curve (PPC) of WH. Each symbol denotes observations from different observatories and instruments. A green dotted line shows a fitted line using a trigonometric function (TF, Eq. (5)), and a blue dashed line shows a fitted line using a linear-exponential function (LE, Eq. (6)).

Table 2

Summary of nightly-averaged linear polarization degrees.

thumbnail Fig. 3

Comparison in polarization phase curves (PPCs) of WH (red circles) with fitted curve using Eq. (5) and other asteroids: two B-type asteroids (3200) Phaethon (blue squares, Ito et al. 2018; Shinnaka et al. 2018; Devogèle et al. 2018), (155140) 2005 UD (cyan diamonds, Ishiguro et al. 2022), a C-type asteroid (162173) Ryugu (cyan upper triangles, Kuroda et al. 2015), two S-type asteroids (1566) Icarus (green crosses, Ishiguro et al. 2017), and (4179) Toutatis (lime pluses, Lupishko et al. 1995; Mukai et al. 1997), and a L-type asteroid (85989) 1999 JD6 (orange lower triangles, Kuroda et al. 2021).

thumbnail Fig. 4

Comparison in polarization phase curves (PPCs) of WH (red circles) with fitted curve using Eq. (5), a nucleus of Jupiter family comet 209P/Linear (black stars, Kuroda et al. 2015), dust particles in comae of Jupiter Family Comets (grey circles, Kiselev et al. 2005), 1P/Halley (cyan squares, Kikuchi et al. 1987; Chernova et al. 1993), and polarization degrees from dust continuum of 2P/Encke (green upper triangles, Jockers et al. 2005; Kwon et al. 2018).

Table 3

Summary of albedo derivation.

3.2 Photometric results

From the photometry, we find that the nuclear magnitudes in 1949 are 14.86 ± 0.20 in the B band and 14.05 ± 0.20 in the RC band. We compared our RC-band magnitude of the nucleus to the phase function given in Ishiguro et al. (2011a) using the heliocentric distance rh = 1.148 au, the geocentric distance Δ = 0.025 au, and the phase angle α = 40.1°, where WH was active in 1949. From the phase function, the expected RC-band reduced magnitude at the time of 1949 observation is 14.4 ± 0.1. We find that this RC-band reduced magnitude is ~0.4 magnitude brighter than the expected nuclear brightness based on the phase function of the WH’s nucleus (Ishiguro et al. 2011a). Considering the magnitude error (0.2), we conclude that WH’s nucleus was brightened when it showed the comet-like tail.

In addition, we find B − RC of the nucleus in 1949 was 0.80 ± 0.29, consistent with previous reports by Lowry & Weissman (2003) (0.81 ± 0.06) and Urakawa et al. (2011) (0.80 ± 0.04, with the magnitude converted from SDSS g′, r′ and i′ to Johnson-Cousins system using Jester et al. 2005) when it was inactive. We also derive the total signal within the tail region in the B band and the RC band to be (1.04 ± 0.09) × 10−15 W m−2 and (1.78 ± 0.09) × 10−15 W m−2, respectively. The corresponding surface brightnesses of the tail are 25.96 ± 0.22 in the blue plate and 25.44 ± 0.21 magnitudes arcseconds−2 in the red plate. We confirm that the B − RC = 0.52 ± 0.30 color of the tail is consistent with the nucleus within a 1-sigma confidence interval. Although the tail color is bluer than the Sun, the similarity in color between the tail and the nucleus may suggest that the tail consists of dust grains with a similar optical color to the nucleus, and was detectable via the scattered sunlight.

thumbnail Fig. 5

Synchrone (green dotted lines) and syndyne (violet solid lines) network drawn on the blue plate image on November 19, 1949. This image has a standard orientation, that is, north is up and east to the left. The numbers on the synchrone curves correspond to days of dust ejection before the observed day, while the numbers on the syndyne curves indicate the corresponding β values.

3.3 Results of dust tail analysis

As shown in Sect. 3.2, the color of the tail is similar to the Sun and the nucleus. We also find that the nuclear magnitude in the 1949 images was brighter than the expected magnitude of the nucleus. This evidence suggests that WH ejected dust particles around the time of the 1949 observation, and the associated dust tail was imaged.

Based on the idea above, we investigated the ejection epoch and size of the dust particles from the tail morphology (see Sect. 2.3 for model description). Fig. 5 compares the observed images with a synchrone–syndyne network. Although the synchrone–syndyne analysis is a simple model that assumes a zero initial velocity, this analysis derives a rough estimate of the particle size and the ejection time. The position angle of the observed tail is 105.5° ± 6.5°. The best-match synchrone curve corresponds to 25–32 days before the observed date (November 19, 1949). In other words, if the ejection velocity was small, the dust particles were likely ejected within 6 days on October 18–25, 1949. This period of activity is likely shorter than 7 days in the realistic case (i.e., non-zero ejection velocity). Moreover, the intersection of the synchrone and syndyne curves indicates β is ranging from ~3 × 10−4 to ~1 × 10−2, which correspond to a=44 μm −1.5 mm when we assume the Ryugu-like dust mass density (ρd=1300 kg m−3, Miyazaki et al. 2023). This maximum radius (a=1.5 mm) is a lower limit. It would be possible that larger dust particles might be present near the nucleus and enhance the nuclear magnitude.

Figure 6 shows the dust simulation results, where we take into account the dust ejection velocity. Since our synchrone analysis suggests that the ejection duration was short, for only 7 days or less, we considered an impulsive dust ejection. In this model, we assumed that dust ejection happened for only Δt = 1 day on October 29.5, 1949 (the middle day from our synchrone analysis) from the entire Sun-lit hemisphere. We also assumed a power index q = −3.5 for the differential size distribution of the dust particles. These parameters (Δt and q) cannot be determined from the observed images not only because of the faintness of the tail but also because of the lack of time-series images. In Fig. 6, four different terminal velocities (V0=10, 20, 30, and 40 m s−1) were tested. By visual inspection, the model image with V0 < 20 m s−1 matches the observation. The visible region of the tail has a width of ≈8″, which matches the model image with V0 ~10 m s−1. With Eq. (4) and V0 ~10 m s−1, our model suggests a terminal ejection velocity of 1 mm-sized dust particles (β = 4.4 × 10−4) would be 0.2 m s−1. This velocity is even slower than the escape velocity from WH (a bulk density of 1190 kg m3 and a diameter of ~4000 m are assumed, Watanabe et al. 2019; Bach et al. 2017).

Finally, we derived the dust mass. We obtained the flux ratio of the dust tail with respect to the nucleus as Itail IWH=0.139$\[\frac{I_{\text {tail }}}{I_{\mathrm{WH}}}=0.139\]$ from the blue plate. Substituting maximum (1.5 mm) and minimum (44 μm) dust sizes and assuming q = −3.5, we find the dust total mass in the tail was 9.27 × 105 kg using Eqs. (1)(2). The estimated dust mass is only 1.8 × 10−6% of WH nucleus mass (when we assume that the bulk density of WH is 1190 kg m3, similar to another C-complex asteroid (162173) Ryugu (Watanabe et al. 2019). We note that this estimated dust mass would be a lower limit since large particles residing near the nucleus were not included in the calculation.

thumbnail Fig. 6

Simulated images of WH with different V0. We assumed that dust ejection happened for one day impulsively around 29.5 days before the observation. The field-of-view of each image is identical to the one of Fig. 1b. The comet-like ejection velocity, V(a)=V0×β$\[V(a)=V_0 \times \sqrt{\beta}\]$, was used for this model. We assumed hemispherical dust ejection with fixed β range (3 × 10−4β ≤ 1 × 10−2) and q for the differential size distribution.

4 Discussion

We have investigated the WH’s origin and its mysterious activity in 1949 from multiple perspectives based on photometry, spectroscopy, and dust dynamical properties. As we mentioned in Sect. 1, the difference between comets and asteroids has become unclear nowadays because of recent findings of objects having the hybrid natures of comets and asteroids (Jewitt & Hsieh 2022). However, we adopt a conventional classification based on the origins in the following discussion. We thus refer to a comet as an object originating from the trans-Neptunian region (Levison & Duncan 1997), while an asteroid is an object in or from the main belt. In the following subsections, we discuss each feature and consider the possible mechanism for the 1949 activity.

4.1 The optical nature

First, we discuss the optical nature. We derived the colors of the nucleus B-RC = 0.80 ± 0.29 and the tail B − RC = 0.52 ± 0.30. Although these errors are not small enough to discuss the origins, it is safe to say that the color of the tail is consistent with the nucleus. From this consistency, we conclude that the tail is dust-scattered light. The 0.4 magnitude enhancement of the nucleus can be explained by large dust particles distributed around the Hill radius.

From previous research, it is known that the nucleus color is slightly bluer than that of the Sun (Chamberlin et al. 1996). In general, comets exhibit red spectral characteristics compared to the solar spectrum (Jewitt 2002). However, the blue color of the nucleus is not necessarily a condition to reject the possibility that WH is a cometary origin. In fact, some comets (e.g., 95P/Chiron) indicated blue properties (Luu 1993; Boehnhardt et al. 1999). More comprehensive spectroscopic research in the optical and near-infrared wavelengths suggested that the WH spectrum is in agreement with carbonaceous chondrite meteorites (Kareta & Reddy 2023), which originate from C-type asteroids. Accordingly, from the reflectance color and spectrum, WH likely originated from the main belt.

We also investigated polarization properties. No polarimetric information was reported before our work. Based on our data, we estimated the range of geometric albedo (pV) to be 0.066–0.112. The derived geometric albedo is marginally consistent with or slightly higher than the previous estimates: pV = 0.055 ± 0.012 (Ishiguro et al. 2011a), 0.040–0.055 (Bach et al. 2017), and 0.059 ± 0.011 (Licandro et al. 2009). This slight discrepancy may be due to a slight underestimation of the inherent errors in the models used in each study (including our research). Geem et al. (2022a) proposed a technique for discriminating between comet-like and asteroid-like objects by combining a polarimetric slope (h) and color index (V − RC). In Fig. 7, we applied the method in Geem et al. (2022a) to our target. Indeed, WH is consistent with C- and B-type asteroids. However, the albedo of comets and some dark asteroids (C-complex and D-types) are largely overlapped with each other (Lamy et al. 2004; Usui et al. 2011; Mainzer et al. 2014), implying the geometric albedo tightly related to the polarimetric slope (h in Geem et al. 2022a) cannot be a ‘conclusive’ determinant. Also, as noted above, some exceptional cometary samples have unusually blue color indices. In other words, although the method of Geem et al. (2022a) is a useful tool that can make rough discrimination of comets and asteroids, it is not necessarily possible to conclude the origin with only h and V − RC. Further careful consideration is needed, as given in Sect. 4.2.

Furthermore, we show that the WH’s large polarization degrees at large phase angles (α > 40°). These values are higher than those of the nucleus of a nearly dormant comet 209P/LINEAR. It is, however, true that the number of polarization observation samples of cometary nuclei at large phase angles is insufficient because the nuclei should be enclosed by thick dust comae at the large phase angles (i.e., the small heliocentric distances). In addition, Kwon et al. (2018) speculated that large dust might have accumulated on the old comet nuclei, and the effect is responsible for large polarization degrees at high phase angles. Indeed, WH has been in a stable orbit for the past 4000 years or more, maintaining a solar distance of ~1 au (Kareta & Reddy 2023). Therefore, it may be possible that the surface materials are thermally metamorphosed or sorted to extract only large dust particles.

Summarizing the discussion regarding the optical nature, WH has a fairly high probability of being of main-belt origin. This fact is also in accordance with dynamical studies (Bottke et al. 2002; Granvik et al. 2018), although dynamical research has an intrinsic difficulty due to the chaotic orbit of near-Earth objects.

thumbnail Fig. 7

Polarimetric slope h and V − RC plot of WH in comparison with different types of asteroids and comet nuclei. The comparison data are taken from Geem et al. (2022a). Points with S, C, B, and D denote S-, C-, B-, and D-type asteroids.

4.2 The cause of the dust ejection in 1949

We subsequently discuss the essential question of why WH exhibited comet-like activity in 1949 based on the abovementioned fact that WH is likely the main belt origin. Fernández et al. (1997) pointed out that the color of the tail was remarkably blue. Our analysis shows that the color of the tail is consistent with the color of the nucleus within the error bar. Actually, our intensity estimate of the tail (1.20 × 10−15 W m−2 and 1.86 × 10−15 W m−2) is consistent with Fernández et al. (1997) within a factor of 2. The difference would be caused by background detection since the signal of the tail is very weak compared to the background. For reference, we recorded our method for photometric analysis in Appendix A. Since the details of the tail color analysis are not provided in Bowell et al. (1992) (where the authors describe the very blue color), it is difficult to compare our results to their report, and we thus proceed with the discussion based on our findings.

Because of the color similarity between the tail and the Sun, we assume that the tail was responsible for the scattered sunlight by the ejected dust particles. Under this assumption, we investigated the particle size, total mass, and epoch/duration of dust ejection. From synchrone–syndyne analysis, we found that the particle size is significantly larger than the wavelength, having the size parameter X=2πaλ100$\[X=\frac{2 \pi a}{\lambda} \gtrsim 100\]$. This large dust size is also consistent with the neutral tail color, which is similar to sunlight.

After the publication of Fernández et al. (1997), a new concept of active asteroids emerged. Asteroids eject their mass through lofting via electrostatic repulsion or thermal radiation pressure, thermal fatigue, impact, ice sublimation, and rotational disruption (Jewitt 2012; Bach & Ishiguro 2021; Molaro et al. 2020). Various recent observations widely support these ideas. Therefore, it is natural to assume that WH is also an active asteroid of main-belt origin, and it indicated an activity in 1949 for a reason.

First, we consider the possibility of dust lofting by either electrostatic repulsion or thermal radiation pressure. Jewitt (2012) estimated dust size that can be ejected from a 1 km asteroid around 1 au as 1.5–5 μm, suggesting that only ≲1 μm-sized dust can be lifted on the WH’s surface by electrostatic repulsion. In addition, Bach & Ishiguro (2021) examined dust lofting via thermal radiation pressure from the heated surface with various sizes at various solar distances. For the WH’s size asteroid at its perihelion, no dust particles can be lifted up according to their calculation. Therefore, the hypothesis of dust lofting via these mechanisms can be ruled out since the particle size from our estimation is larger by at least an order of magnitude.

Second, thermal fatigue can break surface boulders and trigger exfoliation to launch some fragments. This mechanism was suggested as one of the most probable causes of the (101955) Bennu’s activity (Molaro et al. 2020). However, WH was active only once in 1949, although there were > 15 observational opportunities (i.e., the perihelion passages) since 1949. Furthermore, it is difficult to explain WH activity in terms of the mass of the ejecta by thermal fatigue. We estimated that 9.27 × 105 kg of dust was ejected from the WH within about one day. This is equivalent to a dust ejection rate of 10.7 kg s−1. Since the mass ejection rate observed at Bennu is 10−7 kg s−1 (Hergenrother et al. 2020), WH ejected nearly 8 orders of magnitude more dust than Bennu. The short duration of the ejection is also quite different from the nature of dust ejections observed at Bennu. Therefore, thermal fatigue is a mechanism less likely to cause the WH’s 1949 activity.

Third, an impact may trigger the activity of an asteroid (see, e.g., Ishiguro et al. 2011b). We estimated the diameter of a hypothetical impactor, which can produce ejecta whose mass is equivalent to tail mass, using a model in Holsapple (2022). We find that a ~1m-sized object should collide with WH to generate the ejected mass. From interplanetary dust particle impact flux by Grün et al. (1985) and surface area estimated from a diameter by Bach et al. (2017), such impact would occur with a timescale of 108 years, longer than a typical lifetime of near-Earth asteroids. Although the impact probability is not zero (once per 108 years), the possibility of impact-triggered activity is considerably low.

Fourth, we consider volatile sublimation as observed in main-belt comets (MBCs, Hsieh & Jewitt 2006). The activities of MBC samples due to ice sublimation usually continued for several months (Hsieh et al. 2011a,b; Jewitt et al. 2014). The dust ejection velocity from MBCs is typically close to the escape velocity (Jewitt et al. 2014), which is similar to the WH’s 1949 event. Although we derived the duration of the dust ejection of < 7 days from our non-zero ejection velocity model, this estimate should be significantly overestimated because the zero ejection velocity is unrealistic. Adopting the non-zero ejection velocity model, the duration of the activity would be ~1 days. Considering such an impulsive dust ejection, the possibility of volatile sublimation for the 1949 activity is less likely.

The remaining possibility is that the WH tail was formed by rotational instability (see e.g., Jewitt et al. 2013). The low ejection velocity close to the escape velocity and the presence of large particles are consistent with the mechanism. We derived the total dust mass of 9.27 × 105 kg, which is only ~10−6% of the nuclear mass (~5 × 1013 kg). Such rotation-induced mass shedding has been observed in various asteroids (Jewitt 2009). In the case of mass shedding, because the centrifugal force accelerates the surface material, large particles can also be ejected with nearly an escape velocity. Therefore, rotation-induced mass shedding is the most likely scenario for dust ejection based on our tail analysis.

Our inference can be supported by the light curve analyses (Harris & Young 1983; Osip et al. 1995; Urakawa et al. 2011). The rotation period varies widely from 3.556 to 7.15 hours in the literature. Harris & Young (1983) reported a rotation period of 3.556 hours, while Osip et al. (1995) reported 6.1 ± 0.2. Urakawa et al. (2011) conducted a comprehensive analysis of the shape model, assuming the derived rotational period of 7.15 hours. However, Urakawa et al. (2011) also suggested the possibility of the solution for 3.58 hours. If the shorter solution is correct, WH still maintains a rapid rotation period. Such fast-rotating asteroids may create top shape, as observed for, for example, (66391) Moshup, (101955) Bennu, (162173) Ryugu (Ostro et al. 2006; Lauretta et al. 2019; Watanabe et al. 2019), and the shape reduces the rotational variability of the magnitude because of the axis-symmetrical shape.

In summary, WH is very likely an active asteroid of main belt origin. Although the possibility of volatile sublimation cannot be ruled out from our analysis, the possibility of rotational mass shedding is most likely a scenario for the 1949 activity based on our data and analyses, as well as the light curve properties derived by other research groups.

Acknowledgements

We deeply appreciate an anonymous reviewer and Dr. Emmanuel Lellouch for the valuable comments and suggestions. This research at SNU was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (No. 2023R1A2C1006180). The Pirka, Nayuta, and Kanata telescopes are operated by the Graduate School of Science, Hokkaido University; the Center for Astronomy, University of Hyogo; and the Hiroshima Astrophysical Science Center, Hiroshima University, respectively. These telescopes are partially supported by the Optical and Infrared Synergetic Telescopes for Education and Research (OISTER) program funded by the MEXT of Japan. SI acknowledges financial support from INAF – Call for fundamental research 2022, Minigrant RSN3.

Appendix A Linearization of photographic plate images

For the linearization of signals from photographic plates, we utilized an empirical relation from Cutri et al. (1993), originally introduced by de Vaucouleurs (1968), I=Aωn.$\[I=A \omega^n.\]$(A.1)

I is an intensity of incident light, ω is an opacitance of the emulsion on the photographic plate, and A and n are constants. The opacitance can be calculated from the density of the emulsion δ as below: ω=10δ1.$\[\omega=10^\delta-1.\]$(A.2)

Accordingly, the opacitance of the region with a zero emulsion density becomes 0. Then, the magnitude of a point source from aperture photometry (m) is given by m=2.5log10Isource dΩ$\[m=-2.5 \log _{10} \int I_{\text {source }} d \Omega\]$(A.3) =2.5log10(Iaper Isky )dΩ$\[=-2.5 \log _{10} \int\left(I_{\text {aper }}-I_{\text {sky }}\right) d \Omega\]$(A.4) =2.5log10A2.5log10(ωaper nωsky n),$\[=-2.5 \log _{10} A-2.5 \log _{10} \int\left(\omega_{\text {aper }}^n-\omega_{\text {sky }}^n\right),\]$(A.5)

where Isource is the intensity from the source, and Iaper, Isky, ωaper n$\[\omega_{\text {aper }}^n\]$ and ωsky n$\[\omega_{\text {sky }}^n\]$ are intensities and opacitances within the aperture and the sky background, respectively.

We first set n in Eq. A.1 to 1 as an initial guess and conducted aperture photometry on the image using an aperture with the size of ~2× full-width half maximum of the point spread function in the image. Then, we compared the magnitudes of stars from plates to those of equivalent stars at the Pan-STARRS DR1 catalog, whose location is consistent within 3″ difference. (Chambers et al. 2016). We converted Pan-STARRS g,r, and i-band magnitudes into Johnson B- and Cousins RC-band using the relation from Kostov & Bonev (2017) to derive catalog magnitudes (mcat). Cutri et al. (1993) showed the consistency between signals measured from Blue and Red plate images and those from photometry on CCD images with Johnson B- and Cousins RC-band, and their difference is up to 0.3 magnitudes. For the blue plate image, we used 79 stars whose magnitude ranges from 17 to 19 magnitudes. Meanwhile, we used 120 stars with a magnitude range of 16 − 18 magnitudes for the red plate. Then, we plotted the catalog magnitude (mcat) derived above on the x-axis and the plate magnitude (mplate) on the y-axis, defined as mplate =2.5log10(ωaper nωsky n).$\[m_{\text {plate }}=-2.5 \log _{10} \int\left(\omega_{\text {aper }}^n-\omega_{\text {sky }}^n\right).\]$(A.6)

We calculated the slopes a(n) and their intercepts b(n) of a linear regression line using the least square approximation. We determined the n value for each plate that makes the slope to be unity using the chi-square minimization technique. We calculated the reduced chi-square value using the formula below, χred2=i=1N(mcat,implate,ibσi)2,$\[\chi_{\mathrm{red}}^2=\sum_{i=1}^N\left(\frac{m_{\mathrm{cat}, i}-m_{\mathrm{plat}, i}-b}{\sigma_i}\right)^2,\]$(A.7)

where mcat,i and mplate,i are catalog and plate magnitudes of the i-th star within the image. 1-sigma confidence intervals of n is estimated by χred 2(n)<χred, min2(n)+1N2$\[\chi_{\text {red }}^2(n)<\chi_{\text {red, } \min }^2(n)+\sqrt{\frac{1}{N{-}2}}\]$.

We derived the A values using the formula in Cutri et al. (1993), A=10b2.5I0Δλeff,$\[A=10^{\frac{b}{-2.5}} I_0 \Delta \lambda_{\mathrm{eff}},\]$(A.8)

where I0 is zero magnitude intensity of a filter, and Δλeff is its effective width of the filter. We used zero magnitude intensities and effective widths of Johnson B and Cousins RC from the Spanish Virtual Observatory (SVO) Filter Profile Service4. 1-sigma confidence intervals of the A values are calculated from those of n. Finally, the linearization formulae derived for blue and red plate images are shown below: IBlue =(9.972±3.682)×1016×ω0.594±0.066,$\[I_{\text {Blue }}=(9.972 \pm 3.682) \times 10^{-16} \times \omega^{0.594 \pm 0.066},\]$(A.9) IRed =(7.564±1.183)×1015×ω0.951±0.029.$\[I_{\text {Red }}=(7.564 \pm 1.183) \times 10^{-15} \times \omega^{0.951 \pm 0.029}.\]$(A.10)

Appendix B Error estimation of photographic plate images

In this study, we considered three sources of uncertainty regarding digitized photographic plate images. The first uncertainty source is from the linearization coefficients (σA, σn) we derived in Sect. A. The second one is the inhomogeneous thickness of the emulsion, which causes background fluctuation in the image. The level of background fluctuation is estimated by background RMS value (σbg,RMS) calculated by Source-Extractor (Bertin & Arnouts 1996). The last one is the Poisson noise of incident photons, σPoisson =N$\[\sigma_{\text {Poisson }}=\sqrt{N}\]$, where N is the number of incident photons to the plate. We estimated N from intensities using the following equation N=Itexphνeff,$\[N=\frac{I t_{\mathrm{exp}}}{h \nu_{\mathrm{eff}}},\]$(B.1)

where texp, h, and veff are the exposure time, Planck constant, and the effective frequency of each band, respectively. From Eq. A.1, the Poisson noise of ω (σω,Poisson) is given as σω, Poisson =ωnN.$\[\sigma_{\omega, \text { Poisson }}=\frac{\omega}{n \sqrt{N}}.\]$(B.2)

With σA, σn, and σω,Poisson, we can derive the RMS values of ωn and I using the following equations σωn=ωn(nωσω,Poisson)2+(lnω σn)2,$\[\sigma_{\omega^n}=\omega^n \sqrt{\left(\frac{n}{\omega} \sigma_{\omega, \text {Poisson}}\right)^2+\left(\ln \omega ~\sigma_n\right)^2},\]$(B.3) σI=I(σAA)2+(σωnωn)2.$\[\sigma_I=I \sqrt{\left(\frac{\sigma_A}{A}\right)^2+\left(\frac{\sigma_{\omega^n}}{\omega^n}\right)^2}.\]$(B.4)

Finally, an uncertainty of a background-subtracted signal (Isig = IIbg), σIsig $\[\sigma_{I_{\text {sig }}}\]$, is given as σIsig=σI2+σIbg2+σIbg, RMS2.$\[\sigma_{I_{\mathrm{sig}}}=\sqrt{\sigma_I^2+\sigma_{I_{\mathrm{bg}}}^2+\sigma_{I_{\mathrm{bg}, ~\mathrm{RMS}}}^2}.\]$(B.5)

Appendix C Result of fitting for PPCs

The results from the fitting of PPCs are summarized in Table C.1.

Table C.1

Summary of best-fit parameters and their 1-sigma uncertainties.

References

  1. Akitaya, H., Moritani, Y., Ui, T., et al. 2014, SPIE Conf. Ser., 9147, 91474O [NASA ADS] [Google Scholar]
  2. Bach, Y. P., & Ishiguro, M. 2021, A&A, 654, A113 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  3. Bach, Y. P., Ishiguro, M., & Usui, F. 2017, AJ, 154, 202 [NASA ADS] [CrossRef] [Google Scholar]
  4. Bach, Y. P., Ishiguro, M., Takahashi, J., et al. 2024, A&A, 684, A81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Bertin, E., & Arnouts, S. 1996, A&AS, 117, 393 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Boehnhardt, H., Rainer, N., Birkle, K., & Schwehm, G. 1999, A&A, 341, 912 [NASA ADS] [Google Scholar]
  7. Bottke, W. F., Morbidelli, A., Jedicke, R., et al. 2002, Icarus, 156, 399 [NASA ADS] [CrossRef] [Google Scholar]
  8. Bowell, E., West, R. M., Heyer, H. H., et al. 1992, IAU Circ., 5585, 1 [NASA ADS] [Google Scholar]
  9. Cellino, A., Bagnulo, S., Gil-Hutton, R., et al. 2015, MNRAS, 451, 3473 [NASA ADS] [CrossRef] [Google Scholar]
  10. Chamberlin, A. B., McFadden, L.-A., Schulz, R., Schleicher, D. G., & Bus, S. J. 1996, Icarus, 119, 173 [NASA ADS] [CrossRef] [Google Scholar]
  11. Chambers, K. C., Magnier, E. A., Metcalfe, N., et al. 2016, arXiv e-prints [arXiv:1612.05560] [Google Scholar]
  12. Chernova, G. P., Kiselev, N. N., & Jockers, K. 1993, Icarus, 103, 144 [Google Scholar]
  13. Coradini, A., Capaccioni, F., Capria, M. T., et al. 1997, Icarus, 129, 317 [Google Scholar]
  14. Cunningham, L. E. 1950, IAU Circ., 1250, 3 [NASA ADS] [Google Scholar]
  15. Cutri, R. M., Low, F. J., & Guhathakurta, P. 1993, PASP, 105, 106 [NASA ADS] [CrossRef] [Google Scholar]
  16. de Vaucouleurs, G. 1968, Appl. Opt., 7, 1513 [NASA ADS] [CrossRef] [Google Scholar]
  17. Devogèle, M., Cellino, A., Borisov, G., et al. 2018, MNRAS, 479, 3498 [Google Scholar]
  18. Fernández, Y. R., McFadden, L. A., Lisse, C. M., Helin, E. F., & Chamberlin, A. B. 1997, Icarus, 128, 114 [CrossRef] [Google Scholar]
  19. Geem, J., Ishiguro, M., Bach, Y. P., et al. 2022a, A&A, 658, A158 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Geem, J., Ishiguro, M., Takahashi, J., et al. 2022b, MNRAS, 516, L53 [NASA ADS] [CrossRef] [Google Scholar]
  21. Granvik, M., Morbidelli, A., Jedicke, R., et al. 2018, Icarus, 312, 181 [CrossRef] [Google Scholar]
  22. Grün, E., Zook, H. A., Fechtig, H., & Giese, R. H. 1985, Icarus, 62, 244 [NASA ADS] [CrossRef] [Google Scholar]
  23. Harris, A. W., & Young, J. W. 1983, Icarus, 54, 59 [NASA ADS] [CrossRef] [Google Scholar]
  24. Hergenrother, C. W., Maleszewski, C., Li, J. Y., et al. 2020, J. Geophys. Res. (Planets), 125, e06381 [Google Scholar]
  25. Holsapple, K. A. 2022, arXiv e-prints [arXiv:2203.07476] [Google Scholar]
  26. Hsieh, H. H., & Jewitt, D. 2006, Science, 312, 561 [NASA ADS] [CrossRef] [Google Scholar]
  27. Hsieh, H. H., Ishiguro, M., Lacerda, P., & Jewitt, D. 2011a, AJ, 142, 29 [Google Scholar]
  28. Hsieh, H. H., Meech, K. J., & Pittichová, J. 2011b, ApJ, 736, L18 [NASA ADS] [CrossRef] [Google Scholar]
  29. Ishiguro, M., Sarugaku, Y., Ueno, M., et al. 2007, Icarus, 189, 169 [NASA ADS] [CrossRef] [Google Scholar]
  30. Ishiguro, M., Ham, J.-B., Tholen, D. J., et al. 2011a, ApJ, 726, 101 [NASA ADS] [CrossRef] [Google Scholar]
  31. Ishiguro, M., Hanayama, H., Hasegawa, S., et al. 2011b, ApJ, 740, L11 [NASA ADS] [CrossRef] [Google Scholar]
  32. Ishiguro, M., Kuroda, D., Watanabe, M., et al. 2017, AJ, 154, 180 [Google Scholar]
  33. Ishiguro, M., Bach, Y. P., Geem, J., et al. 2022, MNRAS, 509, 4128 [Google Scholar]
  34. Ito, T., Ishiguro, M., Arai, T., et al. 2018, Nat. Commun., 9, 2486 [Google Scholar]
  35. Jester, S., Schneider, D. P., Richards, G. T., et al. 2005, AJ, 130, 873 [Google Scholar]
  36. Jewitt, D. C. 2002, AJ, 123, 1039 [NASA ADS] [CrossRef] [Google Scholar]
  37. Jewitt, D. 2009, AJ, 137, 4296 [Google Scholar]
  38. Jewitt, D. 2012, AJ, 143, 66 [CrossRef] [Google Scholar]
  39. Jewitt, D., & Hsieh, H. H. 2022, arXiv e-prints [arXiv:2203.01397] [Google Scholar]
  40. Jewitt, D., Agarwal, J., Weaver, H., Mutchler, M., & Larson, S. 2013, ApJ, 778, L21 [NASA ADS] [CrossRef] [Google Scholar]
  41. Jewitt, D., Ishiguro, M., Weaver, H., et al. 2014, AJ, 147, 117 [NASA ADS] [CrossRef] [Google Scholar]
  42. Jockers, K., Kiselev, N., Bonev, T., et al. 2005, A&A, 441, 773 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  43. Kaasalainen, S., Piironen, J., Kaasalainen, M., et al. 2003, Icarus, 161, 34 [Google Scholar]
  44. Kareta, T., & Reddy, V. 2023, psj, 4, 174 [NASA ADS] [Google Scholar]
  45. Kawakami, A., Itoh, Y., Takahashi, J., Tozuka, M., & Takayama, M. 2021, Stars Galaxies, 4, 5 [Google Scholar]
  46. Kikuchi, S., Mikami, Y., Mukai, T., Mukai, S., & Hough, J. H. 1987, A&A, 187, 689 [NASA ADS] [Google Scholar]
  47. Kiselev, N., Rosenbush, V., Jockers, K., Velichko, S., & Kikuchi, S. 2005, Earth Moon Planets, 97, 365 [NASA ADS] [CrossRef] [Google Scholar]
  48. Kostov, A., & Bonev, T. 2017, Transformation of Pan-STARRS1 gri to Stetson BVRI magnitudes. Photometry of small bodies observations [Google Scholar]
  49. Kuroda, D., Ishiguro, M., Watanabe, M., et al. 2015, ApJ, 814, 156 [Google Scholar]
  50. Kuroda, D., Ishiguro, M., Naito, H., et al. 2021, A&A, 646, A51 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  51. Kwon, Y. G., Ishiguro, M., Shinnaka, Y., et al. 2018, A&A, 620, A161 [EDP Sciences] [Google Scholar]
  52. Laidler, V. G., Rehner, D. M., & Sturch, C. R. 1994, The Digitized Sky Survey, accessed on May 9, 2024 [Google Scholar]
  53. Lamy, P. L., Toth, I., Fernandez, Y. R., & Weaver, H. A. 2004, in Comets II, eds. M. C. Festou, H. U. Keller, & H. A. Weaver (University of Arizona Press), 223 [Google Scholar]
  54. Lauretta, D. S., Dellagiustina, D. N., Bennett, C. A., et al. 2019, Nature, 568, 55 [NASA ADS] [CrossRef] [Google Scholar]
  55. Levison, H. F., & Duncan, M. J. 1997, Icarus, 127, 13 [Google Scholar]
  56. Licandro, J., Campins, H., Kelley, M., et al. 2009, A&A, 507, 1667 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  57. Lowry, S. C., & Weissman, P. R. 2003, Icarus, 164, 492 [NASA ADS] [CrossRef] [Google Scholar]
  58. Lumme, K., & Muinonen, K. O. 1993, in LPI Contributions, 810, Asteroids, Comets, Meteors 1993, edn. LPI Editorial Board, 194 [Google Scholar]
  59. Lupishko, D. F. 2018, Solar Syst. Res., 52, 98 [NASA ADS] [CrossRef] [Google Scholar]
  60. Lupishko, D. F., & Lupishko, T. A. 2001, Solar Syst. Res., 35, 227 [NASA ADS] [CrossRef] [Google Scholar]
  61. Lupishko, D. F., Vasilyev, S. V., Efimov, J. S., & Shakhovskoj, N. M. 1995, Icarus, 113, 200 [NASA ADS] [CrossRef] [Google Scholar]
  62. Luu, J. X. 1993, Icarus, 104, 138 [NASA ADS] [CrossRef] [Google Scholar]
  63. Mainzer, A., Bauer, J., Cutri, R. M., et al. 2014, ApJ, 792, 30 [Google Scholar]
  64. Miyazaki, A., Yada, T., Yogata, K., et al. 2023, Earth Planets Space, 75, 171 [NASA ADS] [CrossRef] [Google Scholar]
  65. Molaro, J. L., Hergenrother, C. W., Chesley, S. R., et al. 2020, J. Geophys. Res. (Planets), 125, e06325 [NASA ADS] [Google Scholar]
  66. Muinonen, K., Piironen, J., Kaasalainen, S., & Cellino, A. 2002, Mem. Soc. Astron. Italiana, 73, 716 [NASA ADS] [Google Scholar]
  67. Mukai, T., Iwata, T., Kikuchi, S., et al. 1997, Icarus, 127, 452 [Google Scholar]
  68. Osip, D., Campins, H., & Schleicher, D. G. 1995, Icarus, 114, 423 [NASA ADS] [CrossRef] [Google Scholar]
  69. Ostro, S. J., Margot, J.-L., Benner, L. A. M., et al. 2006, Science, 314, 1276 [NASA ADS] [CrossRef] [Google Scholar]
  70. Ramírez, I., Michel, R., Sefako, R., et al. 2012, ApJ, 752, 5 [CrossRef] [Google Scholar]
  71. Salvatier, J., Wiecki, T. V., & Fonnesbeck, C. 2016, PyMC3: Python probabilistic programming framework, Astrophysics Source Code Library [record ascl:1610.016] [Google Scholar]
  72. Shinnaka, Y., Kasuga, T., Furusho, R., et al. 2018, ApJ, 864, L33 [Google Scholar]
  73. Tholen, D. J. 1984, PhD thesis, University of Arizona [Google Scholar]
  74. Uehara, M., Nagashima, C., Sugitani, K., et al. 2004, SPIE Conf. Ser., 5492, 661 [NASA ADS] [Google Scholar]
  75. Urakawa, S., Okumura, S.-i., Nishiyama, K., et al. 2011, Icarus, 215, 17 [CrossRef] [Google Scholar]
  76. Usui, F., Kuroda, D., Müller, T. G., et al. 2011, PASJ, 63, 1117 [Google Scholar]
  77. Watanabe, M., Takahashi, Y., Sato, M., et al. 2012, SPIE Conf. Ser., 8446, 84462O [NASA ADS] [Google Scholar]
  78. Watanabe, S., Hirabayashi, M., Hirata, N., et al. 2019, Science, 364, 268 [NASA ADS] [Google Scholar]
  79. Widorn, T. 1967, Ann. Universitaets-Sternwarte Wien, Dritter Folge, 27, 109 [NASA ADS] [Google Scholar]

1

Latitude: 35 01 31 N, Longitude: 134 20 08 E, Elevation: 449 m.

2

Latitude: 34 22 39 N, Longitude: 132 46 36 E, Elevation: 503m.

All Tables

Table 1

Summary of polarimetric observation.

Table 2

Summary of nightly-averaged linear polarization degrees.

Table 3

Summary of albedo derivation.

Table C.1

Summary of best-fit parameters and their 1-sigma uncertainties.

All Figures

thumbnail Fig. 1

Digitized photographic plate images of WH in 1949 used for the analysis. (a) and (c) are the blue and red plate images taken with exposure times of 720 seconds and 2700 seconds, respectively. In these images, the nucleus and tails are stretched out due to the sidereal tracking mode of the telescope mount. The tail in the red plate image is not as clear as the one in the blue plate because of the lower sensitivity of the red plate. (b) and (d) are zoomed images within the white squares of images (a) and (c). These images are rotated to align the WH’s velocity vectors in a horizontal direction. The areas surrounded by red rectangles indicate the region for the nucleus photometry, while those surrounded by the lower left parts of the red rectangles and yellow lines indicate the region for the tail photometry.

In the text
thumbnail Fig. 2

Polarization phase curve (PPC) of WH. Each symbol denotes observations from different observatories and instruments. A green dotted line shows a fitted line using a trigonometric function (TF, Eq. (5)), and a blue dashed line shows a fitted line using a linear-exponential function (LE, Eq. (6)).

In the text
thumbnail Fig. 3

Comparison in polarization phase curves (PPCs) of WH (red circles) with fitted curve using Eq. (5) and other asteroids: two B-type asteroids (3200) Phaethon (blue squares, Ito et al. 2018; Shinnaka et al. 2018; Devogèle et al. 2018), (155140) 2005 UD (cyan diamonds, Ishiguro et al. 2022), a C-type asteroid (162173) Ryugu (cyan upper triangles, Kuroda et al. 2015), two S-type asteroids (1566) Icarus (green crosses, Ishiguro et al. 2017), and (4179) Toutatis (lime pluses, Lupishko et al. 1995; Mukai et al. 1997), and a L-type asteroid (85989) 1999 JD6 (orange lower triangles, Kuroda et al. 2021).

In the text
thumbnail Fig. 4

Comparison in polarization phase curves (PPCs) of WH (red circles) with fitted curve using Eq. (5), a nucleus of Jupiter family comet 209P/Linear (black stars, Kuroda et al. 2015), dust particles in comae of Jupiter Family Comets (grey circles, Kiselev et al. 2005), 1P/Halley (cyan squares, Kikuchi et al. 1987; Chernova et al. 1993), and polarization degrees from dust continuum of 2P/Encke (green upper triangles, Jockers et al. 2005; Kwon et al. 2018).

In the text
thumbnail Fig. 5

Synchrone (green dotted lines) and syndyne (violet solid lines) network drawn on the blue plate image on November 19, 1949. This image has a standard orientation, that is, north is up and east to the left. The numbers on the synchrone curves correspond to days of dust ejection before the observed day, while the numbers on the syndyne curves indicate the corresponding β values.

In the text
thumbnail Fig. 6

Simulated images of WH with different V0. We assumed that dust ejection happened for one day impulsively around 29.5 days before the observation. The field-of-view of each image is identical to the one of Fig. 1b. The comet-like ejection velocity, V(a)=V0×β$\[V(a)=V_0 \times \sqrt{\beta}\]$, was used for this model. We assumed hemispherical dust ejection with fixed β range (3 × 10−4β ≤ 1 × 10−2) and q for the differential size distribution.

In the text
thumbnail Fig. 7

Polarimetric slope h and V − RC plot of WH in comparison with different types of asteroids and comet nuclei. The comparison data are taken from Geem et al. (2022a). Points with S, C, B, and D denote S-, C-, B-, and D-type asteroids.

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.

OSZAR »