Abstract
On 26 September 2022, the Double Asteroid Redirection Test (DART) spacecraft struck Dimorphos, a satellite of the asteroid 65803 Didymos1. Because it is a binary system, it is possible to determine how much the orbit of the satellite changed, as part of a test of what is necessary to deflect an asteroid that might threaten Earth with an impact. In nominal cases, pre-impact predictions of the orbital period reduction ranged from roughly 8.8 to 17 min (refs. 2,3). Here we report optical observations of Dimorphos before, during and after the impact, from a network of citizen scientists’ telescopes across the world. We find a maximum brightening of 2.29 ± 0.14 mag on impact. Didymos fades back to its pre-impact brightness over the course of 23.7 ± 0.7 days. We estimate lower limits on the mass contained in the ejecta, which was 0.3–0.5% Dimorphos’s mass depending on the dust size. We also observe a reddening of the ejecta on impact.
Similar content being viewed by others
Main
Four Unistellar eVscopes captured observations of Didymos during the Double Asteroid Redirection Test (DART) impact into Dimorphos on the night of 26 September 2022. Of all telescopes that observed the DART impact, from the ground and space, the eVscopes were among the smallest with apertures of 112 mm. Three eVscopes were located on Réunion Island and one in Nairobi, Kenya. Figure 1 shows eVscope images taken before and after the impact as well as the ejecta produced by the impact. We observe a fast-moving ejecta plume moving eastwards on the plane of the sky and spreading over a timescale of minutes, as well as slower-moving ejecta that morphologically appear to have formed a coma.
Of the four eVscope data sets that included the moment of impact, three were suitable for photometric analysis because the observers recorded dark frames for image calibration. Thus, we conducted aperture photometry on these three data sets through a circular aperture radius of 13.6′′ or 750 km at the distance of Didymos The resulting apparent magnitudes measured over time are shown in Fig. 2.
Using data from the eVscopes located in L’Étang-Salé and Saint-Paul, Réunion, we calculate an impact time of UTC 23:15:02 ± 4 s on 26 September 2022, which agrees with the reported Earth-observed impact time of 23:15:02.183 UTC1, itself coming 38 s after the actual time of impact on the spacecraft due to light-travel time (private communication with J. Bellerose). Before the impact, we use the observations taken with the eVscope located in L’Étang-Salé, Réunion to measure an apparent visual magnitude, mv = 14.48 ± 0.11 and a minimum magnitude (maximum brightness) mv = 12.18 ± 0.03. As the fast-moving ejecta moved out of the photometric aperture, the magnitude increased to mv = 12.96 ± 0.04. At a geocentric distance of 0.076 AU, a heliocentric distance of 1.05 AU and a phase angle of 53.28° at the time of impact, these correspond to absolute visual magnitudes of Hv = 18.12 ± 0.11, 15.83 ± 0.03 and 16.61 ± 0.04, respectively.
We estimate the effective mass in the fast-moving ejecta by calculating the change in effective cross-sections corresponding to magnitudes Hv = 15.83 ± 0.03 and 16.61 ± 0.04, and assuming an average density and albedo for Dimorphos. This is further detailed in the Mass of the ejecta section of the Methods. We must also assume an average particle radius, so we consider several scenarios.
To begin with, the particles must be small as evidenced by the long tail that developed in the antisolar direction several days after impact. We approximate an antisolar tail length of roughly 7 × 103 km on the plane of the sky roughly 113.7 h after impact as shown in Fig. 1. Measurements from the 4.1-m Southern Astrophysical Research Telescope, at NSF NOIRLab’s Cerro Tololo Inter-American Observatory in Chile revealed the tail length to be more than 104 km on the plane of the sky 2 days after impact4. Our measurement is shorter because of the smaller collecting area of the eVscope and its lesser sensitivity to low surface brightness. We measure a 3σ limiting magnitude of 17.01 ± 0.03 in this image stack consisting of 1,205 4-s exposures.
We then refer to active asteroid (596) Scheila, whose activity was probably caused by an impact, and whose average ejected dust size is predicted to be small on the basis of the observed effect of solar radiation sweeping5. From this, average dust radii spanning \(\bar{a}\) of roughly 0.1–1.0 μm were estimated. We point out that larger dust sizes were also estimated for Scheila such as \(\bar{a}\) ≅ 100 μm based on an approximate modelled particle range of a ≅ 1–104 μm along a power-law distribution with an index q = 3.5 (refs. 6,7).
We first examine an average particle radius \(\bar{a}\) ≅ 1 μm, as particles much smaller than this become less efficient at scattering visible light, and particles much larger than this are expected to persist longer in the photometric aperture. This particle radius then results in a mass of mfe ≅ (7.0 ± 1.2) × 103 kg contained in the fast-moving ejecta plume, respectively. We measured the speed of this fast-moving ejecta on the plane of the sky through the photometry aperture and find this fast-moving ejecta has a speed of vfe ≅ 970 ± 50 m s−1. The resulting mass then corresponds to a kinetic energy of KEfe ≅ (3.3 ± 0.6) × 109 J. The relative kinetic energy of the DART spacecraft at the time of impact is \({{\rm{K}}{\rm{E}}}_{{\rm{D}}}\,\cong \,(1.094\,\pm \,0.001)\times {10}^{10}\) J (ref. 1). This indicates that the observed fast-moving ejecta plume carried away roughly 30 ± 6% of the kinetic energy delivered by the DART spacecraft. This is comparable to impact simulations, which have shown that kinetic energy can be transferred from the impactor to the ejecta on the order of tens of percent8,9,10. Average particle sizes an order of magnitude larger than this would exceed the kinetic energy introduced by the DART spacecraft, so we do not further consider larger particles sizes as making up a substantial portion of the observed fast-moving ejecta plume. The lower end of the dust size estimations based on solar radiation pressure, \(\bar{a}\) ≅ 0.1 μm, results in an estimated mass of mfe ≅ (7.0 ± 1.2) × 102 kg, corresponding to roughly 3.0 ± 0.6% of DART’s kinetic energy. These values probably represent underestimates on ejecta mass, given the assumed average dust radii, as only the particles with velocities high enough to escape Dimorphos (ν > ve ≅ 0.087 ± 0.01 m s−1) will contribute to the measurable increase in cross-section1,11,12.
To estimate the mass of ejecta contained within the coma, we measure Didymos’ fading rate. The absolute magnitudes of Didymos after impact are plotted in Fig. 3. An error-weighted linear fit indicates that it took roughly 23.7 ± 0.7 days for the dust to move out of the photometric aperture with velocity vce ≅ 0.37 ± 0.01 m s−1. We use an approximated impact relation (equation (6) in Methods)13 to relate this with the mass, density and speed of the DART impactor, and estimate the mass contained in the coma that resulted from the impact to be mce ≅ (1.3 ± 0.1) × 107 kg. Considering the change in effective cross-section before impact and after the fast-moving ejecta dissipated, equation (3) in the Methods section results in an average dust radius of \(\bar{a}\) ≅ 1.7 ± 0.3 mm.
We also estimate the particle radius in the observed coma by relating solar radiation pressure to the turn-around distance of the particles in the coma along the comet–Sun line, and the bulk velocity of the particles, vce (detailed in the Mass of the ejecta section of the Methods). It was found that particles reached distances 𝓁 of roughly 150–250 km in the solar direction before turning around14. We, then find average dust radii \(\bar{a}\) roughly 2.8 ± 0.3–3.8 ± 0.4 mm in the coma, corresponding to a mass range of mce of roughly (1.3–2.2) × 107 kg, in good agreement with our estimate above.
In all scenarios, we find that whereas the impact led to a significant increase in apparent magnitude, the overall mass loss in the observed fast-moving ejecta plum or slower-moving ejecta in the coma created by DART’s impact into Dimorphos was not totally disruptive.
We also measure the colour before, during and after DART’s impact into Dimorphos as seen in Fig. 4 to show a significant reddening at the time of impact. The colours appear to return to their original colour as the fast-moving ejecta plume dissipates. This initial reddening was also seen on 9P/Tempel 1 after the impact of NASA’s Deep Impact spacecraft and was determined to be caused by different sized particles having a range of velocities, causing the particle size distribution and the ejecta optical depth to change over time15.
If the reddened colour is interpreted as a proxy for material composition, we must consider this in the context of the colours of active and inactive small bodies in the Solar System. Typically, active bodies appear bluer in colour on average than their inactive counterparts, such as short-period comets versus Kuiper belt objects16. Some of these redder observed surface colours may be due to irradiation of organics17, which can efficiently mantle the surface of bodies such as the Kuiper belt objects18,19. Furthermore, in general, the highest velocity ejecta from an impact is sourced from the material closest to the site of that impact20. The fast-moving ejecta plume may be more representative of the surficial material of Dimorphos than the material in the slower-moving coma. Whereas much of the fine-grain surface regolith on Dimorphos has probably been depleted through processes such as electrostatic removal21, it is possible that the remaining small particles have experienced some irradiation. Spectra indicate the presence heterogeneity of surface materials on Didymos, with various concentrations of probably hypersthene with a grain radius a < 25 μm, and olivine with a grain radius a < 45 μm (ref. 22). Laboratory simulations of space weathering showed that irradiation can cause significant reddening of olivine’s reflectance spectrum23. We emphasize that these posed interpretations are a few of possibly many.
Methods
Observations
The measurements in this work are derived from 23,240 images taken by 17 different eVscopes in the Unistellar network of eVscopes. At the time of this study, the Unistellar fleet of eVscopes contained three variations: the eVscope 1, eVscope eQuinox and the eVscope 2. All eVscopes have 112-mm diameter primary mirrors. The eVscope 1 and eQuinox both use a Sony IMX224LQR sensor and have a 37′ × 28′ field of view and pixel scale of 1.72′′ per pix, whereas the eVscope 2 uses a Sony IMX347LQR sensor and has a 45′ × 34′ field of view and pixel scale of 1.33′′ per pix. The detectors have RGB photosensors arranged as Bayer filter matrices. Exposure times of individual images varied between observers from 1.0 to 4.0 s but remained constant throughout each individual data set. Images were taken consecutively over ranges of 10–80 min. In this work, we report observations taken from 26 September to 31 October 2022 UT. Over this time range, the observing parameters of Didymos ranged from geocentric distances of 0.076 to 0.109 AU, heliocentric distances of 1.046 to 1.018 AU and phase angles of 53.2° to 73.8°. Figure 1 shows two of the tails that formed after impact as seen on 1 October 2022, when the position angle was 60.4° in the projected antisolar direction and 27.6° in the projected anti-velocity direction. In each data set, the limiting magnitude is between 15 and 16 mag, except for the stacked image that shows the tails in Fig. 1, which consists of 1,250 4-s images. For this data set, we measure a 3σ limiting magnitude of 17.01 ± 0.03. We further summarize the observation circumstances in Extended Data Table 1.
Aperture photometry
To measure the magnitude of Didymos in the images, we first reduce the data submitted by Unistellar citizen astronomers. Unistellar citizen astronomers submit science images with their chosen parameters (as listed above) as well as dark frames for calibration that are recorded immediately after the science observations. The dark frames are median combined and subtracted from the science images. The Unistellar network is not prompted at this point to take flat frames to further calibrate the data. In general, we find the errors on photometric measurements are dominated by photon noise and sky rather than the lack of a flat-field correction. We also aim to keep Didymos in the middle of the detector when conducting observations to further mitigate the lack of a flat frame calibration. Before conducting aperture photometry, the science images are aligned and median combined in time; however, the observations conducted during the DART impact shown in Fig. 2 were measured from images that were not stacked because of the speed of brightening on impact. The photometry measurements on the observations conducted after the impact were conducted on median-combined images that vary from 17 to 30 images. The measurements are then averaged and plotted at the midpoint in time of the conducted observation. We choose an aperture radius of 13.6′′ corresponding to 75 km at the distance of Didymos through which we measure the flux. The point spread function (PSF) of Didymos after the impact depends on the data set. Extended Data Table 1 lists the average full-width at half-maximum of the gaussian-fitted PSFs of the stars in each data set. The most egregious cases were due to out of focus eVscopes. We find the full-width at half-maximum of the PSF of Didymos is roughly 9.0 ± 1.0% larger than that of the stellar PSFs. Our chosen aperture is large enough to measure all the flux from the Didymos system throughout all the submitted data sets. We subtract the flux of the sky background measured through an annulus centred on the target that ranges from 22.21′′ to 30.75′′ in radius, corresponding to roughly 1,218 to 1,686 km at the distance of Didymos. We use stars in the field with known Gaia magnitudes to convert our instrumental flux values to apparent magnitudes, mV. These magnitudes are plotted in Fig. 2. When measuring the magnitude before impact and after the fast-moving ejecta dissipated, we use observations from the eVscope in L’Étang-Salé, Réunion on 26 September 2022. This eVscope was able to reach the greatest sensitivities as it was contained in a protective dome that blocked wind and stray light.
To obtain the colours of the Didymos binary system, the colour channels of the detectors must be isolated. In a Bayer filter matrix of blue (B), green (G) and red (R) photosensors, the first 2 × 2 section of pixels is arranged as [R, G; G, B]. This pattern repeats to fill the size of the detector. We isolate each colour channel on the basis of this known pattern of the photosensors and conduct aperture photometry as described above. We calibrate the instrumental fluxes using calibrator stars with known Gaia magnitudes in the background of each image. We measure the aperture fluxes of those calibrator stars in each colour channel and, combined with their known Gaia magnitudes, use them to convert the Didymos aperture fluxes in those channels to B, G and R magnitudes24. We plot the resulting B-G, G-R and B-R colours, the B-G and B-R colours in Fig. 4.
Mass of the ejecta
Apparent magnitudes mV are converted to absolute magnitudes, HV, which represents the magnitude of the object with heliocentric and geocentric distance (rH and ∆, respectively) of 1 AU and at phase angle, α = 0°. The correction is
where βα is the phase function that represents the dependence on sunlight scattering by the dust particle at an angle α in degrees. We assume a linear phase function with phase coefficient, β = 0.035, a typical value for S-type asteroids such as Didymos25,26,27,28,29. With this, our measured absolute magnitude before the impact of HV = 18.12 ± 0.11 agrees with past measurements of HV = 18.16 ± 0.04 of the binary system30. Absolute magnitudes are plotted in Fig. 3. Next, we estimate the effective cross-sections with
where pv ≅ 0.15 ± 0.02 is the albedo of the Didymos system measured by the DRACO camera1, HV is the absolute magnitude we measure for the Didymos system and V⊙ ≅ −26.77 is the apparent magnitude of the Sun. For Didymos, we find an effective scattering cross-section Ce ≅ 0.53 ± 0.06 km2 before impact, Ce ≅ 4.35 ± 0.13 km2 at Didymos’ peak in brightness just after impact and Ce ≅ 2.17 ± 0.09 km2 after the fast-moving ejecta moved out of the photometric aperture. A cross-section Ce ≅ 0.53 ± 0.06 km2 indicates an effective radius re ≅ 411 ± 22 m, which is consistent with previous radar measurements of Didymos finding a volume equivalent radius of re ≅ 390 ± 15 m (ref. 11). The mass is related to the effective cross-section by
where ρ ≅ 2,400 ± 250 kg m−3 is the bulk density of the Didymos system1, and we adopt \(\bar{a}=\sqrt{{a}_{\min }{a}_{\max }}\) for the mean dust radius among particles having a size range \({a}_{\min }\le a\le {a}_{\max }\). We examine particles with mean radii \(\bar{a}\) ≅ 0.1–1 μm in the fast-moving ejecta as was found for the ejecta in the impacted asteroid Scheila5. The change in effective cross-section measured at peak brightness and at the levelled brightness after the fast-moving ejecta dissipated then allows us to measure the change in mass, or the mass contained in the fast-moving ejecta plume with equation (3). We do the same for the mass lost in the coma of slower-moving particles by examining the change in cross-section from before impact and after impact, when the fast-moving ejecta dissipated from the photometric aperture. However, instead of assuming a dust size within the coma, we estimate the dust size in two ways. First, we estimate the mass in the coma on the basis of the fading time of Didymos after impact and the speed of the particles through a process explained in the following section. Equation (3) then gives an estimate on the average dust radius. We also estimate the dust radius by connecting the distance a particle can reach in the Solar direction before being turned around by Solar radiation pressure to its initial speed using
where B is a dimensionless radiation pressure factor, G = 6.67 × 10−11 m3 kg−1 s−2 is the gravitation constant, M is the mass of the Sun, rH ≅ 1.04 AU is the heliocentric distance of Dimorphos on 1 October 2022, when the turn-around distance 𝓁 was measured to reach roughly 150–200 km (ref. 14) and u is the initial velocity of the particles that we measure as u ≅ vce ≅ 0.37 ± 0.01 m s−1 from the fading rate of Didymos. B is the ratio of acceleration due to Solar radiation pressure to the acceleration due to Solar gravity expressed as
where K = 5.7 × 10-4 kg m−2 is a constant, Qpr is the radiation pressure coefficient often assume to be 1, ρ is the particle density and we examine the average dust a = \(\bar{a}\). Again, we examine dust densities equivalent to the bulk density of the Didymos system1. Assuming ρ ≅ 2,400 ± 250 kg m−3 gives B ≅ (0.24 ± 0.02)/\(\bar{a}\) μm. We point out that using a density ρ ≅ 3,480 ± 80 kg m−3, which is the average density of LL ordinary chondrite material31 as associated with S-type asteroid Didymos32, also results in millimetre-sized particles corresponding to masses within the range we estimate when assuming ρ ≅ 2,400 ± 250 kg m−3.
Speed and energy of the ejected dust
To estimate the energy carried by the mass in that fast-moving ejecta plume and slower-moving coma, we obtain their speeds. The fast-moving ejecta can be measured visually, following the plume on the detector over time. Furthermore, we measure the crossing time, tc, of the particles in the photometric aperture. This is the time between the moment of impact and the moment the magnitude settled to mv = 12.96 ± 0.04. We determine the peak time and dissipation time of the fast-moving ejecta by analysing the magnitudes binned into rolling bins of five images to determine peak time and 15 images for the settling time. We then choose times when the residuals were within the respective measured errors on the peak magnitude and magnitude after fast-moving ejecta dissipation. We find a crossing time tc ≅ 775 ± 40 s over a photometric aperture radius of 13.67′′, which is equivalent to roughly 750 km at the distance of Didymos. Therefore, we obtain a speed of v ≅ 970 ± 50 m s−1.
In the slower-moving ejecta that makes up the coma, we estimate the particle speed from the fading time of Didymos after impact. This is the time between the moment of impact and the moment the magnitude increased back to Didymos’ original absolute magnitude, HV = 18.12 ± 0.11. Then, assuming equal projectile and target densities, we make a simple estimation of the mass of the ejecta me by the impact relation
where UP ≅ 6,144.9 ± 0.3 is the impact speed of the DART spacecraft1, MP ≅ 579.4 ± 0.7 is the mass of the DART spacecraft on impact1, u ≅ vce ≅ 0.37 ± 0.01 m s−1 is the bulk velocity of the particles in the ejecta, and for consistency with past works A = 0.01 is a constant and s is an index that that we approximate to s = −1.5 but depends on the material7,13. The density of Dimorphos is assumed to be the same as the bulk density of the Didymos system in the calculations, however, we emphasize that the density of Dimorphos alone is not measured. The density of the main body of the DART spacecraft (without the solar panels) was roughly 270 kg m−3 at impact time33. For the sake of this simple approximation, we take the densities of the spacecraft and Dimorphos to be similar enough, because we expect lower densities with decreasing diameter for S-type asteroids34 and assume that the spacecraft was stopped by Dimorphos rather than flying straight through it. We also emphasize that this approximation comes with the caveat that impact physics on small asteroids are still not well-understood, so this impact relation serves as a rough estimate34.
With the estimated masses, m and speeds, v of the ejecta, we can estimate the kinetic energy, KE carried by the initial fast-moving ejecta as well as the coma by \({\rm{K}}{\rm{E}}=(1/2)m{v}^{2}\). As a tool to choose appropriate dust size test-cases, we compared the estimated kinetic energies to the kinetic energy introduced by the DART spacecraft where mass on impact is MP ≅ 579.4 ± 0.7 kg and velocity on impact is v = UP ≅ 6,144.9 ± 0.3 m s−1. We also compare this to the orbital energy, EO of Dimorphos around Didymos before and after the orbital period change of −33.0 ± 1.0 min (ref. 35) using
where \({M}_{{\rm{D}}{\rm{i}}{\rm{d}}}\,\cong \,(5.6\,\pm \,0.5)\times {10}^{11}\) kg, \({m}_{{\rm{D}}{\rm{i}}{\rm{m}}}\,\cong \,4.3\times {10}^{9}\) kg and r is the semimajor axis of the orbit of Dimorphos around Didymos1,36. Before the impact, the semimajor axis was measured to be r = 1.206 ± 0.035 km1. We can estimate the original orbital energy of Dimorphos to be EO ≅ −(6.6 ± 0.6) × 107 J. With the orbital period decrease, Kepler’s third law gives a new semimajor axis of r ≅ 1.2 ± 0.1 km resulting in a change of orbital energy of \(\Delta {E}_{{\rm{O}}}\cong \,(2.1\pm \,0.6)\times {10}^{6}\,{\rm{J}}\). These simple approximations are sufficient for the scope of this work, which aims to use these estimates as a means of providing a check on the reasonability of the estimated masses of the ejecta.
Data availability
The Unistellar network of citizen astronomers have the option to upload their FITS (Flexible Image Transport System) images to an Amazon Web Services server rented by the Unistellar Corporation. This data are then available upon request. The resulting photometry used in the data analyses are made available on the corresponding author’s public GitHub repository in the form of CSV files corresponding to the figures and the Extended Data Table. The repository also contains the FITS images used in Fig. 1. The data are located in the corresponding author’s public GitHub repository (https://github.com/Ariel-Graykowski/DART_Unistellar)37. Source data are provided with this paper.
Code availability
The SETI/Unistellar pipeline used to dark-subtract and stack the astronomical images and conduct aperture photometry is currently located on Unistellar’s private GitHub. These codes are available on request. Python codes used to create the figures and conduct the main data analysis are located in the corresponding author’s public GitHub repository (https://github.com/Ariel-Graykowski/DART_Unistellar)37. Source data are provided with this paper.
References
Daly, R. T. et al. Successful kinetic impact into an asteroid for planetary defence. Nature https://doi.org/10.1038/s41586-023-05810-5 (2023).
Fahnestock, E. G. et al. Pre-encounter predictions of DART impact ejecta behavior and observability. Planet. Sci. J. 3, 206 (2022).
Richardson, D. C. et al. Predictions for the dynamical states of the Didymos system before and after the planned DART impact. Planet. Sci. J. 3, 157 (2022).
Blue, C. SOAR telescope catches Dimorphos’s expanding comet-like tail after DART impact. NOIR Lab https:// noirlab.edu/public/news/noirlab2223/ (2022).
Jewitt, D. et al. Hubble Space Telescope Observations of main-belt comet (596) Scheila. Astrophys. J. 733, L4 (2011).
Bodewits, D. et al. Collisional excavation of asteroid (596) Scheila. Astrophys. J. Lett. 733, L3 (2011).
Jewitt, D. The active asteroids. Astron. J. 143, 66 (2012).
Braslau, D. Partitioning of energy in hypervelocity impact against loose sand targets. J. Geophys. Res. 75, 3987–3999 (1970).
Gault, D. E. The partition of energy for hypervelocity impact craters formed in rock. In Proc. Sixth Hypervelocity Impact Symposium Vol. 2, 419–456 (Defense Technical Information Center, 1963).
Okeefe, J. D., & Ahrens, T. J. Impact-induced energy partitioning, melting, and vaporization on terrestrial planets. In Proc. Lunar and Planetary Science Conference Vol. 8, 3357–3374 (LPI, 1977).
Naidu, S. P. et al. Radar observations and a physical model of binary near-Earth asteroid 65803 Didymos, target of the DART mission. Icarus 348, 113777 (2020).
Zubko, E. et al. Expected brightness of ejecta caused by the DART impact. Res. Notes AAS 6, 186 (2022).
Housen, K. R. & Holsapple, K. A. Ejecta from impact craters. Icarus 211, 856–875 (2011).
Li, J.-Y. Ejecta from the DART-produced active asteroid Dimorphos. Nature https://doi.org/10.1038/s41586-023-05811-4 (2023).
D. G. Schleicher, D. G. Photometry and imaging results for Comet 9P/Tempel 1 and deep impact: gas production rates, postimpact light curves, and the ejecta plume morphology. Astron. J. 131, 1130 (2006).
Jewitt, D. Color systematics of comets and related bodies. Astron. J. 150, 201 (2015).
Brunetto, R. et al. Modeling asteroid surfaces from observations and irradiation experiments: the case of 832 Karin. Icarus 184, 327–337 (2006).
Cooper, J. F. Proton irradiation of Centaur, Kuiper Belt, and Oort Cloud objects at plasma to cosmic ray energy. Earth Moon Planets 92, 261–277 (2003).
M. E. Brown, M. E. et al. A hypothesis for the color diversity of the Kuiper Belt. Astrophys. J. Lett. 739, L60 (2011).
Melosh, H. J. Impact Cratering: A Geologic Process (Oxford Univ. Press, 1989).
Hsu, H.-W. et al. Fine-grained regolith loss on sub-km asteroids. Nat. Astron. 6, 1043–1050 (2022).
Ieva, S. et al. Spectral rotational characterization of the Didymos system prior to the DART impact. Planet. Sci. J. 3, 183 (2022).
Yamada, M. et al. Simulation of space weathering of planet-forming materials: nanosecond pulse laser irradiation and proton implantation on olivine and pyroxene samples. Earth Planets Space 51, 1255–1265 (1999).
Cardiel, N. et al. RGB photometric calibration of 15 million Gaia stars. Mon. Not. R. Astron. Soc. 507, 318–329 (2021).
Veverka, J. The physical meaning of phase coefficients. Int. Astron. Union Coll. 12, 79–90 (1971).
Grehels, T. & Tedesco, E. F. Minor planets and related objects. XXVIII-Asteroid magnitudes and phase relations. Astron. J. 84, 1079–1087 (1979).
Schober, H. J. Rotation period of 234 Barbara, a further slowly spinning asteroid. Astron. Astrophys. 96, 302–305 (1981).
De Leon, J. et al. Observations, compositional, and physical characterization of near-Earth and Mars-crosser asteroids from a spectroscopic survey. Astron. Astrophys. 517, A23 (2010).
Cheng, A. F. et al. AIDA DART asteroid deflection test: planetary defense and science objectives. Planet. Space Sci. 157, 104–115 (2018).
Pravec, P. et al. Photometric observations of the binary near-earth asteroid (65803) Didymos in 2015–2021 prior to DART impact. Planet. Sci. J. 3, 175 (2022).
Britt, D. T. & Consolmagno, G. J. S. J. Stony meteorite porosities and densities: a review of the data through 2001. Meteorit. Planet. Sci. 38, 1161–1180 (2003).
Dunn, T. L. et al. Mineralogies and source regions of near-Earth asteroids. Icarus 222, 273–282 (2013).
Rainey, E. S. G. et al. Impact modeling for the Double Asteroid Redirection Test (DART) mission. Int. J. Impact Eng. 142, 103528 (2020).
Carry, B. Density of asteroids. Planet. Space Sci. 73, 98–118 (2012).
Cheng, A. F. et al. Momentum transfer from the DART mission kinetic impact on asteroid Dimorphos. Nature https://doi.org/10.1038/s41586-023-05878-z (2023).
Fang, J. & Margot, J.-L. Near-Earth binaries and triples: origin and evolution of spin-orbital properties. Astron. J. 143, 24 (2011).
Graykowski, A. Ariel-Graykowski/DART_Unistellar: DART unistellar (v1.0.0). Zenodo https://doi.org/10.5281/zenodo.7613581 (2023).
Acknowledgements
This work was supported by a generous donation from the Gordon and Betty Moore Foundation. We thank J. Bellerose for their communication on the observed timing of the DART impact. Citizen astronomers, who are listed as co-authors, voluntarily observed this event and uploaded their data to an Amazon Web Services server rented by the Unistellar Corporation. The analyses presented in this work were conducted locally at the SETI (search for extraterrestrial intelligence) Institute, while the observations used in this work were conducted by citizen astronomers globally. This global network of eVscopes was a necessary tool to coordinate the proper timing and location to observe the DART impact and to continuously monitor the Didymos system over time.
Author information
Authors and Affiliations
Contributions
A.G., R.A.L. and F.M. initiated this work. F.M. conceptualized and organized the contribution of the Unistellar network for the observations of the DART impact. A.G. and R.A.L. conducted photometric measurements on the images. A.G. led the analysis of the photometric data sets. R.A.L. and F.M. made a significant contribution to the photometric data. R.A.L., F.M., D.C., P.A.D., T.M.E., D.O’C.P. and L.A.S. contributed to the interpretation, discussion and the writing. F.M., A.M., L.M. and A.B. contributed to the creation of the Unistellar network. G.B. contributed to the establishment of the photometric pipeline. T.M.P. contributed to discussion of impact physics. P.Huet, M.L., B.P., C.C., S.M., D.C.O. and R.W. contributed observations of Didymos taken before, during and after the impact of DART. M.M., K.F., T.G., B.G., P.Huth, S.I., R.K., M.P., J.R., D.A.R., M.R., M.S., J-P.T., S.W., and W-C.Y. contributed observations of Didymos in the weeks following the impact in this work. M.C., K.G., R.J., D.J., S.K., M.L., J.W.P., B.A.S. and I.M.T. contributed observations of Didymos before the DART impact, which were used to prepare observation parameters for impact and prepare the analyses that were used in this paper.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature thanks Marcel Popescu and Derek Richardson for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Graykowski, A., Lambert, R.A., Marchis, F. et al. Light curves and colours of the ejecta from Dimorphos after the DART impact. Nature 616, 461–464 (2023). https://doi.org/10.1038/s41586-023-05852-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-023-05852-9
This article is cited by
-
The First Lucy Earth Flyby (EGA1)
Space Science Reviews (2024)
-
The Dimorphos ejecta plume properties revealed by LICIACube
Nature (2024)
-
Physical properties of asteroid Dimorphos as derived from the DART impact
Nature Astronomy (2024)
-
The impact of citizen scientist observations
Nature Astronomy (2023)
-
DART’s data verify its smashing success at deflecting asteroid moon Dimorphos
Nature (2023)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.