Content uploaded by José M. Madiedo
Author content
All content in this area was uploaded by José M. Madiedo on Aug 14, 2018
Content may be subject to copyright.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
FIRST DETERMINATION OF THE
TEMPERATURE OF A LUNAR IMPACT FLASH
AND ITS EVOLUTION
José M. Madiedo
1
, José L. Ortiz
2
, Nicolás Morales
2
1
Facultad de Ciencias Experimentales, Universidad de Huelva. 21071
Huelva (Spain).
2
Instituto de Astrofísica de Andalucía, CSIC, Apt. 3004, Camino Bajo de
Huetor 50, 18080 Granada, Spain.
ABSTRACT
We report the first analysis of a flash produced by the impact of a meteoroid
on the lunar surface and recorded both in the near-infrared and in the
visible. Despite the fact that similar data have been recently published by
other team during the refereeing process of our manuscript (Bonanos et al.
2018), our result still forms the first measurement of the temperature of a
telescopic lunar impact flash (Madiedo and Ortiz 2016, 2018). The flash
exhibited a peak magnitude of 5.1 ± 0.3 in the near-infrared I band and 7.3 ±
0.2 in the visible, and the total duration of the event in these bands was 0.20
s and 0.18 s, respectively. The origin of the meteoroid was investigated, and
we inferred that the most likely scenario is that the impactor that belonged
to the sporadic background. The analysis of this event has provided for the
first time an estimation of the emission efficiency in the near-infrared η
I
for
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
sporadic meteoroids impacting the Moon. We have determined that this
efficiency is around 56% higher than in the visible band and we have found
a maximum impact plume temperature of ~4000 K at the initial phase
followed by temperatures of around 3200 K after the peak brightness. The
size of the crater produced as a consequence of this impact is also
calculated.
KEYWORDS: Impact processes, impact flash, Moon, meteoroids, meteors
1. INTRODUCTION
Different researchers have studied the impacts of meteoroids on the lunar
surface by analyzing the flashes produced during these collisions (Ortiz et
al. 1999; Bellot Rubio et al. 2000a,b; Ortiz et al. 2000; Yanagisawa and
Kisaichi 2002; Cudnik et al. 2002; Ortiz et al. 2002; Yanagisawa et al. 2006;
Cooke et al. 2006; Ortiz et al. 2006; Swift et al. 2011; Madiedo et al. 2014a,
Suggs et al. 2014). In this way different parameters can be determined, such
as the energy of the impactor, its mass, and the size of the resulting crater.
From the frequency of these events, paramount information related to the
impact hazard for Earth can be also derived (Ortiz et al. 2006; Madiedo et
al. 2014a,b; Suggs et al. 2014). The lunar impact flash monitoring technique
has the advantage that a single detector covers a much larger area (typically
with an order of magnitude of 10
6
km
2
) than that monitored by ground-
based systems that analyze meteor and fireball activity in the Earth’s
atmosphere. However, the results derived from this method depend strongly
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
on the value adopted for the so-called luminous efficiency. This parameter
is the fraction of the kinetic energy of the impactor that is emitted as visible
light during the impact.
In the last decades, flashes produced by the collision of shower and sporadic
meteoroids have been identified in the framework of several monitoring
programmes by means of small telescopes and high-sensitivity CCD
cameras (Ortiz et al. 2000; Yanagisawa and Kisaichi 2002; Yanagisawa et
al. 2006; Cooke et al. 2006; Ortiz et al. 2006; Ortiz et al. 2006; Madiedo et
al. 2014a; Suggs et al. 2014; Madiedo et al. 2015a,b). However, the
detection of these impact flashes has so far been performed in the visible
range. Our team started in 2009 a lunar impact flashes monitoring program
named MIDAS (Madiedo et al. 2010; Madiedo et al. 2015a,b), which is the
continuation of the lunar impact flashes monitoring project started by the
second author in 1999 (Ortiz et al. 1999). Since 2015, in addition to the
observations performed in visible band, we are also conducting a monitoring
of the night side of the Moon in the near-infrared (NIR) by using a specific
NIR filter. In this way, we can analyze the behaviour of these impact flashes
in different spectral bands. By following this approach, MIDAS became the
first system that can determine the temperature of these impact flashes
(Madiedo and Ortiz 2016, 2018). The usefulness of performing observations
in the near-infrared was addressed in Cudnik (2010). In this work we focus
on a lunar impact flash identified by several of our telescopes on 2015
March 25. It was simultaneously recorded in both visible and NIR bands.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
The analysis of this event has provided an estimation of the emission
efficiency for flashes produced by sporadic sources in the NIR. Here we use
the term emission efficiency in the near Infrared (η
I
)
to distinguish it from
the luminous efficiency concept (η), which is applicable to the whole CCD
sensitivity wavelength range of 400nm to 900 nm as defined in the initial
papers on lunar impact flashes (Ortiz et al. 1999, Bellot et al. 2000a).
2. INSTRUMENTATION AND METHODS
The event analyzed here was recorded from our observatory in Sevilla
(latitude: 37.34611 ºN, longitude: 5.98055 ºW, height: 23 m above the sea
level), during a lunar monitoring campaign conducted on 2015 March 25.
Three f/10 Schmidt-Cassegrain telescopes manufactured by Celestron and
with diameters of 0.36, 0.28 and 0.24 m were employed. Each telescope
used a high-sensitivity CCD video camera (model 902H Ultimate,
manufactured by Watec Corporation, Japan). These devices generate an
interlaced black and white analogue video signal at a rate of 25 frames per
second. This signal is digitized and stored on the hard disk of a PC computer
under AVI video files. These video files have a resolution of 720x576
pixels, with each pixel being represented by 8-bits. Thus, when expressed in
device units, pixel values range between 0 and 255. The gamma setting of
the CCD video camera was adjusted to provide a linear response
(gamma=1). GPS time inserters are used to stamp date and time information
on every video frame with an accuracy of 0.01 seconds. f/3.3 focal reducers
manufactured by Meade are also used in order to increase the area
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
monitored by these devices. The telescopes are tracked at lunar rate, but
they are manually recentered when necessary, since perfect tracking of the
Moon at the required precision is not feasible with this equipment.
The observations conducted with the 0.36 m and the 0.28 m telescopes were
performed without any filter. These provided images in the wavelength
range between, approximately, 400 and 1000 nm. However, a NIR filter
(model Baader IR-pass) was employed for the camera attached to the 0.24 m
telescope. As a result, the images taken by this telescope corresponded to
wavelengths raging from 685 to 1000 nm.
Our dates of monitoring did not coincide with the activity period of any
major meteor shower. So, the telescopes were aimed to an arbitrary but
common area of the lunar disk. Of course, the terminator was avoided in
order to avoid an excess of light from the illuminated side of the Moon in
the telescopes. For the identification and analysis of impact flashes we have
employed the MIDAS software (Madiedo et al. 2010, 2015a).
3. OBSERVATIONS
The monitoring campaign on 2015 March 25 was conducted from 19h 15m
to 23h 45m UT, with a 5.9 day-old waxing crescent Moon and an
illuminated fraction of the lunar disk of about 34 % (Figure 1). The effective
monitoring time was 4.5 hours. As a result of this campaign one impact
flash was identified (Figure 2). The event took place at 21h 00m 16.80 ±
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
0.01 s UT and was simultaneously recorded by the 3 telescopes. It lasted
0.18 s according to the observations performed in visible band with the 0.28
m telescope. However, the images taken in the near IR with the 0.24 m
telescope revealed that the duration of the event was of around 0.20 s. To
calculate the duration of the flash we have employed the MIDAS software.
This software tool measured the average luminosity of the area on the lunar
surface enclosed by the event, and took into account only those video
frames where this luminosity was above the sum of the average luminosity
of the same area on the Moon immediately before the event took place and
the corresponding standard deviation of that luminosity. It is important to
notice that the diameter of the telescope employed to obtain images in the
NIR was smaller. This means that the ability of this device to collect light
(and hence its sensitivity to record dimmer events) is lower. So the larger
duration of the flash in the near-infrared can be explained on the basis of the
enhanced contrast that the NIR filter provides between the impact flash and
the Earthshine (Figures 2a and 2b).
The main parameters of the impact flash are listed in Table 1. The impactor
hit the lunar surface at the selenographic coordinates 11.3 ± 0.1 ºN, 21.6 ±
0.1 ºW, which corresponds to a position close to the northwest wall of crater
Copernicus.
4. RESULTS AND DISCUSSION
4.1. Flash brightness
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
To estimate the magnitude in our NIR band we have followed the procedure
described in Madiedo et al. (2014a). Thus, the brightness of the flash was
compared with the brightness of reference stars recorded during the same
observing session whose visual magnitude is known. This was done by
using the stars magnitudes in I-band, which is the closest band to our filter
in the Bessel Cousins photometric system. Eight calibration stars with
similar airmass and apparent brightness to that of the flash were employed.
Even though the transmission of our filter does not exactly match the shape
and width of the standard I filter, it is reasonably close, given that the
sensitivity of the CCD camera quickly decays in this wavelength range. The
response of the Johnson-Cousins I filter transmission and the Baader IR-
pass filter transmission convolved with the spectral response of the camera
CCD is shown in (Figure 3). From this plot we have obtained that the
effective wavelength for the Johnson-Cousins I filter and the Baader IR-pass
filter is similar (798.0 and 761.5 nm, respectively). As a result of this
procedure we have obtained that the magnitude of the flash in the near
infrared was m
I
= 5.1 ± 0.3. The V-band magnitude m
V
of the event was
also obtained by employing reference stars recorded during the same
observing session with known magnitude in V band. In previous works we
have never used colour terms in the flux to V-magnitude conversion
equations, mainly because we did not know the colour of the impact flashes.
But since in this case observations in the NIR are available, by using the
information from the rough V-I colour determined from our data we can do
that, at least approximately. Given that the sensitivity of the CCD camera is
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
not solely centered in the V band, and we did not use a V-band filter, NIR
radiation can enter the detector and so it is reasonable to expect that this
effect should be taken into account. In fact, we believe that a large fraction
of the flux in our unfiltered observations includes NIR flux. Thus color term
corrections seem necessary. The V-band magnitude was calculated by using
the standard relationship
m
V
=m
Vo
-2.5log(S)+K
I
(m
V
-m
I
)-KX (1)
where m
Vo
is the zero-point for the V filter, S the measured signal, K
I
the
colour term transformation coefficient, K the extinction coefficient, and X
the airmass. We employed calibration stars, with known m
V
and m
I
, to
obtain the values of K
I
, m
Vo
and KX. These stars were observed with our
telescopes to measure their corresponding signal S. We considered eight
calibration stars with the same airmass as the impact flash, so that in this
calibration we simultaneously determined m
Vo
and KX in a single constant
by performing a least-squares fit. The value of K
I
resulting from this
calibration is 0.18, and the peak magnitude of the flash in V band yields 7.3
± 0.2. We have assumed a typical error for K
I
of about 30 %. This error was
provided by the above-mentioned fit. It is important to point out that we
should have used a similar transformation equation for the I magnitude as
we did for the V magnitude in Equation (1). But unfortunately we could not
determine color correction terms for the I filter. For this reason, the
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
uncertainty for the flash magnitude in V band is lower than the uncertainty
in I band.
4.2. Meteoroid source
It is a well-known fact that it is not possible to unambiguously establish the
source of the impactors from the monitoring of lunar impact flashes.
However, it is possible to determine the most likely source of these
projectiles by calculating the probability that an impact flash is associated to
a given meteoroid source (Madiedo et al. 2015a,b). Following this approach,
the impact flash can be linked to the meteoroid source that provides the
highest value for this probability.
At the time of the impact flash detection no major meteor shower was active
on Earth. Thus, we have considered that the event could be produced either
by a sporadic meteoroid or by a particle belonging to a minor meteoroid
stream. From March 24-26 our meteor observing stations (Madiedo and
Trigo-Rodríguez 2008; Madiedo et al. 2013; Madiedo 2014) recorded
activity from the Virginids (VIR) and the γ-Normids (GNO), both with a
zenithal hourly ZHR of below 1 meteor h
-1
. The impact geometry of both
streams was compatible with that of the impact flash discussed here. So, we
have assumed that the impactor could be linked to one of these two
meteoroid streams or to the sporadic background. The corresponding
association probabilities, labelled by p
VIR
, p
GNO
and p
SPO
, respectively, were
calculated according to the methods described in Madiedo et al. (2015a,b).
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
Thus, by using Eq. (15) in Madiedo et al. (2015b) with ZHR = 1 meteors h
-1
for both streams and by assuming an hourly rate of 10 meteors h
-1
for
sporadics (Dubietis and Arlt 2010), p
VIR
and p
GNO
yield 0.04 and 0.08,
respectively. For the calculation of these probabilities we have considered a
population index r = 3.0 for the Virginids, r = 2.4 for the γ-Normids, and r =
3.0 for sporadics. An average value of 17 km s
-1
(Ortiz et al. 1999) has been
assumed for the impact velocity on the Moon of sporadic meteoroids. For
the Virginids and the γ-Normids we have taken these impact velocities as 30
and 56 km s
-1
, respectively. An impact angle of 45º with respect to the local
vertical was considered for sporadic events, while according to the impact
geometry this angle would be of about 40.2º and 46.2º for the Virginids and
the γ-Normids, respectively. These results show that the most likely scenario
is that the meteoroid was associated with the sporadic background, since the
largest probability parameter is obtained for this source, with p
SPO
= 1 - p
VIR
- p
GNO
= 0.88.
4.3. Emission efficiency in the near infrared and impactor mass
The lightcurve of the impact flash in visible and NIR bands is shown in
Figure 4. As can be noticed, and within the time resolution provided by our
cameras, the flash peaks at the same instant in both bands, but it remains
brighter in the near-infrared during the whole duration of the event.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
The radiated energy on the Moon corresponding to an impact flash with a
magnitude m can be obtained by integrating the radiated power P defined by
the following equation:
2)5.2/m(8
Rf10·10·75.3P λ∆π=
−−
(2)
where P is given in watts, 3.75·10
-8
is the flux density in V band of a
magnitude 0 star in Wm
-2
µm
-1
, ∆λ is the wavelength range, expressed in
µm, in which P is calculated, R is the Earth-Moon distance at the instant of
the meteoroid impact in m, and f is a dimensionless factor that measures the
degree of anisotropy of light emission. Thus, for those impacts where light
is isotropically emitted from the surface of the Moon f = 2, while f = 4 if
light is emitted from a very high altitude above the lunar surface. In this
analysis we have assumed f = 2.
Thus, by integrating Eq. (2) we have obtained the radiated energy on the
Moon in the luminous range (the CCD sensitivity range) by employing the
visual magnitude m
V
= 7.3 and a passband ∆λ = 0.5 µm (Bellot Rubio et al.
2000a,b; Ortiz et al. 2000). Also the corresponding energy radiated in the
near-infrared by employing m = 5.1 and ∆λ = 0.315 µm. Note that the zero
magnitude flux density for I band is 9.76 · 10
-9
Wm
-2
µm
-1
according to
Bessel (1998) and not 3.75 ·10
-8
Wm
-2
µm
-1
. The Earth-Moon distance at the
instant of the impact was R= 389134.5 km. According to our calculations
these energies yield E
V
= 1.46·10
6
J and E
I
= 1.44·10
6
J, respectively. This
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
smaller value mainly comes from the smaller bandpass of our near-infrared
observations. If we used a normalized passband of 0.5 microns for the two
cases, or use the flux density, the energy emitted in a normalized passband
in the infrared would be 0.5/0.315 times larger or E
I
= 2.28·10
6
J. Hence
there is a significant increase of the efficiency in the near-infrared compared
to that in the visible.
To estimate the kinetic energy of the impactor, the meteoroid mass and the
emission efficiency in the infrared, we have assumed that the luminous
efficiency in V band for sporadic impact flashes is 2·10
-3
(Ortiz et al. 2006,
2015). However, it must be taken into account that this efficiency in V band
was obtained by employing f = 3 in Eq. (2). Since in this work we have
taken f = 2, we must multiply this efficiency by a factor 3/2, so that the
assumed luminous efficiency in V band for the event considered here yields
η
V
= 3·10
-3
. The kinetic energy E
k
of the impactor is then given by the
equation
E
k
= E
V
/η
V
(3)
and yields E
k
= 4.86·10
8
J.
Once the kinetic energy is known, the emission efficiency in our NIR band
η
NIR
can be estimated from
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
η
I
= E
I
/E
k
(4)
This parameter yields η
I
= 4.7·10
-3
. Nevertheless, this result depends
critically on the value adopted for η
V.
For η
V
= 5·10
-4
and η
V
= 5· 10
-3
the
kinetic energy of the impactor yields 2.91· 10
9
J and 2.91· 10
8
J, respectively,
while the resulting emission efficiency in the infrared yields 7.8·10
-4
and
7.8·10
-3
, respectively. According to this calculation, the emission efficiency
for sporadic events in this spectral band is higher than in V band by a factor
of about 56%. Notice that this factor does not depend on the value adopted
for η
V
. This factor shows that presumably a large part of the
electromagnetic energy released as a consequence of the impact is radiated
in the infrared.
For an impact velocity V for sporadics of 17 km s
-1
the kinetic energy E
k
calculated by assuming η
V
= 3·10
-3
corresponds to an impactor mass M =
3.4 ± 0.3 kg. To obtain the impactor size we have considered a bulk density
of 1.8 g cm
-3
for sporadic meteoroids (Babadzhanov and Kokhirova 2009).
According to this assumption, the impactor diameter would be D
P
= 15.3 ±
0.4 cm. For η
V
= 5·10
-4
and η
V
= 5·10
-3
this calculation yields M = 20 ± 3
kg and M = 2.0 ± 0.3 kg, respectively, and the corresponding meteoroid
diameter yields D
P
= 27 ± 1 cm and D
P
= 12.8 ± 0.6 cm, respectively.
4.4. Impact plume temperature
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
From the magnitudes shown in Figure 4, the evolution with time of the
energy flux density measured on Earth for both V and I bands (denoted by
F
V
and F
I
, respectively) can be obtained by employing the following
relationships:
)5.2/m(
8
V
V
10·10·75.3F
−
−
= (5)
)5.2/m(
9
I
I
10·10·76.9F
−
−
= (6)
where, as mentioned above, 3.75· 10
-8
and 9.76·10
-9
are the irradiance of a
magnitude 0 star in Wm
-2
µm
-1
for V and I bands, respectively (Bessel 1998).
If we assume that the intensity distribution of the impact flash follows
Planck's law, the ratio of these flux densities must satisfy the equation
1
1
F
F
e
e
Tk/hc
Tk/hc
5
V
I
I
V
I
V
−
−
λ
λ
=
λ
λ
(7)
In this relationship λ
V
= 0.55 µm and λ
I
= 0.798 µm are the effective
wavelengths for V and I bands, respectively (Bessel 2005), h is Planck's
constant, c the speed of light in vacuum, and k is Boltzmann's constant. The
evolution with time of the impact plume temperature T estimated by solving
Eq. (7) is shown in Figure 5. This plot shows that this temperature reached a
maximum value of around 4000 K at the beginning of the flash, and then
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
after a sudden decrease to around 3200 K it remains practically constant for
around 0.1 s. This suggests that during this phase the condensation process
gives rise to equilibrium in the impact plume, so that the temperature
remains constant as a consequence of the release of evaporation energy
(Nemtchinov 1998). After that time the plume temperature slowly decreases
to a final value of ~2900 K by the end of the event.
4.5. Crater size
The results obtained from the analysis of the crater resulting from the
impact of the meteoroid on the lunar surface are summarized in Table 2. We
have employed the crater-scaling equation for the Moon given by Gault,
which is valid for craters with a diameter of up to about 100 meters in loose
soil or regolith (Gault 1974, Melosh 1989):
(
)
3/129.0
k
5.0
t
6/1
p
sinE25.0D θρρ=
−
(8)
Magnitudes in this relationship must be entered in mks units. D is the crater
diameter, E
k
the kinetic energy of the impactor, ρ
p
and ρ
t
are the impactor
and target bulk densities, respectively, and θ is the impact angle with respect
to the horizontal. Since this angle is unknown for sporadic meteoroids (the
most likely source of the impact flash), we have considered θ = 45º, which
is the value of the most likely impact angle. The rim-to-rim crater diameter
derived from Eq. (8), with the value of E
k
obtained with η
V
= 3· 10
-3
, yields
D = 6.4 ± 0.2 m for an impactor bulk density of 1.8 g cm
-3
. For the target
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
bulk density we have taken ρ
t
= 1.6 g cm
-3
. As an alternative formula to
obtain the rim-to-rim diameter of this crater we have employed the
following equation (Holsapple 1993):
.
m
K6.2D
3/1
t
v
r
ρ
π
=
(9)
where the adimensional factor π
v
is obtained from
µ+
µ−
µ+
µ
−ν
µ
µ−−ν
ρ
ρ
θρ
+
ρ
ρ
θ
=π
2
3
2
2
3
26
P
t
2
t
2
3
26
P
t
2
1v
))sin(V(
Y
K
))sin(V(
ag
K
(10)
In Equations (9) and (10) physical quantities are entered in mks units, with
K
1
=0.2, K
2
=0.75, K
r
=1.1, µ=0.4, ν=0.333 and Y = 1000 Pa. V, a, and m are
the impactor velocity, radius and mass, respectively. The gravity on the
lunar surface is g = 0.162 m s
-2
. By using Equations (9) and (10) one obtains
D = 6.8 ± 0.2 m for ρ
p
= 1.8 g cm
-3
and V= 17 km s
-1
.
It must be also taken into account that despite ρ
p
= 1.8 g cm
-3
has been
adopted above, the density of sporadic meteoroids can range from 0.3 g cm
-3
(the density of soft cometary materials) to 3.7 g cm
-3
(the density of
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
ordinary chondrites). For these densities the diameter of this crater
calculated from Eq. (8) would be 4.8 ± 0.1 m and 7.3 ± 0.2 m, respectively.
The other two possible sources of the projectile are the Virginids, which
follow an asteroidal orbit, and the γ-Normids, which have a cometary nature
(Jenniskens 2006). So, for the Virginids we have considered projectile
densities ranging between 2.4 g cm
-3
(the corresponding to carbonaceous
chondrites) and 3.7 g cm
-3
(the density of ordinary chondrites). And for the
γ-Normids we have assumed that this density can range between 0.3 g cm
-3
(the density of soft cometary materials) and 1.8 g cm
-3
. As mentioned above,
the impact velocity for the Virginids is 30 km s
-1
, with an impact angle of
40.2º with respect to the local vertical. And for the γ-Normids the impact
velocity is 56 km s
-1
, with an impact angle with respect to the local vertical
of 46.2º. According to this, the Gault model yields for the Virginids a crater
diameter ranging from 6.9 ± 0.2 m (for ρ
p
= 2.4 g cm
-3
) to 7.4 ± 0.2 m (for
ρ
p
= 3.7 g cm
-3
), and for the γ-Normids this diameter would range from 4.7
± 0.1 m (for ρ
p
= 0.3 g cm
-3
) to 6.4 ± 0.2 m (for ρ
p
= 1.8 g cm
-3
).
Despite this crater is too small to be directly observed with instruments from
Earth, it could be observed by means of probes orbiting the Moon (with
before and after images taken under similar illumination conditions), such as
for instance the Lunar Reconnaissance Orbiter (LRO) (Robinson 2015). In
this way, the actual crater size could be compared with the predicted
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
diameter in order to test the validity of the assumptions employed in this
analysis.
5. CONCLUSIONS
We have analyzed a lunar impact flash recorded on 25 March 2015 in both
visual and NIR bands. Despite the fact that estimations of the temperature of
lunar impact flashes have been published by other team during the
refereeing process of our manuscript (Bonanos et al. 2018), our result still
forms the first measurement of the temperature of a telescopic lunar impact
flash (Madiedo and Ortiz 2016, 2018). The peak visual magnitude of this
event was 7.3 ± 0.2 and the peak magnitude in I band was 5.1 ± 0.3. Our
calculations show that the most likely origin of the source meteoroid is the
sporadic background, with a probability of 88 %. By assuming a luminous
efficiency in V-band of 3·10
-3
, the impactor mass for this sporadic projectile
yields 3.4 ± 0.3 kg. And the estimated diameter D of the resulting crater
would range, according to the Gault model, from 4.8 ± 0.1 m (for ρ
p
= 0.3 g
cm
-3
) to 7.3 ± 0.2 m (for ρ
p
= 3.7 g cm
-3
), with D = 6.8 ± 0.2 m for ρ
p
= 1.8 g
cm
-3
. This fresh crater could be observed by means of probes orbiting the
Moon, such as LRO.
The emission efficiency in the NIR for this sporadic event has been also
inferred. The value of this parameter yields 4.7· 10
-3
, which is higher, by
around 56%, than the luminous efficiency in visible band. This shows that
presumably a large part of the electromagnetic energy radiated as a
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
consequence of the impact is emitted in the near infrared. The temperature
of the impact plume, which has been obtained from the energy flux densities
for V and I bands, is of around 3200 K during most of the light curve, which
suggests that condensation gives rise to equilibrium in the impact plume
during this stage. But the very initial flash is even hotter (about 4000 K),
and this temperature decreases to ~2900 K by the end of the event.
ACKNOWLEDGEMENTS
The authors acknowledge support from projects AYA2014-61357-EXP
(MINECO), AYA2015-68646-P (MINECO/FEDER), Proyecto de
Excelencia de la Junta de Andalucía, J.A. 2012-FQM1776, and from
FEDER.
REFERENCES
Babadzhanov P.B. and Kokhirova G.I., 2009, A&A, 495, 353.
Bellot Rubio L.R., Ortiz J.L., Sada P.V., 2000a, ApJ, 542, L65.
Bellot Rubio L.R., Ortiz J.L., Sada P.V., 2000b, Earth Moon Planets, 82–83,
575.
Bessell M.S., Castelli F., Plez B., 1998, Astron. Astrophys. 333, 231.
Bessel M.S., 2005, Annual Review of Astronomy & Astrophysics, 43, 293.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
Bonanos A.Z., Avdellidou C., Liakos
A., Xilouris, E.M. et al., 2018,
Astronomy & Astrophysics, 612, id.A76.
Cudnik B., 2010. Lunar meteoroid impacts and how to observe them.
Springer.
Cooke W.J., Suggs R.M., Swift W.R., 2006, Lunar Planet. Sci. 37. Abstract
1731.
Cudnik B.M., Dunham D.W., Palmer D.M., Cook A.C., Venable J.R., Gural
P.S., 2002, Lunar Planet. Sci. 33. Abstract 1329C.
Dubietis A, Arlt R, 2010, EMP, 106, 105.
Gault D.E., 1974, In: R. Greeley, P.H. Schultz (eds.), A primer in lunar
geology, NASA Ames, Moffet Field, p. 137.
Holsapple K.A., 1993, Annu. Rev. Earth Pl. Sc. 21, 333.
Jenniskens P., 2006, Meteor Showers and their Parent Comets. Cambridge
University Press.
Madiedo J. M., 2014, Earth, Planets and Space, 66, 70.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
Madiedo J. M., Trigo-Rodríguez J. M., 2008, Earth Moon Planets, 102, 133.
Madiedo J.M., Trigo-Rodríguez J.M., Ortiz J.L., Morales N., 2010,
Advances in Astronomy, doi:10.1155/2010/167494.
Madiedo J.M., Trigo-Rodríguez J.M., Konovalova N., Williams I.P., Castro-
Tirado A.J., Ortiz J.L., Cabrera-Caño J., 2013, MNRAS, 433, 571.
Madiedo J.M., Ortiz J.L., 2016, https://www.cosmos.esa.int/documents/
653713/1000954/08_ORAL_Madiedo.pdf/d64b2325-8a37-432a-98f1-
28784da94a40
Madiedo J.M., Ortiz J.L., 2018, MIDAS System. In: Cudnik B. (eds)
Encyclopedia of Lunar Science. Springer, Cham, doi:10.1007/978-3-319-
05546-6_128-1
Madiedo J.M., Ortiz J.L., Morales N., Cabrera-Caño J., 2014a, MNRAS,
439, 2364.
Madiedo J.M., Ortiz J.L., Trigo-Rodríguez J.M., Zamorano J., Konovalova
N., Castro-Tirado A.J., Ocaña F., Sánchez de Miguel A., Izquierdo J.,
Cabrera-Caño J., 2014b, Icarus, 233, 27.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
Madiedo J.M., Ortiz J.L., Morales N., Cabrera-Caño J., 2015a, PSS,
111,105.
Madiedo J.M., Ortiz J.L., Organero F., Ana-Hernández L., Fonseca F.,
Morales N., Cabrera-Caño J., 2015b, A&A, 577, A118.
Melosh H.J., 1989. Impact Cratering: A Geologic Process. Oxford Univ.
Press, New York.
Nemtchinov I.V., Shuvalov V.V., Artemieva N.A., Ivanov B.A., Kosarev
I.B., Trubetskaya I.A., 1998. Light impulse created by meteoroids impacting
the Moon. Lunar Planet. Sci XXIX. Abstract 1032.
Ortiz J.L., Aceituno F.J., Aceituno J., 1999. A&A, 343, L57.
Ortiz J.L., Sada P.V., Bellot Rubio L.R., Aceituno F.V., Aceituno J.,
Gutierrez P.J., Thiele U., 2000, Nature, 405, 921.
Ortiz J.L., Quesada J.A., Aceituno J., Aceituno F.J., Bellot Rubio L.R.,
2002, ApJ, 576, 567.
Ortiz J.L. et al., 2006, Icarus, 184, 319.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
Ortiz J.L., Madiedo J.M., Morales N., Santos-Sanz P., and Aceituno F.J.,
2015. MNRAS, 454, 344.
Robinson M.S, Boyd A.K., Denevi B.W., Lawrence S.J., McEwen A.S.,
Moser D. E., Povilaitis R.Z., Stelling R.W., Suggs R.M., Thompson S.D.,
Wagner R.V., 2015, Icarus, 252, 229.
Suggs R.M., Moser D.E., Cooke W.J., Suggs R.J., 2014, Icarus 238, 23.
Swift W.R., Moser D.E., Suggs R.M., Cooke W.J., 2011, In: Meteoroids:
The Smallest Solar System Bodies, Edited by W.J. Cooke, D.E. Moser, B.F.
Hardin, and D. Janches, NASA/CP-2011-216469, p 125.
Yanagisawa M., Kisaichi N., 2002, Icarus, 159, 31.
Yanagisawa M., Ohnishi K., Takamura Y., Masuda H., Ida M., Ishida M.,
2006, Icarus, 182, 489.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
TABLES
Date and time 2015 March 25 at 21h 00m 16.80 ± 0.01s UT
Peak brightness (magnitude) 7.3±0.2 in V band; 5.1±0.3 in I-band
Selenographic coordinates Lat.: 11.3±0.1 ºN, Lon.: 21.6±0.1 ºW
Duration (s) 0.18 (V band); 0.20 (NIR)
Table 1. Characteristics of the impact flash discussed in this work
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
Meteoroid
source
Model Impact
angle (º)
Meteoroid
Density
(g cm
-3
)
Meteoroid
Mass
(kg)
Impact
Velocity
(km s
-1
)
Crater
Diameter
(m)
Gault 45 0.3 3.4±0.3 17 4.8±0.1
Gault 45 1.8 3.4±0.3 17 6.4±0.2
Gault 45 3.7 3.4±0.3 17 7.3±0.2
Holsapple 45 0.3 3.4±0.3 17 6.8±0.2
Holsapple 45 1.8 3.4±0.3 17 6.8±0.2
Sporadic
Holsapple 45 3.7 3.4±0.3 17 6.8±0.2
Gault 40.2 2.4 1.1±0.1 30 6.9±0.2
Gault 40.2 3.7 1.1±0.1 30 7.4±0.2
Holsapple 40.2 2.4 1.1±0.1 30 6.2±0.2
Virginids
Holsapple 40.2 3.7 1.1±0.1 30 6.2±0.2
Gault 46.2 0.3 0.31±0.05 56 4.7±0.1
Gault 46.2 1.8 0.31±0.05 56 6.4±0.2
Holsapple 46.2 0.3 0.31±0.05 56 5.2±0.2
γ-Normids
Holsapple 46.2 1.8 0.31±0.05 56 5.2±0.2
Table 2. Rim-to-rim crater diameter predicted by the Gault and the
Holsapple models, by assuming η
V
=3·10
-3
. The impact angle is measured
with respect to the local vertical.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
FIGURES
Figure 1. The lunar disk as seen from Earth on 2015 March 25. White
region: area illuminated by the Sun. Gray region: night side. Cross: impact
flash position.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
Figure 2. Impact flash detected from Sevilla on 2015 March 25 as recorded
by the 0.28 cm (a), the 0.36 m (b) and the 0.24 cm (c) SC telescopes
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
operating at this observatory. Images (a) and (b) were recorded with
unfiltered cameras. Image (c) was obtained by means of a NIR filter.
0
10
20
30
40
50
60
70
80
400 500 600 700 800 900 1000 1100 1200
Wavelength (nm)
Relative response
Johnson-Cousin I filter
Baader IR pass filter
Figure 3. Response of the Johnson-Cousins I filter transmission and the
Baader IR-pass filter transmission convolved with the spectral response of
the camera CCD.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
4
5
6
7
8
9
10
0 0,04 0,08 0,12 0,16 0,2 0,24
time (s)
Magnitude
Visual
nIR
Figure 4. Lightcurve of the impact flash in visible light and in the NIR.
Visual data correspond to the observation performed with the 0.36 m
telescope.
Accepted for publication on Monthly Notices of the Royal Astronomical Society
on 2018 July 7
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 0,05 0,1 0,15 0,2
time (s)
T (K)
Figure 5. Evolution with time of the temperature of the impact plume, with a
spline curve fitted through the points.
