© ESO, 2015

1. Introduction

Complex observations of individual meteoroids allow us a better understanding of the relation between meteoroids and their parent bodies. During the short luminous phase when they are called meteors, which is when a meteoroid enters the Earth’s atmosphere, we have the opportunity to measure the meteoroid velocity, trajectory, and orbit with high precision. In addition, spectroscopic observations of meteors reveal the chemical composition of cometary and asteroidal meteoroids.

The study of meteor spectra started in the 1860s with A. S. Hershel’s visual observations (Millman 1963). Since the end of the nineteenth century, photographic techniques were used to observe meteor spectra (Millman 1980). Except for the systematic work by Peter Millman since the 1930s (Tors & Orchiston 2009), there was not much general activity in this field of astronomical photography before World War II. From the 1950s onward, transmission gratings largely replaced prisms in meteor spectrographs. Extensive spectroscopic programs were carried out in the USA, Canada, the former USSR, and former Czechoslovakia (Ceplecha et al. 1998). Hemenway et al. (1971) used a sensitive video technique to observe meteor spectra for the first time. Different TV techniques and the TV data reduction methods were described by Millman & Clifton (1975), Mukhamednazarov & Maltseva (1989), Borovička & Boček (1995), and Zender et al. (2004). Meteor spectra mostly consist of atomic emission lines, accompanied with some molecular bands and continuous radiation. Halliday (1961), Ceplecha (1971), and Borovička (1994a) provided extensive lists of line identifications in high-dispersion photographic spectra. Borovička (1994b) revealed that lines can be divided into two components − the low- and the high-temperature component.

Despite this long history of meteor spectroscopy, very few general surveys of meteor spectra have been presented. Harvey (1973) published statistics of visual inspection of 500 photographic meteor spectra. Borovička et al. (2005) presented a survey of 97 spectra of mainly sporadic meteors mostly in the magnitude range from +3 to 0 that were obtained by sensitive video technique. The spectra were classified according to the relative line intensities of Mg, Na, and Fe, and three distinct populations of Na-free meteoroids were identified. We here extended that work by presenting a catalogue of 84 new video spectra of both sporadic and shower meteors. All these meteors were captured by our sensitive video technique.

Compared to photographic spectra of bright meteors, video spectra have a lower resolution and only contain a few lines. Only four meteoritic elements (Na, Mg, Ca, and Fe) can be measured in most video spectra. Other observed emissions (O, N, and N₂) are of atmospheric origin. On the other hand, the video technique is able to record spectra of relatively weak meteors (compared to photographic spectra). Thus we can obtain a more comprehensive sample of meteors for future analysis. Limits for our technique are +2 mag (the spectrum sufficiently bright for further analysis) and −3.5 mag (the signal is not yet saturated), corresponding to meteoroid sizes in the range of 1−10 mm.

The meteor sample presented here contains members of all major meteor showers as well as sporadic meteors observed in different parts of the year and different parts of the night. Our catalogue is therefore representative in the sense that it contains all common types of spectra in the given range of meteor brightness. In comparison with the work of Borovička et al. (2005), we used an image intensifier that provides a larger field of view, so that the spectral coverage was better. Our catalogue is intended to serve as a reference work for future spectral surveys of meteors.

2. Observations and equipment

Most observations were performed during the periods of activity of major meteor showers (Quadrantids, Lyrids, η Aquarids, Perseids, Orionids, Leonids, and Geminids) in the years between 2006 and 2012. Most observations were carried out from the base Ondřejov−Kunžak. The distance between the stations is 92.5 km. But several meteors were observed during observation campaigns in Tajikistan and Italy. For more details see Table 1.

Each station was equipped with S-VHS-C camcorders with the second-generation image intensifiers Mullard XX1332. One direct camera and one spectral camera was operated from the first station and one direct camera was operated from the second station. The spectral grating with 600 grooves/mm and the Arsat 1.4/50 mm lens (FOV ≈ 50°) was used for all spectral observations, except for the 2011 Draconid campaign, when the lens Jupiter 2/85 mm (FOV ≈ 30°) was used. The resulting dispersion was 30 Å  pixel^-1 for the Arsat lens and 15 Å  pixel^-1 for the Jupiter lens. The spectral sensitivity extends from 3800 Å  to 9000 Å. The sensitivity curve for the whole system (camera, image intensifier, and lens) is given in Fig. 1. The image intensifier affects the spectral sensitivity of the system by far the most, the individual differences between different lenses can be neglected. The intensity was normalized to unity at the wavelength 5500 Å. The Arsat 1.4/50 mm and the Jupiter 2/85 mm lenses were used for direct observations.

For our standard observations (from the base Ondřejov−Kunžak), video data were recorded by S-VHS video recorders on S-VHS video tapes (until the year 2008). After 2009, all videos were recorded straight onto the PC hard drive. Direct recording to the DV cassette was used during expeditions. These records were inspected using the automatic meteor-detection software MetRec (Molau 1999). Parts of the record, with detected meteors, were then saved as uncompressed AVI files with a resolution of 768 × 576 pixels × 8 bit. The frame rate was 25 images per second, from which we obtained a time resolution of 0.04 s. These 8 bit AVI files were then used for all measurements and data analysis presented in this paper.

We only selected meteors that were recorded from both stations for this paper and for which we managed to obtain good spectra. By “good” spectra we mean spectra where the entire meteor (or at least a significant part of it for very long meteors) appeared inside the field of view of the camera and at least the most important part of the spectrum (5000−8000 Å) was covered. For the major showers we did not include all good spectra, but only representative ones (several meteors of different brightness).

3. Data reduction

The date of the observation of the meteor is encoded in the number of the corresponding meteor. The number is given in the form YYMDDXXX, where YY are the last two digits of the year. M is the month of the year: 1 to 9 − January to September and A − October, B − November and C − December. The digits DD indicate the day when the observation started (the evening date). This means that the actual day of the observation is equal to DD if time T> 12 UT and DD+1 if T < 12 UT. The last three digits XXX indicate the number of the meteor; the counting starts at the beginning of the night. For example, meteor 06C13137 was the 137th metor observed on 13 December 2006. The only exception is the single Draconid meteor DRA06. This meteor was observed during the Draconid observation campaign in Northern Italy on 8 October of 2011. The geocentric radiant, the zenith distance of the apparent radiant, the beginning height, the terminal height, the maximum brightness, the photometric mass, the entry velocity, and the spectral type for each meteor are also included.

The name of the meteor spectra contains the prefix SX followed by the number of the spectrum. The numbering is carried out continuously in chronological order.

During the data reduction, we excluded the weakest meteors with a low signal-to-noise ratio. Finally, we used a total number of 84 meteors, 54 of which were sporadic meteors and 30 were showers meteors. Some of the sporadic meteors may be members of minor showers, but the association is unclear, and we counted them as sporadic in the following discussion.

The brightness and positions of all meteors were measured manually in each video frame with a special software developed by Borovička et al. (2005). By combining data from two direct cameras, we determined atmospheric trajectories and heliocentric orbits of meteoroids by the least-squares method (Borovička 1990). The Southworth-Hawkins D-criterion (Southworth & Hawkins 1963) was used to determine the meteor shower membership. The threshold value for the D-criterion was arbitrarily chosen to be 0.2, although none of the determined shower meteors had a value of the D-criterion higher than 0.15. By knowing the meteoroid trajectories in the atmosphere, we were able to determine their absolute magnitude and photometric mass.

Video files with meteor spectra were processed with our standard procedure (Borovička et al. 2005). Positional and photometric calibration was made using the zero-order images of stars in the field. When each frame in which the spectrum appeared was scanned, we refined the wavelengths using well-known atomic lines of the meteor emission.

In Figs. 2 and 3 we present one typical spectrum from the catalogue. In Fig. 2 the uncalibrated spectrum SX336 is given. The spectrum is calibrated for spectral sensitivity of the system in Fig. 3. Below 4000 Å  and above 9000 Å, the final spectra can be strongly influenced by the noise after the calibration due to the low sensitivity of the equipment at these peripheral parts.

Statistical analysis of the noise in video frames was performed for selected recordings. We determined the average value of obtained standard deviations of the Gaussian noise for our equipment, σ = 14 (in device arbitrary units). We then used this value to compute uncertainties for all spectra presented in the catalogue. Since spectra presented in the catalogue are the sums of individual frames, we determined the spectral uncertainties as the multiplication of the standard deviation and the square root of the number of frames used for a given spectrum. We applied the same relative uncertainties for the calibrated spectra.

For the basic analysis presented here, we also measured the spectral line intensities. This procedure is challenging since several components such as continuous emission, nitrogen bands, and other lines can contribute to the intensity of the peak at the line position. For this reason, we estimated intensities of all of these components that contribute to the resulting shape of the spectra to fit each spectrum completely. The fit parameters (e.g. spectral component intensities, temperature of the Planck emission, and the vibrational and rotational temperature of nitrogen bands) were changed manually by the operator of the software to fit the spectrum as well as possible. This procedure was done on each video frame, except for the temperature of Planck continuum and nitrogen bands, where one temperature for all frames was sufficient. In this way, we obtained spectral line intensities for each frame. The individual element intensities were then summed to find the total intensity for the corresponding multiplet.

We must point out that the spectral line intensities estimated by this procedure are meant to have rather illustrative purpose. We are not able, at this stage of the research, to determine uncertainties for line strengths or their ratios.

4. Description and classification of the spectra

4.1. Description of the spectra

The observed spectrum usually consists of the continuum, the emission from the heated atmosphere (specifically the oxygen and nitrogen lines and the nitrogen molecule bands) and the emission that originated in the evaporated material of the meteoroid.

In Table 2 we show the list of lines that significantly contribute to the meteor spectra. According to their origin, these lines are divided into several groups. The lines of the first group are part of a low-temperature (≈4500 K) spectral component. The second group consists of lines of a high-temperature spectral component (≈10 000 K). However, the lines of the second group are usually visible in bright and fast meteors and are of low importance for meteors within the magnitude range presented here. Another group consists of lines emitted just behind the meteoroid; they are prominent in meteor wakes. These lines only last a fraction of a second and are of the low-energy excitation intercombination origin and cannot be fitted by a thermal model (Borovička & Jenniskens 2000). Although they are present typically in the meteor wake, they may also occur in the meteor heads, especially if the collisional deexcitation rate is low (Borovička et al. 2005). A significant part of meteor spectra is formed by lines and bands of atmospheric origin (O and N at ≈10 000 K). In the last group, there is only one line, the forbidden green oxygen line at 5577 Å. This line persists for nearly a second after the meteor disappears and forms a short-duration trail. This line is often noticeable in fast meteors.

The lines of the four meteoric elements can be recognized in the video spectra: magnesium (Mg I), sodium (Na I), iron (Fe I), and calcium (Ca I). However, the calcium lines are only sufficiently intense in bright meteors because our intensifier only has a low image sensitivity in the blue part of the spectrum (see Fig. 1). The Cr I lines are blended with Fe I and cannot be measured.

In meteors that are bright and fast enough, the lines of the high-temperature component (Ca II, Mg II, and Si II) may be present. In Figs. 4 and 5 spectra of two Perseid meteors (≈60 km s^-1) of different brightness are compared. The maximum absolute magnitude of the meteor that formed the spectrum SX1802 in Fig. 4 was − 1.6 mag. The only distinguishable meteoric lines are the low-temperature component lines Mg I, Na I, and Fe I. As an example of a bright Perseid meteor, the spectrum SX1837 is shown in Fig. 5. Because of the high maximum brightness of − 9.2 mag, the first order spectrum was oversaturated in video frames during the brightness maximum, therefore we did not include this spectrum in the catalogue (for more details about this exceptional meteor see Spurný et al. 2014). Moreover, the approach for the classification of spectra that we use below was only developed for small meteroids in the 1−10 mm size range. We chose only one frame from the sequence for this example where the spectra were not yet oversaturated. Low-temperature atomic lines are visible (Mg I, Na I, and Fe I), but the high-temperature atomic lines (Mg II, Ca II) are also sufficiently bright in this case.

The appearance of the spectrum depends not only on the temperature, but also on the speed of the meteor. As an example of a spectrum of a fast (50.7 km s^-1) meteor, the spectrum SX457 is given in Fig. 6. In contrast, in Fig. 7 the example of a slow (24.5 km s^-1) meteor is presented (the spectrum SX1206). Both meteors had a similar visual magnitude (≈+ 1 mag).

The atmospheric lines of O I, N I, the nitrogen bands N₂, and the forbidden green oxygen line [O I] are much brighter in fast meteors and cause brighter red and infrared parts of the spectrum of a fast meteor than in the spectrum of a slow meteor. The only atmospheric line, eminent in the spectrum SX1206, is the brightest oxygen line O I at 7774 Å.

4.2. Classification of spectra

Only three meteoric lines (Na I, Mg I, and Fe I) can be examined in typical video spectra. Nevertheless, the differences in the intensities of these lines can reveal the different composition of individual meteoroids. We used the multiplet 15 at 5270−5450 Å to measure Fe because our equipment is very sensitive in this spectral region (compared to the low sensitivity for Fe lines near 4400 Å), and we summed contributions of all lines of the multiplet. For Na and Mg the bright lines of multiplet 1 and multiplet 2, respectively, were used.

For this analysis we did not take the differential ablation into account. The contributions of individual multiplets were summed along the whole meteor path, and we only worked with total intensities.

To visualize the spectral classification we used the Mg-Na-Fe ternary diagram (Figs. 8 and 9) and the diagram of the dependence of the Mg/Na ratio on speed (Fig. 10).

We used the spectral classification of meteor populations suggested by Borovička et al. (2005). The meteoroids were accordingly divided as follows.

• Iron meteroids No sharp lines are present. Two bands formed by unresolved Fe multiplets (at 4200−4500 Å and 5100−5500 Å). The Mg line at 5180 Å  can contribute to the intensity of the second band, but is much fainter than in normal spectra. The Na line is missing.

• Na-free meteoroids No Na line but, not classified as iron meteorids. The Fe/Mg ratio varies widely.

• Na-rich meteoroids The spectrum is dominated by the Na line. The lines of Mg and Fe are present, but they are faint.

• Mainstream meteoroids Mainstream meteoroids form the majority of meteoroids. Their spectra are closer to the expected chondritic spectra. But there are strong variations in the Na line intensity. Almost the whole range between the Na-free and Na-rich meteoroids is covered. The mainstream meteoroids are divided into four subclasses.

  • Normal meteoroids Normal meteoroids are defined as those lying near the expected position for chondritic bodies in the ternary diagram or with somewhat lower Fe intensity.

  • Na-poor meteoroids The Na line is weaker than expected for the given speed, but still reliably visible in contrast to Na-free meteoroids.

  • Na-enhanced meteoroids The Na line is significantly brighter than expected for the given speed, but not as dominant as for Na-rich meteoroids.

  • Fe-poor meteoroids Meteoroids with the expected Na/Mg ratio, but with the Fe lines too faint to be classified as normal.

The spectrum SX1101 with dominant atmospheric lines and bands was excluded from the classification (see Fig. 12). The lines of Na, Mg, and Fe were not bright enough for the spectral analysis.

The decision to classify between Na-poor and Na-free meteoroids from the ternary graph was not always obvious (Fig. 8). For this purpose, Fig. 10 can be used, which shows the dependence of the Na/Mg ratio on the velocity of the meteorid, where only intensities of Mg and Na are taken into account. A similar situation occurs between Fe-poor and normal meteoroids.

The intensities of the spectral lines correspond to different meteoroid compositions. But as we can see from spectral classification, the meteor speed also needs to be taken into account. According to Fig. 10, the Na/Mg ratio increases for speeds below 35 km s^-1. For speeds higher than 35 km s^-1, the ratio is speed independent. Meteors classified as iron meteorids were excluded from Fig. 10, lines of Na and Mg are absent or weak for iron meteorids, and thus the determination of the Na/Mg ratio is very inaccurate or even impossible.

In Fig. 11 we show the O/Mg ratio as a function of velocity. The ratio of the O to Mg line intensities increases with the velocity. As mentioned before, fast meteors are characterized by more dominant atmospheric lines. For speeds below 30 km s^-1 the scatter is large, mainly due to the faintness of the O line. Meteors classified as iron meteorids were excluded from Fig. 11.

This work is based on the sample of 84 meteors, which includes sporadic meteors and minor and major showers. When compared to the work of Borovička et al. (2005), which was based on sample of 97 sporadic meteors and minor shower meteors, the results are similar.

Figure 9 shows Na-Mg-Fe ternary graph as well, but symbols of individual meteors represent the association with the shower. Most of the major shower meteoroids have been classified as normal. The only exceptions are meteoroids belonging to the Geminid shower and to the Southern δ Aquariids shower. Both Southern δ Aquariids meteoroids (SX731, SX738) were classified as Na-free. The Geminids have members of the Na-free, Na-poor, and normal spectral groups. The reason of Na depletion in both showers is solar heating at low perihelion distances (Čapek & Borovička 2009). The perihelion distance of Geminids is somewhat larger and the degree of Na depletion probably depends on the meteoroid size and structure, especially porosity (Borovička 2010).

It is well known that different strength categories of meteoroids in the millimeter-size range have different beginnings of the meteor luminous path (Ceplecha 1988). For a given speed, a meteoroid composed of stronger material has a lower beginning height than the meteoroid formed by weaker material. As we can see in Fig. 13, the iron meteoroids and the Na-free meteoroids started to ablate at lower heights than most meteoroids, while Fe-poor meteoroids started higher. Although Koten et al. (2003, 2004) found that the beginning height of cometary meteoroids also depends on the meteoroid mass, the dependence of the beginning height on the speed is sufficient for our purpose, and we can distinguish the variance of the material strength of different spectral classes of meteoroids.

5. Meteoroid orbits

The heliocentric orbits are known for all 84 meteors from double-station observations. The catalogue has a representative sample of orbits with a wide variety of orbital elements (see Fig. 14). Orbital elements such as the inclination i, the perihelion distance q, the argument of periapsis ω, and the ascending node Ω are presented in the entire range of possible values. Higher values of eccentricity prevail mostly in the range from 0.6 to 1.0, but this is obvious for bodies originating from comets and asteroids with usually higher values of eccentricity. There is also wide range of values of the semi-major axis a. Semi-major axes near the parabolic limit are hard to determine because they are sensitive to the determination of velocity, and so the accuracy is somewhat lower. One orbit is formally hyperbolic, but we believe this is due to a measurement error.

5.1. Relation of meteor orbit to spectral classification

We can use different sets of orbital parameters to classify meteor orbits: the perihelion distance q, the aphelion distance Q, the inclination i, and the Tisserand parameter relative to Jupiter , where a_J = 5.2 AU is the semi-major axis of Jupiter. These parameters are frequently used to determine typical orbits of asteroids, Jupiter family comets, and Halley-type comets.

Five classes of meteoroid orbits were defined by Borovička et al. (2005):

• (SA) Sun-approaching orbits: q< 0.2 AU. Orbits with small perihelion distances are defined as a separate class.

• (ES) Ecliptic shower orbits: Members of ecliptical meteor showers. For example, the Taurid meteors derived from the comet 2P/Encke and other showers with orbits close to the boundary between asteroids and Jupiter family comets.

• (HT) Halley-type orbits: T_J< 2 or 2 < T_J< 3 and i> 45°.

• (JF) Jupiter-family orbits: 2 <T_J< 3 and i< 45° and Q> 4.5 AU.

• (A-C) Asteroidal-chondritic orbits: T_J> 3 or Q< 4.5 AU.

Spectral classes of meteoroids from the catalogue and their positions within the orbit classification schemes are shown in Figs. 15 and 16. Figure 15 shows the perihelion and the aphelion positions of meteoroids. Figure 16 shows the Tisserand parameter versus inclination.

We can now combine the spectral, orbital, and material strength classification and discuss the individual spectral classes separately.

5.2. Iron meteorids

The orbital elements of all six meteors from the catalogue that were classified as iron meteorids are shown in Table 3. Four of them (SX393, SX689, SX692, and SX1114) satisfy the condition for the asteroidal-chondritic class, their aphelion distances Q are below 4.5 AU and their Tisserand parameters T_J are greater than 3. The meteoroid SX001 with a perihelion of only 0.11 AU can be classified as a Sun-approaching meteoroid. The orbit of meteoroid SX1194 can be classified as a Jupiter-family orbit, but asteroidal origin cannot be excluded, since the inclination is only 3 degrees and the aphelion of 4.7 AU is not particularly high.

5.3. Na-free meteoroids

We can clearly distinguish two different populations in the Na-free meteoroids: the Sun-approaching population with small perihelions and the Halley-type population with high inclinations.

5.3.1. Sun-approaching meteoroids

Most of meteoroids with a perihelion distance q < 0.2 are Na-free or Na-poor. The only exception is the iron meteroid-type meteoroid SX001 mentioned earlier, which, nevertheless, does not contain Na either. This corresponds to the conlcusion of Borovička et al. (2005), that frequent approaches to within ≈0.2 AU to the Sun lead to the loss of Na from meteoroids in the millimeter-size range, irrespective to their origin (see also Čapek & Borovička 2009).

The material of Na-free meteoroids also tends to have a greater strength (see Fig. 13).

The members of δ Aquariids (q = 0.07) can be found among the Na-free meteoroids. One member of the Geminid shower (the SX337) was also classified as Na-free. However, members of this stream (with a small perihelion distance q = 0.14 AU) have a wide spread of Na-line intensities. We can find Na-poor and also normal types of spectra within this meteor stream. The explanation is, as suggested by Borovička et al. (2005), that the Na content is correlated with the age of the meteoroid. Younger meteoroids that have suffered fewer passages close to the Sun retain more Na, which implies that the Geminid meteoroid stream was not formed in one instant. A later analysis of Borovička et al. (2010) suggested that differences in porosity may be the main reason of the different Na content in the Geminids.

5.3.2. Cometary Na-free meteoroids

The close approach to the Sun is not the only process that causes depletion of Na in meteoroids. In our sample we have three meteoroids without a Na line in their spectra (SX350, SX696, and SX1104), but their orbits are different from those of Sun-approaching meteoroids. Their perihelion distances are closer to 1 AU, and they have high inclinations or even retrograde orbits. These orbits are of Halley type. According to Borovička et al. (2005), the reason for Na depletion in these types of orbits might be the long exposure to cosmic rays on the comet surface during their residence in the Oort cloud. This process can lead to the formation of Na-free refractory crust. The gradual or sudden disintegration of the crust during the cometary passage through the inner solar system then produces millimeter-sized compact Na-free meteoroids.

5.4. Na-rich meteoroids

There is only one Na-rich meteoroid (SX150) in our catalogue. The body has a Jupiter-family orbit.

5.5. Normal meteoroids

Both cometary and asteroidal orbits are found among meteorids classified as normal. But only part of these meteoroids have a typically chondritic composition, many of them show somewhat fainter Fe lines. According to the computation of the chondritic composition and the Halley dust composition, based on the data of Leonid observations, Borovička et al. (2005) assumed that the sample of meteoroids classified as normal is a mixture of normal chondritic material and cometary material similar to Leonids. Three Taurid meteors (SX263, SX1122, and SX1128) with ecliptic shower orbits have a normal composition.

5.6. Fe-poor meteoroids

Three of the meteoroids in the catalogue (SX211, SX1064, and SX1802) were classified as Fe-poor. The iron lines were too faint to classify the meteoroids as normal, although the boundary is somewhat arbitrary. All of the Fe-poor meteoroids have cometary Halley-type orbits.

Fe-poor meteorids have low material strength, their beginnings of ablation are usually high (see Fig. 13).

5.7. Na-poor meteoroids

Na-poor meteoroids are the transition between normal and Na-free meteoroids. Like the Na-free meteorids, some of them have low perihelia, others have cometary orbits. Thus they probably have the same two origins as the Na-free meteoroids.

5.8. Enhanced-Na meteoroids

Five meteoroids (SX143, SX500, SX785, SX820, and SX983) were classified as enhanced-Na meteoroids. SX500 has a typical asteroidal-chondritic orbit. The other four meteoroids can be classified as ecliptical or Jupiter-family meteoroids. SX143, SX820, and SX983 were sporadic meteors. Their orbits were similar to the Na-rich meteoroid SX150. Borovička et al. (2008) studied enhanced-Na sporadic meteoroid SX498, and according to their analysis, this meteoroid had a Jupiter-family orbit.

6. Catalogue

The atmospheric trajectories and orbital elements for all meteors of the catalogue are shown in Tables 4 and 5.

The second row in Table 4 for each meteor contains the corresponding errors of the least-squares regression. The standard error for meteor heights are of about 10^-2 km. For the absolute magnitude M_max, the standard error for our technique is ≈0.5 mag. The photometric mass is highly affected by the selected luminous efficiency. We used the luminous efficiency of Pecina & Ceplecha (1983).

Heliocentric orbits are known for all meteors. In Table 5 we present orbital elements (reciprocal semi-major axis 1 /a, eccentricity e, inclination i, perihelion distance q, aphelion distance Q, argument of periapsis ω, longitude of the ascending node Ω, Tisserand parameter T_J , and geocentric velocity v_g) for each meteor of the catalogue. Angular elements are given for the equinox J2000.0. The shower membership is given in the last column, SPO stands for sporadic meteor.

The spectra are plotted in Figs. 17−30 (available in electronic version only). The measurements of all frames of the spectra were combined, and the intensities summed along the whole meteor paths are given as a function of wavelength. Calibrated spectra are plotted. Calibrated and uncalibrated spectra of all 84 meteors are available at the CDS. The total intensities of the multiplets Mg1-2, Na1-1, and Fe1-15 are given explicitly in a separate file.

7. Conclusions

We presented a survey of 84 meteors in the magnitude range from +2 to −3, corresponding to meteor sizes from 1 mm to 10 mm. Using parallel double-station observations, we also computed heliocentric orbits for all of these 84 meteors. We classified the meteor spectra according to the classification suggested by Borovička et al. (2005) based on the intensities of the low-temperature emission lines of Na, Mg, and Fe. Only part of meteoroids were found to have a typical chondritic composition. We found a variety of Na depletion, Fe depletion, or Na enhancement. Approximately 20% of the whole sample of millimeter-sized meteors was found to contain no sodium in the spectra. Three populations can be distinguished among the Na-free meteoroids: the iron meteoroids with an asteroidal origin, the Sun-approaching meteoroids with Na depleted by frequent approaches within ≈0.2 AU to the Sun, and the cometary Na-free meteoroids with Na depletion that might be caused by long exposure to cosmic rays of the surface of comets in the Oort cloud.

Most of the major shower meteors have been classified as normal. Some members of the Geminid shower and the members of Southern δ Aquariids were classified as Na-free. The solar heating at low perihelia distances is the reason for the Na depletion of these showers (Čapek & Borovička 2009). The variations of Na content among the Geminid meteor stream was confirmed. The differences in porosity may be the main reason of this effect (Borovička 2010).

Most of the meteoroids on the asteroidal-chondritic orbits were found to be iron meteoroids. This is in contrast to centimeter-sized and larger meteoroids where chondritic composition prevails, but it agrees with results of Borovička et al. (2005), who reported that millimeter-sized stony debris is rare. One iron meteoroid has a typical sun-approaching orbit.

Meteoroids with cometary origin had a heterogeonous composition, from Na-free, Na-poor, and Fe-poor for Halley-type orbits to Na-rich and enhanced-Na for Jupiter-family orbits.

Acknowledgments

This work was supported by grant No. P209/11/1382 from GA ČR and by the project SVV-260089 by the Charles University in Prague.

References

All Tables

All Figures