The earliest datable noctilucent cloud observation (Parma,Italy,AD 1840)

Abstract

Noctilucent clouds (NLCs) are an uncommon phenomenon that provides information about the conditions and dynamics of the mesosphere. The first observation of NLCs was recorded in 1884/1885, following Krakatoa’s eruption in 1883. The literature speculates that this observation was trigged by the injection of millions of tons of H₂O by the Krakatoa into the stratosphere. We have discovered that 43 years before Krakatoa, Antonio Colla observed an NLC in Parma. He was a meticulous astronomer and meteorologist with special interest in astronomical and atmospheric phenomena occurring during twilight. On 18 June 1840, from 21:00 to 22:15 (Local Mean Sideral Time), Antonio Colla observed a ‘phosphoric cloud’. Analysis of the Colla’s description, the local sky and the condition of the observation proves that he was recording an NLC. This finding forces to develop a new hypothesis to explain the early NLC observations and encourages the rescue of NLC observations from documentary sources.

Keywords

Krakatoa eruption mesosphere noctilucent cloud

Introduction

Noctilucent clouds (NLCs) are the highest clouds on Earth. NLCs are formed in the mesosphere at altitudes ranging from 80 to 90 km. The NLCs are composed of ice crystals with radii of 0.1 µm and concentrations ~1–10 cm (Turco et al., 1982). Taking onto account the very low concentration of water vapor at this altitude (1 to 5 ppmv), the formation of water crystals requires very low temperatures (around −130°C). This temperature only can be reached in the mesopause during the summer, when the mesosphere is colder. The light from NLC is scattered sunlight, but this light is so tenuous that it is only visible during the twilight when the sun in around 6° below the horizon (Fogle and Haurwitz, 1966). NLCs are a proxy for mesosphere conditions and dynamics (Gary, 1991). The structure of an NLC can be generally complex with some similarities with cirrus, and its appearance is brilliant. Generally, naked-eye NLC observations are possible only in a specific region of the sky under particular conditions, that is, when the sun is some degrees under the horizon (generally more than 6°) but reach to illuminate the clouds in the mesosphere during clear days of summer at high latitudes (Gadsden and Schröder, 1989).

Although the first NLC observation was described more than 130 years ago, many questions concerning the formation mechanism and observational trends of these phenomena have not yet been fully explained. Moreover, some authors have suggested that NLCs could be useful as indicators of climate change (Hervig et al., 2016; Lübken et al., 2018; Russell et al., 2014; Thomas, 1996, 2003) although few is known about the long-term variability of these clouds (von Zahn, 2003; von Zahn et al., 2004).

Traditionally, the earliest datable NLC observation is assigned in 1885 (Backhouse, 1885; Jesse, 1885, 1889; Tserasky, 1890). A year earlier (1884), Robert C. Leslie described a possible NLC observation in Nature as ‘weird small cloud forms, at times very regular, like ripple-marks in sand, or the bones of some great fish or saurian embedded on a slab of stone’ (Leslie, 1884) and in 1885 he further described a possible NLC again in Nature as ‘a sea of luminous silver white cloud’ (Leslie, 1885). Several authors linked these first observations to the Krakatoa eruption (1883), arguing that, before this date, not enough condensation nuclei and water vapor were present in the high atmosphere for allowing the NLC brilliance to exceed that of the twilight sky (Thomas et al., 1989). Dalin et al. (2012) highlighted the lack of data in the literature to support these assumptions. First, the amount of water vapor reaching the mesopause after very large volcanic eruptions is still unknown; second, there is still no agreement on why NLCs activity increases after some volcanic eruptions (e.g. Krakatoa in 1883, Bezymianny in 1956, Agung in 1963), but not all (e.g. Okataina in 1886, Mount Pelée and Santa Maria in 1902, Tarumai in 1909, Taal in 1911, Katmai in 1912).

However, it is unquestionable that since 1884/1885, frequent NLCs observations have been recorded. The first published sketch of an NLC was done in 1887 (Jesse, 1887); the first photographs were taken by Tseraskii at the end of the 19th century (Dalin et al., 2012), it has been even suggested that the sky in The Scream, the famous painting by Edvard Munch, was inspired by an NLC, while it is still controversy (Fikke et al., 2017; Prata et al., 2018 c.f., Odenwald, 2015). The first systematic NLC observational networks were developed in 1960s: in the Soviet Union (Bronshten and Grishin, 1970; Vasilyev, 1967; Villmann, 1968), in Poland (Kosibowa and Pyka, 1970), in England and in North America (Fogle and Haurwitz, 1966). Nowadays, space weather (http://spaceweather.com/) and the Noctilucent Cloud Observers (http://www.mcewan.co.uk/nlc/) website archive NLC observations and photo galleries worldwide. The Polar mesospheric clouds (PMCs), that is, the space-observed manifestation of NLCs, have been observed from space, over the last two decades, by various instruments designed initially for other purposes, such as airglow or ozone studies. More recently, various satellite missions have provided continuous information on PMCs observations, for example, SCIAMACHY, AIM, Odin, OSIRIS and SBUV (Llewellyn et al., 2004; Petelina et al., 2007, 2006).

Despite a significant increase in space-based PMC research in recent years, relatively little is known about their frequency changes over time as the data time coverage is limited. Recent studies point out that PMCs have increased in brightness and occurrence while decreasing in altitude over time (Lübken et al., 2018).

In ground-based research, time coverage of NLCs records starts earlier from the date of their first discovery. Prior to 1884, some observations of unusual and unknown phenomena were proposed as NLCs; however, because of brief descriptions, lack of details or quantifiable information, they were largely neglected or, after thorough analysis, discarded to be attributed to other phenomena (Butler, 2006; Schröder, 1999).

We have discovered a manuscript by Antonio Colla in which he described the observation of a ‘Phosphoric cloud’ with enough details and quantifiable information to assess whether he could have observed an NLC 45 years before the currently recognized date of the first NLC observation.

The observer

Antonio Colla (6 December 1806 to 10 March 1857) was a professor of Astronomy and Meteorology at the University of Parma, the city where he spent all his life. In 1834, he started as a meteorologist for the ‘Scuole Superiori di Parma’ (Anonymous, 1843), and later, in 1841, he became director of the Specola (an observatory institution mainly focused on meteorological studies) founded in 1757 by the Jesuit Jacopo Belgrado (1704–1789). Colla devoted his life to the observation of astronomical and meteorological phenomena, with diligence and accuracy that enabled him to register the smallest peculiarities. He was a skilful and patient comet observer and discoverer. He is reported to have discovered his first comet in the morning of 2 June 1845, the C/1845 L1 (Colla, 1845). The attribution of the discovery is correct, although, officially, the comet does not bear his name. After 10 days, this comet was observed by his colleagues in Geneva, Munich and Leipzig. He discovered a second comet on 7 May 1847, the C/1847 J1 (Colla, 1847a). It appeared as a small nebula without a tail, not easily perceptible via telescope, so the astronomer J. R. Hind, in Astronomische Nachrichten on 15 July 1847 (Hind, 1847), is said to have been surprised that Colla succeeded in revealing an object so weak (Colla, 1847b). Finally, Colla discovered a third comet, the C/1854 Y1 Winnecke-Dien, but due to delayed communication during the Christmas period, he was not included among the official discoverers (Colla, 1855).

Colla was recognized by contemporary scientists. He was a member of many scientific societies and academies such as the Accademia Fisico-Medico-Statistica in Milan and the Sciences Academy of the Bologna Institute (Anonymous, 1844). He was in contact and corresponded with the most famous scientists and astronomers of his time, for example, J. R. Hind, K. C. Bruhns, F. W. Herschel, G. Santini, G. Plana, G. Bianchi and F. Carlini (Colla, 1844). He was a well-known scientist and published internationally in scientific journals (average of four articles per year), preferring the French language and the l’Institut journal. His scientific production increased after he was appointed director of the Specola (i.e. after 1841), as well as in conjunction with his comet discoveries.

Colla subdivided his time studying astronomical and meteorological phenomena. He disseminated results from his daily meteorological observations to the general public in Gazzetta Ufficiale di Parma and in his Giornale Astronomico ad uso comune di Antonio Colla (Astronomical Journal for Common Use by Antonio Colla, 1840a) among others. The last series of Giornale Astronomico followed the scope and format of the previous Giornale Astrometeorologico published by Giuseppe Toaldo in Venice (and later in Padua, Italy) from 1773 to 1797 (Pigatto, 2000; Toaldo, 1780). Colla collected meteorological data with high-quality instruments and after 1844 – when the Central Italian Meteorological Service was established – he started an intense exchange with the founder Vincenzo Antinori, sending him observations made in Parma (Colla, 1844).

In short, it is clear that he was an up-to-date scientist and he had the knowledge and skills to make his own discoveries. He was a very experienced, patient and meticulous observer.

Analysis of the observation

The original manuscript entitled Notice about an extraordinary bright phenomenon (Figure 1), in which Colla describes in detail the NLC observation from 18 June 1840, is preserved at the University of Florence in the Physics department’s library (Colla, 1840b). The most important fragments of the NLC description are reported in the following in English. Moreover, Figure 2 shows a diagram of the observation to clarify its geometry:

. . . I want to talk about a kind of bright cloud of phosphoric light

. . . happened the evening of the 18th last June [1840], since in that moment, I was able to observe it together with a Professor of Physics, my correspondent, who was passing through this city [Parma].

The observation made by us was the following:

At 9 p.m. in the evening of the 18th of June of this year, when the sky was partially covered by storm clouds, blowing [for a quite long period of time] a south-westerly wind, and ranging in temperature between 18 and 19 degrees over an eighty-degree scale, we observed from the observatory’s handrail in the southward direction some gleams of very bright intensity of yellow color and almost all in zig-zag shape rising upward – in all directions – from a point in the sky occupied by a kind of bright cloud of phosphoric light and round shape; its apparent diameter could range from 15 to 20 degrees, and its elevation from 4 to 5 on the top of the Apennine mountains.

The glitters manifest themselves at some arcsecond intervals and they rose immutably from the phosphoric cloud described above, and, without changing shape, remained bright continuously until 10 and a quarter, i.e. up to the moment in which the sky – in that direction and toward south-east – was covered with dark clouds with elongated shapes and variable heights, that very soon was the place of a very noisy thunderstorm that continued until a quarter past midnight.

The suspicion that this permanent light that lasted for one and a quarter hours could have originated from a northern light, from the twilight, from the moon or from the Milky Way vanishes very quickly, when considering that on this evening there was no northern light of any type in our horizon, that in contrast [i.e. the phosphoric light was observed as an unchanged luminous phenomenon] the twilight was already in considerable luminous decrease, that the moon was below the horizon, and finally, that the Milky Way was covered by clouds [i.e. tropospheric clouds] and distant eastward – many degrees –from the luminous phenomenon.

Figure 1.

(a) First page of the manuscript reporting the NLC observation made by Antonio Colla on 18 June 1840 in Parma (Foglio 5 in Manuscript 4-80:3-7. Inventory number 16621 owned by Library of the Physics Institute of the Florence University). Available at https://bibdig.museogalileo.it/Teca/Viewer?an = 000000010570. (b) Photographic portrait of Antonio Colla made ca. in 1850 (photo in public domain, rearranged by the portrait in ‘I grandi astronomi di Parma’, ed. La Famija Pramzana, Parma 2009).

Figure 2.

Visual reproduction of the NLC observation made by Antonio Colla, 18 June 1840 in Parma.

This description fits perfectly with an NLC observation. A ‘bright cloud of phosphoric light’ is an accurate description of an NLC. The description of ‘some gleams of very bright intensity of yellow color and almost all in zig-zag shape’ . . . ‘the glitters, manifest themselves at some arcsecond intervals [i.e. the spatial interval between the glittering lasted “some arcseconds”] and they rose immutably’ could be interpreted as the typical ripples structures of NLCs. Moreover, Colla discards most of the phenomena that have been traditionally misunderstood as being NLC: ‘northern light’, ‘twilight’, ‘moon’ or ‘Milky Way’. They were well-known phenomena for Colla and frequently recorded in his astronomical and meteorological diaries. Fortunately, Colla provided quantifiable information on date, time, duration and direction of observation. All this quantifiable information has been analyzed in relation to the known best conditions for observing an NLC in order to evaluate the reliability of his observation.

Seasonality

In the literature, the traditional seasonal boundary for observing an NLC ranges from −40 days from summer solstice (DFS; i.e. after 11 May) to about +65 DFS (i.e. 25 August) with a maximum around 4–7 July (Gadsden, 1998). Colla’s observation was made the 18 June on −3 DFS, perfectly fitting within this range and close to the maximum.

Depression angle and twilight sky area

Colla describes the starting and ending hour of the NLC observation as local solar time (LST) called Ore italiane or Italians hours. In Italy, after the 15th century, this measure of time entered in general use and lasted until 1860. The Italian hours were given in LST without considering the summer time change of 1 h. This time needs to be corrected to the currently used time in NLCs observations, that is, the universal time (UT), a time standard based on Earth’s rotation. UT is the mean solar time on the Prime Meridian at Greenwich. The correction equation is

UT = CET - 1 = (LST + Et + C) - 1

(1)

where CET is one time zone east of Greenwich (UT + 1) and C is the local constant or the local meridian correction of the place where the observer is located, which corresponds to 18 min 44 s at the Specola in Parma (44° 48′ N, 10° 19′ E), while Et is the equation of time that can be neglected as the day of observation is very close to the summer solstice.

Therefore, the starting time of the observation 21:00 (LST) corresponds to 20:18:44 (UT) and the end of ‘permanent light that lasted for one and a quarter hours’ to 21:33:44 (UT). If we correct this measure for the error as calculated by Dominici and Marcelli for horary measurements in Italy before 1866 (Dominici and Marcelli, 1979), the Colla’s observational time becomes 20:03:44 (UT) and 21:18:44 (UT).

The twilight hours for 18 June 1840 were simulated, for the Parma site, using a tool provided by the data service of the Astronomical Applications Department of the U.S. Naval Observatory (https://www.esrl.noaa.gov/gmd/grad/solcalc/azel.html). According with this simulation, Colla’s observation was mainly during the civil and nautical twilight when the sun was from 0.5° to 10.1° below the horizon. NLCs are usually observed with these solar angles, because dark is enough and lower Cirrus clouds are not illuminated by the sun (Gadsden and Schröder, 1989).

Following the methodology of Gadsen (1998), we have computed the region of the sky, which is possible to observe an NLC for the day and times of the Colla’s observation.

Two coordinates define the local twilight area that an NLC can occupy at various degrees of solar depression angles (β): the degrees (°) from the sunset direction (i.e. sun position minus Azimuth angle from the North (Δa)) and the maximum height above the horizon (i.e. the elevation angle (e)). The elevation angle (i.e. y-axis in Figure 3) is calculated from equation (4) in Gadsen (1998)

\cos e = \frac{\sin (β - ϑ)}{\sqrt{1 + \cos^{2} ϑ - 2 \cos ϑ \cos (β - ϑ)}}

(2)

In which \cos ϑ = \frac{R}{R + H}

(3)

Figure 3.

Upper edge of the twilight sky determining the illuminated area of an NLC (i.e. plan Δa–e) for solar depression angles of −0.5° and −9.1°, corresponding to the start (20:03 UT) and end (21:10 UT) of the observation made by Antonio Colla on 18 June 1840 and the sky area in which Colla saw the NLC.

R being the radius of the local sphere taking into account an ellipsoidal model for the Earth’s surface (6378 km) and H being the estimated height of the NLC (i.e. 85 km).

The angle Δa (x-axis in Figure 3) is calculated from equation (4) that refers to the geometry of an NLC observation from the Earth as reported in Gadsen (1998)

S^{2} \cos^{2} a = {(Y - R \sin β)}^{2} + {[R \cos β - (R + h) \cos ϑ]}^{2}

(4)

If an observer is at 0, where the solar depression angle is β, he sees the NLC as a point on the edge of the illuminated area of NLC at a slant range, S. The Earth (radius R) projects a cylindrical shadow which is increased in radius by the screening height h (i.e. 5 km). A point on the edge of the shadow lies at a distance (R + H) from the center of the Earth where H is the height of the noctilucent cloud.

S and Y are calculated from equations (5) and (6)

S^{2} = {(R + H)}^{2} - 2 Y R \sin β + R^{2} - 2 R (R + h) \cos ϑ \cos β

(5)

Y^{2} = {(R + H)}^{2} - {(R + h)}^{2}

(6)

Figure 3 shows the upper edge of the twilight sky determining the illuminated area of an NLC at 85 km for two extreme cases with solar depression angles of −0.5° (Sun position angle is 304.7°) and −10.1° (Sun position angle is 318.3°), corresponding to the start (20.03:44 UT) and end (21:18:44 UT) time of the observation made by Antonio Colla on 18 June 1840. The sky area in which Colla saw the NLC is clearly reported, that is, ‘we observed from the observatory’s handrail in the southward direction’, ‘5º on the top of the Apennine mountains’ and an ‘apparent diameter could range from 15 to 20 degrees’, this area in the local sky corresponds to (180°–150.5°) < h < (180°–174.5°) and 168° < a < 192°, considering 1° the elevation of the Apennines regarding to the observer and defining southward as a direction of 20° ± 2° centered on South (i.e. 180°). This region has been represented as a red square in Figure 3. As we can see, the area in which Colla saw the phosphoric cloud fall into the region in which an NLC can be illuminated by the sun at the time of the observation. For the night of observation, if the sun had been more than 12° below the horizon, an NLC could be seen only close to the horizon in the direction of the sun (e.g. for Colla’s observational conditions, 321°, that is, NW direction). However, if the sun had been less than 9° below the horizon, the clouds can be seen at the zenith and beyond, in the antisolar hemisphere, that is, within the Colla’s observational conditions, Southern direction. At the beginning of the Colla’s observation (20:03 CET), when the sun is 0.5° below the horizon – in the local sky in Parma – an NLC can be observed 133° from the sunset direction (i.e. Southern), and it has an elevation angle of 179.6° that extends above the horizon up to 89.6° past the zenith (i.e. ≈0.4° above the horizon southern direction; Table 1). Therefore, it could be possible to observe an NLC in the area carefully described by Colla. He could see an NLC in southward direction, from 20:03 up to 21:10 (i.e. up to a depression angle β = 9.1). The phenomenon lasting for 1 h and 6 min and not for 1 h and 15 min as Colla describes. Nevertheless, Colla’s description of the observation time is not highly detailed, that is, ‘at 9 p.m.’ ‘until 10 and a quarter’, so differences till 15 min can be accepted as systematic error (Dominici and Marcelli, 1979).

Table 1.

Depending on degrees of solar depression angle, the table reports the portions of the local sky that an NLC at H = 85 km can occupy in the various twilight areas. The twilight area coordinates are (1) elevation angle (°) or maximum height above the horizon and (2) sun position minus azimuth angle from the north (°). The degrees (°) from the sunset direction and the time (UT) of the solar depression angle are also reported.

h _NLC	Solar depression angle (°)	Elevation angle (°)	Azimuth angle (°) screen heights = 5 km	Solar direction	Time (UT)
85	0.5	179.6	47.0	304.7	20:03
85	2.3	177.4	53.6	307.0	20:18
85	6.0	168.8	71.3	311.8	20:45
85	7.0	163.0	77.0	313.1	20:58
85	8.0	150.6	83.0	314.0	21:01
85	9.1	105.7	90.3	316.4	21:10
85	10.1	43.4	95.8	318.3	21:18

A non-expert observer can misinterpret a tropospheric clouds (cirrus) reflected by the sun as NLC. Although the great experience of Colla observing the sunset sky could discard this possibility, we have simulated the Colla observation, in the case the phenomenon observed by Colla was a Cirrus at an height ranging from 6 to 12 km. Cirrus at an elevation of 6 km could be observed, in the Colla’s conditions only for about 10 min from 20:06 to 20:18; Cirrus at 9 km up to 20:20 and at 12 km up to 20:26. The duration of such observational phenomenon in such a case should be described as lasting 10 or 20 min at maximum and not an hour and a quarter as Colla reported in his observations.

Disturbing factors (e.g. aurorae, twilight, moon, Milky Way)

Colla describes other astronomical phenomena that, to a non-expert observer, might look like the luminescent cloud he was observing. He analyzed the following: (1) northern light, (2) twilight, (3) moon and (4) the Milky Way. As an astronomer and meteorologist, he disproved them, one by one, very easily as follows: (1) Colla reports that on 18 June 1840, ‘there was no northern light of any type in our horizon’. This description is consistent with the absence of known auroral records on that night (Angot, 1896; Krivsky and Pejml, 1988), as aurora is a global phenomenon and is expected simultaneously observed somewhere else as well (Hayakawa et al., 2019a; Silverman, 2006). (2) Colla observes that ‘in contrast the twilight was already in considerable luminous decrease’. This is confirmed by the simulation we made in the previous section. (3) Colla reports that ‘the moon was below the horizon’. This assertion is also corroborated by the simulation of the moon phase, moonrise and moonset. The moon, on 18 June 1840, rose at 21:00 (UT) and at the end of Colla’s NLC observation. (4) Finally, Colla reports ‘the Milky Way was covered by clouds [i.e. tropospheric clouds] and distant eastward – many degrees – from the luminous phenomenon’. This statement is confirmed by the Milky Way’s position observed in Parma during the summer solstice, that is, an ‘arch’ of stars that was several degrees far (eastward) from his direction of observation (i.e. in the southward direction toward the zenith over the twilight).

Unusual features of the Colla report

It is difficult to imagine an explanation different to an NLC for the phosphoric cloud observed by Colla, taking into account the description, the perfect fit in the local sky illuminated by the twilight and the phenomena discarded by the observer. Nevertheless, there are some unusual points on the interpretation of the phenomenon in the Colla description as an NLC, which we want to clarify: (1) the description ‘without changing shape’ is very unusual for the NLC dynamics, which is highly variable in time and space; (2) the ‘bright intensity of yellow color’, yellow is absent in the twilight sky due to Rayleigh scattering and typical aerosol loading in the troposphere; and (3) the ‘round shape’ is uncommon in NLC. To interpret these points is important to take into account that Colla described a phenomenon unknown for him and for the science at the time of the observation. Therefore, we cannot expect that the description provides all the information that nowadays we consider relevant to identify an NLC. Because simply he did not know the main characteristics of an NLC. Nowadays, this circumstance can happen with still no well-defined atmospheric phenomena, for example, ball lightings (Domínguez-Castro, 2018; Stenhoff, 1999). Taken this into account, when Colla reported that the glitters ‘rose immutably from the phosphoric cloud described above, and, without changing shape’ probably he wanted to discriminate this phenomenon from other highly dynamic atmospheric phenomena as lightning, dynamical aurorae borealis or from other clouds as we do nowadays describing the NLC as dynamics. About the shape and the color, it is important to note that Colla saw the NLC within small solar depression angles. In these circumstances, the contrast of NLCs is very low with related difficulties in distinguishing their characteristic silver color and undistorted pattern by oblique viewing (Gadsden and Parviainen, 1995). Moreover, it is important to note that in his description of the phenomenon, Colla mentioned three times phosphoric color and only one time the yellow color. Therefore, the ‘yellow’ appearance can be explained considering the early time of the observation with a low depression angle (i.e. $β < 3 °$ ) which makes it difficult to distinguish their characteristic bluish and whitish luminescent colors through the reddening by atmospheric absorption in the twilight luminosity on the background.

Another anomalous aspect of the Colla’s observation is the low latitude of observation. Colla observed the NLC at 44° 48′ N, that is, more than 5° below the traditional equatorward boundary of NLC observations (50°–75°). Midlatitude NLCs are unusual, but since 1998 more than 280 NLC sightings have been recorded below 50°N by Noctilucent Cloud Observers (http://ed-co.net/nlcnet/). NLCs have been observed as far south as Calar Alto (37° 13′ N; Baumgarten et al., 2009), which is more than 10° below the latitude of Parma. The factors that control the appearance of NLCs in low latitudes are still under discussion. Recent studies (Gerding et al., 2013; Russell et al., 2014) based on satellite observations and simulations consider the dynamics (e.g. planetary wave anomaly and wind advection) and the water (in a minor role that controls the cloud frequency variation) as the most influencing factors for midlatitude (40° N–55° N) NLC observations rather than the mesospheric temperature.

Discussion and conclusion

We have found a record of an NLC observation on 18 June 1840. The NLC was observed by Antonio Colla and described as a ‘bright cloud of phosphoric light’. This observation is the most accurate of all those proposed in the literature as a possible NLC prior to 1885 (Butler, 2006; Leslie, 1884; Schröder, 1999). Thanks to the accuracy of Colla’s report, we have been able to analyze quantitatively his reported information and facts that contribute in confirming he was able to observe an NLC that night. The date of the observation fits the seasonal range when it is more likely to observe NLCs. The observational time coincides with a solar depression angle ranging from 0.5° to 9.1°. The altitude and direction of the observation are in the range in which it is possible to observe an NLC. In conclusion, we can state that an NLC was observed on the night of 18 June 1840 in Parma by Antonio Colla.

Colla’s NLC observation took place 45 years before the first previously known recorded NLC observation. This confutes the recurrent mentioned hypothesis in literature that there are no suitable conditions in the mesosphere to observe NLCs prior to the Krakatoa eruption.

The natural forces (solar and volcanic) in 1840 were not extraordinary (Figure 4). In many works (Dubietis et al., 2010; Gerding et al., 2013; Robert et al., 2009), a slight anti-correlation between solar activity (through sunspot number analysis) and NLCs has been detected as the solar activity influences the temperature of the mesosphere. The descending phase of the solar cycle is when NLCs are more frequent. The year 1840 is in the middle of the decrease phase of solar cycle 8. This confirms that Colla had good conditions to observe an NLC phenomenon, although not exceptional ones. The injection of high quantities of H₂O in the mesosphere by high explosive tropical eruptions has been proposed as the main factor to increase the luminosity of the NLC and made it eye naked visible from the ground in the pre-industrial period. This is the most common argument to explain the absence of NLC observations previous to the Krakatoa eruption. Nevertheless, the amount of water vapor injected by the historical eruptions in the mesosphere is unknown (Dalin et al., 2012). For this reason, there is a great uncertainty in the lag between volcanic eruptions and NLCs observations or the time range of NLC visibility after these eruptions.

Figure 4.

Annual sunspot numbers (Version No. 2.0, WDC-SILSO, Royal Observatory of Belgium, Brussels; Clette and Lefèvre, 2016; Clette et al., 2015, 2014) and large volcanic eruptions from 1800 to 2015.

The volcanic activity during the Colla observation is low. The closest explosive eruption was that of the Cosigüina volcano (Nicaragua) 5 years earlier (22 January 1835). This was a highly explosive eruption (Volcanic Explosivity Index (VEI) = 5) but lower than Krakatoa (VEI = 6). The Cosigüina eruption had been considered an eruption with low climate impact (Self et al., 1981), but recently found that the eruption had a climate impact similar to Pinatubo (VEI = 6) with a decrease of northern hemisphere temperature of about 0.4°C for 2–3 years after the eruption (Longpré et al., 2014). However, the 5 years that separate the Cosigüina eruption and Colla’s observation reduce the volcano’s possible influence on triggering an NLC phenomenon. Moreover, precisely the Mount Pinatubo eruption coincides with a decrease of NLC nights in the following years (Gadsden, 1998; Thomas and Olivero, 2001). To sum up, we cannot identify exceptional solar or volcanic forces during 1840 and the previous years.

As with the novelty of Colla’s observation in his epoch, many climate events have been detected recently and cataloged as new events; this is the case, for example, with the 2005 Vince Hurricane (Franklin, 2006). Vaquero and collaborators undertook intense archival works to find events that had parallels in the past and found similar hurricanes in 1840 (Vaquero et al., 2008) and 1724 (Domínguez-Castro et al., 2013). We consider that systematic archive research can bring to light new NLCs observations prior to the established discovery date in literature (1885) or even prior to 1840. Great efforts have been done recently in the search and digitalization of early meteorological observations both in terms of European-funded research projects, such as IMPROVE (Improved understanding of past climatic variability from early daily European instrumental) and MILLENNIUM (European climate of the last millennium) and as publications (Allan et al., 2011; Brönnimann et al., 2019; Camuffo and Bertolin, 2012; Camuffo and Jones, 2002; Domínguez-Castro et al., 2014, 2017) with a focus mainly on pressure, temperature and rainfall measurements. Nevertheless, other records such as the descriptions of the sky have not been systematically explored. These observational fields in the records are a jumble and difficult to process, but frequently include extraordinary sky events such as atmospheric optics and aurorae (Domínguez-Castro et al., 2016; Hayakawa et al., 2019b). In our opinion, descriptions of the state of the sky in meteorological records from the 18th and 19th centuries are good sources to start a systematic search of other possible NLC observations prior to 1885.

Footnotes

Acknowledgements

The sunspot records are courtesy of WDC-SILSO, Royal Observatory of Belgium, Brussels.

Funding

This work was supported through the financial support guaranteed by the Onsager Fellowship – Research Excellence Program at the Norwegian University of Science and Technology (NTNU) in Trondheim, Norway and CLICES project (CGL2017-83866-C3-1-R) financed by the Ministry of Economy and Competitiveness of the Spanish Government.

ORCID iD

Fernando Domínguez-Castro

References

Allan

Brohan

Compo

, et al. (2011) The international atmospheric circulation reconstructions over the earth (ACRE) initiative. Bulletin of the American Meteorological Society 92(11): 1421–1425:. doi:10.1175/2011BAMS3218.1

Angot

(1896) The Aurora Borealis. Kegan Paul, Trech, Truber, and Company.

Anonymous (1843): Indice Cronologico della raccolta generale delle leggi per gli stati di Parma [Chronological Index of the general collection of laws for the states of Parma].

Anonymous (1844) Atti della Quinta Unione degli scienziati italiani [Acts of the Fifth Union of Italian Scientists].

Backhouse

(1885) The luminous cirrus clouds of June and July. Met. Mag 20: 33.

Baumgarten

Gerding

Kaifler

, et al. (2009) A trans European network of cameras for observation of noctilucent clouds from 37° N to 69° N. 9th ESA Symposium on European Rocket and Balloon Programmes and Related Research,. Bad Reichenhall.

Brönnimann

Allan

Ashcroft

, et al. (2019) Unlocking pre-1850 instrumental meteorological records: A global inventory. Bulletin of the American Meteorological Society. doi: 10.1175/BAMS-D-19-0040.1

Bronshten

Grishin

(1970) Serebristye Oblaka (Noctilucent Clouds). Moscow: Nauka.

Butler

(2006) Possible observations of noctilucent clouds by Thomas Romney Robinson. Weather 61(5): 143–144.

10.

Camuffo

Bertolin

(2012) The earliest temperature observations in the world: The Medici Network (1654–1670). Climatic Change 111(2): 335–363.

11.

Camuffo

Jones

(2002) Improved Understanding of past Climatic Variability from Early Daily European Instrumental Sources. Dordrecht: Kluwer Academic Publishers.

12.

Clette

Lefèvre

(2016) The New Sunspot Number: Assembling All Corrections. Solar Physics 291: 2629–2651.

13.

Clette

Cliver

Lefèvre

, et al. (2015) Revision of the Sunspot Number(s). Space Weather 13: 529–530.

14.

Clette

Svalgaard

Vaquero

, et al. (2014) Revisiting the Sunspot Number. A 400-Year Perspective on the Solar Cycle. Space Science Reviews 186: 35–103.

15.

Colla

(1840a) Nogiornale Astronomico Ad Uso Comune [Astronomical Journal for Common Use]. Tipografia Ferrari.

16.

Colla

(1840b) Nota intorno ad un fenomeno luminoso straordinario [Notice about an extraordinary bright phenomenon].Manuscript 4-80:3-7. Inventory number 16621 owned by Library of the Physics Institute of the Florence University.

17.

Colla

(1844) Letters 1–65 in the Central Italian Meteorological Archive: AMCI fund. AMCI – Registri Della Corrispondenza. Parma: 1844–1849.

18.

Colla

(1845) Sopra la cometa scoperta a Parma nel giorno 2 Giugno 1845 [Over the comet discovered in Parma on June 2, 1845]. Annali Delle Scienze Del Regno Lombardo-Veneto 14: 225–230.

19.

Colla

(1847a) Sur la nouvelle comete telescopique decouverte a Parme le 7 Mai 1847 [On the new telescopic comet discovered in Parma on May 7, 1847]. L’Institut 15: 295–296.

20.

Colla

(1847b) Nachrichten uber die von Colla entdeckten Cometen [News about the comets discovered by Colla]. Astronomische Nachrichten 25: 395–398.

21.

Colla

(1855) Observations de la quatrieme comete de 1854 [Observations of the Fourth Comet of 1854]. Astronomische Nachrichten 39: 333–334.

22.

Dalin

Pertsev

Romejko

(2012) Notes on historical aspects on the earliest known observations of noctilucent clouds. History of Geo- and Space Sciences 3: 87–97.

23.

Domínguez-Castro

(2018) An early record of ball lightning: Oliva (Spain), 1619. History of Geo- and Space Sciences 9(1): 79–83.

24.

Domínguez-Castro

Trigo

Vaquero

(2013) The first meteorological measurements in the Iberian Peninsula: Evaluating the storm of November 1724. Climatic Change 118(2): 443–455:. doi:10.1007/s10584-012-0628-9

25.

Domínguez-Castro

Vaquero

Bertolin

, et al. (2016) Aurorae observed by Giuseppe Toaldo in Padua (1766-1797). Journal of Space Weather and Space Climate 6: A21. doi: 10.1051/swsc/2016016

26.

Domínguez-Castro

Vaquero

Gallego

, et al. (2017) Early meteorological records from Latin-America and the Caribbean during the 18th and 19th centuries. Scientific Data 4: 170169. doi:10.1038/sdata.2017.169

27.

Domínguez-Castro

Vaquero

Rodrigo

, et al. (2014) Early Spanish meteorological records (1780-1850). International Journal of Climatology 34(3): 593–603:. doi:10.1002/joc.3709

28.

Dominici

Marcelli

(1979) Historical evolution of the horary measurements in Italy. Annals of Geophysics 32(1): 131–212.

29.

Dubietis

Dalin

Balčiunas

, et al. (2010) Observations of noctilucent clouds from Lithuania. Journal of Atmospheric and Solar-terrestrial Physics 72(14–15): 1090–1099.

30.

Fikke

Kristjánsson

Nordli

(2017) Screaming clouds. Weather 72(5): 115–121.

31.

Fogle

Haurwitz

(1966) Noctilucent clouds. Space Science Reviews 6: 279–340.

32.

Franklin

(2006) Tropical Cyclone Report: Hurricane Vince, 8–11 October 2005.

33.

Gadsden

(1998) The north-west Europe data on noctilucent clouds: A survey. Journal of Atmospheric and Solar-terrestrial Physics 60(12): 1163–1174.

34.

Gadsden

Parviainen

(1995) Observing Noctilucent Clouds. Available at: www.iaga-aiga.org/data/uploads/pdf/guides/onc.pdf

35.

Gadsden

Schröder

(1989) Noctilucent Clouds. Berlin: Springer, Heidelberg.

36.

Gadsen

(1998) Can I see a noctilucent clouds? Journal of British Astronomical Association 108(1): 35–38.

37.

Gary

(1991) Mesospheric clouds and the physics of the mesopause region. Reviews of Geophysics 29: 553–575.

38.

Gerding

Höffner

Hoffmann

, et al. (2013) Noctilucent cloud variability and mean parameters from 15 years of lidar observations at a mid-latitude site (54-n, 12-e). Journal of Geophysical Research Atmospheres 118(2): 317–328.

39.

Hayakawa

Ebihara

Cliver

, et al. (2019b) The extreme space weather event in September 1909. Monthly Notices of the Royal Astronomical Society 484: 4083–4099.

40.

Hayakawa

Ebihara

Willis

, et al. (2019a) Temporal and Spatial Evolutions of a Large Sunspot Group and Great Auroral Storms around the Carrington Event in 1859. Space Weather 17(11): 1553–1569. doi: 10.1029/2019SW002269

41.

Hervig

Berger

Siskind

(2016) Decadal variability in PMCs and implications for changing temperature and water vapor in the upper mesosphere. Journal of Geophysical Research 121(5): 2383–2392.

42.

Hind

(1847) Schreiben des Herrn J.R. Hind an den Herausgeber. Astronomische Nachrichten 26(4): 49–50.

43.

Jesse

(1885) Auffallende Abenderscheinungen am Himmel im Juni und Juli [Striking evening appearances in the sky in June and July]. Meteorol. Zeitschr 3(21): 8–15.

44.

Jesse

(1887) Die Hohe der leuchtenden (silbernen) Wolken [The height of the glowing (silver) clouds]. Meteorol 4: 424.

45.

Jesse

(1889) Die leuchtende Nachwolken [The bright afterglow]. Himmel Und Erde 1: 263–286.

46.

Kosibowa

Pyka

(1970) Wyniki obserwacji nad ob okami mezosferyczynmi w Polsche w 1969 [Observations of clouds on mesosphere in Poland in 1969]. Przegl. Geofiz 15: 357–366.

47.

Krivsky

Pejml

(1988) Solar Activity, Aurorae and Climate in Central Europe in the Last 1000 Years. Publications of the Astronomical Institute of the Czechoslovak Academy of Sciences.

48.

Leslie

(1884) The sky-glows. Nature 30: 583.

49.

Leslie

(1885) Sky Glows. Nature 32: 245.

50.

Llewellyn

Lloyd

Degenstein

, et al. (2004) The OSIRIS instrument on the Odin spacecraft. Canadian Journal of Physics 82(6): 411–422.

51.

Longpré

M-A

Stix

Burkert

, et al. (2014) Sulfur budget and global climate impact of the A.D. 1835 eruption of Cosigüina volcano, Nicaragua. Geophysical Research Letters 41(19): 6667–6675.

52.

Lübken

F-J

Berger

Baumgarten

(2018) On the Anthropogenic Impact on Long-Term Evolution of Noctilucent Clouds. Geophysical Research Letters 45(13): 6681–6689.

53.

Odenwald

(2015) Solar Storms: 2000 Years of Human Calamity. San Diego, Createspace.

54.

Petelina

Llewellyn

Degenstein

(2007) Properties of polar mesospheric clouds measured by Odin/OSIRIS in the northern hemisphere in 2002-2005. Canadian Journal of Physics 85(11): 1143–1158.

55.

Petelina

Llewellyn

Degenstein

, et al. (2006) Odin/OSIRIS limb observations of polar mesospheric clouds in 2001-2003. Journal of Atmospheric and Solar-terrestrial Physics 68(1): 42–55.

56.

Pigatto

(2000) Giuseppe Toaldo e il suo tempo nel bicentenario della morte [Giuseppe Toaldo and his time in the bicentennial of death]. Scienze e lumi tra Veneto e Europa [Science and knowledge between the Veneto region and Europe]. Padova: Bertoncello, 1033.

57.

Prata

Robock

Hamblyn

(2018) The sky in Edvard Munch’S the scream. Bulletin of the American Meteorological Society 99(7): 1377–1390.

58.

Robert

von Savigny

Burrows

, et al. (2009) Climatology of noctilucent cloud radii and occurrence frequency using SCIAMACHY. Journal of Atmospheric and Solar-terrestrial Physics 71(3–4): 408–423.

59.

Russell

III Rong

Hervig

, et al. (2014) Analysis of northern midlatitude noctilucent cloud occurrences using satellite data and modeling. Journal of Geophysical Research 119(6): 3238–3250.

60.

Schröder

(1999) Were Noctilucent Clouds Caused by the Krakatoa Eruption? A Case Study of the Research Problems before 1885. Bulletin of the American Meteorological Society 80(10): 2081–2085.

61.

Self

Rampino

Barbera

(1981) The possible effects of large 19th and 20th century volcanic eruptions on zonal and hemispheric surface temperatures. Journal of Volcanology and Geothermal Research 11(1): 41–60.

62.

Silverman

(2006) Comparison of the aurora of September 1/2, 1859 with other great auroras. Advances in Space Research 38: 136–144.

63.

Stenhoff

(1999) Ball Lightning: An Unsolved Problem in Atmospheric Physics. New York: Kluwer Stott Academic Publisher.

64.

Thomas

(1996) Is the polar mesosphere the miner’s canary of global change? Advances in Space Research 18(3): 149–158.

65.

Thomas

(2003) Are noctilucent clouds harbingers of global change in the middle atmosphere? Advances in Space Research 32(9): 1737–1746.

66.

Thomas

Olivero

(2001) Noctilucent clouds as possible indicators of global change in the mesosphere. Advances in Space Research 28(7): 937–946.

67.

Thomas

Olivero

Jensen

, et al. (1989) Relation between increasing methane and the presence of ice clouds at the mesopause. Nature 338(6215): 490–492.

68.

Toaldo

(1780) Giornale astro-meteorologico [Astro-meteorological Journal]. Gaspare Storti alla Fortezza

69.

Tserasky

(1890) Sur les nuages lumineux [On the bright clouds]. Annales De L’observatoire De Moscou 2: 176–180.

70.

Turco

Toon

Whitten

, et al. (1982) Noctilucent clouds: Simulation studies of their genesis, properties and global influences. Planetary and Space Science 30: 1147–1181.

71.

Vaquero

García -Herrera

Wheeler

, et al. (2008) A historical analog of 2005 Hurricane Vince. Bulletin of the American Meteorological Society 89(2): 191–201.

72.

Vasilyev

(1967) Astrofizi eskie issledovanija serebristych oblakov [Astrophysical investigations of noctilucent clouds]. Inf. Soobš. (Moskva) 4.

73.

Villmann

(1968) Statisti eskie dannye o pojavlenii serebristych oblakov za period MGSS (1964–1965 gg.) [Extensional data on the appearance of noctilucent clouds during the period of the MGSS (1964-1965)]. Astron. Vestn. Mosk 2: 161–169.

74.

von Zahn

(2003) Are noctilucent clouds truly a ‘Miner’s canary’ for global change? Eos 84.

75.

von Zahn

Baumgarten

Berger

, et al. (2004) Noctilucent clouds and the mesospheric water vapour: The past decade. Atmospheric Chemistry and Physics 4(11–12): 2449–2464.

76.

VV. AA. (Various Authors – group of amateur astronomers and argonauts) (2009) I grandi Astronomi di Parma all’epoca del ducato. La Famija Pramzana Editor: pp131.