- Email: [email protected]
Mesospheric cloud observations at unusually low latitudes
Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
www.elsevier.com/locate/jastp
Mesospheric cloud observations at unusually low latitudes M.J. Taylora; ∗ , M. Gadsdenb , R.P. Lowec , M.S. Zalcikd , J. Brausche a Space
Dynamics Laboratory and Department of Physics, Utah State University, 4415 Old Main Street, Logan, UT 84322-4415, USA b 12 Keir Street, Perth, Scotland, UK c Department of Physics, University of Western Ontario, London, Ont., Canada d 14130 80th Street, Edmonton, Alberta, Canada e 208 South G Street, Glen Ullin, ND, USA
Abstract Noctilucent clouds (NLC) are a beautiful, high-latitude, summertime phenomenon that was 2rst reported over 100 years ago. They are seen during the hours of twilight by the scattering of sunlight from sub-micron-sized ice particles that form in the vicinity of the cold mesopause region. NLC are quite distinctive, often appearing silvery-blue in color. In recent years there has been a marked increase in their frequency of occurrence, possibly due to an increase in mesospheric water vapor and=or to a cooling of the mesopause region, prompting speculation that they are “harbingers” of potentially serious changes in the mesospheric climate. In concert with this trend there are also a growing number of ground-based NLC sightings at signi2cantly ◦ lower latitudes than expected. Here we report two unusual NLC displays photographed from Logan, UT, USA (∼42 N) ◦ during June 1999, well over 10 lower in latitude than expected and implying a major, yet temporary, departure from normal mid-latitude summertime conditions. These data provide new evidence for the occasional expansion of NLC to unusually low latitudes possibly due to exceptional dynamical forcing. Alternatively, they may be an early indicator of signi2cant long-term c 2002 Elsevier Science Ltd. All rights reserved. changes taking place in the upper mesospheric summertime environment. Keywords: Noctilucent clouds; Mesospheric clouds; Mesospheric climate
1. Introduction From their very 2rst observation in the twilight sky noctilucent clouds (NLC) were accepted as being a most unusual phenomenon. Well over a 100 years ago Leslie (1884) described their appearance as “weird small cloud forms, at times very regular, like ripple marks in the sand, or the bones of some great 2sh” seen late in the evening twilight. Within a few years of their 2rst sightings, ground-based photographic triangulation measurements by Jesse (1896) had determined that NLC were by far the highest clouds on earth, residing in the upper mesosphere at heights close to 82 km. Subsequent research has shown that these mesospheric clouds are most often seen from the ground at middle to high latitudes ∗
Corresponding author. Fax: +1-435-713-0054. E-mail address: [email protected] (M.J. Taylor).
(∼55–65◦ ) and that they are a product of the intense cooling that occurs in the summertime mesopause region at polar latitudes (e.g. Gadsden and SchrFoder, 1989). Observations from satellites (e.g. Donahue et al., 1972; Thomas, 1991; Evans et al., 1995) have signi2cantly extended the realm of NLC studies and have revealed the frequent presence of mesospheric clouds (sometimes referred to as polar mesospheric clouds, PMC), in the perpetually sunlit summer polar regions, at latitudes extending up to 80◦ and often encompassing the entire polar cap (e.g. Carbary et al., 1999). Mesospheric clouds occur most frequently during the months of June and July in the northern hemisphere and December and January in the southern hemisphere. Detailed observational and modeling studies indicate that NLC consist of microscopic ice particles (e.g. Witt, 1967) that nucleate and grow to sub-micron sizes over periods of several hours to a few days when the ambient temperature
c 2002 Elsevier Science Ltd. All rights reserved. 1364-6826/02/$ - see front matter PII: S 1 3 6 4 - 6 8 2 6 ( 0 2 ) 0 0 0 5 3 - 6
992
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
falls well below 150 K (e.g. Gadsden, 1981; Jensen and Thomas, 1988). Such harsh conditions usually only exist in the high-latitude summer mesopause region. As the clouds are very tenuous, they can only be seen from the ground by the scattering of sunlight during the hours of twilight, when the observer and the atmosphere below them are in darkness while the clouds themselves remain sunlit (this condition occurs for solar depression angles, ∼6–16◦ ). In particular, ground-based measurements have provided important information on the occurrence, morphology and lower latitude limit of the clouds observed over the past century and it is generally agreed that mesospheric clouds are most frequently observed down to latitudes of ∼55–60◦ (Gadsden, 1998). This lower latitude limit represents a statistical bound to the clouds southerly extension. Optimum locations for observing mesospheric clouds now and over the past 100 years are Scandinavia=northern Europe, central Asia and Canada in the northern hemisphere and the southern tip of South America in the southern hemisphere, where they are readily detected during the prolonged twilight conditions.
2. Occurrence frequency In recent years, a marked increase in the frequency of occurrence of NLC has been reported (Gadsden, 1990). The number of NLC sightings has also risen substantially in the past few years, primarily due to increased public awareness. However, the ongoing cloud occurrence measurements are based on an established set of long-term professional and amateur observers and indicates a signi2cant secular increase. This increase has not been accompanied by an extension in the length of the “season” during which the clouds may be seen but in the probability of seeing a noctilucent cloud on a particular date. It seems too that the height of the clouds measured over the past 100 years has also not changed (Gadsden and Taylor, 1994). This said, there are beginning to be indications that the increased frequency of occurrence is being accompanied by an increase in the number of mesospheric cloud sightings at relatively low latitudes. As the number of these sightings are still few, it is not yet possible to show this result statistically. However, observers in Canada, Europe and Asia have all reported NLC at unexpectedly low latitudes in recent years. For example, in the UK mesospheric clouds are traditionally seen from Scotland (∼56–59◦ N); however, several sightings have recently been reported from southern England (∼51◦ N) (Livesey, 1998). Twenty years ago the chance of such a sighting was considered vanishingly small. Yet corroborative measurements by lidar from Kuhlungsborn, Germany (∼54◦ N) have clearly now established this fact (e.g. von Cossart et al., 1996). The speculation is that these sightings may result from an actual expansion “of NLC to more southerly latitudes as a result of global climate changes” (Zalcik, 1998).
3. Mid-latitude NLC observations 3.1. Logan, UT On 22=23 and 23=24 June 1999 two most unexpected NLC displays were seen, and photographed, over the northwestern USA from Logan, UT (41:7◦ N, 111:8◦ W). The display of 22=23 June was immediately recognizable and consisted of two distinct cloud “patches” containing diLuse and 2ne-scale wave structures. The two patches were located in the northwestern (azimuth ∼320◦ ) and northern (azimuth ∼355◦ ) sky and subtended elevations of ¡15◦ at the time of their initial sighting at 04:00 UT (22:00 Mountain Daylight Time (MDT)) when was ∼9◦ . No data are available prior to this time as this was a chance sighting. Photographs and video images of the NLC were obtained over the next 30 min (from 04:10 to 04:40 UT) ( ∼10– 14◦ ) during which time the elevation of the visible region of the NLC reduced signi2cantly. By 04:50 UT (∼15◦ ), the clouds were evident only in a very thin strip of northern sky (¡2◦ wide) adjacent to the horizon. Throughout this period, the NLC appeared bluish-white in color and were easily visible against the foreground twilight glow and in the presence of local streetlights. Fig. 1 shows a remarkable image of this NLC display consisting of two photographs electronically joined together to provide a panoramic view of the clouds at 04:30 UT (∼13◦ ) looking over Utah State University campus. At this time the visible upper border of the two cloud patches was ¡6◦ in elevation (the radio tower in the foreground subtends 3:5◦ in elevation). Both NLC patches are clearly evident, the one to the left of the tower was fainter and characterized by diLuse structures whilst the patch to the right was more contrasted and contained many wave-like billow patterns. NLC are also evident at the tower azimuth but at much lower elevations (∼2◦ ) suggesting that both patches were equatorward expansions of a much larger NLC display extending well to the north of Logan. A narrow bank of tropospheric cloud (dark silhouette) was also present at this time partially obscuring the upper elevations of the display with clear sky at higher elevations. A series of photographs using exposure times of initially 2, 5 and 10 s (later 20 s) was obtained during this period from three local vantage points to document the display to its fullest. On the following night (23=24 June) a more extensive, yet visibly weaker mesospheric cloud display was again photographed and videoed from Logan. On this occasion, the clouds were observed as early as 03:50 UT (∼8◦ ) and consisted of up to eight faint, irregularly shaped bands of much larger horizontal scale-sizes than the billows seen on the previous night, with diLuse structures visible initially at ∼40◦ elevation. Over the next hour the elevation of the display was observed to decrease systematically as twilight faded. Fig. 2 shows six example photographs of this NLC display taken at ∼10 min intervals during the period 03:50 – 04:40 UT (21:50 –22:40 MDT) during which time
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
993
Fig. 1. Composite image showing NLC observed from Logan, UT (41:7◦ N) on the night of 22=23 June 1999 at 04:30 UT when the sun was ∼13◦ below the horizon. The photographs were obtained using a Minolta SLR camera 2tted with a 50 mm, f=1.7 lens and Kodak 100 ASA color 2lm exposed for ∼10 s and show two distinct cloud “patches” containing faint diLuse structures and 2ne billow waves in the northwestern and northern sky respectively at elevations up to ∼6◦ .
Fig. 2. Six examples of the NLC display photographed on the following night (23=24 June) from Logan, UT during the period 03:50 – 04:40 UT. The bands initially subtended elevations of higher than 40◦ (Fig. 2a). Subsequent images show the display at ∼10 min intervals as twilight diminished.
increased from ∼8 to 14◦ . On this occasion coordinated measurements using a powerful Raleigh lidar were also made from Logan campus (Wickwar et al., 2002). Initial results from the lidar measurements show a strong backscatter signal centered at ∼82 km (range 81.6 –82:4 km) indicating the presence of an unlit NLC overhead (around 04:30 UT) at the same time as they were photographed (Fig. 2) in the twilight sky to the northwest. These data are consistent with previous lidar soundings of NLC at higher latitudes (e.g. Thayer et al., 1995) and provide independent, corroborative
evidence for the existence of unusually low-latitude mesospheric clouds at this time. Although favorable observing conditions persisted throughout the summer there were no further NLC sightings from Logan during the period up to mid-August. 3.2. Concurrent image data Although mesospheric cloud sightings are relatively common throughout Canada, observations from the USA (with
994
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
Fig. 3. Photographs of the NLC display of 22=23 June, 1999 from (a) Namao, Alberta, Canada (53:7◦ N) at local midnight (06:00 UT), and (b) Glen Ullin, ND, USA (46:8◦ N) at 04:00 UT. Both photographs show extensive NLC in the northwestern sky and were captured using 400 ASA 2lm and an ∼5 s exposure. Low elevation meteorological cloud obscures the Namao NLC observations which reached over 50◦ elevation. The bright “star” in the wide-angle Glen Ullin data is the planet Venus, NLC were evident up to ∼22◦ elevation and extended over 65◦ across the twilight sky.
the exception of Alaska) have until recently been considered extremely rare. However, over the past decade, there have been several unmistakable sightings (and photographs) from the northwestern USA, most notably from North Dakota (46:8◦ N) and from northern Montana (44:7◦ N), showing NLC extending south of the US–Canadian border, below the 48th parallel (Zalcik, 1998). Until now these were the most southerly sightings of NLC in the USA. However, examination of the NLC display of 22=23 June (Fig. 1) indicates that the clouds had penetrated much further south and were located over central Idaho and southern Montana (assuming a cloud altitude of 82 km). This display was also observed and photographed from Namao, near Edmonton, Alberta, Canada (53:7◦ N) approximately 1300 km due north of Logan where it was described as spectacular, stretching all the way into the southwestern sky, reaching ∼50◦ in elevation, and from Glen Ullin, ND (46:8◦ N) where extensive band- and billow-type clouds
reaching up to ∼22◦ in elevation were photographed in the western sky that apparently extended overhead as far south as southern Montana. Fig. 3a and b shows two examples of this spectacular display recorded from Namao and Glen Ullin, respectively. Moreover, the same display was also sighted in the northwestern sky from Craig, CO (40:5◦ N) by R. Moore (private communication) and from Coal Creek Canyon near Boulder, CO (40◦ N) by R. Keen (as reported in the Denver Post by A. Schrader). To put all of these sightings in perspective a map showing the locations of these sites is shown in Fig. 4. Together these observations suggest a large latitudinal extent for this NLC display (40 –55◦ N) with an unusual southerly extension. 4. PSMOS 2000 NLC observations The unusual NLC sightings of 22=23 and 23=24 June, 1999 were presented at the Second International Workshop
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
995
Fig. 4. Map showing the location of several sites in USA and Canada from which the NLC display of 22=23 June 1999 was observed. These sites span a latitudinal range of ∼14◦ and suggest an extensive NLC cloud was present over the central and western USA. Simultaneous optical measurements made from Delaware Observatory, ∼2500 km to the East of Logan, indicate unusually low mesospheric temperatures but no NLC were detected.
on the Planetary Scale Mesosphere Observing System (PSMOS) Project held in Toronto, Canada during May 2000. As a result of the ensuing discussions, it was decided to establish a network of PSMOS and other interested observers (including the well-established NLC CAN-AM amateur network in Canada) to help coordinate mid(and high) latitude NLC studies from the USA, Canada, Greenland and Europe. An e-mail alert was set up to convey NLC sighting information rapidly to participants from 32 institutions. A range of instrumentation was made available for these observations including Rayleigh and resonant lidars, visible and near-infrared interferometers, spectrometers, CCD imagers, and photographic cameras. One of us (M.S.Z.) has compiled a report summarizing the NLC CAN-AM observations for the summer of 2000 indicating distinct NLC displays on as many as 35 nights during the period May 28–Aug 27. The most extensive NLC displays may have occurred on June 27=28 and July 5=6 when they were observed from Europe and the USA. Many of the sightings, particularly after mid-July were from higher latitude sites but several of the earlier displays were seen from as far south as Vancouver, Canada (49:5◦ N) and Glen Ullin, ND (46:8◦ N). To our knowledge, the only potential NLC sighting at unusually low latitudes was made on 30 June=1 July from a mountain site near Ft. Collins, CO (∼41◦ N). On this occasion, one of us (M.J.T.) observed several faint luminous bands at low elevations in the evening twilight sky towards the northwest around 03:50 – 04:10 UT. The “clouds” appeared milky white in color and consisted of a single
elongated luminous band accompanied by several shorter wave-like structures. Separate observations of these clouds were also made from the same site by T. Nakamura (private communication). They were evident well after sunset under near new moon conditions and well away from city lights. Observations of meteorological cloud later that night (at similar low elevations) were quite diLerent in appearance. Unfortunately, although attempted, there are no photographic data to support these sightings. On the same night, but unknown to us at that time, an extensive, bright NLC display was observed over much of the Canadian Prairies. Fig. 5 shows a photograph of this display as seen from Namao, Alberta (53:7◦ N) at 06:30 UT. This event was also evident in the dawn twilight from Glen Ullin, ND indicating a signi2cant southerly extension. As a result of this potential NLC sighting, mesospheric temperature soundings were made on the following night from Ft. Collins (∼41◦ N) using the Colorado State University Na lidar system. Temperatures as low as 159 K were recorded in hourly averaged data with a mean nocturnal value of 168 K for the 83–84 km level (C.Y. She, private communication). These values are somewhat lower than expected from the seasonal climatology but signi2cantly above those necessary for sustained NLC cloud existence (see Section 5). Thus, it is possible that the unusual luminous clouds observed from northern Colorado on 30 June–1 July may have been a transient southerly extension of the large NLC display observed over Canada and the northern USA. However, without supporting photographic or other scienti2c evidence this must remain speculative.
996
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
Fig. 5. Example image of the bright, extensive NLC display of 30 June=1 July, 2000 photographed from Namao, Alberta, Canada (53:7◦ N) at 06:30 UT looking towards the NNW using 100 ASA 2lm with a ∼15 s exposure. This display was also observed from Vancouver, Canada and North Dakota and possibly from Ft. Collins, CO.
5. Discussion Modeling studies indicate that mesospheric ice particles can only nucleate and grow to an observable size (¿50 nm) if the ambient temperature in the region in which they form remains well below 150 K for periods of many hours (e.g. Gadsden, 1981; Jensen and Thomas, 1988). Due to potentially strong background winds in the upper mesosphere the geographic regions of nucleation, growth and detection can often be separated by many hundreds of kilometers. As the lifetime of an NLC particle is only a few seconds at temperatures ¿150 K, the background temperature throughout this region is a critical factor in controlling both the altitudinal and latitudinal extent of the observed clouds. Favorable conditions for NLC formation often exist in the polar latitude summer mesosphere (e.g. von Zahn and Meyer, 1989) but not normally at mid-latitudes where measured mesospause temperatures at ∼41◦ N typically range from 180 –190 K (Yu and She, 1995; She et al., 2000). Thus, the detection of NLC at mid-latitudes implies a major departure from normal mesospheric conditions. The simplest explanation for our 1999 NLC observations is that, for a limited period (∼2 days) close to the summer Solstice (when the cold polar air mass near the mesopause is expected to be well developed), an extremely cold air mass penetrated southward well into the mid-latitude mesosphere and persisted long enough for existing NLC particles to be transported equatorward from higher latitudes and=or for new particles to continue to nucleate and grow to form the
observed cloud patches at much lower latitudes than normal. This “dynamically driven” hypothesis is supported by the apparently large latitudinal extent of the NLC display observed on 22=23 June (¿15◦ ), as indicated in Fig. 4. Additional evidence for unusual mesospheric conditions on the night of June 22=23 comes from observations of the hydroxyl (OH) Meinel (3,1) and (4,2) band nightglow emission measured by a University of Western Ontario Michelson interferometer located at the Delaware Observatory, Canada (42:9◦ N, 81:4◦ W). (The location of this site is also indicated in Fig. 4.) The nocturnal OH emission originates from a layer of half-width ∼6–8 km, and She and Lowe (1998) have shown that measurements of its rotational temperature are a good proxy for the ∼87 km atmospheric temperature. Although 87 km is higher than the NLC base altitude, the summer-time OH emission layer encompasses the mesopause region where NLC nucleation is expected. In a series of combined imaging and spectrometric observations from central Alaska (∼65◦ N), Taylor et al. (1995) found that during overhead NLC displays, the column integrated OH rotational temperature was always ¡154 K. Fig. 6 shows the zenith nocturnal OH rotational temperature data (open circles) for June 22=23, 1999. Shortly after sunset the OH temperature was high around 185 K, however, within the next 2 h it dropped rapidly to a remarkably low value of ∼155–160 K (around 04 – 05 UT). This unusual behavior is exempli2ed by comparison with the mean (hourly averaged) nocturnal behavior of the OH rotational
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
997
Fig. 6. OH rotational temperature variation over the Delaware Observatory, Canada (41:9◦ N, 81:4◦ W) on the night of 22=23 June 1999. The temperature is the mean of the OH (3–1) and (4 –2) band temperatures. The mean variation (hourly average) on nine nights within 14 days of the 1999 summer solstice is shown for comparison. The standard deviations of the hourly means lie in the range 6 –8 K. The data indicate unusually low temperatures at the ∼87 km level during the period 04 – 05 UT when NLC were imaged well to the West.
temperature averaged over nine nights during the summer solstice period of 1999 (indicated by the solid circles). Comparison of these datasets shows that the OH temperature was about 10 K higher than average in the early evening of the 22=23 June, but was 10 –15 K below average for much of the night, and as much as ∼20 K below average at times during the period 04 – 05 UT (when the NLC were sighted from Utah). The Delaware observations ceased at about 07:00 UT due to meteorological cloud. In concert with these temperature changes the integrated zenith OH intensity also fell dramatically by a factor of over 4, from well above average to well below. These OH observations are consistent with either the passage of a very large amplitude wave with a period of at least several hours, or the incursion of a large volume of low temperature air into the mid-latitude mesosphere, possibly from the north. Unfortunately, meteorological cloud prevented observations from the Delaware Observatory on 23=24 June and several subsequent nights which would have helped clarify this unusual behavior. However, even at their lowest, the Delaware OH temperatures are considered too high for NLC existence. This fact is supported by independent measurements from the Delaware Observatory using a Rayleigh lidar. Although the lidar temperatures for this (and the previous) night show values as low as 156 K over the altitude range ∼83–85 km there was no evidence for an NLC scattering layer (R.J. Sica, private communication). Nevertheless, these two data sets provide strong evidence for a large deviation below the average summer-time temperature behavior during this period, from a similar mid-latitude site to Logan, UT, but ∼2500 km to the east of the con2rmed
NLC volume. Thus, it is feasible that the critical temperature limit of ∼150 K at which particle sublimation becomes very rapid was shifted signi2cantly southward over the western USA during this period. However, cold temperatures alone are not usually enough and there is a critical need for suNcient mesospheric water vapor for the successful development of an NLC. Thomas et al. (1989) and later Thomas (1996) have postulated that increases in lower atmospheric methane and carbon dioxide, possibly due to anthropogenic sources (biological and industrial), may have modi2ed signi2cantly the temperature and water vapor concentration at mesopause heights over the past century. If so, this would cause a substantial increase in the number and brightness of mesospheric clouds which could account for some (but not all) of the strong secular increase in NLC occurrence observed in recent years (Gadsden, 1990). Thomas (1996) further conjectured that a doubling of methane and carbon dioxide could “extend the NLC region to mid-latitudes, where they would be observable by a large fraction of the world’s population”. Whether the growing number of NLC sightings at unusually low latitudes is linked to the observed secular increase, and hence possibly to long-term changes in mesospheric temperature or water vapor content (perhaps of anthropogenic origin), has yet to be determined. The NLC sightings in the American Sector this past year (close to solar maximum conditions) indicate signi2cant activity despite the fact that the occurrence of NLC is expected to reduce during enhanced solar conditions. In this context, it is interesting to note that the incidence of NLC over Europe has fallen markedly over the past 3– 4 years from a maximum
998
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
frequency in 1995 (Gavine, 2000). However, the NLC observations discussed herein suggest that the conditions responsible for the lower latitude displays observed over North America may be becoming more prevalent with time. This said, coordinated visual and lidar observations from several mid-latitude sites during the summer 2000 have revealed no further de2nitive cases of NLC at unexpectedly low latitudes. 6. Summary The ground-based sightings reported here provide new evidence for the occasional expansion of NLC to unusually low latitudes. Whether they are due to unusual dynamical forcing, or perhaps are an early indicator of signi2cant long-term changes in the climate of the summertime high-latitude mesosphere, remains to be seen. However, it is clear that such perturbations can have a large, but so far temporary, impact on the local mid-latitude mesospheric conditions. Further, coordinated ground-based and satellite observations are essential to help quantify changes in their occurrence frequency, to determine the mesospheric conditions under which they can exist at mid-latitudes and, to measure their geographical extent. The NASA Aeronomy of Ice in the Mesosphere (AIM) mission, which is currently under Phase-A study, may address many of these questions. Such investigations should also be broad enough to consider that mesospheric clouds might also become an occasional feature of the undisturbed low-latitude=tropical mesosphere. For example, Summers and Siskind (1999) have recently suggested that an enhancement of water vapor observed in the equatorial mesosphere and extending up to ±30◦ in latitude may well result from the association of atomic oxygen and molecular hydrogen onto the solid surface of meteoric debris. Under favorable conditions this meteoric dust, covered in water, could lead to the formation of a visible, low-latitude, mesospheric cloud layer. Possible evidence for such low-latitude cloud layers may already exist in the records of the cosmonauts onboard the Salyut-6 spacecraft who reported seeing mesospheric clouds within 40◦ of the equator on a number of occasions (Lazarev et al., 1987). Anomalous mesospheric clouds over Iraq and Syria (¡40◦ N) have also been reported from an aircraft Pying over the Caspian Sea (Vasiliev et al., 1974). However, it is also possible that high-altitude aerosol layers, resembling NLC (to the untrained observer), may be produced by exhaust gases from sounding rocket or satellite launches. Unfortunately, to date there have been no corroborative ground-based sightings or photography of such low-latitude events. Thus, while it seems clear that the two NLC displays reported here were outstanding examples associated with an extensive high-latitude cloud system, apparently penetrating deep into the mid-latitude mesosphere, the chance of sighting even lower latitude NLC, including “equatorial” mesospheric clouds, cannot be disregarded in future studies.
Acknowledgements We gratefully acknowledge the valuable reports by R.J. Sica, V.B. Wickwar, R. Moore and R. Keen supporting these unusual observations and the comments from the two referees. We thank D. Jereb for his help in obtaining the photographic data on 22=23 June and B. Atkinson for assistance with formatting the NLC imagery presented here. Support for the collation and analysis of these data and for optical observations during the summer 2000 has been provided by NSF Grant ATM-0003684. Note added at proof stage On 2/3 July 2001 NLC were again observed and photographed from Logan, UT at low elevations in the dawn twilight. Corroborative photographic measurements were also obtained from Glen Ullin, ND. These new data provide further evidence for the existence of NLC at unexpectedly low latitudes, and suggest that their occurrence rate may be increasing. References Carbary, J.F., Romick, G.J., Morrison, D., Paxton, L.J., Meng, C.I., 1999. Altitudes of polar mesospheric clouds observed by middle ultraviolet imager. Journal of Geophysical Research 104, 10,089–10,100. Donahue, T.M., Guenther, B., Blamont, J.E., 1972. Noctilucent clouds in daytime: circumpolar particulate layers near the summer mesopause. Journal of Atmospheric Science 29, 1205. Denver Post, 25 July, 1999, article by A. Schrader. Evans, W.F.J., Laframboise, L.R., Sine, K.R., Wiens, R.H., Shepherd, G.G., 1995. Observation of polar mesospheric clouds in summer, 1993 by the WINDII instrument on UARS. Geophysical Research Letters 22, 2793–2796. Gadsden, M., 1981. The silver-blue cloudlets again: nucleation and growth of ice in the mesosphere. Planet and Space Science 29, 1079–1087. Gadsden, M., 1998. The North-West Europe data on noctilucent clouds: a survey. Journal of Atmospheric and Solar-Terrestrial Physics 60, 1163–1174. Gadsden, M., 1990. A secular change in noctilucent cloud occurrence. Journal of Atmospheric and Solar-Terrestrial Physics 52, 247–251. Gadsden, M., SchrFoder, W., 1989. Noctilucent Clouds. Springer, Berlin. Gadsden, M., Taylor, M.J., 1994. Measurements of noctilucent cloud heights: a bench mark for changes in the mesosphere. Journal of Atmospheric and Solar-Terrestrial Physics 56, 461–466. Gavine, D., 2000. Noctilucent clouds over Britain and Western Europe, 1995 –1997. Journal of the British Astronomical Association 110, 4. Jensen, E.J., Thomas, G.E., 1988. A growth-sedimentation model of polar mesospheric clouds: comparison with SME measurements. Journal of Geophysical Research 93, 2461–2473.
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999 Jesse, O., 1896. Die hFohe der leuchtenden nachtwolken. Astron. Nacht. 140, 161–168. Lazarev, A.I., Kovalyenok, V.V., Avakyan, S.V., 1987. Investigation of Earth from Manned Spacecraft. Issledovanye zemli c pilotir xhyenich kosmicheskikh korablyei Gidrometeoizdat, Leningrad, pp. 316 –319. Leslie, R., 1884. The sky-glows. Nature 30, 583. Livesey, R.J., 1998. British observations of noctilucent clouds in 1998 (provisional). Journal of the British Astronomical Association 108, 241. She, C.Y., Lowe, R.P., 1998. Seasonal temperature variations in the mesopause region at mid-latitude: comparison of lidar and hydroxyl rotational temperatures based on WINDII=UARS OH height pro2les. Journal of Atmospheric and Solar-Terrestrial Physics 60, 1573–1583. She, C.Y., Chen, S., Hi, Z., Sherman, J., Vance, J.D., Vasoli, V., White, M.A., Yu, J., Krueger, D.A., 2000. Eight-year climatology of nocturnal temperature and sodium density in the mesopause region (80 –105 km) over Fort Collins, CO (41◦ N, 105◦ W). Geophysical Research Letters 27, 3289–3292. Summers, M.E., Siskind, D.E., 1999. Surface recombination of O and H2 on meteoric dust as a source of mesospheric water vapor. Geophysical Research Letters 26, 1837–1840. Taylor, M.J., Lowe, R.P., Baker, D.J., 1995. Hydroxyl temperature and intensity measurements during noctilucent cloud displays. Annales Geophysicae 13, 1107–1116. Thayer, J.P., Nielsen, J.P., Jacobsen, N., 1995. Noctilucent cloud observations over Greenland by Rayleigh lidar. Geophysical Research Letters 22, 2961–2964.
999
Thomas, G.E., 1991. Mesospheric clouds and the physics of the mesopause region. Reviews of Geophysics 29, 553–575. Thomas, G.E., 1996. Is the polar mesosphere the miner’s canary of global change? Advances in Space Research 18, 149–158. Thomas, G.E., Olivero, J.J., Jensen, E.J., SchrFoder, W., Toon, O.B., 1989. Relation between increasing methane and the presence of ice clouds at the mesopause. Nature 338, 490–492. Vasiliev, O.B., Grishin, N.I., Zinoviev, V.V., Chapurskii, L.I., 1974. Observations of anomalous appearances of noctilucent clouds. Solar System Research 8, 153–156. von Cossart, G., HoLman, P., von Zahn, U., Keckhut, P., Hauchecorne, A., 1996. Mid-latitude noctilucent cloud observations by lidar. Geophysical Research Letters 23, 2919–2922. von Zahn, U., Meyer, W., 1989. Mesopause temperatures in polar summer. Journal of Geophysical Research 94, 14,647–14,651. Wickwar, V.B., Taylor, M.J., Herron, J.P., Martineau, B.A., 2002. Visual and lidar observations of noctilucent clouds above Logan, Utah, at 41:7◦ N. Journal of Geophysical Research, in press. Witt, 1967. Optical characteristics of mesospheric aerosol distributions in relation to noctilucent clouds. Tellus 20, 98. Yu, J.R., She, C.Y., 1995. Climatology of a midlatitude mesopause region observed by lidar at Fort Collins, Colorado (40:6◦ N, 105◦ W). Journal of Geophysical Research 100, 7441–7452. Zalcik, M.S., 1998. A possible increase in mid-latitude sightings of noctilucent clouds. Journal of the Royal Astronomical Society of Canada 92, 197–200.
www.elsevier.com/locate/jastp
Mesospheric cloud observations at unusually low latitudes M.J. Taylora; ∗ , M. Gadsdenb , R.P. Lowec , M.S. Zalcikd , J. Brausche a Space
Dynamics Laboratory and Department of Physics, Utah State University, 4415 Old Main Street, Logan, UT 84322-4415, USA b 12 Keir Street, Perth, Scotland, UK c Department of Physics, University of Western Ontario, London, Ont., Canada d 14130 80th Street, Edmonton, Alberta, Canada e 208 South G Street, Glen Ullin, ND, USA
Abstract Noctilucent clouds (NLC) are a beautiful, high-latitude, summertime phenomenon that was 2rst reported over 100 years ago. They are seen during the hours of twilight by the scattering of sunlight from sub-micron-sized ice particles that form in the vicinity of the cold mesopause region. NLC are quite distinctive, often appearing silvery-blue in color. In recent years there has been a marked increase in their frequency of occurrence, possibly due to an increase in mesospheric water vapor and=or to a cooling of the mesopause region, prompting speculation that they are “harbingers” of potentially serious changes in the mesospheric climate. In concert with this trend there are also a growing number of ground-based NLC sightings at signi2cantly ◦ lower latitudes than expected. Here we report two unusual NLC displays photographed from Logan, UT, USA (∼42 N) ◦ during June 1999, well over 10 lower in latitude than expected and implying a major, yet temporary, departure from normal mid-latitude summertime conditions. These data provide new evidence for the occasional expansion of NLC to unusually low latitudes possibly due to exceptional dynamical forcing. Alternatively, they may be an early indicator of signi2cant long-term c 2002 Elsevier Science Ltd. All rights reserved. changes taking place in the upper mesospheric summertime environment. Keywords: Noctilucent clouds; Mesospheric clouds; Mesospheric climate
1. Introduction From their very 2rst observation in the twilight sky noctilucent clouds (NLC) were accepted as being a most unusual phenomenon. Well over a 100 years ago Leslie (1884) described their appearance as “weird small cloud forms, at times very regular, like ripple marks in the sand, or the bones of some great 2sh” seen late in the evening twilight. Within a few years of their 2rst sightings, ground-based photographic triangulation measurements by Jesse (1896) had determined that NLC were by far the highest clouds on earth, residing in the upper mesosphere at heights close to 82 km. Subsequent research has shown that these mesospheric clouds are most often seen from the ground at middle to high latitudes ∗
Corresponding author. Fax: +1-435-713-0054. E-mail address: [email protected] (M.J. Taylor).
(∼55–65◦ ) and that they are a product of the intense cooling that occurs in the summertime mesopause region at polar latitudes (e.g. Gadsden and SchrFoder, 1989). Observations from satellites (e.g. Donahue et al., 1972; Thomas, 1991; Evans et al., 1995) have signi2cantly extended the realm of NLC studies and have revealed the frequent presence of mesospheric clouds (sometimes referred to as polar mesospheric clouds, PMC), in the perpetually sunlit summer polar regions, at latitudes extending up to 80◦ and often encompassing the entire polar cap (e.g. Carbary et al., 1999). Mesospheric clouds occur most frequently during the months of June and July in the northern hemisphere and December and January in the southern hemisphere. Detailed observational and modeling studies indicate that NLC consist of microscopic ice particles (e.g. Witt, 1967) that nucleate and grow to sub-micron sizes over periods of several hours to a few days when the ambient temperature
c 2002 Elsevier Science Ltd. All rights reserved. 1364-6826/02/$ - see front matter PII: S 1 3 6 4 - 6 8 2 6 ( 0 2 ) 0 0 0 5 3 - 6
992
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
falls well below 150 K (e.g. Gadsden, 1981; Jensen and Thomas, 1988). Such harsh conditions usually only exist in the high-latitude summer mesopause region. As the clouds are very tenuous, they can only be seen from the ground by the scattering of sunlight during the hours of twilight, when the observer and the atmosphere below them are in darkness while the clouds themselves remain sunlit (this condition occurs for solar depression angles, ∼6–16◦ ). In particular, ground-based measurements have provided important information on the occurrence, morphology and lower latitude limit of the clouds observed over the past century and it is generally agreed that mesospheric clouds are most frequently observed down to latitudes of ∼55–60◦ (Gadsden, 1998). This lower latitude limit represents a statistical bound to the clouds southerly extension. Optimum locations for observing mesospheric clouds now and over the past 100 years are Scandinavia=northern Europe, central Asia and Canada in the northern hemisphere and the southern tip of South America in the southern hemisphere, where they are readily detected during the prolonged twilight conditions.
2. Occurrence frequency In recent years, a marked increase in the frequency of occurrence of NLC has been reported (Gadsden, 1990). The number of NLC sightings has also risen substantially in the past few years, primarily due to increased public awareness. However, the ongoing cloud occurrence measurements are based on an established set of long-term professional and amateur observers and indicates a signi2cant secular increase. This increase has not been accompanied by an extension in the length of the “season” during which the clouds may be seen but in the probability of seeing a noctilucent cloud on a particular date. It seems too that the height of the clouds measured over the past 100 years has also not changed (Gadsden and Taylor, 1994). This said, there are beginning to be indications that the increased frequency of occurrence is being accompanied by an increase in the number of mesospheric cloud sightings at relatively low latitudes. As the number of these sightings are still few, it is not yet possible to show this result statistically. However, observers in Canada, Europe and Asia have all reported NLC at unexpectedly low latitudes in recent years. For example, in the UK mesospheric clouds are traditionally seen from Scotland (∼56–59◦ N); however, several sightings have recently been reported from southern England (∼51◦ N) (Livesey, 1998). Twenty years ago the chance of such a sighting was considered vanishingly small. Yet corroborative measurements by lidar from Kuhlungsborn, Germany (∼54◦ N) have clearly now established this fact (e.g. von Cossart et al., 1996). The speculation is that these sightings may result from an actual expansion “of NLC to more southerly latitudes as a result of global climate changes” (Zalcik, 1998).
3. Mid-latitude NLC observations 3.1. Logan, UT On 22=23 and 23=24 June 1999 two most unexpected NLC displays were seen, and photographed, over the northwestern USA from Logan, UT (41:7◦ N, 111:8◦ W). The display of 22=23 June was immediately recognizable and consisted of two distinct cloud “patches” containing diLuse and 2ne-scale wave structures. The two patches were located in the northwestern (azimuth ∼320◦ ) and northern (azimuth ∼355◦ ) sky and subtended elevations of ¡15◦ at the time of their initial sighting at 04:00 UT (22:00 Mountain Daylight Time (MDT)) when was ∼9◦ . No data are available prior to this time as this was a chance sighting. Photographs and video images of the NLC were obtained over the next 30 min (from 04:10 to 04:40 UT) ( ∼10– 14◦ ) during which time the elevation of the visible region of the NLC reduced signi2cantly. By 04:50 UT (∼15◦ ), the clouds were evident only in a very thin strip of northern sky (¡2◦ wide) adjacent to the horizon. Throughout this period, the NLC appeared bluish-white in color and were easily visible against the foreground twilight glow and in the presence of local streetlights. Fig. 1 shows a remarkable image of this NLC display consisting of two photographs electronically joined together to provide a panoramic view of the clouds at 04:30 UT (∼13◦ ) looking over Utah State University campus. At this time the visible upper border of the two cloud patches was ¡6◦ in elevation (the radio tower in the foreground subtends 3:5◦ in elevation). Both NLC patches are clearly evident, the one to the left of the tower was fainter and characterized by diLuse structures whilst the patch to the right was more contrasted and contained many wave-like billow patterns. NLC are also evident at the tower azimuth but at much lower elevations (∼2◦ ) suggesting that both patches were equatorward expansions of a much larger NLC display extending well to the north of Logan. A narrow bank of tropospheric cloud (dark silhouette) was also present at this time partially obscuring the upper elevations of the display with clear sky at higher elevations. A series of photographs using exposure times of initially 2, 5 and 10 s (later 20 s) was obtained during this period from three local vantage points to document the display to its fullest. On the following night (23=24 June) a more extensive, yet visibly weaker mesospheric cloud display was again photographed and videoed from Logan. On this occasion, the clouds were observed as early as 03:50 UT (∼8◦ ) and consisted of up to eight faint, irregularly shaped bands of much larger horizontal scale-sizes than the billows seen on the previous night, with diLuse structures visible initially at ∼40◦ elevation. Over the next hour the elevation of the display was observed to decrease systematically as twilight faded. Fig. 2 shows six example photographs of this NLC display taken at ∼10 min intervals during the period 03:50 – 04:40 UT (21:50 –22:40 MDT) during which time
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
993
Fig. 1. Composite image showing NLC observed from Logan, UT (41:7◦ N) on the night of 22=23 June 1999 at 04:30 UT when the sun was ∼13◦ below the horizon. The photographs were obtained using a Minolta SLR camera 2tted with a 50 mm, f=1.7 lens and Kodak 100 ASA color 2lm exposed for ∼10 s and show two distinct cloud “patches” containing faint diLuse structures and 2ne billow waves in the northwestern and northern sky respectively at elevations up to ∼6◦ .
Fig. 2. Six examples of the NLC display photographed on the following night (23=24 June) from Logan, UT during the period 03:50 – 04:40 UT. The bands initially subtended elevations of higher than 40◦ (Fig. 2a). Subsequent images show the display at ∼10 min intervals as twilight diminished.
increased from ∼8 to 14◦ . On this occasion coordinated measurements using a powerful Raleigh lidar were also made from Logan campus (Wickwar et al., 2002). Initial results from the lidar measurements show a strong backscatter signal centered at ∼82 km (range 81.6 –82:4 km) indicating the presence of an unlit NLC overhead (around 04:30 UT) at the same time as they were photographed (Fig. 2) in the twilight sky to the northwest. These data are consistent with previous lidar soundings of NLC at higher latitudes (e.g. Thayer et al., 1995) and provide independent, corroborative
evidence for the existence of unusually low-latitude mesospheric clouds at this time. Although favorable observing conditions persisted throughout the summer there were no further NLC sightings from Logan during the period up to mid-August. 3.2. Concurrent image data Although mesospheric cloud sightings are relatively common throughout Canada, observations from the USA (with
994
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
Fig. 3. Photographs of the NLC display of 22=23 June, 1999 from (a) Namao, Alberta, Canada (53:7◦ N) at local midnight (06:00 UT), and (b) Glen Ullin, ND, USA (46:8◦ N) at 04:00 UT. Both photographs show extensive NLC in the northwestern sky and were captured using 400 ASA 2lm and an ∼5 s exposure. Low elevation meteorological cloud obscures the Namao NLC observations which reached over 50◦ elevation. The bright “star” in the wide-angle Glen Ullin data is the planet Venus, NLC were evident up to ∼22◦ elevation and extended over 65◦ across the twilight sky.
the exception of Alaska) have until recently been considered extremely rare. However, over the past decade, there have been several unmistakable sightings (and photographs) from the northwestern USA, most notably from North Dakota (46:8◦ N) and from northern Montana (44:7◦ N), showing NLC extending south of the US–Canadian border, below the 48th parallel (Zalcik, 1998). Until now these were the most southerly sightings of NLC in the USA. However, examination of the NLC display of 22=23 June (Fig. 1) indicates that the clouds had penetrated much further south and were located over central Idaho and southern Montana (assuming a cloud altitude of 82 km). This display was also observed and photographed from Namao, near Edmonton, Alberta, Canada (53:7◦ N) approximately 1300 km due north of Logan where it was described as spectacular, stretching all the way into the southwestern sky, reaching ∼50◦ in elevation, and from Glen Ullin, ND (46:8◦ N) where extensive band- and billow-type clouds
reaching up to ∼22◦ in elevation were photographed in the western sky that apparently extended overhead as far south as southern Montana. Fig. 3a and b shows two examples of this spectacular display recorded from Namao and Glen Ullin, respectively. Moreover, the same display was also sighted in the northwestern sky from Craig, CO (40:5◦ N) by R. Moore (private communication) and from Coal Creek Canyon near Boulder, CO (40◦ N) by R. Keen (as reported in the Denver Post by A. Schrader). To put all of these sightings in perspective a map showing the locations of these sites is shown in Fig. 4. Together these observations suggest a large latitudinal extent for this NLC display (40 –55◦ N) with an unusual southerly extension. 4. PSMOS 2000 NLC observations The unusual NLC sightings of 22=23 and 23=24 June, 1999 were presented at the Second International Workshop
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
995
Fig. 4. Map showing the location of several sites in USA and Canada from which the NLC display of 22=23 June 1999 was observed. These sites span a latitudinal range of ∼14◦ and suggest an extensive NLC cloud was present over the central and western USA. Simultaneous optical measurements made from Delaware Observatory, ∼2500 km to the East of Logan, indicate unusually low mesospheric temperatures but no NLC were detected.
on the Planetary Scale Mesosphere Observing System (PSMOS) Project held in Toronto, Canada during May 2000. As a result of the ensuing discussions, it was decided to establish a network of PSMOS and other interested observers (including the well-established NLC CAN-AM amateur network in Canada) to help coordinate mid(and high) latitude NLC studies from the USA, Canada, Greenland and Europe. An e-mail alert was set up to convey NLC sighting information rapidly to participants from 32 institutions. A range of instrumentation was made available for these observations including Rayleigh and resonant lidars, visible and near-infrared interferometers, spectrometers, CCD imagers, and photographic cameras. One of us (M.S.Z.) has compiled a report summarizing the NLC CAN-AM observations for the summer of 2000 indicating distinct NLC displays on as many as 35 nights during the period May 28–Aug 27. The most extensive NLC displays may have occurred on June 27=28 and July 5=6 when they were observed from Europe and the USA. Many of the sightings, particularly after mid-July were from higher latitude sites but several of the earlier displays were seen from as far south as Vancouver, Canada (49:5◦ N) and Glen Ullin, ND (46:8◦ N). To our knowledge, the only potential NLC sighting at unusually low latitudes was made on 30 June=1 July from a mountain site near Ft. Collins, CO (∼41◦ N). On this occasion, one of us (M.J.T.) observed several faint luminous bands at low elevations in the evening twilight sky towards the northwest around 03:50 – 04:10 UT. The “clouds” appeared milky white in color and consisted of a single
elongated luminous band accompanied by several shorter wave-like structures. Separate observations of these clouds were also made from the same site by T. Nakamura (private communication). They were evident well after sunset under near new moon conditions and well away from city lights. Observations of meteorological cloud later that night (at similar low elevations) were quite diLerent in appearance. Unfortunately, although attempted, there are no photographic data to support these sightings. On the same night, but unknown to us at that time, an extensive, bright NLC display was observed over much of the Canadian Prairies. Fig. 5 shows a photograph of this display as seen from Namao, Alberta (53:7◦ N) at 06:30 UT. This event was also evident in the dawn twilight from Glen Ullin, ND indicating a signi2cant southerly extension. As a result of this potential NLC sighting, mesospheric temperature soundings were made on the following night from Ft. Collins (∼41◦ N) using the Colorado State University Na lidar system. Temperatures as low as 159 K were recorded in hourly averaged data with a mean nocturnal value of 168 K for the 83–84 km level (C.Y. She, private communication). These values are somewhat lower than expected from the seasonal climatology but signi2cantly above those necessary for sustained NLC cloud existence (see Section 5). Thus, it is possible that the unusual luminous clouds observed from northern Colorado on 30 June–1 July may have been a transient southerly extension of the large NLC display observed over Canada and the northern USA. However, without supporting photographic or other scienti2c evidence this must remain speculative.
996
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
Fig. 5. Example image of the bright, extensive NLC display of 30 June=1 July, 2000 photographed from Namao, Alberta, Canada (53:7◦ N) at 06:30 UT looking towards the NNW using 100 ASA 2lm with a ∼15 s exposure. This display was also observed from Vancouver, Canada and North Dakota and possibly from Ft. Collins, CO.
5. Discussion Modeling studies indicate that mesospheric ice particles can only nucleate and grow to an observable size (¿50 nm) if the ambient temperature in the region in which they form remains well below 150 K for periods of many hours (e.g. Gadsden, 1981; Jensen and Thomas, 1988). Due to potentially strong background winds in the upper mesosphere the geographic regions of nucleation, growth and detection can often be separated by many hundreds of kilometers. As the lifetime of an NLC particle is only a few seconds at temperatures ¿150 K, the background temperature throughout this region is a critical factor in controlling both the altitudinal and latitudinal extent of the observed clouds. Favorable conditions for NLC formation often exist in the polar latitude summer mesosphere (e.g. von Zahn and Meyer, 1989) but not normally at mid-latitudes where measured mesospause temperatures at ∼41◦ N typically range from 180 –190 K (Yu and She, 1995; She et al., 2000). Thus, the detection of NLC at mid-latitudes implies a major departure from normal mesospheric conditions. The simplest explanation for our 1999 NLC observations is that, for a limited period (∼2 days) close to the summer Solstice (when the cold polar air mass near the mesopause is expected to be well developed), an extremely cold air mass penetrated southward well into the mid-latitude mesosphere and persisted long enough for existing NLC particles to be transported equatorward from higher latitudes and=or for new particles to continue to nucleate and grow to form the
observed cloud patches at much lower latitudes than normal. This “dynamically driven” hypothesis is supported by the apparently large latitudinal extent of the NLC display observed on 22=23 June (¿15◦ ), as indicated in Fig. 4. Additional evidence for unusual mesospheric conditions on the night of June 22=23 comes from observations of the hydroxyl (OH) Meinel (3,1) and (4,2) band nightglow emission measured by a University of Western Ontario Michelson interferometer located at the Delaware Observatory, Canada (42:9◦ N, 81:4◦ W). (The location of this site is also indicated in Fig. 4.) The nocturnal OH emission originates from a layer of half-width ∼6–8 km, and She and Lowe (1998) have shown that measurements of its rotational temperature are a good proxy for the ∼87 km atmospheric temperature. Although 87 km is higher than the NLC base altitude, the summer-time OH emission layer encompasses the mesopause region where NLC nucleation is expected. In a series of combined imaging and spectrometric observations from central Alaska (∼65◦ N), Taylor et al. (1995) found that during overhead NLC displays, the column integrated OH rotational temperature was always ¡154 K. Fig. 6 shows the zenith nocturnal OH rotational temperature data (open circles) for June 22=23, 1999. Shortly after sunset the OH temperature was high around 185 K, however, within the next 2 h it dropped rapidly to a remarkably low value of ∼155–160 K (around 04 – 05 UT). This unusual behavior is exempli2ed by comparison with the mean (hourly averaged) nocturnal behavior of the OH rotational
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
997
Fig. 6. OH rotational temperature variation over the Delaware Observatory, Canada (41:9◦ N, 81:4◦ W) on the night of 22=23 June 1999. The temperature is the mean of the OH (3–1) and (4 –2) band temperatures. The mean variation (hourly average) on nine nights within 14 days of the 1999 summer solstice is shown for comparison. The standard deviations of the hourly means lie in the range 6 –8 K. The data indicate unusually low temperatures at the ∼87 km level during the period 04 – 05 UT when NLC were imaged well to the West.
temperature averaged over nine nights during the summer solstice period of 1999 (indicated by the solid circles). Comparison of these datasets shows that the OH temperature was about 10 K higher than average in the early evening of the 22=23 June, but was 10 –15 K below average for much of the night, and as much as ∼20 K below average at times during the period 04 – 05 UT (when the NLC were sighted from Utah). The Delaware observations ceased at about 07:00 UT due to meteorological cloud. In concert with these temperature changes the integrated zenith OH intensity also fell dramatically by a factor of over 4, from well above average to well below. These OH observations are consistent with either the passage of a very large amplitude wave with a period of at least several hours, or the incursion of a large volume of low temperature air into the mid-latitude mesosphere, possibly from the north. Unfortunately, meteorological cloud prevented observations from the Delaware Observatory on 23=24 June and several subsequent nights which would have helped clarify this unusual behavior. However, even at their lowest, the Delaware OH temperatures are considered too high for NLC existence. This fact is supported by independent measurements from the Delaware Observatory using a Rayleigh lidar. Although the lidar temperatures for this (and the previous) night show values as low as 156 K over the altitude range ∼83–85 km there was no evidence for an NLC scattering layer (R.J. Sica, private communication). Nevertheless, these two data sets provide strong evidence for a large deviation below the average summer-time temperature behavior during this period, from a similar mid-latitude site to Logan, UT, but ∼2500 km to the east of the con2rmed
NLC volume. Thus, it is feasible that the critical temperature limit of ∼150 K at which particle sublimation becomes very rapid was shifted signi2cantly southward over the western USA during this period. However, cold temperatures alone are not usually enough and there is a critical need for suNcient mesospheric water vapor for the successful development of an NLC. Thomas et al. (1989) and later Thomas (1996) have postulated that increases in lower atmospheric methane and carbon dioxide, possibly due to anthropogenic sources (biological and industrial), may have modi2ed signi2cantly the temperature and water vapor concentration at mesopause heights over the past century. If so, this would cause a substantial increase in the number and brightness of mesospheric clouds which could account for some (but not all) of the strong secular increase in NLC occurrence observed in recent years (Gadsden, 1990). Thomas (1996) further conjectured that a doubling of methane and carbon dioxide could “extend the NLC region to mid-latitudes, where they would be observable by a large fraction of the world’s population”. Whether the growing number of NLC sightings at unusually low latitudes is linked to the observed secular increase, and hence possibly to long-term changes in mesospheric temperature or water vapor content (perhaps of anthropogenic origin), has yet to be determined. The NLC sightings in the American Sector this past year (close to solar maximum conditions) indicate signi2cant activity despite the fact that the occurrence of NLC is expected to reduce during enhanced solar conditions. In this context, it is interesting to note that the incidence of NLC over Europe has fallen markedly over the past 3– 4 years from a maximum
998
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999
frequency in 1995 (Gavine, 2000). However, the NLC observations discussed herein suggest that the conditions responsible for the lower latitude displays observed over North America may be becoming more prevalent with time. This said, coordinated visual and lidar observations from several mid-latitude sites during the summer 2000 have revealed no further de2nitive cases of NLC at unexpectedly low latitudes. 6. Summary The ground-based sightings reported here provide new evidence for the occasional expansion of NLC to unusually low latitudes. Whether they are due to unusual dynamical forcing, or perhaps are an early indicator of signi2cant long-term changes in the climate of the summertime high-latitude mesosphere, remains to be seen. However, it is clear that such perturbations can have a large, but so far temporary, impact on the local mid-latitude mesospheric conditions. Further, coordinated ground-based and satellite observations are essential to help quantify changes in their occurrence frequency, to determine the mesospheric conditions under which they can exist at mid-latitudes and, to measure their geographical extent. The NASA Aeronomy of Ice in the Mesosphere (AIM) mission, which is currently under Phase-A study, may address many of these questions. Such investigations should also be broad enough to consider that mesospheric clouds might also become an occasional feature of the undisturbed low-latitude=tropical mesosphere. For example, Summers and Siskind (1999) have recently suggested that an enhancement of water vapor observed in the equatorial mesosphere and extending up to ±30◦ in latitude may well result from the association of atomic oxygen and molecular hydrogen onto the solid surface of meteoric debris. Under favorable conditions this meteoric dust, covered in water, could lead to the formation of a visible, low-latitude, mesospheric cloud layer. Possible evidence for such low-latitude cloud layers may already exist in the records of the cosmonauts onboard the Salyut-6 spacecraft who reported seeing mesospheric clouds within 40◦ of the equator on a number of occasions (Lazarev et al., 1987). Anomalous mesospheric clouds over Iraq and Syria (¡40◦ N) have also been reported from an aircraft Pying over the Caspian Sea (Vasiliev et al., 1974). However, it is also possible that high-altitude aerosol layers, resembling NLC (to the untrained observer), may be produced by exhaust gases from sounding rocket or satellite launches. Unfortunately, to date there have been no corroborative ground-based sightings or photography of such low-latitude events. Thus, while it seems clear that the two NLC displays reported here were outstanding examples associated with an extensive high-latitude cloud system, apparently penetrating deep into the mid-latitude mesosphere, the chance of sighting even lower latitude NLC, including “equatorial” mesospheric clouds, cannot be disregarded in future studies.
Acknowledgements We gratefully acknowledge the valuable reports by R.J. Sica, V.B. Wickwar, R. Moore and R. Keen supporting these unusual observations and the comments from the two referees. We thank D. Jereb for his help in obtaining the photographic data on 22=23 June and B. Atkinson for assistance with formatting the NLC imagery presented here. Support for the collation and analysis of these data and for optical observations during the summer 2000 has been provided by NSF Grant ATM-0003684. Note added at proof stage On 2/3 July 2001 NLC were again observed and photographed from Logan, UT at low elevations in the dawn twilight. Corroborative photographic measurements were also obtained from Glen Ullin, ND. These new data provide further evidence for the existence of NLC at unexpectedly low latitudes, and suggest that their occurrence rate may be increasing. References Carbary, J.F., Romick, G.J., Morrison, D., Paxton, L.J., Meng, C.I., 1999. Altitudes of polar mesospheric clouds observed by middle ultraviolet imager. Journal of Geophysical Research 104, 10,089–10,100. Donahue, T.M., Guenther, B., Blamont, J.E., 1972. Noctilucent clouds in daytime: circumpolar particulate layers near the summer mesopause. Journal of Atmospheric Science 29, 1205. Denver Post, 25 July, 1999, article by A. Schrader. Evans, W.F.J., Laframboise, L.R., Sine, K.R., Wiens, R.H., Shepherd, G.G., 1995. Observation of polar mesospheric clouds in summer, 1993 by the WINDII instrument on UARS. Geophysical Research Letters 22, 2793–2796. Gadsden, M., 1981. The silver-blue cloudlets again: nucleation and growth of ice in the mesosphere. Planet and Space Science 29, 1079–1087. Gadsden, M., 1998. The North-West Europe data on noctilucent clouds: a survey. Journal of Atmospheric and Solar-Terrestrial Physics 60, 1163–1174. Gadsden, M., 1990. A secular change in noctilucent cloud occurrence. Journal of Atmospheric and Solar-Terrestrial Physics 52, 247–251. Gadsden, M., SchrFoder, W., 1989. Noctilucent Clouds. Springer, Berlin. Gadsden, M., Taylor, M.J., 1994. Measurements of noctilucent cloud heights: a bench mark for changes in the mesosphere. Journal of Atmospheric and Solar-Terrestrial Physics 56, 461–466. Gavine, D., 2000. Noctilucent clouds over Britain and Western Europe, 1995 –1997. Journal of the British Astronomical Association 110, 4. Jensen, E.J., Thomas, G.E., 1988. A growth-sedimentation model of polar mesospheric clouds: comparison with SME measurements. Journal of Geophysical Research 93, 2461–2473.
M.J. Taylor et al. / Journal of Atmospheric and Solar-Terrestrial Physics 64 (2002) 991 – 999 Jesse, O., 1896. Die hFohe der leuchtenden nachtwolken. Astron. Nacht. 140, 161–168. Lazarev, A.I., Kovalyenok, V.V., Avakyan, S.V., 1987. Investigation of Earth from Manned Spacecraft. Issledovanye zemli c pilotir xhyenich kosmicheskikh korablyei Gidrometeoizdat, Leningrad, pp. 316 –319. Leslie, R., 1884. The sky-glows. Nature 30, 583. Livesey, R.J., 1998. British observations of noctilucent clouds in 1998 (provisional). Journal of the British Astronomical Association 108, 241. She, C.Y., Lowe, R.P., 1998. Seasonal temperature variations in the mesopause region at mid-latitude: comparison of lidar and hydroxyl rotational temperatures based on WINDII=UARS OH height pro2les. Journal of Atmospheric and Solar-Terrestrial Physics 60, 1573–1583. She, C.Y., Chen, S., Hi, Z., Sherman, J., Vance, J.D., Vasoli, V., White, M.A., Yu, J., Krueger, D.A., 2000. Eight-year climatology of nocturnal temperature and sodium density in the mesopause region (80 –105 km) over Fort Collins, CO (41◦ N, 105◦ W). Geophysical Research Letters 27, 3289–3292. Summers, M.E., Siskind, D.E., 1999. Surface recombination of O and H2 on meteoric dust as a source of mesospheric water vapor. Geophysical Research Letters 26, 1837–1840. Taylor, M.J., Lowe, R.P., Baker, D.J., 1995. Hydroxyl temperature and intensity measurements during noctilucent cloud displays. Annales Geophysicae 13, 1107–1116. Thayer, J.P., Nielsen, J.P., Jacobsen, N., 1995. Noctilucent cloud observations over Greenland by Rayleigh lidar. Geophysical Research Letters 22, 2961–2964.
999
Thomas, G.E., 1991. Mesospheric clouds and the physics of the mesopause region. Reviews of Geophysics 29, 553–575. Thomas, G.E., 1996. Is the polar mesosphere the miner’s canary of global change? Advances in Space Research 18, 149–158. Thomas, G.E., Olivero, J.J., Jensen, E.J., SchrFoder, W., Toon, O.B., 1989. Relation between increasing methane and the presence of ice clouds at the mesopause. Nature 338, 490–492. Vasiliev, O.B., Grishin, N.I., Zinoviev, V.V., Chapurskii, L.I., 1974. Observations of anomalous appearances of noctilucent clouds. Solar System Research 8, 153–156. von Cossart, G., HoLman, P., von Zahn, U., Keckhut, P., Hauchecorne, A., 1996. Mid-latitude noctilucent cloud observations by lidar. Geophysical Research Letters 23, 2919–2922. von Zahn, U., Meyer, W., 1989. Mesopause temperatures in polar summer. Journal of Geophysical Research 94, 14,647–14,651. Wickwar, V.B., Taylor, M.J., Herron, J.P., Martineau, B.A., 2002. Visual and lidar observations of noctilucent clouds above Logan, Utah, at 41:7◦ N. Journal of Geophysical Research, in press. Witt, 1967. Optical characteristics of mesospheric aerosol distributions in relation to noctilucent clouds. Tellus 20, 98. Yu, J.R., She, C.Y., 1995. Climatology of a midlatitude mesopause region observed by lidar at Fort Collins, Colorado (40:6◦ N, 105◦ W). Journal of Geophysical Research 100, 7441–7452. Zalcik, M.S., 1998. A possible increase in mid-latitude sightings of noctilucent clouds. Journal of the Royal Astronomical Society of Canada 92, 197–200.