Laser-Induced Fluorescence - OSA Publishing

3 downloads 0 Views 445KB Size Report
OPN September 2007 | 35. Chiao-Yao She and David A. Krueger. Laser-Induced. Fluorescence: Spectroscopy in the Sky. Laser beams shoot into the. Colorado ...
Laser-Induced Fluorescence: Spectroscopy in the Sky Chiao-Yao She and David A. Krueger

Scientists are using laserinduced fluorescence to assess temperature and wind in a largely unstudied layer of the Earth’s atmosphere. Their results are crucial to our understanding of solarterrestrial relations and lay the groundwork for studies of global warming and atmospheric turbulence.

Laser beams shoot into the Colorado night sky, probing the secrets of the mesopause region of the atmosphere. The long exposure revealed the location of the North Star as well as low laser power (about 0.5 W per beam).

OPN September 2007 | 35

I

Using laser-induced fluorescence from atmospheric Na atoms to measure temperature and wind is a technical challenge, but the concept is straightforward. A pulsed, monochromatic laser beam illuminates Na atoms in the atmosphere and fluorescence is observed. The random motion of atoms in local thermal equilibrium gives rise to a Dopplerbroadened spectrum, and the background wind pushes atoms collectively, causing an overall Doppler shift; this principle

Illustration by Marko Batulan

Annual mean temperatures across each layer of the Earth’s atmosphere. Interestingly, summer is colder than winter in the mesopause. It reaches as low as -150° C in the polar region, when tiny ice particles form noctilucent clouds. At pressures of 2 x 10-6 atm., the air is dense enough to ablate meteoroids.

[ Doppler-free fluorescence spectroscopy and calibration curves ]

1.3

1.5 1.0

1.0

ν−

νa

0.5 0.0 -1.5

0.8 -0.8

ν+

-0.7

-0.6

νc νb

-1

-0.5

0

0.5

1

1.5

Frequency [GHz]

-0.5

0.6 0.5 0.4 0.3 0.2

-100 m/s

0.7

60 m/s

1.5

20 m/s

1.8

2.0

Calibration curves 0.8

-20 m/s

2.0

-100 m/s

N2D2 doppler-free spectrum

-60 m/s

2.5

Temperature ratio [(I + +I – )/2I a ]

Measurement principles

[ Mesopause: The coldest place on Earth ]

Fluorescence [a.u.]

t has long been a challenge for scientists to make in situ measurements of the mesopause region of the Earth’s atmosphere (80 to 110 km in altitude), because it is too high for airplanes and balloons, but too low for satellites. Meteorological rockets reach altitudes up to about 90 km. Although sounding rockets can reach well beyond that, they are rather infrequent. Exploring the thermal structure of the mesopause by remote sensing from space has also proven to be difficult—and in some ways even problematic—owing to its limited local time coverage. This scarcity of observations had earned for the region the nickname “the Ignorosphere.” Our work and that of others are helping to make the mesopause ignored no more. We use laser-induced fluorescence spectroscopy to study a naturally occurring layer of sodium atoms in the mesopause, which is formed by the atomization of meteorites plunging into the Earth’s atmosphere. Our data can be processed to determine the Doppler broadening and shift by the atoms, thereby allowing us to deduce atmospheric temperature and line-ofsight (LOS) wind. Our work validates satellite data and contributes to a fuller understanding of the dynamics of the upper atmosphere.

300 K 260 K 220 K 180 K 140 K

100 K

0.1 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6

LOS wind ratio [(I + –I – )/I a ]

(Left) The lidar transmits three frequencies sequentially at νa, ν+, and ν– , determined from the Doppler-free fluorescence spectrum of the Na D2 transition. (Right) The temperature and LOS wind are determined by the look-up table from the measured intensity ratios.

was first demonstrated for temperature measurement in England in 1979 by Lance Thomas’ group (Nature 281, 131-2). In the mid-1980s, profiles of

mesopause temperatures were measured for a number of years by scanning a tunable dye laser across the Na D2 spectrum in Andoya, Norway by Ulf von Zhan’s

>> Our work validates satellite data and contributes to a fuller und 36 | OPN September 2007

www.osa-opn.org

group in Germany (J. Geophys. Res. 96, 20841-57). Our system was deployed for the first time in 1989. It consisted of a singlefrequency, continuous wave (cw) tunable ring dye laser seeding a dye amplifier, which was pumped by a pulsed singlemode YAG laser. Doppler-free fluorescence spectroscopy at 589 nm provides a real-time marker of absolute frequency and allows identification of the absolute frequency of pre-selected Doppler-free features to within 1 to 2 MHz (1.0 MHz = 0.6 m/s). This allows us to lock the transmitter laser in real-time, leading to a robust semi-automated lidar operation. The Na D2 transition is between the ground state, 2S1/2 (two levels separated by about 1.77 GHz), and the excited state, 2P3/2 (four closely spaced levels within about 100 MHz). Thus, there are two groups of transitions, D2a around νa, and D2b around νb, depending on which ground state level is involved. In the laboratory, we counter-propagate a pair of cw beams from the ring dye laser through a laboratory Na cell, and observe the Doppler-free fluorescence spectrum of the Na D2 transition in a direction perpendicular to the beams. As the laser frequency is tuned, we observe features at the frequencies νa and νb and at the cross-over resonance at νc = 0.5(νa + νb). The sharp dip at νa may be used to lock the laser to within 1-2 MHz in absolute frequency. The incorporation of Doppler-free spectroscopy is, in our view, the critical innovation that enables both accuracy and robustness of this frequency agile narrowband lidar (Appl. Opt. 34, 1063-75). To measure temperature and LOS wind simultaneously, we lock the cw laser at the sharpest dip at νa in the D2a transition, and perform the threefrequency operation by using a dual-pass acousto-optic modulator to cyclically up-and-down shift the laser frequency

History of the project

T

he existence of sodium atoms in the Earth’s atmosphere has been known since the early 1920s, and clear emission lines from them were reported in the 1930s. At a 1987 Remote Sensing Conference in Cape Cod, we first learned about the sodium layer from a paper on the possible creation of a laser guided star. Almost instantly, we realized that if we tune the narrowband laser system in our lidar facility to sodium absorption (Na D2 line at 589 nm), we could detect laserinduced fluorescence from mesopause sodium atoms. These data could be processed to determine their Doppler broadening and shift, thereby allowing us to deduce atmospheric temperature and line-ofGroup photo commemorating the first North American sight (LOS) wind, respecobservation of mesopause region temperatures. From tively. Chet Gardner of the left: Hamid Latifi, Chiao-Yao She, Chet Gardner, Richard University of Illinois at UrBills and Jirong Yu. bana (UIUC), who already had a broadband sodium lidar for observing atmospheric sodium density and deducing the activity of atmospheric buoyancy wave (termed gravity waves by atmospheric scientists), directed us to the Aeronomy/CEDAR (Coupling, Energetics and Dynamics of Atmospheric Regions) Program in the NSF/Geoscience Directorate, which was interested in obtaining temperature and wind measurements in the mesopause region. In August 1989, Richard Bills, then a graduate student at UIUC, brought a 1.2-m-diameter Fresnel lens and receiver equipment, including his data-processing electronics, to Fort Collins, Colo. Within a week, Richard, along with CSU’s graduate student Jirong Yu and postdoc Hamid Latifi, had combined the UIUC receiver with the CSU transmitter. With the help of Raul Avarez, our resident laser expert and graduate student working on a tropospheric lidar project, we took measurements of mesospheric temperatures—the first in North America—over three consecutive nights. That same year, we received our first research grant on high-spectral-resolution sodium lidar development and geophysical observation. This new lidar and its initial observational results were published in 1990 (Geophys. Res. Lett. 17, 929-32). We began making nighttime temperature measurements on a regular basis in May 1991. Now in our 17th year, we are poised to conduct climate change studies. Between 1994 and 1999, we developed capabilities for wind measurement, and for measurement under sunlit conditions. Since 2002, with additional support from the NASA TIMED (Thermosphere, Ionosphere, Mesosphere and Electrodynamics) project, we upgraded from a one-beam to a two-beam system. The data we obtained were used to validate limb observations aboard the TIMED satellite. This type of lidar data has produced unique results on solar tides and buoyancy waves, and is accessible from the NSF/CEDAR database.

derstanding of the dynamics of the upper atmosphere. OPN September 2007 | 37

[ Monthly mean temperatures ] Temperature

T [K]

105 215 211

100

Altitude [km]

207 203 95

199 195 191

90

187 183 179

85

175

80 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month Monthly mean temperatures show the counter-intuitive thermal structure of the mesopause region, with cold summer and warm winter, as well as sharp wintersummer-winter transition.

from νa to ν+ = νa + 630 MHz, and then to ν– = νa – 630 MHz. We then obtain two intensity ratios from the lidar returns at the three frequencies; the ratios RT = (I+ + I–)/2Ia and RW = (I+ – I–)/Ia are most sensitive to temperature and wind changes, respectively. With the two measured ratios, one can retrieve both temperature and LOS wind from calibration curves derived from the theoretical resonant scattering spectrum of the Na D2 transition. Since the dwell time through a cycle of three frequencies is 3/50 s for a 50 Hz system, the ratios and thus temperature and wind retrieval will not be affected by the slower changes of the atmospheric Na density. The use of intensity ratios is of great practical importance, as it removes an unknown multiplicative factor due to effects of the atmosphere from the Earth’s surface to the sodium layer.

Counter-intuitive thermal structure In recent years, scientists have begun taking nocturnal temperature observations by sodium and potassium lidars deployed at different longitudes and latitudes. With 417 nights of temperature observations over eight years at Fort Collins (40.6 N, 105 W), we have compiled a climatological temperature map (Geophys. Res. Lett. 27, 3289–92), which is shown in the figure above. Two characteristics stand out. First, the altitudes of the mesopause (i.e., the altitude of the coldest temperature in the region) abruptly change in mid-May from near 100 km in winter to near 85 km in summer and are then reversed in mid-August. This two-level mesopause behavior appears to have global validity (Geophys. Res. Lett., 23, 141 4). Second,

between 80 and 100 km, the thermal structure is counter-intuitive, with colder summers and warmer winters. Below 80 km, Rayleigh lidar shows that this counter-intuitive thermal structure exists down to about 65 km. Radiative balance between solar heating and radiative cooling of the Earth cannot account for this behavior. Atmospheric physicists explain it in terms of the ubiquitous buoyancy waves that are generated in the troposphere. These waves are parcels of air oscillating due to an imbalance in buoyancy and gravity forces. As long as these waves propagate upward in an energy conserving manner (i.e., constant ½ ρν2), their amplitudes manifested in temperature, density and wind perturbations increase to compensate for the decrease in ambient air density, ρ. Between 65 and 100 km, the amplitudes of these zonally propagating waves become large enough for them to break, and the momentum is transported as a body force to the ambient air. This could change the wind direction and at the same time induce a meridional flow (drift) from the summer pole to the winter pole with the associated Coriolis force balancing the body force in the steady state. The continuity of this meridional drift induces a circulation, with air rising in the summer hemisphere and sinking in the winter hemisphere, leading to cold summers and warm winters in these altitudes.

Full diurnal cycle observation and solar tides This account of the counter-intuitive thermal structure suggests the importance of atmospheric waves and wave forcings. Since the details of sources and spectra of the buoyancy waves are not known, the detailed and quantitative pictures will help to constrain the many possible atmospheric models and will require

>> With additional support from the NASA TIMED project, we are measuring si

38 | OPN September 2007

www.osa-opn.org

[ A sodium vapor Faraday filter ]

Transmission

1

Na

P1

B

P2

Na vapor Faraday filter 1800 G, 4 cm 440 K, 343 K

0.8 0.6 0.4 0.2 0 -15 -10

-5

0

5

10

15

Frequency [GHz] (Left) Schematic of a sodium cell between two crossed polarizer in an axial magnetic filed, and (right) transmission function with roughly 90 percent peak transmission and narrow bandwidth, FWHM, of about 2 GHz.

[ Time series from a nine-day continuous campaign ] Temperature [K] —UT day 264-272, 2003

Altitude [km]

100

240 220 200

90

180

85

160 140

265

100 Altitude [km]

K

95

80 264

266 267 268 269 270 271 272 Zonal wind [m/s] —UT day 264-272, 2003

273 m/s 160 100

95

40

90

-20

85 80 264 100

Altitude [km]

more and continued observations with a time scale much shorter than a number of hours. On the other hand, solar radiation, which varies with season and during a day, is predictable. The tidal waves resulting from solar heating of H2O and O3 often dominate atmospheric perturbations. Thus, their definition and variability are of great importance, as are interactions with the mean states and other internally generated waves, local buoyancy waves and global planetary waves. To assess the full tidal influence, simultaneous, 24-hour continuous (i.e., full diurnal cycle) observations of temperature, zonal wind and meridional wind are required. We have developed and deployed a dual-path acousto-optic modulator, which allows simultaneous measurements of temperature and LOS wind, as well as a Faraday filter, which rejects sky background by a factor of 6,000 to 8,000, allowing daytime operation. In a sodium Faraday filter, the linearly polarized light at the atmospheric sodium frequency passes through the first polarizer and then propagates through the Na cell in an axial magnetic field where the polarization axis of the light is rotated by 90°—just right to pass the second polarizer. If small passive losses are ignored, the transmission of our Faraday filter has a maximum transmission of roughly 90 percent, and a full-width half-maximum (FWHM) of about 2 GHz (1.2 x 10-3 nm). Background light at other frequencies is blocked by the crossed polarizer. With additional support from the NASA TIMED project, we are measuring simultaneous temperature, zonal and meridional wind during full diurnal cycles. Our longest record has been a 14-day campaign in September 2003 with a nine-day continuous observation. The initial results of this dataset were

-80 -140

265 266 267 268 269 270 271 272 Meridional wind [m/s] —UT day 264-272, 2003

273 m/s 160 100

95

40

90

-20

85 80 264

-80 -140

265

266

267

268 269 270 Time [UT days]

271

272

273

The series shows cold temperatures and high winds in the mesopause region with solar tidal waves (24-hr and 12-hr periods), and shorter period buoyancy waves with downward phase progression and definite phase relations between temperature, zonal and meridional winds.

imultaneous temperature, zonal and meridional wind during full diurnal cycles.

OPN September 2007 | 39

Other ongoing and future efforts The mature narrowband Na lidar can monitor dynamic motions in the mesopause region. Our data have confirmed a general and qualitative understanding of upper atmospheric dynamics. On the other hand, the dataset from a long-period observation includes variability that challenges modelers and provides constraints to current models. Our continued observation of nocturnal temperatures contributes to climate change studies, a topic of considerable public interest. After seven years of observation, we reported episodic warming in response to the Mt. Pinatubo eruption, and, using the 11-year dataset, we deduced an altitude-dependent solar response,

[ Arctic temperatures and sodium densities ] Lidar-CONE temperature [ALOMAR]

Weber Na lidar signal photon counts

100

100

96

July 02, 02 1.83 UT Resolution: 0.5 km and 60 min 200 m

92 88 84

1.83 UT_West Beam 1.83 UT_East Beam CONE Temperature

80 100 120 140 160 180 200 220 240

Temperature [K]

Altitude [km]

Altitude [km]

published in December 2004 (Geophys. Res. Lett. 31, L24111, doi:10.1029/ 2004GL021165); this work was highlighted by the journal as “the longest continuous middle atmospheric lidar observation recorded to date.” According to the journal, “the data allow detailed study of the atmospheric influences of disturbances produced by gravity waves and short-period planetary waves and the causes behind the recurring tidal wind and temperature changes observed in this region.” In addition to the tidal perturbations that can be observed with 24- and 12-hr periods in all three fields, one can also notice tidal variability, reflecting the interaction of tides with buoyancy waves and planetary waves, with a considerable tidal enhancement on Julian day 267. This also illustrates the downward phase progression of the tidal perturbations opposite to the expected upward vertical group velocity. Sporadically, one also observes shorter period atmospheric buoyancy waves, more clearly in the wind field on day 265 below 95 km and on day 272 throughout the altitude range.

July 02,02 1.83 UT

96 92 88

1.83 UT 1.83 UT SNR [dB] BR [NLC]

84 80 0

20

40

60

80 100 120

Photon counts/75 m–1 min; PMSE/SNR; NLC/B.R.

(Right) Dramatic conditions exist in the summer mesopause region inside the Arctic, as shown here with the very low temperature and extreme temperature gradient, and (left) the interaction between ice particle and sodium atoms in summer.

with 0.04 K/SFU [Solar Flux Unit = 10-22Wm-2Hz-1, corresponding to ~7 K change between solar extremes] between 92 and 99 km, decreasing to nearly zero at 84 km and 103 km (Adv. Space Phys. 34, 330-6). After the natural variability is removed, the temperature cooling (negative) trend deduced from the 11-year dataset is 0.3 K/Y at 84 km, increasing to a maximum of 0.7 K/Y at 100 km, and then reverting to a positive trend of 0.1 K/Y at 105 km. The deduced cooling, like most other observational results, appears to be larger than model predictions, though the 2σ uncertainty is comparable to the trend for most altitudes. This large uncertainty reflects the difficulty in assessing temperature trends from a single site. Clearly, we need to observe more than two solar cycles in order to more robustly remove solar variability and reliably deduce temperature trends. Ground-based (GB) instruments with good local time coverage and space-borne instruments with global coverage are complementary. The NASA/TIMED

satellite, launched on December 7, 2001, has made unprecedented global measurements of not only temperature and winds, but many important chemical species such as O3, H2O, NO, etc. Our lidar measurements over Fort Collins provided initial validation of TIMED temperature measurements. Analysis of the TIMED temperatures of July 4, 2002, clearly demonstrated the global validity of the two-level, counter-intuitive thermal structure. Joint TIMED and GB lidar datasets have revealed tidal variability resulting from buoyancy wave interactions and planetary wave modulations. With five years of data, TIMED is poised for the evaluation of solar cycle effects with a global perspective. Community campaigns, including the 2002 joint Rocket, Lidar and Radar Campaign to study the polar summer mesosphere, observed the same volume above Andoya, Norway (69 N, 12 E). A graph that includes data from ALOMAR (Arctic Lidar Observatory for Middle Atmospheric Research) is shown in the graph to the left in the figure above; it

>> With five years of data, TIMED is poised for the evaluation of s 40 | OPN September 2007

www.osa-opn.org

was featured on the December 28, 2004, cover of Geophysical Research Letters. It shows hourly mean temperature measurements from two lidar beams, and an ionization gauge temperature observation on board a rocket at the middle of the hour. The polar summer mesopause temperature may be as low as 120 K, but the most spectacular feature of this figure was that both instruments observed a sharp temperature change of 40 K within 1 km near 90 km (Geophys. Res. Lett. 31, L24S06, doi:10.1029/ 2003GL01938). According to the lidar observation, this extreme temperature gradient actually lasted for about three hours. The graph on the right in the figure shows a simultaneous observation of Na density profile by Na lidar, a noctilucent cloud layer (which consists of ice particles >20 nm) by a Rayleigh lidar, and polar mesosphere summer echo (with ice particles about 10 nm in size) by radar. Taken together, they indicate that Na atoms are scavenged by ice particles, partially by small particles and totally by large particles (J. Atmos. Solar-Terres. Phys. 68, 93-101), making measurements in the polar summer mesopause by a sodium lidar more difficult. In order to understand the dynamics in the mesosphere and lower thermosphere, it is important to measure the momentum flux of buoyancy waves (J. Geophys. Res. 112, D09113, doi:10.1029/2005JD006179). In October 2006, we upgraded the CSU Na lidar from a two-beam geometry to a three-beam geometry, with East and West beams 20° off zenith, and the North beam 30° off zenith, allowing the measurement of zonal momentum flux, when the signal is strong at night. It also allows 24-hour continuous temperature, zonal and meridional wind observation to deduce solar tides, thus elucidating tidal-buoyancy wave interactions.

For more extended lidar observations and at different geophysical locations, a technical center and the three U.S. supported Na lidar facilities—the Norwegian Arctic, the one in Fort Collins, and the one in a still-to-be-determined equatorial location—have recently formed a Consortium of Resonance and Rayleigh Lidars (CRRL) sponsored by the National Science Foundation. In the next few years, the CRRL lidar sites will be poised to serve as the nucleus of cluster instrumentation sites for community-wide mesosphere and lower thermosphere campaigns. We would like to further extend the geophysical coverage to include remote locations

using an all-solid-state transportable narrowband sodium lidar.  We acknowledge the many excellent contributions of our students, postdocs and visitors over the past two decades. Without their dedication, the story of our odyssey would have been very limited indeed. We also express our appreciation for the encouragement from a number of colleagues in the University, notably Steve Lundeen, Steve Reising and Azer Yalin. [ Chiao-Yao She ([email protected]. edu) and David A. Krueger are with the physics department at Colorado State Member University. ]

[ References and Resources ] >> A.J. Gibson et al. “Laser observation

of the ground-state hyperfine structure of sodium and of temperature in the upper atmosphere,” Nature 281, 131-2 (1979). >> C.Y. She et al. “Two-Frequency

Lidar Technique for Mesospheric Na Temperature Measurements,” Geophys. Res. Lett. 17, 929-32 (1990). >> F.-J. Lübken and U. von Zahn. “Thermal

structure of the mesopause region at polar latitudes,” J. Geophys. Res. 96, 20841-57 (1991). >> C.Y. She and J.R. Yu. “Doppler-free

saturation fluorescence spectroscopy of Na atoms for atmospheric applications,” Appl. Opt. 34, 1063-75 (1995). >> U. von Zahn and J. Höffner.

“Mesopause temperature profiling by potassium lidar,” Geophys. Res. Lett. 23, 141-4 (1996). >> C.Y. She et al. “Eight-year climatology

of nocturnal temperature and sodium density in the mesopause region (80 to 105 km) over Fort Collins, CO (41° N, 105° W),” Geophys. Res. Lett. 27, 3289-92 (2000).

campaign,” Geophys. Res. Lett. 31, L24S06, doi:10.1029/2003GL019389 (2004). >> C.Y. She and D.A. Krueger. “Impact

of natural variability in the 11-year mesopause region temp erature observation over Fort Collins, CO (41 N, 105 W),” Adv. Space Phys. 34, 330-6 (2004). >> C.Y. She et al. “Tidal perturbations

and variability in the mesopause region over Fort Collins, CO (41 N, 105 W): Continuous multi-day temperature and wind lidar observations,” Geophys. Res. Lett. 31, L24111, doi:10.1029/ 2004GL021165 (2004). >> C. Y. She et. al. “Observation of anti-

correlation between sodium atoms and PMSE/NLC in summer mesopause at ALOMAR, Norway (69 N, 12 E),” J. Atmos. Solar-Terres. Phys. 68, 93-101 (2005). >> C.S. Gardner and A.Z. Liu. “Seasonal

variations of the vertical fluxes of heat and horizontal momentum in the mesopause region at Starfire Optical Range, New Mexico,” J. Geophys. Res. 112, D09113, doi:10.1029/ 2005JD006179 (2007).

>> D.C. Fritts et al. “Observations of

extreme temperature and wind gradients near the summer mesopause during the MaCWAVE/MIDAS rocket

solar cycle effects with a global perspective. OPN September 2007 | 41