South Pole Michelson Interferometer
The South Pole Michelson Interferometer is operated by the Space Physics Research Laboratory of Embry-Riddle Aeronautical University (ERAU) with support from the Office of Polar Programs at the National Science Foundation (NSF). It is a designated Ground Based Instrument for the TIMED satellite mission, with additional funding from TIMED/CEDAR at NASA.
A near infrared (NIR) BOMEM MB160 Michelson interferometer (MI) has operated at South Pole Station (90.00S, 2835 m) in Antarctica since January 1992.
The MI scans the dark sky NIR airglow emissions between ~5000-10000 cm-1. A periscope directs the airglow into the MI sequentially from three locations at an elevation angle of 25 deg at azimuth angles of 0 (0 E), 120 (120 E), and 240 (240 E) deg. The periscope dwells at each position for about 4-15 min. Airglow hydroxyl Meinel (OH-M) band emissions peak at 87 km and are located between about 83 and 89 km for an elevation angle of 25 deg, the 87 km peak is located at a latitude spacing of 1.6 degrees, or at 88.4 S.
At the end of 2001 at 87 km above South Pole, the apex magnetic lat,lon are (-74.2, 18.6) deg. The magnetic inclination and declination angles are -73.1 deg and -28.2 deg (or 151.8 E toward the compass south pole), while the geographic coordinates of the apex magnetic south pole are at (74.3 S, 125.9 E). The magnetic local time at 0 UT is 2007 MLT.
The spectral resolution can be set either manually or through a keyboard command to any one of eight different values in the range 1 cm-1 to 128 cm-1 in binary steps. The normal resolution is 1.92847 cm-1 (~2 cm-1) in 2495 resolution steps between 4998.59 cm-1 and 10,001.04 cm-1. Wavelength is:
- wavelength (nm) = 1.e+7/wavenumber (cm-1)
The wavelengths thus go from 2000.56 nm with a spacing of about 0.8 nm to 999.896 nm with a spacing of about 0.2 nm.
The MI has 2 inch optics, a field-of-view (fov) of ~2 deg and a throughput (optical collecting area times fov in steradians) of ~0.06 cm2 steradian). The MI is fitted with a three-stage thermoelectrically (TE) colled 1 mm diameter InGaAs (Indium Gallium Arsenide) detector with Noise Equivalent Power (NEP) of less than 1.e-14 W. Calibrations are performed at the beginning and in the middle of each observing season. The MI scans the NIR in a few seconds. The spectra are summed over several minutes to increase the signal to noise ratio. The dwell time at each look direction can be up to 15 min, with cycle times up to one hour for 4 look directions.
Absolute brightness calibration of the MI is accomplished with the aid of a blackbody source operating at 1273 K and illuminating a Lambertian screen. There is a different calibration number for each wavenumber, and after calibration, brightness is in kR/cm-1. The brightness in kR/nm is:
- brightness(kR/nm) = brightness(kR/cm-1) * wavenumber(cm-1) / wavelength(nm)
Selected wavenumber portions of the calibrated brightness in kR/cm-1 may be available in kindat 10001 or the raw spectra in kindat 7001. The brightness is converted to kR by summing over the wavenumbers or wavelengths in a way to account for the resolution of the MI. However, the variations in brightness are good, but the absolute numbers should not be used.
Eleven S OH Meinel (v',v'') bands (from (2,0) to (9,5) with delv=2, 3 and 4) are between wavenumbers 5882-10,000 cm-1 or 1700-1000 nm. The upper and lower vibrational states are designated as v' and v''. Each band has 6 branches (Q1, Q2, R1, R2, P1 and P2). Each branch consists of a large number of rotational lines. The rotational lines of P1 and P2 of various OH-M bands are used to determine the rotational population distribution of each OH-M (v') state. This distribution is thermalized to the ambient kinetic temperature of the air in the mesopause region [Sivjee and Hamwey, 1987].
The temperature and the brightness are related by a Boltzman distribution as described in equations 1 and 2 of Walterscheid and Sivjee . The temperature values depend on several molecular parameters including Einstein's radiative transition probability A(v',J) coefficients which are derived from theoretical quantum mechanical calculations. Systematic errors in these A coefficients can result in temperature differences up to 40 K derived from different OH-M bands. However, studies of planetary, tidal and gravity waves require only measurements of changes in temperature, or delta(Tn)/Tn, which is about the same for each band. The brightest bands are usually the (4,2) and the (3,1) bands. The (3,1) band is least affected by water vapor absorption. Thus the OH-M (3,1) band is preferred for derivation of the modulations of the mesopause kinetic temperature and air density by planetary, tidal and gravity wave actitivites. The latest A values of Nelson et al.  and F values of Coxon  are used in deriving mesopause kinetic temperatures from the relative brighntess of 3 P1 and 3 P2 rotational lines of the OH-M (3,1) airglow emission. Clouds reduce the brightness of the OH-M bands, but the temperature is unaffected since it can be found from the ratio of the brightnesses from 2 lines, and 5-8 lines are used for better statistics. However, scattering of airglow emissions by clouds smears out some of the directional information.
The 8 brightest peaks of the P branch of the OH-M (3,1) and (4,2) bands are listed below with their wavelength and wavenumber peaks, their angular momentum (orbit plus spin) of state J, rotational term values F(J), and Einstein's radiative transition probability A(v',J) values.
OH-M(3,1)----------------------> OH-M(4,2)----------------------> P line J Wvl(nm) Wvn(cm-1) F(cm-1) A(s-1) Wvl(nm) Wvn(cm-1) F(cm-1) A(s-1) P2(2) 0.5 1518.30 6586.3 128.1 18.6 1596.9 6262.1 P1(2) 1.5 1523.67 6563,1 0.0 11.8 1602.7 6239.5 P2(3) 1.5 1528.37 6542.9 181.9 17.4 1607.6 6220.5 P1(3) 2.5 1532.81 6524.0 74.9 14.5 1612.4 6201.9 P2(4) 2.5 1539.10 6497.3 271.1 17.1 1619.1 6176.3 P1(4) 3.5 1542.79 6481.8 180.4 15.6 1623.1 6161.0 P2(5) 3.5 1550.5 6449.5 1631.3 6130.1 P1(5) 4.5 1553.6 6436.7 1634.7 6117.3
The wavelengths and F(J) are from Coxon , while the A(v',J) are from Nelson et al. . The half-width of each rotational line, measured from each rotational line profile at two points where the brightness of the profile is half the maximum (peak) brightness, is referred to as Full Width at Half Maximum (FWHM) of each rotational line. The BOMEM MB160 MI is normally operated with this parameter set at ~4 cm-1. The FWHM is referred to as the spectroscopic resolution of the MI. It is not the same as the spacing of the wavenumbers for the MI discussed earlier, which represents the interval in sampling the Fouriergram generated by the scaiing of the MI's etalons.
The brightness is found one of two ways. A least squares synthetic profile fit to the observed OH-M band brightness profile, coupled with the absolute brightness calibration yields the total band brightness. Alternatively, the brightness of the first six rotational lines of the P1 and P2 branches given above, coupled with the temperature determined from these same rotational lines yield the total OH-M band brightness [Walterscheid and Sivjee, 2001]. The two methods yield consistent brightness values. The first method is more involved and computationally intensive compared to the method using the 5-8 brightness peaks. Thus the latter is usually preferred. Which is used depends on what the conditions are (ie, aurora, etc).
The rotational (thermal) temperature and the band brightness are given in kindat 17001 for old OH (3,1), which is the preferred band, but analyses are made for using other bands which may have more usable times. New kindats 17031 and 17042 are for OH (3,1) and (4,2), respectively. The error bars in temperature represent relative errors determined from the variance in the least-squares fit of the rotational brightness [Sivjee and Hamwey, 1986], while error bars in band brightness represent one sigma value for random emission processes. The absolute brightness is not well validated and should not be used, but variations in the brightness due to gravity waves or other sources can be used with great confidence.
References for the instrument and data processing procedures
Coxon, J. A., Optimum molecular constants and term values for the X2Pi(v<=5) and A2Sigma+(v<=3) states of OH, Canadian J. Phys., 58, 933-949, 1980.
D.D. Nelson Jr., A. Schiffman, D.J. Nesbit, J.J. Orlando and J.B. Burkholder, H+O3 Fourier-transform infrared emission and laser absorption studies of the OH (X2Pi) radical: An experimental dipole moment function and state-to-state Einstein A coefficients, J. Chem. Phys., 93, 7003-7019, 1990.
Sivjee, G. G. and R. M. Hamwey, Temperature and chemistry of the polar mesopause OH, J. Geophys. Res., 92, 4663-4672, 1987.
Sivjee, G. G. and R. L. Walterscheid, Six-hour zonally symmetric tidal oscillations of the winter mesopause over the South Pole Station, Planet. Space Sci., 42, 447-453, 1994.
Sivjee, G. G., R. L. Walterscheid and D. J. McEwen, Planetary wave disturbances in the Arctic winter mesopause over Eureka (80 N), Planet. Space Sci., 42, 973-986.
Walterscheid, R. L. and G. G. Sivjee, Zonally symmetric oscillations observed in the airglow from South Pole station, J. Geophys. Res., 106, 3645-3654, 2001.
Summary Plots for South Pole Michelson Interferometer [OH] Airglow
Summary plots of the brightness and rotational (ambient) temperature from airglow hydroxyl Meinel (OH-M) bands (preferably the (3,1) band) that peak around 87 km are shown combining all look directions. The P1 and P2 branches of the OH-M (3,1) and OH-M (4,2) bands are between about 1517-1544 nm and 1595-1636 nm, respectively. The dashed line is the position of local midnight.
NOTE: The brightness is missing for 1992-1999.
- Apr 27-May 26, 1992 All data for this period
- May 27-Jun 25, 1992 All data for this period
- Apr 10-May 09, 1995 All data for this period
- May 10-Jun 08, 1995 All data for this period
- Jun 09-Jul 08, 1995 All data for this period
- Jul 09-Aug 07, 1995 All data for this period
- May 09-Jun 07, 1996 All data for this period
- Jun 08-Jul 07, 1996 All data for this period
- Jul 08-Aug 06, 1996 All data for this period
- Apr 01-Apr 30, 1997 All data for this period
- May 01-May 30, 1997 All data for this period
- May 31-Jun 29, 1997 All data for this period
- Jun 30-Jul 29, 1997 All data for this period
- Apr 10-May 09, 1998 All data for this period
- May 10-Jun 08, 1998 All data for this period
- Jun 09-Jul 08, 1998 All data for this period
- Jul 09-Aug 07, 1998 All data for this period
- Apr 01-Apr 30, 1999 All data for this period
- May 01-May 30, 1999 All data for this period
- May 31-Jun 29, 1999 All data for this period
- Jun 30-Jul 29, 1999 All data for this period
- OH-M(3,1) or OH-M(4,2) for Apr 17-May 16, 2002 All data for this period
- OH-M(3,1) or OH-M(4,2) for May 17-Jun 15, 2002 All data for this period
- OH-M(3,1) or OH-M(4,2) for Jun 16-Jul 15, 2002 All data for this period
- OH-M(3,1) or OH-M(4,2) for Jul 16-Aug 14, 2002 All data for this period
- OH-M(3,1) or OH-M(4,2) for Aug 15-Sep 13, 2002 All data for this period
- OH-M(3,1) or OH-M(4,2) for Apr 10-May 09, 2003 All data for this period
- OH-M(3,1) or OH-M(4,2) for May 10-Jun 08, 2003 All data for this period
- OH-M(3,1) or OH-M(4,2) for Jun 09-Jul 08, 2003 All data for this period
- OH-M(3,1) or OH-M(4,2) for Jul 09-Aug 07, 2003 All data for this period
- OH-M(3,1) or OH-M(4,2) for Aug 08-Sep 06, 2003 All data for this period
-Revised 12 Sep 2005 by Barbara Emery