Instrument/Model Description

The Aerospace 4 channel filter photometer at Poker Flat, Alaska (65.12N, 212.57E) was developed to estimate the energy flux of auroral electrons, their average energy Eo (usually from 0.1 to 30 keV using a modified Gaussian or Maxwellian shape), and a scale factor fo for the atomic oxygen densities [O] from an MSIS model atmosphere. The [O] densities using the scaling factor fo can be compared to GUVI/TIMED estimates. The technique is only valid for clear night conditions where the solar zenith angle (sza) is greater than 102 deg (code 4102=1), and there is enough auroral emission (427.8 nm brightness, code 2421). Poker Flat apex magnetic coordinates are (65.44, -95.56) at 110 km in 2001, with 0 UT corresponding to 12.93 MLT. The magnetic inclination and declination angles are 77.41 deg and 24.89 deg. The photometer looks up the magnetic field line with a field of view (fov) of about 1 deg. The data rate can be programmed, but typically each channel is integrated for 1 sec, where the whole cycle including filter moves takes about 8 sec.

The 4 channels correspond to:

1. N2+ (427.8 nm, blue) first negative group (1NG) 0,1 molecular band
2. OI (630.0 nm, red) forbidden
3. OI (844.6 nm, eight) permitted
4. N2 (871.0 nm or 871.4 nm) first positive group (1PG) 2,1 molecular band

A correction for atmospheric scattering is made after the backgrounds are removed using the ratio of 4278/8710 or 4278/8714. The nominal clear night sky diffuse aurora ratio of 4278/8710(4) is 2.60 for Poker Flat (code 2600). This ratio depends on the filters, band portions, and calibrations, and can be different using different filters. It is assumed to be constant otherwise, i.e., to be the same for different locations, seasons, magnetic activity and solar flux conditions. The normalized ratio (code 2601) is also the scattering correction factor (Icorr). 871.0(4) nm and 844.6 nm are assumed to have the same scattering (much smaller than the other two filters), and are not corrected for scattering. Icorr>1 means 4278 is being scattered into the fov, perhaps from a bright arc just outside the fov. Icorr<1 means 4278 is being scattered out of the fov, perhaps from a bright arc in the fov. Thus, the final 427.8 nm brightness must be divided by Icorr (code 2601), or the normalized 4278/8710(4) ratio, where the normalization is with respect to diffuse auroral conditions. Similarly, 630.0 nm emission is also corrected, but only half as much. The equations are:

1. blue = 427.8 new = (427.8 brightness minus background) / Icorr
2. red = 630.0 new = (630.0 brightness minus background)*(Icorr+1)/(2*Icorr)

The auroral electron energy flux is related to the corrected 4278 brightness:

The estimated auroral energy flux is found for all conditions, but the electron mean energy (Eo) and the scale factor for atomic oxygen (fo) are only found if there is sufficient 427.8 brightness. Eo and f0 are complex functions of the model conditions and the ratios of the calculated red/blue (rb, 6300/4278) and eight/blue (eb, 8446/4278) emissions. Plots of Eo and fo are given in Hecht et al (1989, 1999) and Strickland et al (1989) for various optical emission ratios where the x-axis is red/blue, and the y axis can be several different ratios. The red/blue ratio increases with smaller Eo, while the eight/blue ratio increases with larger fo. Increased solar flux or magnetic activity increases the neutral temperatures in models like MSIS, and decreases the O/N2 ratios at auroral emission heights, thus decreasing the eb ratio usually more than the rb ratio. The rb and eb ratios are also corrected for scattering after background emissions have been removed. The formulae are: rb=(6300/4278)*((Icorr-1)/2+1) and eb=8446/4278*Icorr.

The Strickland model (Strickland et al, 1989) takes an incoming auroral electron energy spectrum of a modified Maxwellian (code 2156=1) or Gaussian (code 2156=2) shape with average energy Eo, and transports it through a neutral atmosphere such as MSIS-1990 (Hedin, 1991) and a Chapman function ionosphere with a specified peak density, height and scale height. It then calculates the resulting auroral emissions. The average energy Eo is the energy flux over the number flux between 0.1 (formerly 1 keV) and 30 keV. The model is run at a particular location (codes 153, 156), for a specified day (code 22), solar local time (code 44) or UT, solar flux (code 350), and magnetic activity (code 340), with varying values of the auroral electron mean energy (Eo) in an atmosphere where the atomic oxygen is multiplied by various values of a scale factor (fo). The varying values of Eo and fo are interpolated in tables of the calculated red/blue (630.0 nm/427.8 nm) versus eight/blue (844.6 nm/427.8 nm) emission ratios provided the uncorrected 427.8 nm brightness (code 2421) is greater than 400 R. When Eo ~<0.3 keV or fo ~<0.2, the algorithm can fail and give negative (unphysical) values.

The estimated fo atomic oxygen scale factor must be used in conjuction with the original MSIS model atomic oxygen densities, or as an indication of change during the night from Joule or particle heating events that can cause local upwelling and decreased O/N2 ratios. The fo scaling factor applies to the whole MSIS [O] profile, which is given in kindat=18001 along with the temperature and molecular densities of [N2] and [O2] as a function of height between 82 and 390 km. There is a greater sensitivity to E region changes between 100 and 150 km where the auroral precipitation peaks, although the red line emission peaks around 200 km in aurora. This method really reflects changes in the O/N2 ratio, although only [O] is changed, not [N2]. This is approximately correct for the E region.

References for the instrument and data processing procedures:

Hecht, J. H., A. B. Christensen, D. J. Strickland, R. R. Meier, Deducing composition and incident electron spectra from ground-based auroral optical measurements: Variations in oxygen density, J. Geophys. Res., 94, 13,553-13,563, 1989.

Hecht, J. H., A. B. Christensen, D. J. Strickland, T. Majeed, R. L. Gattinger, A. Vallance Jones, A comparison between auroral particle charateristics and atmospheric composition inferred from analyzing optical emission measurements alone and in combination with incoherent scatter radar measurements, J. Geophys. Res., 104, 33-44, 1999.

Hecht, J. H., D. L. McKenzie, A. B. Christensen, D. J. Strickland, J. P. Thayer and J. Watermann, Simultaneous observations of lower thermospheric composition change during moderate auroral activity from Kangerlussuaq and Narsarsuaq, Greenland, J. Geophys. Res., 105, 27,109-27,118, 2000.

Hedin, A. E., Extension of the MSIS thermospheric model into the middle and lower atmosphere, J. Geophys. Res, 96, 1159-1172, 1991.

Strickland, D. J., D. L. Book, T. P. Coffey and J. A. Fedder, Transport equation techniques for the deposition of auroral electrons, J. Geophys. Res., 81, 2755-2764, 1976.

Strickland, D. J., J. R. Jasperse and J. A. Whalen, Dependence of auroral FUV emissions on the incident electron spectrum and neutral atmosphere, J. Geophys. Res., 88, 8051-8062, 1983.

Strickland, D. J., R. R. Meier, J. H. Hecht and A. B. Christensen, Deducing composition and incident electron spectra from ground-based auroral optical measurements: Theory and model results, J. Geophys. Res., 94, 13,527-13,539, 1989.

Summary Plots for Model MSIS-1990 Values

Jan 31, 2001 (SLT=3.17, F10.7=175, Ap=10) All data for this period

Summary Plots for Poker Flat 4 channel photometer at the Poker Flat website

Summary Plots for Poker Flat 4 channel photometer

All these nights used the same model background. Nights with bad moon background light, bad clouds, or no aurora are not submitted to the CEDAR DB, although they are plotted on the web page http://gedds.pfrr.alaska.edu/aerospace/pokerflatdata. There is an overall nightly quality code for the data that are submitted. Good nights are marked 'good', while nights with some problem like moonlight, weak aurora, or some clouds, are marked 'OK'. Summary plots of the observed 427.8 nm brightness, the corrected 427.8 nm brightness (corrected = (observed - background)/Icorr), and the Icorr ratio (corrected 427.8/871.4 divided by the Poker Flat diffusive value 2.6) show that the corrected 427.8 nm emission is increased over the observed if Icorr is less than 1. The Icorr ratio corresponds to the scattering correction from the diffuse auroral condition. Also plotted are the auroral electron energy flux and mean electron energy, and the atomic oxygen scaling factor for a standard MSIS profile are given. The time when the solar zenith angle is 102 degrees (12 degrees below the horizon, or nautical twilight) is given as vertical dashed lines, and the calculated values are not shown beyond those times since they are not valid. The periods when it is moonlit are shown as a dark line at the top of each day. When the background correction for 427.8 nm emission is high and the emission is low, the calculated flux can be small negative (essentially zero). However, there are times when the 427.8 nm brightness is larger than 400 R, and yet the calculated Eo and fo are large negative as in Nov 6, 2001, or only fo is negative as in Oct 28, 2001. The algorithm fails for low Eo or low fo, but can be re-run in a different mode on a case-by-case basis in collaboration with the data supplier.

Apr 15-Apr 24, 2001 All data for this period

Oct 03-Oct 12, 2001 All data for this period

Oct 13-Oct 22, 2001 All data for this period

Oct 23-Nov 01, 2001 All data for this period

Nov 02-Nov 11, 2001 All data for this period

Nov 12-Nov 21, 2001 All data for this period

Nov 22-Dec 01, 2001 All data for this period

Dec 02-Dec 11, 2001 All data for this period

Dec 12-Dec 21, 2001 All data for this period

Dec 22-Dec 31, 2001 All data for this period

Jan 01-Jan 10, 2002 All data for this period

Jan 11-Jan 20, 2002 All data for this period

Jan 21-Jan 30, 2002 All data for this period

Jan 31-Feb 09, 2002 All data for this period

Feb 10-Feb 19, 2002 All data for this period

Mar 02-Mar 11, 2002 All data for this period

Mar 12-Mar 21, 2002 All data for this period

Mar 22-Mar 31, 2002 All data for this period

Apr 01-Apr 10, 2002 All data for this period

Apr 11-Apr 20, 2002 All data for this period

Apr 21-Apr 30, 2002 All data for this period

Return to CEDAR homepage.

-Revised 30 Jan 2003 by emery@ucar.edu