Instruments:son

From CedarWiki

Sondrestrom IS Radar

The incoherent scatter transmitter station at Kangerlussuaq (Sondrestrom), Greenland (66.99N, 50.95W, 185m ASL) has been in operation since February, 1983. It was moved from Chatanika, Alaska in 1982-1983, where it operated from July 1971 to March 1982. On day 359 of 2001 at 110 km altitude, the apex magnetic coordinates were (73.0, 40.5) degrees. The magnetic inclination and declination angles were 80.1 deg and -35.2 deg. The magnetic local time at 0000 Universal Time (UT) is about 2137 MLT. The solar local time (SLT) is UT minus 3 hours and 24 minutes (-50.95/15.=-3.397).

There are many pulse types used at Sondrestrom, and the modes that have been used for world day measurements are briefly described here. A more detailed description of the mode used for each experiment is given in that experiment's data (CMODEXP).

Longpulse, usually 320 microsec (us) or 48 km resolution, is best in the F region. In the auto-correlation analysis, the data are integrated in range 21 km up to ~200km, integrated 42 km from ~200-400km, and integrated 63km above ~400 km. The longpulse data are available as 'uncorrected' Ne data at 2.1 km resolution (oversampled, e.g. kindat=5100, 9001), and as auto-correlation (AC) data (~20-65km resolution, e.g. kindat=5121,5501). The uncorrected ('raw' or 'power profile') data are uncorrected for the plasma effects on the radar cross section, which includes the plasma temperature and the deBye length of our transmitted frequency. These 'raw' data contain uncertainties in the F region due to Tr effects and they greatly range-smear the E-region densities. They can, however, be useful for contextual information.
Alternating code (ac), which is a sequence of coded 320us pulses with 20us baud used to improve the range resolution to 3 km, is best in the E region. The 'raw' data (kindat=9001) are not appropriate because they represent a range smearing of 48km, equivalent to that of an uncoded pulse. The ac data have poorer signal-to-noise ratios (SNR) than long pulse measurements because of they measure aurocorrelation functions at much higher actual range resolution. In the auto-correlation analyses (e.g. kindat=5521), the 3 km resolution is used to ~140 km, is integrated to ~9 km from ~140-220 km, and to ~42 km above ~220 km. The short pulse used for the zero lag of the ACF is 20us long (to match the range resolution of the other lags, but it contains about 16 times less power (3 km range smearing vs 48 km) than the long pulse.
Barker-coded pulse, although 50us, is 5 baud, resulting in 1.5km range resolution. The data exist as 'raw' (uncorrected) Ne data (~1.5 km resolution, kindat=9001) and the data are not sampled for auto-correlation functions. Consequently the auto-correlation (AC) data are meaningless. However, these 'raw' data are useful for E region and lower measurements of thin layers. The first barker-coded World Day data were in Aug and Nov 2005.

The data are routinely calibrated and are considered to be well within 10% accuracy. The calibration of received power (which is related to Ne) is done after each pulse by injecting a noise source of a known amplitude into the front end of the receiver chain. When the F-region electron densities are sufficiently high, typically near the maximum of the sun spot cycle, additional calibration is accomplished using the plasma line portion of the IS spectrum. The plasma line frequencies contain independent information about the electron density as well as the electron temperature and cross comparison with ion line power yields the desired calibration.

Because of the complexities of the modes and various assumptions, users should ALWAYS contact the data providers before using the data to get a good idea of the limitations of each kind of data. This requirement and the obligation to offer co-authorship are part of the [/catalog/Rules.html 'Rules of the Road'] for the CEDAR Database. A list of the kinds of data analyses for the Sondrestrom ISR are [#KINDAT listed] below the references.

Nonlinear least square fits of theoretical autocorrelation functions were made to observed autocorrelation fuctions (Lejeune, 1979; Wickwar et al, 1981). The ion composition is assumed to be 75% NO+ and 25% O2+ in the E region, and to be O+ in the F region, with a smooth transition from molecular to atomic ions in-between. At times, the proportion of molecular ions is underestimated (and thus both Te and Ti are underestimated). The effects of H+ are neglected, since these appear to be small at high latitudes.

From the 11 position scan, velocity components perpendicular and parallel to B are found. For each of the 3 pairs of observing directions furthest from the direction of the magnetic field, the velocity parallel to B is assumed zero, and pairs of measurements are combined to find the velocity components perpendicular to B. For each of the 2 remaining pairs of observing directions, the measurement along B is included in the derivation of the velocity components perpendicular to B. For different scans, such as the 3 position scan, data from 5 positions are used to find the 3 vector components at the time of the middle position, assuming linear changes in time. For a single azimuth position, the electric field, or ion drift velocity, can be determined as in de la Beaujardiere et al [1977]. The ion velocities can be very spiky at time resolutions of a few seconds (e.g. kindat=15610), or fairly smooth at time resolutions of a few minutes (e.g. kindat=15011), but one or the other analysis is done. F region averages over 2-12 min are also calculated (e.g. kindat=15520 and 15630), but not plotted.

LTCS Description

The experiments that produce E region neutral dynamics (cwinds), observe at least 3 positions in the near vertical (elevation angles 70 or above) with 4 kindats. The cwinds experiments include the official Lower Thermosphere Coupling Study (LTCS) campaigns.

kindat=15413 Resolved ion drifts from the longpulse data (~90 to >1000km, but best data in F region) along with ion and electron densities and temperatures.
kindat=15414 Resolved ion drifts from the alternating code data (~90 to ~500 km, but best data in E region) along with ion and electron densities and temperatures
kindat=15115 Neutral dynamics in the E region (~90-200 km) from the combined pulses (assume a model neutral atmosphere)
kindat=15116 Various electrodynamics calculations, electron densities and neutral dynamics in the E region (~90-200 km) from the combined pulses (assume a model neutral atmosphere)

The neutral atmosphere model used is given in parameter code 3574 where

-1 = Banks and Kockarts
0 = Jacchia 1971
1 = MSIS83 (Hedin, 1983), injected Tinf, iterated on Ap
2 = MSIS83, model Tinf
3 = MSIS83, injected Tinf, iterate on F10.7
5 = MSIS90 (Hedin, 1991), model Tinf

Most CWINDS data should have a value of 5 since MSIS90 is being used up to the present (2006).

Azeem and Johnson (1997) describe the CWINDS analysis. The standard SRI analysis for LTCS assumes:

a single ion species with mean ion mass of 30.5 amu (75% NO+, 25% O2+)
ion-electron thermal equilibrium below 110 km
neutral densities and temperatures from an MSIS model
ion-neutral collision frequencies assuming the neutral model from (3), and assuming 75% NO+ and 25% O2+ (1) ratios to the total ion content using Schunk and Nagy (1980) formulae

Above the ion-electron thermal equilibrium cut-off height (>110 km or >105 km) the auto-correlation functions are simultaneously solved for Te, Ti, Ne and Vi.

Azeem and Johnson relax the second assumption to:

ion-electron thermal equilibrium below 105 km

They also combine the gates differently and revise the neutral model values, and thus the ion-neutral collision frequency, in order to derive better neutral winds below 105 km especially. Their revised winds and model neutral temperatures and ion-neutral collision frequencies are available in kindat=30024 as a supplement to the original SRI outputs. The Azeem/Johnson neutral winds should be superior to the original CWINDS neutral winds.

Early LTCS runs prior to LTCS-8 (MLTCS-4) in July 1992 used 4*20usec multipulse data instead of the alternating code data. The multipulse scheme is generally applicable only up to about 125 km. The alternating code data has better statistics and lag spacing that is better suited for the decorrelation times of the E region, enabling measurements to be resolved throughout the E and F region. HOWEVER, ALL NEUTRAL WINDS ARE BAD ABOVE ABOUT 130 KM!

Please check the other CWINDS kindats to find out which neutral atmosphere model was used, and for plasma parameters etc. Be aware that some calculated parameters like Joule heating are estimates only (sometimes using different assumptions). Before using any of these parameters, including the neutral winds, please get in touch with the contact persons to find out about caveats and proper usage.

LTCS plots show both SRI CWINDS neutral winds and the revised Azeem/Johnson neutral winds. Some SRI LTCS periods are already in the DB, and old versions may be replaced. Older revised Azeem/Johnson neutral winds may also be replaced. After April 2002, the LTCS plots are only the SRI CWINDS neutral winds with plots of the MODEL neutral temperature and the MODEL ion-neutral collision frequency. Sometimes, there are 'LTCS' winds when there are high wind conditions at the site (e.g. 28-29 March 2005) although the basic mode is not LTCS.

References for the instrument and data processing procedures:

Azeem, S. M. I. and R. M. Johnson, Lower thermospheric neutral winds at Sondre Stromfjord: A seasonal analysis, J. Geophys. Res., 7379-7397, 1997.

de la Beaujardiere, O., R. Vondrak and M. Baron, Radar observations of electric fields and currents associated with auroral arcs, J. Geophys. Res., 82, pp 5051-5062, 1977.

Lejeune, G., A program library for incoherent scatter calculation, EISCAT Tech Note 79/18, Kiruna, Sweden, 1979.

Kelly, J.D., C.J. Heinselman, J.F. Vickrey and R.R. Vondrak, The Sondrestrom radar and accompanying ground-based instrumentation, Space Sci. Rev., 71 (1-4), 797, 1995.

Wickwar, V. B., C. Lathuillere, W. Kofman and G. Lejeune, Elevated electron temperatures in the auroral E layer measured with the Chatanika radar, J. Geophys. Res., 86, 4721-4730, 1981.

List of Kindats (kind of data analysis) for the Sondrestrom IS Radar

- - 5000 COMBINED DENSITY FIXED
  - 5001 POWER CHANNEL DENSITY FIXED
  - 5010 MAC (PRISIS) FIXED
  - 5020 OAC (PRISIS) FIXED
  - 5050 COMBINED DENSITY ELSCAN
  - 5051 POWER CHANNEL DENSITY ELSCAN
  - 5060 MAC (PRISIS) ELSCAN
  - 5070 OAC (PRISIS) ELSCAN
  - 5100 PHYNAL LP (TANDEN) FIXED
  - 5101 PHYNAL SP (TANDEN) FIXED
  - 5102 PHYNAL SP+LP (TANDEN) FIXED
  - 5103 VAST POWER CHANNEL DATA FIXED
  - 5106 EISCAT POWER FIELD ALIGNED 3 POS
  - 5110 MAC ZONE 1 (TANDEN/AKIDEN) FIXED
  - 5111 MAC ZONE 2 (TANDEN/AKIDEN) FIXED
  - 5112 MAC ZONE 3 (TANDEN/AKIDEN) FIXED
  - 5113 MAC ZONE 4 (TANDEN/AKIDEN) FIXED
  - 5120 OAC ZONE 1 (TANDEN/AKIDEN) FIXED
  - 5121 OAC ZONE 2 (TANDEN/AKIDEN) FIXED
  - 5122 OAC ZONE 3 (TANDEN/AKIDEN) FIXED
  - 5123 OAC ZONE 4 (TANDEN/AKIDEN) FIXED
  - 5124 ACFIT SPLINE KNOTS (TANDEN) FIXED
  - 5140 EISCAT TEMP, FIELS ALIGNED
  - 5141 EISCAT LP, RESULTAT
  - 5150 PHYNAL LP (TANDEN) ELSCAN
  - 5151 PHYNAL SP (TANDEN) ELSCAN
  - 5152 PHYNAL SP+LP (TANDEN) ELSCAN
  - 5160 MAC ZONE 1 (TANDEN/AKIDEN) ELSCAN
  - 5161 MAC ZONE 2 (TANDEN/AKIDEN) ELSCAN
  - 5162 MAC ZONE 3 (TANDEN/AKIDEN) ELSCAN
  - 5163 MAC ZONE 4 (TANDEN/AKIDEN) ELSCAN
  - 5170 OAC ZONE 1 (TANDEN/AKIDEN) ELSCAN
  - 5171 OAC ZONE 2 (TANDEN/AKIDEN) ELSCAN
  - 5172 OAC ZONE 3 (TANDEN/AKIDEN) ELSCAN
  - 5173 OAC ZONE 4 (TANDEN/AKIDEN) ELSCAN
  - 5174 ACFIT SPLINE KNOTS (TANDEN) ELSCAN
  - 5251
  - 5350 PHYNAL LP (TANDEN) AZSCAN
  - 5351 PHYNAL SP (TANDEN) AZSCAN
  - 5352 PHYNAL SP+LP (TANDEN) AZSCAN
  - 5360 MAC ZONE 1 (TANDEN/AKIDEN) AZSCAN
  - 5361 MAC ZONE 2 (TANDEN/AKIDEN) AZSCAN
  - 5371 OAC ZONE 2 (TANDEN/AKIDEN) AZSCAN
  - 5374 ACFIT SPLINE KNOTS (TANDEN) AZSCAN
  - 5400 PHYNAL LP (AKIDEN) FIXED
  - 5401 PHYNAL SP (AKIDEN) FIXED
  - 5402 PHYNAL SP+LP (AKIDEN) FIXED
  - 5450 PHYNAL LP (AKIDEN) ELSCAN
  - 5451 PHYNAL SP (AKIDEN) ELSCAN
  - 5452 PHYNAL SP+LP (AKIDEN) ELSCAN
  - 5500 ACPORT/MUSCOX - FITTED GATE DATA FROM MANUAL ANTENNA CONTROL
  - 5501 ACPORT/MUSCOX - FITTED GATE DATA FROM DWELLS
  - 5502 ACPORT/MUSCOX - FITTED GATE DATA FROM AZSCANS
  - 5503 ACPORT/MUSCOX - FITTED GATE DATA FROM ELSCANS
  - 5504 ACPORT/MUSCOX - FITTED GATE DATA FROM COMPOSITE SCANS
  - 5505 ACPORT/MUSCOX - FITTED GATE DATA FROM VSR ELSCANS
  - 5506 ACPORT/MUSCOX - FITTED GATE DATA FROM VSR COMPSCANS
  - 5520 ACPORT/MUSCOX - FITTED GATE DATA (A16) FROM MANUAL ANTENNA CONTROL
  - 5521 ACPORT/MUSCOX - FITTED GATE DATA (A16) FROM DWELLS
  - 5522 ACPORT/MUSCOX - FITTED GATE DATA (A16) FROM AZSCANS
  - 5523 ACPORT/MUSCOX - FITTED GATE DATA (A16) FROM ELSCANS
  - 5524 ACPORT/MUSCOX - FITTED GATE DATA (A16) FROM COMPOSITE SCANS
  - 5525 ACPORT/MUSCOX - FITTED GATE DATA (A16) FROM VSR ELSCANS
  - 5526 ACPORT/MUSCOX - FITTED GATE DATA (A16) FROM VSR COMPSCANs
  - 5650 PHYNAL LP (AKIDEN) AZSCAN
  - 5651 PHYNAL SP (AKIDEN) AZSCAN
  - 5652 PHYNAL SP+LP (AKIDEN) AZSCAN
  - 5811 - 5946 Reserved for Pre-Muscox acport
  - 5811 Pre-Muscox MAC Zone 1 DWELLS
  - 5812 Pre-Muscox MAC Zone 1 AZSCANS
  - 5813 Pre-Muscox MAC Zone 1 ELSCANS
  - 5821 Pre-Muscox MAC Zone 2 DWELLS
  - 5822 Pre-Muscox MAC Zone 2 AZSCANS
  - 5823 Pre-Muscox MAC Zone 2 ELSCANS
  - 5921 Pre-Muscox OAC Zone 2 DWELLS
  - 5922 Pre-Muscox OAC Zone 2 AZSCANS
  - 5923 Pre-Muscox OAC Zone 2 ELSCANS
  - 9000 ACPORT - over-sampled raw Ne, from manual antenna control
  - 9001 ACPORT - over-sampled raw Ne, from dwells
  - 9002 ACPORT - over-sampled raw Ne, from azscans
  - 9003 ACPORT - over-sampled raw Ne, from elscans
  - 9004 ACPORT - over-sampled raw Ne, from composite scans
  - 9005 ACPORT - over-sampled raw Ne, from vsr elscans
  - 9006 ACPORT - over-sampled raw Ne, from vsr compscans
  - 15011 EPEC/RVEL3PV - RESOLVED VELOCITY FROM 3 POSITIONS - OAC
  - 15012 EPEC/VAST - RESOLVED VELOCITY FROM 3 POSITIONS, MULTIPULSE - OAC
  - 15013 VAST - OAC
  - 15014 VAST - MAC
  - 15021 EPEC - RESOLVED VELOCITY FROM MULTI-POSITIONS (usually 8, 9 or 11) - OAC
  - 15022 EPEC - 1 CYCLE SUMMARY OF RESOLVED VELOCITY FROM MULTI-POSITIONS (not used)
  - 15023 EPEC - 11 POSITION - GATE JX, JY (Not Used)
  - 15031 IMPEC - AZSCANS - OAC
  - 15032 IMPEC - AZSCANS - MAC
  - 15041 OPEC - ELSCANS - OAC (Not Used)
  - 15042 OPEC - ELSCANS - SCAN SUMMARY
  - 15051 CHAMP
  - 15052 EISCAT VELOCITIES
  - 15061 UNTANGLE - DENSITY LOS
  - 15062 UNTANGLE - DENSITY BLEDEN
  - 15065 UNTANGLE - FLUX LOS
  - 15066 UNTANGLE - FLUX LOS
  - 15071 IMPEC - 3 POSITION - OAC
  - 15072 IMPEC - 3 POSITION - MAC
  - 15081 PREPEC - 11 POSITION - OAC
  - 15082 PREPEC - 11 POSITION - GATE JX, JY
  - 15083 PREPEC - 11 POSITION - 1 CYCLE SUMMARY
  - 15100 MERWIND/FWINDS - F REGION MERIDIONAL NEUTRAL WIND, B PARALLEL ONLY
  - 15011 EPEC/RVEL3PV - RESOLVED VELOCITY FROM 3 POSITIONS
  - 15012 EPEC - RESOLVED VELOCITY FROM 3 POSITIONS, MULTIPULSE
  - 15021 EPEC - RESOLVED VELOCITY FROM MULTI-POSITIONS (usually 8, 9 or 11)
  - 15022 EPEC - 1 CYCLE SUMMARY OF RESOLVED VELOCITY FROM MULTI-POSITIONS
  - 15113 EPEC - OAC AROUND SCAN SCAN4
  - 15114 EPEC - OAC AROUND SCAN ROTATED, INV OF EPEC BLEDEN
  - 15115 CWINDS/EWINDS/MUSCWINDS - NEUTRAL DYNAMICS FROM 3 POSITIONS
  - 15116 CWINDS - E-REGION ELECTRODYNAMICS FROM 3 POSITIONS, COMBINED PULSES
  - 15213 EPEC- OAC BINNED AROUND SCAN SCAN4
  - 15312 BLEDEN ELECTRO DYNAMIC BLEDEN
  - 15313 EPEC - OAC AROUND SCAN ROTATED, INV OF SCAN BLEDEN
  - 15413 CWINDS/MUSCWINDS/MUSCOX - RESOLVED VELOCITY FROM 3 POSITIONS, VAST LONGPULSE DATA
  - 15414 CWINDS/MUSCWINDS/MUSCOX - RESOLVED VELOCITY FROM 3 POSITIONS, VAST A16 OR MULTIPULSE DATA
  - 15510 MUSCOX - BLEDEN VEL (VAR OF DATA ) VEEST
  - 15515 MUSCOX - BLEDEN DEN (VAR OF DATA) VEEST
  - 15520 PVEEST/MUSCOX/VEEST - meridional velocity from elscans (VAR of MEAN)
  - 15525 BLEDEN/MUSCOX/VEEST- FIELD-ALIGNED PARAMETERS FROM ELSCANS, DEN (VAR of MEAN)
  - 15610 VEL2CMPBIN/VEL2CPM - RESOLVED VELOCITY FROM COMPOSITE SCANS, pair of records
  - 15613 CWINDS - RESLOVED VELOCITY LP (3 POS) CWINDS
  - 15614 CWINDS - RESOLVED VELOCIYT MP/A16 (3 POS) CWINDS
  - 15615 CWINDS - NEUTRAL DYNAMICS (3 POS) CWINDS
  - 15616 E-region electro-dynamic part of CWINDS
  - 15620 VEL2CMPB - Conductance record (scrapped - to 140)
  - 15621 VEL2CMPB - Conductance record (scrapped - to 500)
  - 15630 VEL2CMPBIN - RESOLVED F-REGION VELOCITY FROM COMPSCANS BINNED IN LATITUDE
  - 15700 BLECONS Detailed S datpwr (Elscans and Composites)
  - 15701 BLECONS Detailed A16 datuni (Elscans and Composites)
  - 15702 BLECONS Detailed combined (Elscans and Composites)
  - 15700 BLECONS - Field-aligned conductivities etc, longpulse (S datpwr), from scans (elscans and composites)
  - 15701 BLECONS - Field-aligned conductivities etc, A16 (A16 datuni), from scans (elscans and composites)
  - 15702 BLECONS - Field-aligned conductivities etc, combined pulses, from scans (elscans and composities)
  - 19004 ARCDEN - Signal record - Channel 1
  - 19005 ARCDEN - Signal record - Channel 2
  - 19006 ARCDEN - Density record
  - 19007 ARCDEN - Backscatter record
  - 19008 ARCDEN - Temperature record
  - 19009 ARCDEN - Rho Wave record
  - 19020 ARCDEN - Model
  - 19022 ARCDEN - Radiosonde
  - 19100 BLECONS - Field-aligned conductance etc, longpulse (S datpwr at 2 km intervals), from scans (elscans and composites)
  - 19101 BLECONS - Field-aligned conductance etc, A16 (A16 datuni at 2 km intervals), from scans (elscans and composites)
  - 19102 BLECONS - Field aligned conductance etc, combined pulses (S+A16 at 2 km intervals), from scans (elscans and composites)
  - 19120 PVEEST/CONDEN - Resolved velocity from elscans and conductance
  - 19650 BLECOND - Conductivity etc from dwell, longpulse (S datpwr) at 2 km intervals
  - 19651 BLECOND - Conductivity etc from dwell, A16 (A16 datnui) at 2 km intervals
  - 19652 BLECOND - Conductivity etc from dwell, combined (S+A16) pulses at 2 km
  - 19653 BLECOND - Conductivity etc from dwell, longpulse (S datuni) at fitted gate resolution
  - 19654 BLECOND - Conductivity etc from dwell, A16 (A16 datuni) at fitted gate resolution
  - 19901 BCompare Fitted A16
  - 19902 BCompare Raw Ne A16 corrected with LP temperatures
  - 19903 BCompare Raw Ne A16 corrected with A16 temperatures
  - 21710 BLE2CMP (Blecons + Vel2cmp) for Ennio

Summary Plots for Sondrestrom IS Radar

There are 4 areas in the summary plots for image plots as a function of height and time of the electron density, electron temperature, ion temperature, and either the line-of-sight velocity, or line plots of the perpendicular and parallel ion velocities closest to 300 km. The ion velocities from kindat=15011 are averaged over 1-3 min and look smooth, while the ion velocities from kindat=15610 starting in 1997 are 2-15 sec resolution and are spiky (e.g. Dec 1998 with 12 sec minimum resolution), or very spiky (e.g. Sep 1997 with ~2 sec minimum resolution). Analyses with kindat=15610 do not contain analyses with kindat=15011, so the ion velocity line plots are either smooth or spiky. The F region ion velocities averaged over 2-12 min (kindat=15630 starting in 2000 as an adjunct to 15610, or earlier as kindat=15520) are not plotted. The image plots are binned in 30 minutes and 40 km, except for the electron density, which has a height bin of 20 km for corrected densities, and 10 km for uncorrected. The data are averaged above 60 degrees of elevation. There are 2 days per page.

LTCS plots are neutral horizontal winds from SRI CWINDS (kindat 15115) or from the revised study of Johnson and Azeem (kindat 30024). After April 2002, they are just the SRI CWINDS with the MODEL neutral temperature and ion-neutral collision frequency.