The Incoherent Scatter Radar (ISR) at Sondrestrom (66.99N, 50.95W) is at the
approximate apex magentic latitude of 73.3 magN with IGRF inclination of 79.995
deg, declination of -31.867 deg (NW), and Bmag at 300 km of 0.4865 gauss, compared
with the actual Bmag at 298 km of 0.4884 gauss.  The radar was moved from 
near Fairbanks, Alaska at the Chatanika site (65.1N, 147.4W, ~65.1 magN)
by Stanford Research Institute International in 1982-1983.  A link to the
operations schedule, list of archived data, and summary plots of all the 
recent data is located at http://isr.sri.com.  All the data are in the SRI
madrigal database at http://isr.sri.com/madrigal.  The contact person
is Anja Stromme (anja.stromme@sri.com, (650)-859-5343, Fax (650)-322-2318,
SRI International, 333 Ravenswood Avenue, Menlo Park, CA 94025).  Please
contact her early to confirm correct use of the ISR data and to offer
co-authorship for any publications.
  ACKNOWLEDGMENTS:  The Sondrestrom incoherent scatter radar is supported
by the National science Foundation.
  Some of the data for the joint GEM-CEDAR Challenge of 2011 were taken from
kindat=5501, which is a long-pulse (LP) experiment of 320 microsec with height 
resolution of ~46km.  The short-pulse (SP) 20 microsec (~3km height resolution)
kindat=5521 experiment with better height resolution in the E-region is not used.
Kindat=5501 is a 3-position cycle of ~4.5 minutes at the zenith position, and then 
at 75 degrees elevation angle to the NE and NW.  The zenith position only was used
which had 42 altitudes from 70-578 km.  Only 10 altitudes between 140-500 km were 
retained in files labelled son_.._20m_profiles.  The observations below ~250km from  
the LP are smeared in altitude, and may not represent the real conditions, especially 
after they are binned in 20 min time segments for this Challenge.
  Corrected electron densities (Ne in m-3), ion (Ti in K) and electron (Te) 
temperatures, and the line-of-sight (LOS) velocities (Viz or Wi for the zenith position)
and their error bars were available for each look direction.  The zenith data were 
retained if the temperature error bars were <200 K, if Ti>500K for heights>300km, 
if the LOS error bar were <200 m/s, and if the Ne error bar was less than the Ne 
value. The medians of approximately 4 to 5 for 14-15 May 2005 and 0 to 2 for 14-15
December 2006 1-min profiles were found every 20 min to improve the data quality, 
where the number of retained data points in each time/height bin was kept. For 
bins with only 1 point, the error bar was the original error bar, while if there 
were 2 or more in a bin, the error bar was the standard deviation from the mean. 
The GEM-CEDAR Challenge included the values of Viz (Wi), Ne, Te, and Ti near 300 km 
(at 297.95+/-~23km) from kindat 5501.  
  The maximum F-region Ne value (Nmax in m-3) was found in each profile between 
140-500 km at height hmax (km) using a polynomial fit if the maximum was away from 
the edges.  The total electron content (TEC in m-2) was found between 140-500 km.
These parameters with the computed foF2 from Nmax were stored in files called
son_..._20m_nmax.  However, TEC should include the E region from the SP experiment, and 
hmax and Nmax were not always in the F-region, and so none of the variables in 
the son_..._20m_nmax files were used in the GEM-CEDAR Challenge.  
  Additional values of the resolved perpendicular east and north ion drifts ~300 km
using all 3 positions from the EPEC program with kindat=15011 were computed from fitted
electric fields.  The electric fields are independant of height, but we used Bmag of 
0.4884 gauss to derive the ExB/B^2 ion drifts.  These ExB drifts were similar to 
computed ion drift averages between 176-563 km.  These variables from kindat=15011 were
averaged over 20 min bins and save in files called son_epec_.._20m_ExB files where 
ViExBn (vperpN from EperpE/Bmag) and ViExBe (vperpE from -EperpN/Bmag) were used in 
the CEDAR-GEM Challenge.
  The analyst for making these 20-min data files and plots was Barbara Emery of HAO/NCAR.