Comparison between Sodium Lidar,
Medium Frequency Radar and
Meteor Radar Wind Observations at


ALOMAR, Norway

B. P. Williams, J. Vance, K. Arnold,
and C. Y. She
Department of Physics
Colorado State University
Fort Collins, CO
biffw@lamar.colostate.edu



D. E. Gibson-Wilde and D. C. Fritts
Colorado Research Associates Division
Northwest Research Associates Inc.
Boulder, CO

W. Singer and R. Latteck
Leibniz-Institute for Atmospheric Physics
Kuhlungsborn, Germany.

Introduction

• Interest in MLT wind velocities and temperatures measured simultaneously and at high temporal and spatial resolution has motivated the continued development of the suite of instruments at the ALOMAR site (69$^{\circ}$N) in northern Norway (See Figure 1).
• Horizontal winds in the mesopause region have been measured using the MF radar technique at ALOMAR since 1998 [Singer et al., ESA, 1997, Hoffmann et al., JASTP, 2001].
• The all-sky SKiYMET meteor radar system installation was completed in 2001.
• The installation of Weber sodium resonance lidar commenced in June 2000 [She et al., EOS, 2002] with first winds observed in March 2001.

Comparison

• On January 17, 2002 the sodium lidar measured zonal winds with high signal levels and good stability for 5 hours using coplanar 30E and 30W beams. Signal levels at night in the winter are good (130 counts/shot), so the photon noise is generally less than 1m/s and 1K from 78 to 98km altitude for 1 hour time average and 3.1 km vertical resolution.
• The MF and meteor radars were also making routine wind measurements on this night with the meteor radar also inferring a nightly mean temperature from meteor trail diffusion rates.
• The lidar temperatures for the two beams show similar structure with a 2-5K offset. The offset is probably due to an error in telescope/beam overlap causing problems with accounting for differential absorption of backscattered light by the sodium layer. The temperatures are much more sensitive to this than winds. The meteor radar temperature is reasonable but many more days are needed to make a significant comparison.
• The lidar radial winds measured in the east and west directions are mirror images with a 8.5m/s zero-point offset (Fig2b). This offset is constant most of the night and height invariant, so it is likely due to a frequency offset in the laser. The 2 beam nightly average from the lidar agrees to within a few m/s with the meteor radar. The MF radar shows a similar altitude gradient with a 5-15 m/s offset from 80-90km (Fig2c).
• Figures 3 and 4 break down the comparison by altitude and time with similar results as the nightly mean.
• Scatter plots of lidar vs. meteor radar and lidar vs. MF radar both show a strong correlation greater than 0.9. The mean difference between the lidar and meteor wind is less than 1 m/s, while the lidar and MF mean difference shows a significant offset of 15 m/s. Both radars measure 10% smaller winds than the lidar, but this may not be statistically significant.

Discussion and Conclusions

• The agreement between the lidar and radar winds is much better than most of the published comparisons [Franke et al., JASTP, 2001, for example] and is very encouraging. Much more lidar data is needed, however.
• We are currently conducting an intensive summer campaign at ALOMAR with east and west beams so we should have many more lidar wind measurements in a few months.
• The signal in January was strong enough to allow us to calculate the variance in the east and west beam and hence the zonal momentum flux for periods between 6 and 90 minutes. This is encouraging but very preliminary due to the smal amount of data.

Figure 1: Beam positions at 90km altitude for lidar (+) and MF radar (solid circle) overlain on a map of the Andoya region of Norway. The lidar beams have abour a 50m diameter at 90km altitude. The MF radar has a beamwidth of about 40 degrees. The meteor radar beam is all aky and is larger than the map. Ground location of ALOMAR and radars is marked by the * at the center.

Figure 2: a: Nightly mean temperature for both beams plus hourly means spaced by 20K/hour. The meteor radar 12-hr layer averaged temperature is shown by the diamond. b: Nightly averaged radial wind measured by the lidar for the two beams. c: Average zonal wind for all good data between 22 to 27 UT for the lidar and both radars.

Figure 3: Hourly average zonal wind profiles for the lidar (solid line), meteor radar (blue X), and MF radar (red diamond).

Figure 4: Time series of zonal wind measurements for 8 altitudes from the lidar (solid line), meteor radar (blue X), and MF radar (red diamond).

Figure 5: Scatter plots of winds for a: lidar vs. meteor radar and b: lidar vs. MF radar. Histograms of the difference between winds from lidar and c: meteor radar and d: MF radar.

Figure 6: Variance of the east and west beams and zonal momentum flux for periods between 6 and 90 minutes. This calculation is very preliminary due to short dataset.