Comparison of Vertical Velocity Observations Between the ARM Doppler Lidar and the 915 MHz Radar During MC3E
Rob Newsom | Pacific Northwest National Laboratory |
Richard Coulter | Argonne National Laboratory |
Marc Fischer | Lawrence Berkeley National Laboratory |
Edwin Campos | Argonne National Laboratory |
Category: Instruments
This poster presents a comparison of vertical velocity measurements from a 915-MHz radar and a collocated coherent Doppler lidar. The comparison was conducted at the Atmospheric Radiation Measurement Climate Research Facility Southern Great Plains site during a three-month period spanning the Midlatitude Continental Convective Clouds Experiment (MC3E). The accuracy of the lidar measurements was also assessed by comparison with sonic anemometers at the 25- and 60-meter levels on a nearby meteorological tower.
During this study, the radar was operated exclusively in its low-power, short-pulse mode and sampled only in the vertical direction. In this mode of operation, the radar provided vertical velocity measurements with a height resolution of 121 meters and a nominal temporal resolution of about 6 seconds. The low-power mode provided finer height resolution than the high-power mode but reduced sensitivity in clear-air. To compensate, the radar vertical velocities were reprocessed by averaging raw Doppler spectra over 60-second time intervals. The lidar, on the other hand, was operated with a pulse integration time of 1 second and a height resolution of 30 meters. To enable comparison, the lidar data were averaged and resampled to match the height and time resolution of the radar.
The lidar-radar comparisons were restricted to periods with no precipitation and sorted by surface heat flux and friction velocity measurements as determined from sonic anemometer data. The comparisons were also restricted to heights below 2 kilometers AGL. Above that level, measurements from both systems were dominated by noise.
As expected, the radar vertical velocities exhibited higher noise levels than the lidar measurements. Histograms of the difference between the radar and lidar vertical velocities show a tendency for the radar to be slightly negatively biased relative to the lidar in convective to neutral conditions. Under stable conditions the bias was essentially zero. When averaged over all stability conditions, the mean value of the radar vertical velocity was -9 cm s-1, compared to a mean of 0.04 cm s-1 for the lidar. For heights below 500 meters AGL, the correlation coefficients ranged from about 0.8 under convective conditions to 0.3 under stable conditions. A comparison between the lidar and sonic anemometers showed excellent agreement, with an overall RMS deviation of less than 50 cm s-1 and a linear correlation coefficient of greater than 0.95.
This poster will be displayed at ASR Science Team Meeting.