Temperature profiling capability of the ARM Raman lidar

Description

The Atmospheric Radiation Measurement (ARM) Climate Research Facility has operated a Raman lidar (RL) at the Southern Great Plains (SGP) site since 1996. This turn-key, nearly autonomous system transmits at a wavelength of 355 nm and incorporates 10 detection channels that measure both elastic and Raman-shifted backscatter returns from the atmosphere. The signals from the various detection channels are processed to generate a number of value-added products (VAPs), including water vapor mixing ratio, aerosol scattering ratio, aerosol backscatter, aerosol extinction, and depolarization ratio. In October 2005, two detection channels were added to the system to enable temperature profiling. These channels sense Raman-shifted backscatter arising from rotational energy state transitions in atmospheric N2 and O2 molecules. Each channel measures a slightly different portion of the rotational energy spectrum such that the ratio of the two signals exhibits a very well defined, nonlinear dependence on the air temperature of the scattering volume. Recently, significant progress has been made on the development of a new operational algorithm that derives temperature profiles from these two rotational Raman signals. The goal of this poster is to present results from a study to validate the RL temperature measurements. Calibrated RL temperature data are compared to temperature retrievals from a collocated AERI instrument. Calibration of the RL temperature data and estimation of the RL overlap function is achieved using data from radiosondes launched at the SGP Central Facility. By contrast, the AERI algorithm uses a physical retrieval approach that does not make direct use of the radiosonde data. For altitudes below 3 km AGL, the AERI retrievals provide a source of essentially independent temperature profile measurements, with time and height resolutions comparable to the RL data. Comparisons between the RL and AERI temperature data were performed over the span of one complete annual cycle. Preliminary results indicate good agreement between the two data sets. The correlations between time series of RL and AERI temperatures are better than 0.97 below 3 km, and median relative differences are less than 0.5% in this height layer. We also show that error estimates associated with individual RL data samples provide an effective means of quality control.