LITE Level 1 System Calibration Constants

BACKGROUND INFORMATION

LITE system calibration constants, which include effective receiver aperture, nonvariable system parameters, and losses in the transmitting and receiving optics, are useful for retrieving aerosol data in the form of measurements of scattering ratio, aerosol backscatter, and aerosol extinction. LITE calibration constants were calculated for all nighttime, high-gain data at 532 nm and 355 nm. (LITE data at 1064 nm do not have sufficient signal-to noise (S/N) to calibrate in the same manner.) These constants can also be applied, with less accuracy, to daytime or nighttime, low-gain data for which the S/N is insufficient to perform high-altitude normalization or calibration.

CALCULATION

The LITE system calibration constant for a given wavelength is defined as follows:

S(z) = C* el* gain* ff* af* ba(z)* 10^(-att/20)* tr(z)/range² (1)

where

• S(z) is the background-subtracted signal at altitude z,
• el is the laser energy in Joules,
• gain is the PMT gain,
• ff is the filter function,
• af is the aperture function,
• ba(z) is the aerosol plus molecular backscatter in units of 1/m-sr at altitude z,
• att is the attenuator setting in units of dB,
• tr(z) is the two-way transmission between altitude z and the orbiter altitude,
• range is the distance in meters between altitude z and the orbiter altitude.

The calibration factors were calculated using LITE data between 30 and 34 km, assuming that the backscatter at those altitudes is due almost entirely to molecular scattering, which is proportional to density. Density was obtained from the LITE meteorological data. Two-way transmission above 30 km is assumed to be 1. (Two-way transmission is in fact closer to 0.99, with the major contribution at 355 nm due to Rayleigh extinction and the major contribution at 532 nm due to ozone absorption.) Both ff and af are 1 during normal nighttime configuration.

VARIABILITY

To determine how constant the LITE calibration "constants" really are, average calibration factors were calculated for each LITE orbit with nighttime, high-gain data. The figure below is a plot of these average factors at 532 nm and 355 nm. As can be seen, there is a significant decrease with time. The calibration factors at 532 nm decreased approximately 15% and 20% for laser A and laser B, respectively. The calibration factors at 355 nm decreased approximately 20% and 28% for lasers A and B, respectively. These decreases are assumed to be due, in large part, to decreases in optical throughput over the LITE mission, since changes due to diminishing laser energy have been compensated for.

The figure to the left shows the average calibration factors for each orbit with nighttime, high-gain data. Symbols on the top and bottom rows correspond to calibration factors for 532 nm and 355 nm, respectively. Different symbols were used to differentiate laser A orbits (1-82 and 148-150) from laser B orbits (83-147).

The variability of the LITE calibration factors within each orbit was analyzed by computing the percent difference from the mean for each orbit. The within orbit variability for the 532 nm calibration factors is generally within +/- 5% of the mean. Most of the points outside this range are from later orbits with decreased S/N. The within orbit variability of the 355 nm calibration factors is generally within +/- 3% of the mean, somewhat lower than for the 532 nm calibration factors.

IMPLEMENTATION

As a result of the variability of the LITE calibration factors discussed above, the following procedure was adopted for including calibration factors in the LITE Level 1 data. For the portions of an orbit with nighttime, high-gain data, the system calibration constant is based on the mean of 1000 laser shots. For the daytime or low-gain portions of an orbit, the average calibration factor for that orbit is used. In the case of orbits with no nighttime, high-gain data, a calibration factor is interpolated from average factors for adjacent orbits.

APPLICATION

Once the calibration constant is known, the only two unknowns in equation 1 are ba(z) and tr(z). The transmission can often be adequately modeled, especially for the stratosphere, by using zonal mean measurements of aerosol extinction and ozone absorption from the Stratospheric Aerosol and Gas Experiment (SAGE) II. The remaining unknown, backscatter at altitude z, is usually the measurement being derived. LITE daytime measurements were made with an interference filter in place and a smaller aperture than used during nighttime measurements. The Table below lists values for the reduction in signal due to the filter and smaller aperture.

LITE Filter and Aperture Functions
.	532 nm	355 nm
Filter function	0.65	0.30
Aperture function (Laser A)	0.90	0.95
Aperture function (Laser B)	0.74	0.83

REFERENCES

Osborn, M. T., Calibration of LITE data, ILRC 19^th International Laser Radar Conference, Singh, U., Ismail, S., and Schwemmer, G. K. eds., NASA/CP-1998-207671/PT1, 245-247, 1998.

Russell, P.B., T.J. Swissler, and M.P. McCormick, Methodology for error analysis and simulation of lidar aerosol measurements, Applied Optics, 18, 3783-3797, 1979.