Laboratory, ground-based, and airborne tunable diode laser systems: performance characteristics and applications in atmospheric studies

2

O) used as the target gas. One method utilizes the Allan variance. When our present aircraft system is operated in the laboratory, the Allan variance indicates a CH₂O detection limit of 0.031 ppbv for integration times of 25 s. This sensitivity corresponds to a minimum detectable absorbance of 1.0×10^-6, and this is within a factor of two of that reported by Werle et al. who used high-frequency modulation spectroscopy. The present instrument utilizes conventional low-frequency (2f=100 kHz) wavelength modulation. Instrument performance, obtained from replicate measurements of CH₂O standards acquired over time periods as long as 1.5 hours, on average resulted in a factor of two poorer precision than indicated by the Allan variance. Since replicate measurements precisely simulate the acquisition procedures employed, including the acquisition of sample and background spectra, they present a more meaningful measure of instrument performance. Preliminary evidence suggests that slow drifts in the laser wavelength control during acquisition of replicate measurements may play an important role in the above disparity. The resultant laser wavelength correction voltage, which is applied to counter such drifts, may also be a factor in this disparity. A limited number of replicate measurements with minimal drift in the laser wavelength yield much closer agreement between replicate and Allan variance precisions.