It is shown that, under clear sky conditions, water-vapor mixing-ratio meas
urements by solar-blind Raman lidars can be improved if the differential tr
ansmissivity is calculated by use of a single Raman signal instead of the u
sual Raman differential absorption lidar method, which allows one to exploi
t the large absorption cross section of ozone in this spectral region. We p
resent a discussion of statistical and systematic errors in both methods an
d show the results of a numerical simulation. (C) 2000 Optical Society of A
merica.