Water vapour lidar system

Water vapour is a trace gas that plays a crucial role in the atmosphere. It is the dominant greenhouse gas, so that a detailed knowledge of the variation (temporal and geographical) of concentration is important for climate studies. Measuring and modelling water vapour concentrations with height and time are important aspects of weather forecasting. However, water vapour shows rapid variability both spatially and temporally. Near real-time measurements of the variability would help greatly with weather forecasting, as well as having significant impacts on understanding processes as diverse as the initiation of convection and fog formation.

This project is a joint initiative of the Optics and Laser Physics and Atmospheric Physics groups at the University of Adelaide. It grew out of the desire to understand the role that water vapour plays in determining the strength of VHF (~50 MHz) atmospheric radar echoes from the lower troposphere.

The dual wavelength system known as DIAL (differential absorption lidar) is the route we have chosen. Two laser wavelengths are transmitted, and the light back-scattered by aerosols is detected. The wavelengths are chosen so that one is absorbed by intervening water vapour and the other is not. From the returns at these two wavelengths we can infer the vater vapour concentration in the atmosphere. Our DIAL system is unusual in that it is designed to be low-cost.

A necessary first step is to confirm that there are enough aerosols to produce measurable scattered returns with our low-power lasers. The picture on the right confirms that this is the case, at least over Adelaide. Note that the laser pulse (single wavelength) has only 10% of the energy of the final DIAL system.

The next step is to calibrate the detection system so that the back-scatter coefficient can be measured. This has been done in a preliminary measurement, which then was fed into a DIAL sensitivity analysis. This is just a calculation of the random errors expected when measuring the water concentration. The result of this calculation is shown to the right. At 4km range, the random error remains below 10%, which is a very encouraging result.

Student Projects available!! Contact A/Prof Murray Hamilton or Prof. Bob Vincent.

Personnel