Combining Lidar and Radon-222 to Measure Mixing Height

We have found that near-surface hourly radon-222 measurements can be combined with elastic backscatter lidar to obtain a higher quality time series of mixing height than is possible from lidar alone and a more quantitative measure of mixing height than is possible using only radon.

The usual method to determine mixing height from lidar relies on the optical contrast between aerosol-laden air within the boundary layer and clear air above. However, this approach can be confounded by meteorological conditions that lead to the formation of multiple aerosol layers within or above the boundary layer, or when the contrast between boundary layer air and the overlying air is weak. Furthermore, it is usually not possible to measure the mixing height at night; indeed many lidars are unable to measure anything at all closer than ~100m. Under these circumstances extra information would be helpful, either to incorporate into retrieval algorithms or as an alternative to lidar.

Radon-222 is a nearly ideal passive tracer. Its surface emissions vary slowly, compared with turbulent mixing, and removal from the air column happens over several days by radioactive decay. Assuming horizontal homogeneity, the near-surface concentration time series can be inverted to determine an effective mixing height, which is equal to the true mixing height when the boundary layer is well mixed. This method is applicable from around sunset until part way through the morning transition but, when used alone, it is qualitative.

During the morning transition, on account of vigorous mixing in the boundary layer, the two techniques can be merged by scaling the radon-derived mixing height to best fit the lidar. This simultaneously constrains the lidar-derived mixing heights, excluding false positives from the time series.

We have applied this approach to data from a two-week pilot study at a site with relatively difficult conditions for lidar-based mixing depth detection. Based on this, it seems that time series of mixing heights derived from a combination of lidar and radon observations would have more complete coverage, and therefore be more useful for applications such as model validation or pollution studies under a wider range of meteorological conditions.