Characterising vertical turbulent dispersion by observing artificially released SO 2 puffs with UV cameras

Abstract. In atmospheric tracer experiments, a substance is released into the turbulent atmospheric flow to study the dispersion parameters of the atmosphere. That can be done by observing the substance's concentration distribution downwind of the source. Past experiments have suffered from the fact that observations were only made at few discrete locations and/or at low time resolution. The Comtessa project (Camera Observation and Modelling of 4D Tracer Dispersion in the Atmosphere) is the first attempt at using ultraviolet (UV) camera observations to sample the three-dimensional (3D) concentration distribution in the atmospheric boundary layer at high spatial and temporal resolution. For this, during a three-week campaign in Norway in July 2017, sulfur dioxide (SO 2 ), a nearly passive tracer, was artificially released in continuous plumes and nearly-instantaneous puffs from a 9 m high tower. Column-integrated SO 2 concentrations were observed with six UV SO 2 cameras with sampling rates of several Hertz and a spatial resolution of a few centimetres. The atmospheric flow was characterised by eddy covariance measurements of heat and momentum fluxes at the release mast and two additional towers. By measuring simultaneously with six UV cameras positioned in a half circle around the release point, we could collect a data set of spatially and temporally resolved tracer column densities from six different directions, allowing a tomographic reconstruction of the 3D concentration field. However, due to unfavourable cloudy conditions on all measurement days and their restrictive effect on the SO 2 camera technique, the presented data set is limited to case studies. In this paper, we present a feasibility study demonstrating that the turbulent dispersion parameters can be retrieved from images of artificially released puffs. The 3D trajectories of the centre of mass of the puffs were reconstructed enabling both a direct determination of the centre of mass meandering and a scaling of the image pixel dimension to the position of the puff. The latter made it possible to retrieve the temporal evolution of the puff spread projected to the image plane. The puff spread is a direct measure of the relative dispersion process. Combining meandering and relative dispersion, the absolute dispersion could be retrieved. The turbulent dispersion in the vertical is then used to estimate the effective source size, source time scale and the Lagrangian integral time. In principle, the Richardson-Obukhov constant of relative dispersion in the inertial subrange could be also obtained, but the observation time was not sufficiently long in comparison to the source time scale to allow an observation of this dispersion range. While the feasibility of the methodology to measure turbulent dispersion could be demonstrated, a larger data set with a larger number of cloud-free puff releases and longer observation times of each puff will be recorded in future studies to give a solid estimate for the turbulent dispersion under a variety of stability conditions.