Identifying Horizontal Convective Rolls

Horizontal convective rolls (HCRs) are counter-rotating vortices in the boundary layer with a vertical extent of 1-2 km and a downstream extent of up to 1000 km.  There are updraft regions that transport moist surface air upwards into the boundary layer and downdraft regions that force dry air down. The formation of cloud streets can occur above the updraft regions as the moist air is cooled as it is carried aloft.  Due to the lack of instrumentation, boundary layer structures are difficult to observe and are not represented well by models but are important for the initiation of convection.  Deep convection is unable to occur until inversions are eroded.  Inversions prevent the growth of convection and create a cap to convection.   Radar imagery can be used to identify boundary layer structures. HCRs can be identified on plan position indicator (PPI) scans when the radar remains at the same elevation angle but the azimuth angle changes.  The PPI scans show HCRs as parallel lines of enhanced reflectivity in the updraft regions. Insects actively resist upward vertical motion so there is a higher concentration in the updraft regions therefore there is a higher reflectivity in the updraft regions compared to the downdraft regions. Radial velocity PPI scans show the component of the wind in the direction the radar is pointed. Radial velocity data can be used to plot Velocity-Azimuth displays (VAD) which are used to derive the mean wind speed and direction.  VAD plots can be used to identify HCRs following the method by Banghoffet al. (2020) where a measure of the variance is used to differentiate between different convective structures.  This method still requires manual inspection to differentiate between the presence of HCRs, convective cells and a null case (a case with disorganised convection).  This project aims to measure the wavelength of HCRs but first needs to identify HCRs cases.  Knowing the wavelength is important as it allows us to compare the size of HCRs with those represented in the models. This project uses data from radars based in southern England, taken over a period of several months, to identify boundary layer structures such as HCRs.  The data is provided every 10 minutes and signals are typically seen out to ranges between 20-40 km from the radar depending on the boundary layer depth.  The radar data will be used to measure the spacing between the HCRs (the distance between the updrafts) and their orientation.