PBL Height Retrievals from a Collocated Ceilometer, a Lidar and a Radar Wind Profiler: an Intercomparison

Recent work has established that PBL heights can be reliably retrieved from a widely varying set of instruments that are ubiquitous and autonomous. These include networks of ceilometers, lidars and radar wind profilers. Because none of these instruments alone can provide the spatial density necessary for a global PBL observing system, there is a need for combined multi-instrument surface observations. Consolidating all these estimates requires two important elements. The first is the establishment of common retrieval algorithms and methodologies with high fidelity in order to obtain PBL heights on a global scale. The second is the ability to reconcile differences among PBL heights retrieved from different instruments, which can be done using data from sites with collocated instruments. We present here results from our work on the second of these elements.

For the first element, PBL height datasets exist and have been produced using common methodologies. In particular for the instruments used in our work, these datasets are the GRWP-PBLH: Global Radar Wind Profiler Planetary Boundary Layer Height, available from https://grwp-pblh.hunter.cuny.edu/, that uses data from operational networks located in Australia, Europe and Japan; the Micro Pulse Lidar Network (MPLNET), a global network with data since 2000 available from https://mplnet.gsfc.nasa.gov, and the Unified Ceilometer Network (UCN), with data derived from ceilometers located in the Continental US (see https://www.ucn-portal.org). In addition, PBL height data for 1992-2012 at 26 locations over the US Central Plains are available and were obtained from NOAA Profiler Network (NPN).

In this presentation, we discuss our progress towards the second element: case studies from an intercomparison of PBL height retrievals from a ceilometer, a lidar and a radar wind profiler collocated at the Howard University Beltsville Campus in Maryland. Our case studies focus on observations for the months of July 2011, during DISCOVER-AQ, and February 2020, when a prescribed fire took place in northeastern Virginia. We use ancillary data from a near-by meteorological station and from MERRA-2 reanalysis to perform a detailed analysis of the different retrievals and to assess their potential for more accurately capturing the height of the PBL under varying meteorological conditions.

We discuss a characterization of the different estimates of PBL height as a function of meteorological condition, including cloud cover, surface fluxes and lifting condensation level deficit. Our case studies allow for an analysis of the seasonal and synoptic variability of PBL height. In addition, the results of our case studies will allow for a discussion of the limitations and errors associated with the retrievals from each instrument and will aid in providing recommendations for exploiting the synergy among them