Planetary Boundary Layer Height Estimates From ICESat-2 and CATS Backscatter Measurements

• Main Authors: Stephen P. Palm, Patrick Selmer, John Yorks, Stephen Nicholls, Edward Nowottnick
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2021-09-01
• Series: Frontiers in Remote Sensing
• Subjects: planetary boundary layer height; ICESat-2; CATS; machine learning; LiDAR remote sensing; MERRA-2 PBL height
• Online Access: https://www.frontiersin.org/articles/10.3389/frsen.2021.716951/full

The lowest layer of the atmosphere in which all human activity occurs is called the Planetary Boundary Layer (PBL). All physical interactions with the surface, such as heat and moisture transport, pollution dispersion and transport happen in this relatively shallow layer. The ability to understand and model the complex interactions that occur in the PBL is very important to air quality, weather prediction and climate modeling. A fundamental and physically important property of the PBL is its thickness or height. This work presents two methods to obtain global PBL height using satellite lidar data from the Ice, Cloud and land Elevation Satellite-2 (ICESat-2) and the Cloud-Aerosol Transport System (CATS). The first method is a straightforward backscatter threshold technique and the second is a machine learning approach known as a Convolutional Neural Network. The PBL height retrievals from the two methods are compared with each other and with PBL height from the NASA GEOS MERRA-2 reanalysis. The lidar-retrieved PBL heights have a high degree of spatial correlation with the model heights but are generally higher over ocean (∼400 m) and over northern hemisphere high latitude regions (∼1,000 m). Over mid-latitude and tropical land areas, the satellite estimated PBL heights agree well with model mid-day estimates. This work demonstrates the feasibility of using satellite lidar backscatter measurements to obtain global PBL height estimates, as well as determining seasonal and regional variability of PBL height.