An Approach for Identification, Tracking, and Prediction of Convective Planetary Boundary Layer Phenomena

Low altitude aircraft can conserve energy by flying in regions of ascent within planetary boundary layer (PBL) circulations such as open cell convection and horizontal convective rolls. To achieve this, it is necessary to accurately identify such areas of ascent and predict their motion. These regions are highlighted by areas of enhanced reflectivity within radar clear air returns associated with the concentrating of scatterers such as insects and can therefore be identified and tracked using radar data. In this work, regions of enhanced reflectivity associated with PBL structures are identified by applying a wavelet transform similar to the approach of Hagelberg and Helland (1995). Tracking is accomplished by comparing the location, size, and orientation of features that were identified at prior times, using the Advanced Algorithm for the Tracking of Objects. The motion is then estimated based on the prior motion history of the feature and the mean wind over the depth of the PBL. This allows for prediction of the position of the feature at future times. These predictions are also incorporated back into the identification of features at later times so that the phenomena are more consistently identified from one time to the next. This innovation addresses a known limitation of many tracking algorithms that inconsistent identification of features results in degraded tracking. Output from sample cases incorporating data from WSR-88D radars and the Texas Tech mobile Ka-band radars will be presented. Finally, we discuss an architecture for incorporating the region tracking into path planning for unmanned aircraft and provide several examples where flight planning can extract energy from the atmosphere to extend mission duration.