A statistical analysis of applying Haar wavelets to predict planetary boundary layer decoupling
The modeling of the planetary boundary layer (PBL) is essential to the prediction of local weather phenomena; however, the complexities of the PBL make it very difficult and computationally expensive to model. One of the complexities is the existence of multiple mixing layers beneath the PBL. These mixing layers, which are thought to form under optically-dense clouds, restrict the transport of aerosols, water, and energy throughout the PBL. Measuring the mixing layers through active remote sensing has proven to be a robust method given an adequate algorithm. However, these algorithms tend to be very complicated and computationally dense, and most LiDAR data is too sparse to determine spatial and temporal resolution. Using a Haar wavelet, a simplified algorithm was produced. The algorithm’s output is accurate some days and overestimates other days. The prediction of decoupling using this algorithm is highly aligned with the presence of clouds indicating that the simplified prediction of decoupling could be used even if the raw values need adjusting.