DL-FRONT is a Deep Learning Neural Network (DLNN) that was trained to detect weather fronts using spatial grids of near-surface atmospheric variables. The dataset is composed of various artifacts from training the network with label data from the Coded Surface Bulletin dataset from the National Weather Service Weather Prediction Center (WPC) and input data from the National Aeronautics and Space Administration (NASA) Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2). There are training loss and accuracy logs and the results of a variety of statistical analyses comparing the network outputs with the label data. There is also a movie that shows a side-by-side comparison between the label data and the network output using a year of data that was not used in the training.

{"references": ["The Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2), Ronald Gelaro, et al., 2017, J. Clim., doi: 10.1175/JCLI-D-16-0758."]}