The CONUS1 dataset is a "toy" dataset that is small enough to be able to train a convolutional neural network on a laptop computer to do image translation from geostationary satellite images to ground-based radar images. It provides three input channels from GOES-16 ABI, one input channel from GOES-16 GLM, and one output channel from MRMS.

The methodology is described in detail by Hilburn et al. (2021). The ABI, GLM, and MRMS data sets were resampled to a common 3 km grid. A cloud height of 10 km was used for removing parallax displacements. Satellite and radar samples were matched in time with a maximum time difference of 2.5 minutes. GLM lightning groups were accumulated over 15-minute time periods.

You may read the dataset using any software that can read NetCDF-4.