ScatteredLightBackground Flowchart

Below is a flowchart showing the decisions made in the ScatteredLightBackground class. This flowchart shows how scatterbrain converts TESS FFI images from MAST into a fits file containing weights. In this case we are fitting a 2D model to each TESS FFI frame. We are allowing for correlations in pixels in the spatial dimension. The resultant weight file can be combined with design_matrix objects to create a model of the TESS scattered light at any pixel, at any time. The tool will also extract a set of jitter components. These components are the top principal components of 5000 test pixel time series, chosen to be close to stars in the image. These jitter components can be used to correct pixel time series in TESS data. These components are used in StarScene to remove stars in the TESS FFIs, which is described below.

StarScene Flowchart

Below is a flowchart showing the decisions made in the StarScene class. This flowchart shows how StarScene removes the stars from TESS FFI images, after the scattered light has been modeled with the ScatteredLightBackground. In the case of StarScene we are fitting a 1D model to each TESS FFI pixel. We are allowing for correlations in pixels in the temporal dimension.

StarScene fits each TESS orbit separately. The final weight file fit by this class contains 2 (43 x 2048 x 2048) image stacks, i.e. one 2048 x 2048 pixel image per each of the 43 weights. There are two of these sets, because there are two orbits per sector.