4. Example 2 - Deep observation

This is a Flamingos-2 imaging observation of a rather sparse field but with the objective of going deep. In most cases, the default recipe will work fine and match the science objectives. In some cases, often due to a crowded field or the need to go very deep on a faint target and have a very smooth background, there is a need to do an extra accurate sky subtraction.

The normal recipe reduces the frames that will be used for sky subtraction, identifies sources in them and mask them before creating a stack sky frame to use on a science image.

The ultradeep recipe will do the same, then use the stacked science image to detect sources, mask them and then redo the sky subtraction all over again, ensuring that even the very faint sources have been masked during the creation of the sky frames. This recipe take a lot longer to run and uses more CPU and memory resources.

The recipe can be run in one shot by calling the ultradeep recipe, or in stages by calling three segments of the recipe: ultradeep_part1, ultradeep_part2, and ultradeep_part3. We will use the segments in this tutorial.

• 4.1. Example 2 - Datasets descriptions
• 4.2. Example 2 - Deep observation - Using the “reduce” command line
• 4.3. Example 2 - Deep observation - Using the “Reduce” class