When configuring saving settings for the JPEG format in the Saving Images section of the Configure digiKam window, you can choose between medium and high chroma subsampling, or disable it altogether. But what is chroma subsampling and what does it do? As you may know, JPEG is a lossy format, which means that it trades quality for a smaller size. One of the methods for achieving a smaller size is to store color information (chroma) at a lower resolution than brightness (a.k.a intensity or luma) data. This is possible due to the fact that the human eye is less sensitive to color than brightness, so discarding color information doesn’t produce a perceptible loss of quality.
Most graphical applications do that by using the 2×2 chroma subsampling method that breaks the image into 2×2 pixel blocks and only stores the average color information for each 2×2 pixel group. digiKam deploys two subsampling schemes. The 4:2:2 ratio is used for the Medium chroma subsampling that reduces color resolution by 1/3 with no perceptible difference. The High chroma subsampling uses the 4:1:1 ratio and reduces color resolution by half. This results in higher compression but tends to alter colors. The Chroma subsampling Wikipedia article provides more information about chroma subsampling and various schemes.
Unscientific tests show that the subsampling methods deployed by digiKam can achieve up to 20% size reduction (i.e., a ~19MB JPEG file can be compressed to ~15MB using the high chroma subsampling). For collections containing thousands of photos, this translates into significant disk space and bandwidth savings.