Co-Register >2 SAR products

Hi everyone,

I am following this tutorial:

at page 23 it suggests to analyze a series of images to minimize the impact of atmospheric phase disturbances and unwrapping errors, which seems reasonable.

However, it says “All steps in this tutorial were repeated for each consecutive image pair, for instance 1+2, 2+3, 3+4, and so on…”.
Here I do not understand. I think it would make sense either to co-register everything to a single (properly selected) master or to co-register as: mst-slv1 and then slv1^-slv2, BUT where slv1^ is the slv1 image Co-Registered to mst in the first step, and not the original “raw” product, as seems to be proposed in the tutorial.

I think this has also been argument of discussion in other topics (e.g. Coregistrating more than two Sentinel-1 SLC IW products) but even in this thread it seems no optimal solution has been clearly found.

That post is quite old and probably a lot of things changed in SNAP since then. In particular, I see that the S1 TOPS Coregistration with ESD (which performs Back-Geocoding) can take as input more than 2 products and generate a big stack. I am wondering, is it a good procedure to Co-Register more than 2 products in one step? Does it result in the same as performing several pairs of stacks?