The trim of Titanic as a function of time as the flooding progressed was derived from a list of key eyewitness accounts.  In each case the appropriate waterline
was determined based on their specific observations.  In doing so, the reported list of the vessel at that time was also factored in if the observation was anywhere
away from the ship's centerline. The estimated times of these observations, in minutes past the collision time, is taken either directly from the testimony, or implied
from some immediate action taken in relation to time taken from the testimony. (See Appendix H of my book, Samuel Halpern, et. al.,
Report Into the Loss of
the SS Titanic - A Centennial Reappraisal, The History Press, 2011.)

In the early stages of flooding in the vicinity of the mail room we have a number of eyewitness accounts which allow us to plot the rise of water in No. 3 hold as a
function of time.  This is shown in the diagram below along with names of the witnesses involved.
From other quantifiable observations we were able to establish how far the ship had trimmed down by the head over time.  These results are shown below with
the specific observations identified.  For comparison, we show a theoretical curve marked by conditions C1 through C7 that came from the pioneering work of
Hackett and Bedford.
Below is an animation showing how the vessel trimmed down by head over time.  These profile views do not convey the effect of the vessel's list over time which
was as much as 5 degrees to starboard in the early stages of sinking, to 10 degrees to port in the later stages of sinking. Thus, they represent the profile view
taken at the ship's centerline.