Many thanks to Dr Alec Duncan of Curtin University, who has been very helpful in providing accurate values for the calculations. He has taken the time to customize a program for computing triangulation positions using his own best estimates for parameters, and provided the results. He consulted a tabular atlas for local SOFAR sound speed and provided additional corrections for temperature by latitude along each path. His chosen values of 1487.2 for HA01 and 1487.4 for RCS paths confirm that the 1488 m/s speed used here was a surprisingly good estimate. (His values will be adopted). Despite a slight 0.2 km difference in the center location of HA01 and larger signal arrival timing estimation difference of about 2.6 seconds, his computed path differs from the proposed search path by only 6.5 km in the area of interest.
This improved estimate of the wave speed has reduced the largest remaining error component (previously between 1460 to 1516), significantly reducing the proposed search path width.
Getting confirmation of the calculation methods used in the analysis is a big step forward!