Building off of the curves on https://drupal.star.bnl.gov/STAR/blog/bsumma/radiation-damage-curves-w-fixed-points , I note that the low, medium and high slopes tend to evolve similarly from their starting values (see for example slide 8 at https://drupal.star.bnl.gov/STAR/system/files/FMS_11Oct2017.pdf ), within errors

Therefore, my model has a single evolution from the time evolution for all 3 slopes (created by fitting the slope evolution of the mid-energy slope to a const-expo fit). The results of these fits are:
  Large Cells: slope = initial_slope * ( 4.45576 - exp(1.24338  -  0.00277894*t) )
  Small Cells: slope = initial_slope * ( 3.44334 - exp(0.896651 - 0.00217923*t) )

With initial slope values of {Low, Mid, High}
   Large: {0.0146288, 0.00822771, 0.002269 }
   Small: {0.0197536, 0.0048173,  0.0019957 }

We then need to fold in the effect from different eta regions giving different amounts of radiation. For this I'm fitting to an exponential, based on results from https://drupal.star.bnl.gov/STAR/system/files/FMS_21Sept2017.pdf, slide 9. From this I get a modification to the effective amount of time by:
  Large: exp(-7.366003 + 2.37613*eta)
  Small: exp(-7.43847 + 1.90730*eta )