The spectra contain two Lorentzian peaks. I merged and fit both peaks at each power, then plotted peak intensity and peak position against power with separate allometric fits for each peak. The fitted exponents are:

- **Intensity:** peak 1, \(b=0.933\pm0.009\); peak 2, \(b=0.713\pm0.006\)
- **Peak position:** peak 1, \(b=0.00409\pm0.00006\); peak 2, \(b=-0.00308\pm0.00117\)

The peak 2 fits reach the width limit used in the Lorentzian fit, so those peak 2 estimates are less reliable.

RESULT: b = 0.933 ± 0.009