It looks as a dynamic "impedance" graph (input acceleration/input force) by driving an electrodynamic shaker with a sine sweep, displayed as module only and linear scales
To exploit it, you should better edit modulus in log scales and edit associated phase plots as well, describe the test set-up (in particular the fixture of the PCB on the shaker), the equipment (impedance head???) - and then only we can comment these data !!!
How dare can someone perform shaker test without knowing anything on structural dynamics??? Good luck!
I now understand the PCB is clamped on its whole circumference with a combination of 8 screws and 3 clamps... And the response sensor is in the middle of the board, which presumably is the axis of the shaker.
Unfortunately none of your response graphs is exploitable
It must be plotted in logarithmic scales as a ratio response/reference (measured acceleration/ reference acceleration? force??) , not linear scales,
the reference should be preferably the input force, not the acceleration of the shaker head, assuming a force sensor is present on the shaker head assembly;
with also the frequency in log scale;
the phase must be plotted in +/- 180° scale with the same log scale;
and the scales must be explicit, together with the clear indication of the plotted sensor location
Log/log plots allow direct reading of mass and stiffness curves; phase is needed to confirm the modal character of the peaks.
The only statement possibly made at this stage is that the first dominant resonance (probably the drum response of the PCB) has a Q-factor of about 3.5/0.2 = 17 which is realistic (this 6% damping is provided mainly by the fixture). It is only meaningful if the PCB is mounted similarly on the real system!