You have anisotropic spectra, it means that they are presumably a sum of chaotically oriented particle spectra (generally, g and A depend on orientation with respect to field), so, if you are familiar to this particular type of EPR spectra, you can estimate g and A from particular point positions, but a more general and safe way - is to simulate spectrum and then find model parameters so that model spectrum fits experimental one. There are many software packages to do that, for example EasySpin (requires matlab). Particular procedure - see in the manual of program you choose. So at the end you will get g- and A- tensors
This is a typical "powder" EPR pattern of vanadyl (VO2+) ion in the zeolite matrix generated in trace levels during the processing of the zeolite due to reduction. Vanadium nucleus has spin I = 7/2, 100% natural abundance and hence will give an 8-line isotropic, equally spaced first derivative hyperfine pattern in solution. Unfortunately, being a powder pattern in the solid state where all orientations of the (VO2+) moiety are possible what you have is a superposition of spectra from a million orientations. Fortunately, even in such a case the EPR powder pattern is capable of producing "kinks" at the places where the orientations are unique. What I mean is that in spite of being in random orientations you can extract the principal values of the g- and A (hyperfine) tensors. Vanadyl species in more than 95% of the cases occur with an axially symmetric g and A tensor. In other words gxx=gyy = g-perpendicular and gzz is unique and is called g-parallel. The features of the g-perpendicular oocur usually as first derivative spectra, there being 8 such lines from g-perpendicular orientation. The features from g-parallel occur usually as absorptions (positive in one extreme and negative in the other extreme). I have indicated the 8 perpendicular hyperfine lines by the downward arrows in the attached figure. The center of these 8 lines will correspond to the g-perpendicular and the average separation (measured from end to end and divided by 7) will be the so-called A-perpenduclar. The 'parallel' hyperfine features are, unfortunately masked by the strong perpendicular features. But we are in luck, because the lowest field parallel feature (downward absorption) and the highest field parallel feature, the upward absorption, are visible as indicated by the two thick upward arrows. Measure the distance between them and divide by 7 to get the value of A-parallel. The midpoint between these two parallel feaures will correspond to the g-parallel. You can then compare these values from what is available in the literature and ascertain yourself that indded the spectum you have got in that of VO2+.in zeolite. You could heat the system in the EPR cavity and see whether the vanadyl species will rotate in the large cages and generate an isotropic spectrum!!