I have nothing specific to offer but I'd approach it by looking at options for discriminant validity for each subscale. The validity criteria would vary by subscale.
I agree with Gary that this might best be done at a factor level, since PWB's factors have quite different foci. For example, different aspects of broad P-E fit (P-O fit, P-J fit, P-G fit) may evidence discriminant validity for some of the factors (e.g., arguably P-G fit & mastery), but would be expected to be related to others (e.g., P-G fit & positive relations with others).
I partly agree with the previous answers. It may be better to look at the particular scales; yet since PWB total aims to represent PWB, For divergent/discriminant validity, Subjeective wellbeing indicators like PANAS and SWLS can be utilized, Keyes's Mental Health Continuum scale may also be of benefit. Depression scales, self-esteem scales (e.g Rosenberg) would be functional in criterion validty in general for your purposes. It would be wise howeever to look up for previously validated instruments for the linguistic community from which you gather the data.