Local stress evaluation:

when evaluating local stress at trunnion(pipe to pipe), Longitudinal stress SL and circumferential stress SC are calculated per WRC 107. I am asking how to combine these local stress with CAESAR stress and what are the allowables to use.

IMHO, I evaluated in 3 load cases

1. SUSTAINED
SL(CAESAR)+ SL(WRC) < SH

2. Expansion
SE(CAESAR)+ sqrt(SL(WRC)^2+SC(WRC)^2) < SA + SH - SL(SUSTAINED)

3 OCCASIONAL
SB(CAESAR)+SL(CAESAR)+SL(WRC) < 1.33 SH

What should I check circumferential stress with ?

would like to have some input from you guys. Thanks in advance.

Yes you "can" do this. Note that the WRC stresses (shell stresses) are calculated from CAESAR II loads, but you're right that they will be IN ADDITION TO the calculated CAESAR II pipe stresses (beam stresses).

In reality, a local stress (such as the WRC stress) is not as harmful as a membrane stress (such as the pressure stress), since 1) it is self-limiting to some extent and 2) it doesn't cause the entire section to go plastic. We could probably look to the Section VIII Div 2 code for justification of an increase. For example, local membrane and bending stresses are accorded a 50% increase of the allowable over the allowable for global membrane stresses. And stresses around pipe to cylinder connections (nozzles) are even treated more leniently -- bending stresses in those locations are treated as secondary stresses, counting towards the 3Sm (roughly equivalent to SA) allowable.