Conversion routines between fault plane representations (Ray Buland)
CNFG : CoNjugate Fault Geometry S(3),N(3),B(3) -> S(3),N(3),B(3) (overwrite) B changes in sign
CROSS : CROSS product of two vectors A(3),B(3) -> C(3)
FGF : Fault Geometry Forward STRIKE,DIP,RAKE -> S(3),N(3),B(3) B computed (B = N x S)
FGI : Fault Geometry Inverse S(3),N(3),B(3) -> STRIKE,DIP,RAKE B not used
FGMRTF : Fault Geometry Moment RT Forward S(3),N(3),B(3),M0 -> M(3,3) B not used
FGMRTI : Fault Geometry Moment RT Inverse M(3,3) -> S(3),N(3),B(3),M0 B computed (B = P x T)
FGPAF : Fault Geometry Principal Axes Forward S(3),N(3) -> P(3),T(3)
FGPAI : Fault Geometry Principal Axes Inverse P(3),T(3) -> S(3),N(3)
FGPL : Fault Geometry Planes (PAI) S(3),N(3),B(3) -> ZETA(3),ETA(3) ??? B used! N is already a nodal plane
JACOBI : JACOBI's method
MTOFF : Moment Tensor OF Forward M(3,3) -> G(6)
MTOFI : Moment Tensor OF Inverse G(6) -> M(3,3)
PAF : Principal Axes Forward ZETA(3),ETA(3) -> S(3),N(3),B(3) ZETA(3), ETA(3) recomputed (B)
PAI : Principal Axes Inverse S(3),N(3),B(3) -> ZETA(3),ETA(3) (1=P,2=T,3=B) B used!
vectors are in spherical coordinates: radius up, theta south, and phi east
horizontal = (theta,phi) plane
1 = sin(eta)
2 = cos(zeta) * cos(eta)
3 = -sin(zeta) * cos(eta)
According to Kagan: B = V x U (V = N: pole, normal to fault plane; U = S: slip vector)
B = N x S (same as in Buland subroutines)
B = T x P (in Buland subroutines B = P x T!)
T = (V + U)/sqrt(2) = (N + S)/sqrt(2)
P = (V - U)/sqrt(2) = (N - S)/sqrt(2)