SUBROUTINE INTHYD(W0,X0,Z0,IWL,ILINE) C C Interpolation in temperature and electron density from the C hydrogen odening tables to the actual valus of C temperature and electron density C INCLUDE 'PARAMS.FOR' PARAMETER (TWO=2.) DIMENSION ZZ(3),XX(3),WX(3),WZ(3) C NX=3 NZ=3 NT=NTH(ILINE) NE=NEH(ILINE) BETA=WL(IWL,ILINE)/FXK IF(ILEMKE.EQ.1) THEN BETA=WL(IWL,ILINE)/XK NX=2 NZ=2 END IF C C for values lower than the lowest grid value of electron density C the profiles are determined by the approximate expression C (see STARKA); not by an extrapolation in the HYD tables which may C be very inaccurate C IF(Z0.LT.XNE(1,ILINE)*0.99.OR.Z0.GT.XNE(NE,ILINE)*1.01) THEN CALL DIVSTR(A,DIV) W0=STARKA(BETA,A,DIV,TWO)*DBETA W0=LOG10(W0) GO TO 500 END IF C C Otherwise, one interpolates (or extrapolates for higher than the C highes grid value of electron density) in the HYD tables C DO IZZ=1,NE-1 IPZ=IZZ IF(Z0.LE.XNE(IZZ+1,ILINE)) GO TO 20 END DO 20 N0Z=IPZ-NZ/2+1 IF(N0Z.LT.1) N0Z=1 IF(N0Z.GT.NE-NZ+1) N0Z=NE-NZ+1 N1Z=N0Z+NZ-1 C DO 300 IZZ=N0Z,N1Z I0Z=IZZ-N0Z+1 ZZ(I0Z)=XNE(IZZ,ILINE) C C Likewise, the approximate expression instead of extrapolation C is used for higher that the highest grid value of temperature, C if the Doppler width expressed in beta units (BETAD) is C sufficiently large (> 10) C IF(X0.GT.1.01*XT(NT,ILINE).AND.BETAD.GT.10.) THEN CALL DIVSTR(A,DIV) W0=STARKA(BETA,A,DIV,TWO)*DBETA W0=LOG10(W0) GO TO 500 END IF C C Otherwise, normal inter- or extrapolation C C Both interpolations (in T as well as in electron density) are C by default the quadratic interpolations in logarithms C DO IX=1,NT-1 IPX=IX IF(X0.LE.XT(IX+1,ILINE)) GO TO 40 END DO 40 N0X=IPX-NX/2+1 IF(N0X.LT.1) N0X=1 IF(N0X.GT.NT-NX+1) N0X=NT-NX+1 N1X=N0X+NX-1 DO IX=N0X,N1X I0=IX-N0X+1 XX(I0)=XT(IX,ILINE) WX(I0)=PRF(IWL,IX,IZZ,ILINE) END DO IF(WX(1).LT.-99..OR.WX(2).LT.-99..OR.WX(3).LT.-99.) THEN CALL DIVSTR(A,DIV) W0=STARKA(BETA,A,DIV,TWO)*DBETA W0=LOG10(W0) GO TO 500 ELSE WZ(I0Z)=YINT(XX,WX,X0) END IF 300 CONTINUE W0=YINT(ZZ,WZ,Z0) 500 CONTINUE RETURN END