;+ ; NAME: ; oclc_epq03_bnu ; PURPOSE: (one line) ; calculate bnu from rbot, dtheta ; DESCRIPTION: ; calculate bnu from rbot, dtheta ; CATEGORY: ; Occultation lightcurve (oclc) ; CALLING SEQUENCE: ; bnubot = oclc_epq03_bnu(rbot, rtop, thetamid, deltathetamid, ib, ch) ; INPUTS: ; rbot = planet radius, at shell bottom border ; rtop = planet radius, at shell top border ; rbot[i] = r_{i + 1/2} ; rtop[i] = r_{i - 1/2} ; thetamid = bending angle at shell midpoint ; deltathetamid = change in bending angle over shell ; ib = index of top of inversion region ; ch = 1 if adding chapman top, 0 otherwise ; OPTIONAL INPUT PARAMETERS: ; none ; KEYWORD INPUT PARAMETERS: ; none ; KEYWORD OUTPUT PARAMETERS: ; none ; OUTPUTS: ; snubot = bending angle at shell borders ; COMMON BLOCKS: ; None ; SIDE EFFECTS: ; RESTRICTIONS: ; ; PROCEDURE: ; For the bins, follow ; Elliot, Person and Qu 2003, ; eq 35, 57 - similarity of integrand for Inu and Bnu ; eq. 51, 52 - summation form for Snu approx. to Inu ; For the topmost bin, note: ; ; eq 13 can be written ; ; 1 / inf ; nu(r) = - --- | theta(r')/r' dx ; pi / 0 ; ; We are concerned with the integral above the topmost bin. ; Let's call the radius of the topmost bin R = rtop[0] = r_1/2. ; Then the integral starts at ; X = sqrt(R^2 - r^2) ; ; 1 / inf ; nu_ch(r) = - --- | theta(r')/r' dx ; pi / X ; ; Assuming that theta/r is an exponential with scale height H, this ; integral is ; ; 1 theta(R) ; nu_ch(r) = - --- -------- * H * ch(R/H, asin(r/R) ) ; pi R ; ; where ch is the Chapman grazing incidence integral ; R/H = X_p is the radius/scale_height at base of integral ; asin(r/R) = zenith angle of integration path at R ; ; MODIFICATION HISTORY: ; Written 2006 Aug 23, Leslie Young, SwRI ;- function oclc_epq03_bnu, rbot, rtop, thetamid, deltathetamid, ib, ch imax = n_elements(rbot) bnubot = dblarr(imax) ; get thetatop0 and h, scale height and value at ; top of first bin thetartop0 = (thetamid[0] - deltathetamid[0]/2.)/rtop[0] h_theta = - (rbot[0]-rtop[0]) * thetamid[0]/deltathetamid[0] h = 1/(1/h_theta + 1/rtop[0]) Xp = rtop[0]/h ; first argment to Smith and Smith for i = ib, imax - 1 do begin zm = rtop[0:ib-1]/rbot[i] ; minus zp = rbot[0:ib-1]/rbot[i] ; plus chi = asin(rbot[i]/rtop[0]) ; chi bnubot[i] = -(1/!dpi) * $ total( (1/(zp - zm)) * $ ( ( zp * acosh(zp) - sqrt(zp^2 - 1) ) - $ ( zm * acosh(zm) - sqrt(zm^2 - 1) ) ) * $ deltathetamid[0:ib-1] ) if ch eq 1 then begin bnubot[i] = bnubot[i] - (1/!dpi) * $ ( thetartop0 * h * chapfun(Xp, chi,0.) ) end end return, bnubot end