------------------------------------------------------------------------------- Revision/update history for the SARTA package ------------------------------------------------------------------------------- ******************************************************************************* VERSION 1.00 ******************************************************************************* ------------------------------------------------------------------------------- 16 March 2001: v1.00 01-03-16 Initial package created by Howard Motteler and Scott Hannon. Uses: RTP version 0.96 KLAYERS version Feb 2001 (RTP version) SARTA version Aug00 nom/p16/m16 (RTP version) This version was delivered to Luke Chen/JPL ~20 March 2001 ------------------------------------------------------------------------------- 28 March 2001: v1.00 01-03-28 Minor changes for RTP v0.97 a) HEAD.obflag replaced by HEAD.pfields b) Added PROF.mwobs and PROF.mwcalc ------------------------------------------------------------------------------- 3 April 2001: v1.00 01-03-28 Trivial change: correct CSARTA comment string in "Src_sarta/incFTC_m16.f" to indicate the appropriate version of SARTA (said "Aug00nom" instead of "Aug00m16"). Also a similar correction for the p16 version. Version is unchanged. ------------------------------------------------------------------------------- 16 July 2001: v1.00 01-07-16 Minor update for RTP version 1.01. ------------------------------------------------------------------------------- VERSION 1.01 ------------------------------------------------------------------------------- 13 September 2001: v1.01 01-09-13 Minor changes: a) Change filenames in sarta include files from relative paths to absolute paths. Minor update to readme.txt file to mention absolute pathnames need to be set by the user. b) Check for surface temperature and pressure in input RTP file c) HEAD.ptype can now be type AIRSLAY as well as LAYPRO d) Set PROF.irinst to generic AIRS code number 800 e) Changes to check of input RTP channel freqs; now it first checks that the first element is not "nodata". ------------------------------------------------------------------------------- ******************************************************************************* VERSION 1.02 ******************************************************************************* ------------------------------------------------------------------------------- 28 November 2001: v1.02 01-11-28 Changes to the makefiles for the re-organization of the RTP, KLAYERS, and SARTA packages. Add VSARTA, VSCOEF, and VCLOUD to include file to assist in tracking the versions of the src code, coefficient database, and clouds. Also redo the block of code in opnrtp_sarta.f that assigns the SARTA comment string for the header of the output RTP file. Delete "nwantg" from command line arguments (in rdinfo_sarta.f). ------------------------------------------------------------------------------- 5 December 2001: v1.02 01-11-28 Trivial change to rdinfo_sarta.f to removed unused variable LENNB. Version not changed. ------------------------------------------------------------------------------- 28 January 2002: v1.02 01-11-28 Recompile using rtpV103 (trivial changes make_sarta, test_in.rtp, and comparison.rtp). No changes to sarta code. ------------------------------------------------------------------------------- 8 February 2002: v1.02 01-11-28 Update "readme.txt" to mention that references to the "Data_mie" database should be ignored (it is currently unsupported). No changes to sarta code. ------------------------------------------------------------------------------- 21 February 2002: v1.02 01-11-28 Simplify "Makefile" and "make_sarta" so that all object files are deleted prior to compilation. The dependency rules in the old makefiles were not quite right and did not always redo existing object files when necessary. ------------------------------------------------------------------------------- 18 March 2002: v1.02 01-11-28 Trivial changes for rtpV105. No changes to code. A new test file, test_obs_in.rtp, has been added to the Test dir for use in testing the (new for rtpV105) calflag field. ------------------------------------------------------------------------------- ******************************************************************************* VERSION 1.03 ******************************************************************************* ------------------------------------------------------------------------------- 12 September 2002: v1.03 02-09-12 Water continuum changed from 5 to 7 predictors/coefficients. This change was need to allow accurate modelling of Sergio Machado's revised water continuum in the 1300-1620 cm-1 region. The affected routines are calpar.f and calt1.f thru calt7.f, plus of continuum array dimensions in the incFTC.f include file. A new fast model has been generated for T_grating=155.1325 K with y-offset -13.50 um and fringes with T_filter=156.157 K. This has been named the "m135f" AIRS-RTA. ------------------------------------------------------------------------------- 20 September 2002: v1.03 02-09-20 Modify "opnrtp_sarta.f" to check for a "sarta" header attribute field. If found, the old value is overwritten; if not found, the new field is added. Prevoious to this change, the code always attempted to append the field. This resulted in a warning message and no update if the field already existed. ------------------------------------------------------------------------------- 1 November 2002: v1.03 02-11-01 Modify "rdrtp_sarta.f" to read in satellite zenith angle ("SATZEN") and satellite altitude ("ZSAT") from input RTP file. Modify main program ("sarta_rtp.f") for changes to "rdrtp_sarta.f" call. Main program now checks for a valid satellite altitude ("SALT") rather than always using the default value. The main program now also checks for satellite zenith angle ("SATZEN") and if available uses that instead of the scan angle ("SATANG") to determine the view angle. ------------------------------------------------------------------------------- 19 December 2002: v1.03 02-11-01 No changes to the source code apart from adding include file incFTC_m135f_dec02.f for the new December 2002 fast model coefficients. The include for the September 2002 fast model has been renamed incFTC_m135f_sep02.f. ------------------------------------------------------------------------------- ******************************************************************************* VERSION 1.04 ******************************************************************************* ------------------------------------------------------------------------------- 3 January 2003: v1.04 03-01-03 Added code to allow a fudged calculation of the solar contribution at large solar zenith angles. Previously SARTA would treat sun angles > 80 as night time. The model physics/equations used for the solar calculation breaks down at large angles, but this fudge allows the calculation to smoothly continue up to 89.9 degrees rather than stopping abruptly at 80 degrees as it had previously. BEWARE: I still expect the the solar calculation is POOR QUALITY for solar zenith angles larger than ~80 degrees. To implement this change, a few lines of code and new variables had to be added to sarta_rtp.f and calt[4567].f. ------------------------------------------------------------------------------- 6 February 2003: v1.04 03-02-06 Bug fix in routine calt6.f; ozone was using the wrong coef (21 instead of 23). Max errors were ~0.01 K in the 2394-2544 cm-1 regions. ------------------------------------------------------------------------------- 30 May 2003: v1.04 03-02-06 New coefficient database created using Clough et al's latest(?) continuum model, version 1 ("con1"). Database created 29 May 2003. ------------------------------------------------------------------------------- 26 June 2003: v1.04 03-02-06 New include file for Jun03 CrIS 0.2/0.4/0.8 cm OPD fast model. ------------------------------------------------------------------------------- 02 July 2003: v1.04 03-02-06 New include file for Jul03 version of AIRS fast model. The Jul03 fast model is the same as Dec02 except the set1, set2, optran, and co2 (but not solar or therm) database coefficients for modules 5 & 6 have been redone for channels at -10.5 um instead of -13.5 um. ------------------------------------------------------------------------------- 24 July 2003: v1.04 03-07-24 Correct error in "sarta_rtp.f" calculation of bottom fractional layer temperature. The old code gave the correct bottom layer temperature only when the fraction was one; otherwise it was typically off by a few tenths of a degree. ------------------------------------------------------------------------------- 05 August 2003: v1.04 03-08-05 Correct FIN & FOUT in "rdinfo_sarta_rtp.f" & FIN in "opnrtp_sarta.f" to CHAR*80 (not 70). ------------------------------------------------------------------------------- 25 September 2003: v1.04 03-08-05 New include file "incFTC_iasi_sep03.f" for Sep03 IASI fast model. Also new "iasi_rms.txt" and "iasi_readme.txt" files for Doc dir. ------------------------------------------------------------------------------- ******************************************************************************* VERSION 1.05 ******************************************************************************* ------------------------------------------------------------------------------- 06 February 2004: v1.05 04-02-06 Three new features have been added to SARTA: 1) Optical depth tuning has been made "standard" to SARTA. This is accomplished by reading in a file of tuning multimpliers and applying them to the RTA coefs. New routine "TUNMLT.f" and associated minor changes to "sarta_rtp.f" and "incFTC.f". 2) A runtime command line switch has been added to allow user control of the reflectivity (rho) in the downwelling thermal radiance calculation. The default (set in "rdinfo_sarta.f") is to use the rho found in the input RTP file. The user may force SARTA to use rho=(1-emis)/pi instead by command line argument "lrhot=true". A comment has been added to the header attribute to specify the value of lrhot. Minor changes to "rdinfo_sarta.f", "opnrtp_sarta.f", "setems.f", and "sarta_rtp.f". 3) Code has been written (BUT SO FAR NOT TESTED!) to allow SARTA to process AIRS PGE level2 support profiles (100 layer/psuedo levels). New routine "mean_t.f" and minor changes to "opnrtp_sarta.f" and "sarta_rtp.f". ------------------------------------------------------------------------------- 06 October 2004: v1.05 04-10-06 Add error trap for profile latitude to "rdrtp_sarta.f". ------------------------------------------------------------------------------- 17 December 2004: v1.05 04-12-17 Bug fixes associated with processing RTP "head.ptype=2" profiles: rdrtp_sarta.f: add PTYPE to call; add if/then/else for setting NLAY based on PTYPE. getbot.f: add NLAY to call; add error trap for LBOT>NLAY; add warning for big BLMULT. sarta_rtp.f: update calls to rdrtp and getbot (for new vars listed above). Without these fixes, the code generally returned incorrect radiances, errors typically at the 0.1 K level in window channels but occasionally much large (over 1 K). ------------------------------------------------------------------------------- 20 December 2004: v1.05 04-12-20 Same updates as 04-12-17 above except for so2/co2/cc2 variant codes. ------------------------------------------------------------------------------- 15 March 2005: v1.05 05-03-15 Experimental new and revised code to include the effects of non-Local Thermal Dynamic Equalibrium. New routine "calnte.f" to calculate the non-LTE effect. Revised "rcoef_nte.f" to read in the non-LTE coefficients. Revised "sarta_nte.f" for the new non-LTE variables passed back by rdcoef, and the new call to calnte. This non-LTE code is experimental and is not currently supported. ------------------------------------------------------------------------------- ******************************************************************************* VERSION 1.06 ******************************************************************************* ------------------------------------------------------------------------------- 05 April 2005: v1.06 05-04-05 Substantially revised the algorithm/code used to calculate the reflected downwelling atmospheric thermal emission radiance contribution. rdcoef.f: Revise the block of code that reads in the reflected thermal coefficients. The "layer above" variable is now a dummy. calrad.f: Revise the block of code that calcualtes the reflected thermal radiance contribution. Add a new block of code to calc "RDOWN". ------------------------------------------------------------------------------- 07 April 2005: v1.06 05-04-05 No code changes. Add new tuning multiplier file "tunmlt_apr05.txt". This is similar to the tuning file "tunmlt_jan04deliv.txt" except it adds adjustments to ozone in the 10 um region to approximate the effects of changes in the ozone line parameters between HITRAN2000 and HITRAN2004. This tuning file is the new "standard" tuning file for sartaV106. ------------------------------------------------------------------------------- 09 May 2005: v1.06 05-05-09 Minor changes to "calrad.f" so that RTHERM is always calculated. ------------------------------------------------------------------------------- 18 May 2005: v1.06 05-05-18 Create "hno3" version with 100 layer variable HNO3, and update "so2" version up to bring it up to date for v1.06. ------------------------------------------------------------------------------- 24 June 2005: v1.06 05-06-24 Modify "getbot.f" so the "10" loop starts on the "DELPX" line (ie so DELPX is recal'ed every layer). ------------------------------------------------------------------------------- 11 November 2005: v1.06 05-11-11 Modify "tunmlt.f" (and "sarta_nte.f" call) for non-LTE tuning. The non-LTE tuning replaces the CO2pert tuning previously allowed (but not actually used). ------------------------------------------------------------------------------- 13 December 2005: v1.06 05-12-13 Add "F" limit check to "calrad.f". ******************************************************************************* VERSION 1.07 ******************************************************************************* ------------------------------------------------------------------------------- 29 June 2005: v1.07 05-06-29 Add 100 layer variable "trace" gases CO2, SO2, HNO3, and N2O. Variable CO2 had been included in the previous few versions of SARTA, but it has now been changed to a 100 layer variable instead of a total column variable. Variable SO2 and HNO3 had previously been available only in SARTA variants. The basic SARTA code is mostly unchanged from v1.06 apart from the obvious changes required for the new trace gases. ------------------------------------------------------------------------------- 13 October 2005: v1.07 05-10-13 Bug fix: set TRCPRD 1-7 in sunpar.f (was only setting 1-4). This lead to errors of (typically) less than 0.1K in the reflected solar in the N2O channels . Add non-LTE by making the following changes: incFTC.f - new non-LTE variables rdcoef.f - new variables/code to read in non-LTE coefficients calnte.f - new routine sarta.f - new non-LTE variables and call to "calnte". The non-LTE code had previously been implemented in a variant of SARTA v1.05, so it was a simple matter to add it to v1.07. ------------------------------------------------------------------------------- 08 December 2005: v1.07 05-12-08 Modify "tunmlt.f" for non-LTE tuning, modify call argument list to match "rccoef.f" call, and update tunmlt call in "sarta.f". ------------------------------------------------------------------------------- 13 December 2005: v1.07 05-12-13 Add "F" limit check to "calrad.f". Note: exact same change as made to v1.06 ------------------------------------------------------------------------------- 08 January 2007: v1.07 07-01-08 PCLSAM Major revision of PCLSAM cloudy SARTA variant to allow a more flexible mix of transmissive cloud types, and the inclusion of black clouds. The calrad1 & calrad2 transmissive cloud routines are unchanged, but many other routines and the main sarta_pclsam program have many changes. This is only "v1.07" in the sense the PCLSAM code is built upon that version of standard (non-cloudy) SARTA; the code changes are so extensive a new version number is warranted. ------------------------------------------------------------------------------- 22 January 2007: v1.07 07-01-22 PCLSAM Minor fix of cfrac checks (change "<" in cfrac12 < cfrac1 & 2 to "<=") ------------------------------------------------------------------------------- 23 February 2007: v1.07 07-02-23 PCLSAM_X Major re-write of PCLSAM "calrad1.f" and "calrad2.f" routines for 100 layer cloud profiles. Minor changes to "calrad0.f" and "getmie.f". Major changes to "ccprep.f" and new routine "bkprep.f". Remove much of the cloud housekeeping from "sarta.f" to new routine "getcld.f". Remove RTP attribute adjustments from "opnrtp.f" to new routine "doattr.f". Minor change to "rdcoef.f". The old cloud version info in "incFTC.f" was deleted and replaced by expanded info in "fnmie.f". ------------------------------------------------------------------------------- 02 March 2007: v1.07 07-03-02 PCLSAM_X Bug fix to prevent NaN radiances when two complex clouds have an empty gap somewhere between cloud top and bottom layers and sun term is on. Fixed by adding "LC1L", "LC2L", & "LC12L" to "sarta_pclsam.f" to specify which layers require a complex cloud calcualtion, and pass to "calrad[12].f". Add generic "LCL" input argument to "calrad1.f" and "calrad2.f", and revise "DO" loop to use it. ------------------------------------------------------------------------------- 02 May 2007: v1.07 07-05-02 all versions Replace hardcoded default satellite altitude in "sarta.rtp" with XSALT which is specified in the include file. ------------------------------------------------------------------------------- 03 July 2007: v1.07 07-05-02 Created "calrad_pge.f" variant routine to better mimic PGE code. This PGE variant does not allow negative RSUN, and the limits on the F check are expanded to 0.4 to 2.5 (were 0.696 to 2.09). ------------------------------------------------------------------------------- 10 September 2007: v1.07 07-09-10 Modify the 100 layer cloud variant of sarta_pclsam to allow cloud particle size profile. Also apply a bug fix to the old version (ie with layer independent particle size) for cases where the input cloud amount profile was in reverse (bottom-up) order. ------------------------------------------------------------------------------- 15 November 2007: v1.07 07-11-15 PCLSAM slabcloud Modify main program "sarta_pclsam.f" to move much of the cloud housekeeping into routines "getcld" (different than 100layercloud code) and "bkprep" (same as 100layercloud code). No new capabilities except for improved error traps and handling transmissive clouds above black clouds (the old code gave a segmentation fault). A new ctype=2 was added to "bkprep" for emis=1 and rho=(1-cemis)/pi; this might be helpful for cases where the solar is the most important of the emis/rho parameters. ------------------------------------------------------------------------------- 23 January 2008: v1.07 08-01-23 Modify main program "sarta.f" and routines "rdrtp.f" & "calpar.f" to correct a minor bug that affected the perturbation trace gases (CO2, N2O,SO2,HNO3) when no user profile info for these gases was provided. These same files plus opnrtp.f were also modified to allow use a CO2 profile in units of ppmv dry air mixing ratio (ie HEAD.gunit=10). ------------------------------------------------------------------------------- 31 January 2008: v1.07 08-01-31 Bug fix in "sarta.f": LCO2,LN2O,LSO2,LHNO3 changed to LOGICAL (were mistakenly REAL). Update PCLSAM_slabcloud and PCLSAM_100layercloud for perturbation gas bug fix and CO2 ppmv profile. ------------------------------------------------------------------------------- 30 April 2008: v1.07 08-04-30 co2_5term Create variant "co2_5term" with shortwave CO2 changed from 4 to 5 predictors ("NCO2") in the include file "incFTC.f" and calt routines "calt4_co2_5term.f", "calt5_co2_5term.f", "calt6_co2_5term.f", & "calt7_co2_5term.f". ------------------------------------------------------------------------------- 05 May 2008: v1.07 08-05-05 Add a line to "sarta.f" to set SUNCOS=0 when DOSUN=false; this has no effect except possibly if SUNCOS were NaN. ******************************************************************************* VERSION 1.08 ******************************************************************************* ------------------------------------------------------------------------------- 14 May 2008: v1.08 08-05-14 Implement an approximate adjustment to nonLTE for variable CO2. This involved adding a 7th predictor to the nonLTE coefficients and a new include file variables "CO2NTE" and "NTEBOT". Add new variable "CO2TOP" in sarta.f which is calculated in calpar.f and passed to calnte.f. Moved the last remaining "CO2MLT" adjustment from sarta to calpar.f and this required adding "CO2PPM" to the calpar call. Make 5 term CO2 the default (was a variant in v1.07). ------------------------------------------------------------------------------- 14 May 2008: v1.08 08-05-14 dual-frequency Implement a dual-frequency variant of sarta v1.08 that allows calculation of radiances for variable channel frequencies. Uses routines "rdsun_df.f", "tunmlt_df.f", "rdcoef_df.f", "rdrtp_df.f", and "sarta_df.f" instead of the standard ones. Reads in two sarta databases (A and B) and interpolates them with new routine "sumcof.f". The freqs are specified with a grating model yoffset "YOFF" which is set in rdrtp_df.f with "prof.udef1". Added new include file variles "YOFFA" for database A, "YOFFB" for B, "YOFDEF" for the default yoffset, "YOFMIN" for minimum allowed yoffset, and "YOFMAX" for maximum. Replace almost all database filename variables "FN*" to dual variables "FA*" and "FB*". ------------------------------------------------------------------------------- 16 May 2008: v1.08 08-05-16 Bug fix for nonLTE: add "NJUNK" to "tunmlt.f" (and "tunmlt_df.f") to prevent adjustment of the 7th (ie CO2 adjustment) nonLTE coefficient. ------------------------------------------------------------------------------- 02 Sep 2008: v1.08 08-09-02 PCLSAM Bug fix to calt4/5/6/7 for 5 term CO2. ------------------------------------------------------------------------------- 24 Oct 2008: v1.08 rtpV201 Modify SARTA for new rtpV201. Src split into SrcV105 and SrcV201. The routines modified for SrcV201 are: opnrtp.f, rdrtp.f, rdrtp_df.f, sarta.f. sarta_df.f, setems.f, wrtrtp.f Most of the changes are related to removing "NRHO" and "FRHO", as rtpV201 uses a common NEMIS and FEMIS for all emissivity and reflectance data. Note: The new rtpV201 has been developed partly to address issues related to the upcoming AIRS version-6 Level-2 processing, as that will require spectral emis and rho data for (black) clouds which rtpV105 can not handle. Other changes for rtpV201 are intended to eliminate unwanted old variables and to add a few desired new ones. Support for rtpV105 will likely be dropped when sartaV108 is publically released (estimated spring 2010). ------------------------------------------------------------------------------- 02 Nov 2008: v1.08 08-11-02 rtpV105 & rtpV201 Bug fix to tunmlt.f & tunmlt_df.f related to nonLTE (the modification added 16 May 2008 for "NJUNK" was done wrong). ------------------------------------------------------------------------------- 25 Nov 2008: v1.08 08-11-25 PCLSAM Bug fix to sarta_pclsam.f for cloud1 reflectivity (was using crho2) ------------------------------------------------------------------------------- 26 Nov 2008: v1.08 08-11-26 PCLSAM rtpV201 Update slab cloudy code for rtpV201: partial rewrite of getclg_slab.f, significant changes to setems.f and sarta_pclsam.f, and small changes to rdrtp.f and bkprep.m. The new rtpV201 has dedicated fields for cloud2, and also spectral emis and rho for both clouds. ------------------------------------------------------------------------------- 01 Dec 2008: v1.08 08-12-01 PCLSAM rtpV201 Update/debug slab cloudy code for rtpV201: spectral cloud/surf emissivity & rho for black (non-transmissive) now implemented, along with optional "cstemp" (if cstemp <= 0 then uses T interpolated from ptemp at cprtop). ------------------------------------------------------------------------------- 29 Apr 2009: v1.08 09-04-29 PCLSAM rtpV105 & rtpV201 Bug fix for "calrad2.f" to initialize RDOWN and TDOWNN; the bug occasionally caused NaN radiances. ------------------------------------------------------------------------------- 07 May 2009: v1.08 09-05-07 PCLSAM rtpV201 Bug fix for "getcld_slab.f" to initialize CFRA1X & CFRAC2X; the bug caused bad "clear" radiances (sometime but not always NaN). ------------------------------------------------------------------------------- 12 May 2009: v1.08 09-05-12 rtpV201 ---non-cloudy --- Remove VCLOUD from "incFTC.f" and add new VTUNNG tuning version string. Remove VCLOUD from "sarta" HATT in "opnrtp.f" and replace with VTUNNG. ---PCLSAM--- Bug fix for Mie table reader : old code (in rdcoef_pclsam.f) screwed up if HEAD.ichan was not sorted in assending order. Delete routine "rdcoef_pclsam.f" and replace with non-cloudy "rdcoef.f". The Mie cloud tables reader (was in rdcoef_pclsam) has been moved to new routine "rdcldt.f" which is called by "sarta_pclsam.f". Call to "fnmie.f" (was in rdcoef_pclsam) also moved to "sarta_pclsam.f". Remove VCLOUD from "incFTC.f" and move it to "fnmie.f" with all new VCLOUD related code. Add new VTUNNG tuning version string to "incFTC.f". Add new "clouds" HATT for VCLOUD in "opnrtp_pclsam.f". Also remove VCLOUD from "sarta" HATT and replace with VTUNNG. ------------------------------------------------------------------------------- 03 June 2009: v1.08 09-06-03 swch4 rtpV201 Create shortwave CH4 variant "swch4". This required changes to routines: calt5_swch4.f, calt6_swch4.f, calt7_swch4.f, rdcoef_swch4.f, calpar_swch4.f, sarta_swch4.f as well as the "incFTC*swch4.f" include file. The changes are relatively simple in the sense they mostly repeat existing code for perturbation gases. ------------------------------------------------------------------------------- 06 August 2009: v1.08 09-08-06 "df" rtpV201 Update dual frequency (df) SARTA variant to apply a channel dependent spectral calibration adjustment to the nominal yoffset. This involved changes to the incFTC*f include file and routines: rdcoef_df.f, tunmlt_df.f, sumcof_df.f, and sarta_df.f, and new routine speccal_df.f. A new "speccal" text file is read in at startup, and contains deltas to the nominal yoffset for each channel. ------------------------------------------------------------------------------- 13 April 2010: v1.08 10-04-13 PCLSAM swch4 rtpV201 Create "swch4" variant of the PCLSAM code. This required changes to routines: calt5_od_swch4.f, calt6_od_swch4.f, calt7_od_swch4.f, sarta_swch4_pclsam.f as well as the "incFTC*swch4.f" include file. ------------------------------------------------------------------------------- 14 September 2010: v1.08 10-09-14 all variants Bug fix: Add 5th CO2 coef to calt1.f/calt1_od.f & calt2.f/calt2_od.f. This bug fix only affects IASI; no effect for AIRS April 2008 or CrIS April 2009 RTAs. ------------------------------------------------------------------------------- 22 June 2011: v1.08 2011-06-22 "nh3" and "swch4_nh3" variants Create "nh3" variants of the standard and "swch4" SARTA codes for variable ammonia (NH3). ------------------------------------------------------------------------------- 23 June 2011: v1.08 2011-06-23 "nh3_df" variant Create "nh3" variant of the "df" (dual-frequency) SARTA code for variable ammonia (NH3). ------------------------------------------------------------------------------- 21 October 2011: v1.08 2011-10-21 "AIRS_PGEv6" variant AIRS PGEv6 (PGE v6 equivalent, more or less) version of SARTA. This grew out of the old "df" variant code, and shares much in common with that code. The code uses the RTP "prof.pnote" field to ingest the Yoffset and Doppler shift data. It uses the non-standard DEC extension "UNION/MAP" to convert the last 76 CHARACTERs of pnote into 19 REAL values (the first 4 CHARACTERS of pnote must be "YOFF"). There are 17(+1 unassigned) module Yoffset (um), 1 Doppler shift (ppm), and 11 orbit phase (degrees). The "+1"=18th module is for the fake channels (so they can also be shifted if so desired); currently unassigned. ------------------------------------------------------------------------------- 21 October 2011: v1.08 2011-11-21 "AIRS_PGEv6" variant AIRS PGEv6 version of SARTA. Add line to assign fake channels yoffset "YMODP1(18) = YOFDEF" to routine "rdrtp_df.f" (was zero). ------------------------------------------------------------------------------- 10 February 2012: v1.08 2011-11-21 "AIRS_PGEv6" variant Updated the "inc*.f" include files to use "ABweight_table_20120210.txt" which includes the A/B weight table uploaded to AIRS on 2012-01-21 (epoch6), as well as a correction for the A/B weight of channel 1031 in epoch4. -------------------------------------------------------------------------------