From david@dews1.space.swri.edu Thu Apr 9 12:46:53 1992 Return-Path: Received: from ncar.UCAR.EDU (holycross.scd.ucar.edu) by sting.hao.ucar.edu.hao (4.1/SMI-4.1) id AA11431; Thu, 9 Apr 92 12:46:51 MDT Received: from dews1.space.swri.edu by ncar.ucar.EDU (5.65/ NCAR Central Post Office 04/10/90) id AA06129; Thu, 9 Apr 92 12:47:03 MDT Received: by dews1.space.swri.edu (4.1/0.0) id AA28440; Thu, 9 Apr 92 13:49:34 CDT Date: Thu, 9 Apr 92 13:49:34 CDT From: david@dews1.space.swri.edu (David Winningham) Message-Id: <9204091849.AA28440@dews1.space.swri.edu> To: chris@pemrac.UCAR.EDU, sharb@swri.UCAR.EDU, chenette@agena.space.lockheed.com, rudy@pemrac.UCAR.EDU, foster@ncar.ucar.edu, ngale@pemrac.UCAR.EDU, potemra@aplsp.dnet.nasa.gov, bythrow@ampte.dnet.nasa.gov, anderson@ampte.dnet.nasa.gov, clauer@sprlc.sprl.umich.edu, lumm@ncar.ucar.edu, roble@ncar.ucar.edu, rees@barney.dnet.nasa.gov, jim@swri.UCAR.EDU, gaines@agena.space.lockheed.com Subject: HALOE Production Processing Status: R ----- Begin Included Message ----- >From "HALOE::RUSSELL"@PEM.space.swri.edu Thu Apr 9 12:43:56 1992 Date: Thu, 9 Apr 1992 17:40:44 GMT From: "HALOE::RUSSELL"@PEM.space.swri.edu Subject: HALOE Production Processing To: DAVID@DEWS1.space.swri.edu, SHARB@SWRI.space.swri.edu X-Vmsmail-To: @UPI.DIS Content-Length: 5394 HALOE Production Processing and Data Status As you probably have noted, we are now running the HALOE software through Level 2 in production mode on the CDHF. Also, we have classified this as our Version 4 software since it is the version we used at the Oxford intercomparison meeting. Therefore the results have "some measure of validity" which is the criterion for calling this Version 4. I want to stress however, that even though the data look reasonable, the HALOE team still has much work to do to properly validate the data. One major area of work is in correcting for aerosol effects on gas retrievals. We are making good headway on this but we have a long way to go. The Version 4 data being produced has no correction for aerosol included. The gas filter channels (HCl,HF,CH4,and NO) are least affected but they are not unaffected. We have to carefully study the data in the aerosol region (e.g. 100mb to 10mb) and assess what effect the aerosol has on our data. This will take time, detailed work, and comparisons with correlative measurements. THE HELP OF THE UARS SCIENCE TEAM IN LOOKING AT THE VALIDITY OF THE DATA IN THIS AND OTHER WAYS IS WELCOMED. You will see that some of the time, the direct effect of aerosols on the signal/retrieval and the indirect effect through perturbations to our pointer/tracker, cause obviously invalid results. THEREFORE, VIEW THE HALOE GAS FILTER CHANNEL RESULTS BELOW ABOUT 10mb to 20mb AS VERY PRELIMINARY AND IN SOME OBVIOUS CASES AS INVALID. The effect of aerosols on radiomenter channel results (H2O,NO2, and O3) is clear. Since no correction is being made at present, the gas retrievals show unrealistically large mixing ratios below about 10mb (and sometimes from about 7mb to 10mb) due to the fact that the HALOE retrieval ignores the additional aerosol absorption and considers the total absorption as due only to the gas. We have chosen to still do a retrieval for the lower stratosphere and include it in the data base since (1) it shows the aerosol effect and (2) the altitude where the effect starts varies with channel and geographic location. NOTE HOWEVER, THAT THE HALOE RADIOMETER CHANNEL MIXING RATIO RESULTS AT AND BELOW 10MB (SOMETIMES 7mb TO 10mb) ARE CONTAMINATED BY AEROSOL ABSORPTION AND THEREFORE DO NOT GIVE CORRECT GAS MIXING RATIOS. A series of other improvements also need to be made to the processing software including forward model updates to incorporate the most recent spectral data base (current software uses 1989 data bases in some cases) and modifications to the EGA data base being used. We will be evaluating the accuracy of the HALOE forward model for the radiomenter channels by performing comparisons with line-by-line calulations. Other changes that need to be included are temporal smoothing at the highest altitudes to reduce "top of profile" errors, optimization of the optimal onion peeling constaints and retrieval grid spacing to better capture atmospheric variability, and inclusion of a HALOE temperature retrieval. The NMC temperature data stop at about 0.4mb which requires that climatology be used at present for the higher altitudes. This could have a large effect especially on the nitric oxide results. As a final note on the data status, very few comparisons have been made with correlative data since since the data base has been very sparse. This situation is changing as campaigns are carried out and more data accumulate in the CDHF. This will be an important focus of future HALOE validation efforts. AGAIN THE HELP OF THE UARS SCIENCE TEAM IN THIS EFFORT IS WELCOMED. A description of the HALOE level 1 and 2 files is now on the CDHF in the directory HAL$DISK:[HALCMN.DOC]. The files in this directory are: LEVEL12.TXT - A preliminary description of the file structure. Users should read this before trying to run the following program. READL2B.COM - Command procedure to compile and link the routines in this directory. READL2B.FOR - The main program that calls the following routines. RL2DAT.FOR, RL2SUM.FOR, RWL1HD.FOR Several INCLUDE files and a test version of a level 2 file are contained in these files. The outline in the .TXT file and the .COM file should be enough to enable users to use, and modify, READL2B.FOR. This file is now set to operate in an interactive mode but it can be modified to operate in a batch mode and to write a file in the users own format. The quality data in the files are variances. If the READL2B.FOR is used to read the data however, standard deviation (SD) is returned as agreed at the last UARS science team meeting. Currently, the negative sign attached to the SD does not indicate the point where 50% weighting of the apriori and measured signals occurs as agreed at the meeting; this will be corrected in a future software delivery. The negative SD which is encountered at the lowest altitude point indicates the starting altitude for the retrieval. If there are questions about the mechanics of reading the data please contact Ken Haggard, the HALOE Data Reduction Manager at HALOE::Haggard. If there are questions about data quality, comments on data validation, or other information you feel would be helpful in the validation effort, please contact me. Jim Russell  ----- End Included Message -----