Auxhist streams time variables

[epotter1]

Hi,

I'm trying to output different WRF output variables at different time resolutions using the auxhist streams option, but I am having trouble with the time variables. I have the 'Times' variables, but I would like to include XTIME in all different history streams. When I added this directly into the history streams (e.g. streams 7 and 10 in the attached text file), the unit (and description) of XTIME in the output is "minutes since 0000-00-00 00:00:00", whereas in the main wrfout files the unit is "minutes since 2019-01-01 00:00:00" (beginning of the run). Does anyone know how to fix this?

Three follow-up questions:
I can't find XTIME in Registry.EM_COMMON, is it in a different one?
If I can find it and change the registry to output into the history streams, do I need to rerun real? I'm running multiple runs over large domains so keen to avoid rerunning if possible!
Lastly, I'm using (mostly) the default options for pressure level output in auxhist23, but I am also getting C1H, C1F, C2H etc output in this stream. What are these and why do they appear here?

I've attached my namelist and an example text file with the history streams (identical for all domains).

Any help would be hugely appreciated!

 

[kwerner]

Hi,
Thanks for your patience. I did a test run, only adding the following to my I/O file (+:h:7:XTIME,XLONG,XLAT,TIMES) and when I do an 'ncdump -h' on my aux7 output files, the description is coming out correctly (the time since the beginning of the run). Can you let me know which version of WRF you are using? I was testing with the latest version (V4.6.0).

I can't find XTIME in Registry.EM_COMMON, is it in a different one?
If I can find it and change the registry to output into the history streams, do I need to rerun real? I'm running multiple runs over large domains so keen to avoid rerunning if possible!
Lastly, I'm using (mostly) the default options for pressure level output in auxhist23, but I am also getting C1H, C1F, C2H etc output in this stream. What are these and why do they appear here?

1. I'm able to find 'xtime' in Registry.EM_COMMON - but it's in lowercase. Did you check for the lowercase version?
2. Take a look at davegill's post on this previous thread. It discusses a few options for possibly doing this without having to rerun everything.
3. If you go to the Registry directory and search for these variables (make sure to do a case-insensitive search), you should be able to find them in the registry.hyb_coord file. For e.g.,

state    real      c1h       k     misc      1   -     i02rh0{22}{23}{24}      "C1H"  "half levels, c1h = d bf / d eta, using znw"    "Dimensionless"

 

[epotter1]

Hi,

I'm very sorry for my extremely late reply, I got distracted with other issues but am still keen to solve this! I am using WRF version 4.5.1, and have tried an example with just the line you've suggested in the I/O file, but I still have the same problem.

I've noticed that in the output files, the stream 0 file (wrfout) has a global attribute "SIMULATION_START_DATE" as well as "START_DATE", but in the other output files (stream 7 for example), "SIMULATION_START_DATE" is missing. Could the be the problem? Do you know where the global attributes are added? If you still have this test, would you be able to check the global attributes on the auxhist7 file for me, to see if this is there? If you still have it, I'd also be grateful to see your namelist and I/O text file if you have them.

Many thanks for your help with this.

 

[kwerner]

Hi,
Yes, I do still have the simulation! I am attaching my namelist.input file and the I/O text file for you.

Regarding the SIMULATION_START_TIME and START_TIME, I am also seeing what you are - that SIMULATION_START_TIME is only in the stream 0 history file, but that didn't seem to make a difference in the time for XTIMES. This is the print-out from that file:

> ncdump -h xtimes_d01_2016-03-23_03\:00\:00
netcdf xtimes_d01_2016-03-23_03\:00\:00 {
dimensions:
    Time = UNLIMITED ; // (1 currently)
    DateStrLen = 19 ;
    west_east = 74 ;
    south_north = 69 ;
variables:
    char Times(Time, DateStrLen) ;
    float XLAT(Time, south_north, west_east) ;
        XLAT:FieldType = 104 ;
        XLAT:MemoryOrder = "XY " ;
        XLAT:description = "LATITUDE, SOUTH IS NEGATIVE" ;
        XLAT:units = "degree_north" ;
        XLAT:stagger = "" ;
        XLAT:coordinates = "XLONG XLAT" ;
    float XLONG(Time, south_north, west_east) ;
        XLONG:FieldType = 104 ;
        XLONG:MemoryOrder = "XY " ;
        XLONG:description = "LONGITUDE, WEST IS NEGATIVE" ;
        XLONG:units = "degree_east" ;
        XLONG:stagger = "" ;
        XLONG:coordinates = "XLONG XLAT" ;
    float XTIME(Time) ;
        XTIME:FieldType = 104 ;
        XTIME:MemoryOrder = "0  " ;
        XTIME:description = "minutes since 2016-03-23 00:00:00" ;
        XTIME:units = "minutes since 2016-03-23 00:00:00" ;
        XTIME:stagger = "" ;

 

[epotter1]

Thanks very much for this. I've tried running with exactly your namelist and file.txt (edited only for time and space to match my met files) and still have the same problem, so it must be in my compilation or possibly the Registry. At least I know the problem is not with the .txt files. I can't see any differences in the obvious lines of any of the Registry files and the version 4.6.0 ones, but at some point I'll recompile to see if that helps.

I also notice the XTIME variable is missing from the pressure level diagnostic output (I/O stream 23) by default, which seems strange when XLAT and XLONG are included by default in the registry (XTIME was also not included in the pressure level output in an older version of WRF I ran a few years ago). If I add it in the .txt file, I have the same problem as with the other streams. The Times coordinate is in all files by default and correct, but I can't find it in the registry because there's so many instances of the word 'times' in the various registry files!

Thanks again for all your help, I'll have a go at recompiling or try version 4.6.0.

dimensions:
        Time = UNLIMITED ; // (1 currently)
        DateStrLen = 19 ;
        west_east = 209 ;
        south_north = 259 ;
variables:
        char Times(Time, DateStrLen) ;
        float XLAT(Time, south_north, west_east) ;
                XLAT:FieldType = 104 ;
                XLAT:MemoryOrder = "XY " ;
                XLAT:description = "LATITUDE, SOUTH IS NEGATIVE" ;
                XLAT:units = "degree_north" ;
                XLAT:stagger = "" ;
                XLAT:coordinates = "XLONG XLAT" ;
        float XLONG(Time, south_north, west_east) ;
                XLONG:FieldType = 104 ;
                XLONG:MemoryOrder = "XY " ;
                XLONG:description = "LONGITUDE, WEST IS NEGATIVE" ;
                XLONG:units = "degree_east" ;
                XLONG:stagger = "" ;
                XLONG:coordinates = "XLONG XLAT" ;
        float XTIME(Time) ;
                XTIME:FieldType = 104 ;
                XTIME:MemoryOrder = "0  " ;
                XTIME:description = "minutes since 0000-00-00 00:00:00" ;
                XTIME:units = "minutes since 0000-00-00 00:00:00" ;
                XTIME:stagger = "" ;

 

[hesue]

Hi,
Thanks for your patience. I did a test run, only adding the following to my I/O file (+:h:7:XTIME,XLONG,XLAT,TIMES) and when I do an 'ncdump -h' on my aux7 output files, the description is coming out correctly (the time since the beginning of the run). Can you let me know which version of WRF you are using? I was testing with the latest version (V4.6.0).



1. I'm able to find 'xtime' in Registry.EM_COMMON - but it's in lowercase. Did you check for the lowercase version?
2. Take a look at davegill's post on this previous thread. It discusses a few options for possibly doing this without having to rerun everything.
3. If you go to the Registry directory and search for these variables (make sure to do a case-insensitive search), you should be able to find them in the registry.hyb_coord file. For e.g.,

state    real      c1h       k     misc      1   -     i02rh0{22}{23}{24}      "C1H"  "half levels, c1h = d bf / d eta, using znw"    "Dimensionless"

Hi, I saw a small part of this thread mentioning about installing and compiling WRF version 4.6 and would like to ask about a small problem in my compilation process. It's that when I compile to the last step I can't generate *.exe successfully. What's the reason for this? I would appreciate it if you could tell me something about this.

 

[kwerner]

Hi, I saw a small part of this thread mentioning about installing and compiling WRF version 4.6 and would like to ask about a small problem in my compilation process. It's that when I compile to the last step I can't generate *.exe successfully. What's the reason for this? I would appreciate it if you could tell me something about this.

This thread is not discussing a compiling issue. If you're still experiencing this problem, please start a new thread with an explanation and attach the necessary files. Thanks.