UM-UKCA run error

Unified Model
1

Hi CMS,

My user ID is emxin, and the suite I am working on is u-dv236. I encountered a problem when I tried to add a new sea-salt aerosol Aitken mode to the UM-UKCA, following previous work. Since I could not directly use their branches in my suite, thus I merged the relevant code into my UM and UKCA branches. My model failed in running “atoms-main” with the error message shown below:

? Error code: 702
? Error from routine: ATM_STEP_4A_MOD:ATM_STEP_4A
? Error message: ukca_radaer_get: Diag(s) needed for UKCA are missing from D1.

Could you please help with what caused the failure? Thanks!

Best,

Xin

2

Xin

Why could you not use “their branches” ?

Grenville

3

Hi Xin,

Quick suggestion: When I have had this error in the past it has been because the options codes in STASHmaster_A were such that certain tracers where not “selected” when running.

Try removing the two 1 values in the option code for 34261 / 51261 (3rd line) and setting them to zero, at least for a test.

You will need commit your stashmaster changes and give it a few minutes before you rerun.

Cheers,

James
4

Grenville,

The main reason is that I do not need to add an extra SSA emission tracer that was included in their branches. FYI, I got the same error message when directly using part of their branches. Therefore, I manually added the relevant codes to my ranches. Thus, I think the original codes may contain something incomplete, but I am not sure. This is why I asked for help.

Best,

Xin

5

Hi James,

Thanks for the helpful message. You pointed out something (option code in STASHmaster_A) I never thought of before. However, after some tests, I found the error is not related to tracer 34261 and 51261 (they worked well). In this new run, I added one new prognostics tracer 34127 (and 51127) and one standard tracer 34860 (and 51860), they both are SS aerosols. Were their option codes not correctly given? I listed the codes used below; maybe someone can take a look and see if they need improvement. BTW, I have initialised their values to 1.0000000E-15.

1| 1 | 34 | 127 |AITKEN MODE (SOLUBLE) SEA SALT MMR |

2| 2 | 0 | 1 | 1 | 2 | 40 | 11 | 0 | 0 | 0 | 0 |

3| 100000000000000000001010000011 | 00000000000000000001 | 1 |

4| 1 | 0 | -99 -99 -99 -99 -99 -99 -99 -99 -99 -99 |

5| 0 | 2227 | 0 | 65 | 0 | 0 | 0 | 0 | 0 |

#

1| 1 | 51 | 127 |AITK MODE (SOL) SEASALT ON PRS LEVS |

2| 0 | 0 | 1 | 1 | 3 | 1 | 2 | 0 | 0 | 0 | 1 |

3| 100000000000000000000000000000 | 00000000000000000001 | 3 |

4| 1 | 0 | -99 -99 -99 -99 -99 -99 -99 -99 -99 -99 |

5| 0 | 2227 | 0 | 8 | 0 | 0 | 0 | 0 | 0 |

1| 1 | 34 | 860 |SS PARTIAL VOLUME CONC (AIT-SOL) |

2| 2 | 0 | 1 | 1 | 2 | 40 | 11 | 0 | 0 | 0 | 0 |

3| 100000000000000000005000000000 | 00000000000000000001 | 3 |

4| 1 | 0 | -99 -99 -99 -99 -99 -99 -99 -99 -99 -99 |

5| 0 | 0 | 0 | 65 | 0 | 0 | 0 | 0 | 0 |

#

1| 1 | 51 | 860 |AITK MODE (SOL) SS ON PRESS LEVELS |

2| 0 | 0 | 1 | 1 | 3 | 1 | 2 | 0 | 0 | 0 | 1 |

3| 100000000000000000000000000000 | 00000000000000000001 | 3 |

4| 1 | 0 | -99 -99 -99 -99 -99 -99 -99 -99 -99 -99 |

5| 0 | 0 | 0 | 8 | 0 | 0 | 0 | 0 | 0 |

Other changes are:

in UM (committed revision 132143):

· control/stash/um_stashcode_mod.F90 (2 diffs)

· control/ukca_interface/ukca_radaer_init-ukca1.F90 (4 diffs)

· control/ukca_interface/ukca_setd1defs.F90 (3 diffs)

· utility/qxreconf/rcf_set_data_source_mod.F90 (2 diffs)

in UKCA (committed revision 8135):

· control/core/interface/ukca_fieldname_mod.F90 (1 diff)

· control/core/interface/ukca_ntp_mod.F90 (4 diffs)

· science/core/aerosols/glomap/ukca_aero_ctl.F90 (1 diff)

· science/core/aerosols/glomap/ukca_mode_setup.F90 (1 diff)

· science/core/aerosols/glomap/ukca_prim_ss.F90 (1 diff)

· science/radaer/ukca_check_radaer_coupling_mod.F90 (1 diff)

Best,

Xin

6

Hi Xin,

From memory, I had this issue when I was trying to set up Mode Setup 4, following UKCA Chemistry and Aerosol vn10.4 Tutorial 11 - UKCA where multiple prognostics needed to have their option code updated to work with mode setup 4. I wonder if it is a similar situation here. Can you ensure all new tracers have the same option codes as the tracers which already work for the given mode setup you are using?

In your aerosol changes, have you also ensured that the aerosol setup you intend to use expects Aitken mode sea salt?

Let me know,

James

7

Hi James,

Thanks for the further message. It seems I know what caused the problem: my current suite setup has an I_MODE_SETUP=2 option, which does not work with the new SS tracers in the Aitkin mode. Thus, I changed to I_MODE_SETUP=8 in the new setup. I am testing and will let you know if I have a confirmed result.

Best,

Xin

8

Hi there,

I have made some progress, but am still unsure about a few things that need your advice. For example, is option i_mode_setup=2 supporting the Aithin mode for sea salt? Currently, it only supports accumulation and nucleation modes; I feel it would be easier to work on top of this option, but I may be wrong to do so.

Best,

Xin

i_mode_setup=2

9

Hi there,

Happy New Year to you all!

Just to let you know that I have made it work with the standard option of i_mode_setup=2.

xin

1 Like
10

Hi there,

Since this is a follow-up question, I did not raise a new ticket for it.

The added tracer (34127) seems to work well. However, the outputs for this tracer, both the monthly mean (UPM) and the 3-hourly, are inconsistent, which is strange. I can see that the results in the monthly mean are fine (with a level of 1E-12), but the 3-hourly file (UP8) does not appear to capture the new tracer values, giving ~1E-16, close to the initial value given in “Reconfiguration and Ancillary Control” by initialising its value to a constant value of 1E-15. Could you please instruct me on how to fix this problem?

Cheers,

Xin

11

Please ignore the above item as I raised a new ticket for it.

Xin