How to reduce ISAM output variables

CMAQ-ISAM
1

Hi,

I ran CMAQ/ISAM using CMAQ-ISAM tag definition control file with

TAG CLASSES |OZONE

I noticed that the size of following output files are pretty large due to too many variables related to OZONE formation:

CCTM_SA_ACONC_v533_ISAM* ~ 827M
CCTM_SA_CGRID_v533_ISAM*
CCTM_SA_CONC_v533_ISAM*
CCTM_SA_DRYDEP_v533_ISAM*
CCTM_SA_WETDEP_v533_ISAM*
…

I am wondering if I am able to reduce those files’ size by setting up output variables like CCTM_ACONC files using

setenv AVG_CONC_SPCS “O3 NO NO2 CO ISOP”
setenv ACONC_BLEV_ELEV " 1 1"

For example, I would like to have

O3_REGIONs
O3_EmissionCategories

only in CCTM_SA_* files.

I am wondering if it is possible and how?

Thanks for your idea and help in advance.

Feng

2

Hello Feng,

no, it is currently not possible to output just a subset of the tracked species defined for a particular tag class.

Christian

3

Note, by the way, that these are selective direct access files: while it does cost you output-time to write this data, it does not cost you input-time for data you do not read.

In fact, one of the tests you do when you do the I/O API make tests demonstrates that when you have a year-long hourly-timestep data file and read just the midnight Cristmas eve (timestep 8592) data, the response is essentially instantaneous.

4

Hello Christian,

Thank you very much for your reply and confirmation.

Feng

5

Hello, Carlie,

Thank you very much for your response.

That is true, and interesting. But CCTM_SA_ACONC_v533_ISAM files provide hourly mean data rather than instantaneous one, right?

Thanks.

Feng

6

The point is that it takes exactly the same amount of I/O (and compute-time) to analyze just one variable from a thousand-variable file that it does to analyze one variable from a one-variable file.
(This is in sharp contrast with variables stored in “so-called normal” sequential files.)

7

Carlie,

Thanks, that point is also what ISAM exactly does. Yes, all variables are associated with all species involved ozone chemistry. So the same amount of I/O is understandable.

Thank you again for the key message.