TagO3 cannot obtain the correct vertical distribution map of ozone.

[Yuh2ng]

Name and Institution (Required)

Name:Chen YH Institution:Yunnan University

Confirm you have reviewed the following documentation

• [x] Support guidelines
• [x] User manuals
• [x] Debugging GEOS-Chem and HEMCO errors
• [x] Current and past Github issues

Description of your issue or question

Please provide as much detail as possible. Always include the GEOS-Chem version number and any relevant configuration and log files. I am studying stratospheric invasion: in some special cases, ozone in the stratosphere can enter the troposphere through the tropopause and even reach the ground under appropriate conditions. I want to track the ozone in the stratosphere through the TgO3 model, but after several attempts, I still can't get the correct vertical distribution map of ozone.

Based on the ERA5 data and the analysis results of WACCM, I am certain that there is indeed a stratospheric invasion in the simulation area, but my simulation results cannot accurately reflect this phenomenon. Furthermore, I have also discovered that the vertical distribution of ozone simulated by the TagO3 model differs from that simulated by the fullchem model. How should I address this issue? This is the result obtained through TagO3. This is the result obtained through fullchem. Here are the results of the stratospheric ozone using WACCM model data. Here are my relevant documents： geoschem_config.txt HEMCO_Config.txt

[yantosca]

Thanks for writing @Yuh2ng. This is more of a science question and may be better addressed to the Stratosphere Working Group. Also tagging @ncolombi, who has recently used the tagO3 simulation.

How long did you spin up the stratosphere for? You would probably need to spin up the tagO3 simulation for several years (maybe 10-12) until the stratosphere is in steady state.

Also, which horizontal resolution are you running at? Wondering if the stratospheric intrusion would only be visible at a fine horizontal resolution. You might not see that at e.g. 4x5.

Another thing you can do is to look at the SpeciesConcVV_CLOCK species, which is the age of air. That may give you some idea where stratospheric air is entering the troposphere.

[Yuh2ng]

I have checked my settings, and my horizontal resolution is 0.5x0.625. With this resolution, I should be able to observe the process of stratospheric intrusion. Since this is my first time using the TagO3 model, I would like to ask if such a long preheating time is necessary（several years (maybe 10-12)） even for simulating a process of only half a month?

[ncolombi]

Hi @Yuh2ng . Yes, I have found that the tagged O3 simulation needs to be spun up for about 10 years in order for the stratosphere to reach steady state. Even if the process is only half a month, the stratosphere wont have reached steady state without proper spinup so your results will likely not be able to capture the event. I find that that the fullchem results are similar to the total tagged O3 results when I do this. It might be best to set up a time to meet and discuss this further and I can report back our findings here after we chat a bit. You can reach out to me at ncolombi@g.harvard.edu.

[yantosca]

@Yuh2ng: I wonder if you could spin up for 10-12 years with the 4x5 simualation and then use the initial conditions from that to start your production run at 0.25x0.3125. That would certainly take less time. HEMCO will regrid the restart file automatically.

[Yuh2ng]

@ncolombi @yantosca Thank you both for your replies. Before I strat the spin up simulation, I have one more question regarding the necessity of using "O3_PROD_LOSS" or other emission sources during simulation.

[ncolombi]

Hi @Yuh2ng. You will need prodloss files for the spin up as well as your 0.5x0.625 simulation. However, you can save out one year of Prodloss files and cycle through those for the whole spin up (so you don't have to generate 10 years of Prodloss files). For example, I run my simulation from 2007-2018, but only use Prodloss files for years 2017-2018 to run my tagged simulation as shown in the image below: ![Uploading Screen Shot 2024-06-13 at 4.05.12 PM.png…]()