Create a simulation case

Last modified by Xwiki VePa on 2025/01/08 07:07

This tutorial will guide you through the most important steps of creating a basic simulation case. You can also watch the same tutorial from this video.

After installing ARCA the GUI will first load the default options. When starting for the first time, the defaults do not exist and the hard coded settings are used. You can define the default settings the GUI should load every time it starts, by simply saving the current settings as defaults from the toolbar: File → Save as defaults.

General options

For this task we'll start by loading the minimal settings from the toolbar: File → Load minimal settings. After that the General options tab looks something like this:

Defining the Simulation identifiers

Let's first change the "Simulation identifiers" so that we know where the model will write the output and what naming conventions are used. First, if you are simulating a case which you identify by date, the you should choose "Date" as the numbering option. If on the other hand you are simulating some chamber experiment or other such, the "Index" option might suit better (see also Date and Index). Here we use Date.

Next we choose a Group name (see Group name), which is sort of the overall name for the simulation. You might eventually create many cases and they would all use the Group name in their directory names.

Then we pick a Run name. This will identify different simulations, and the output of each Run will be under <Group name>_<Date or Index>/<Run>. If this sounds confusing, you can always check the current output path from the bottom of the GUI (here we also see that we have not yet saved any INITFILE):

It also doesn't hurt to write a description of the run, it is then easy to see why this run had the options in the first place.

Paths and files

The Common root is the directory which will contain the Group and Run directories. It is by default INOUT in the ARCA main directory, but it could be anywhere on your computer. If the Common out does not exist, you must create it, the GUI won't do this. This is a safety precaution.

Next, I will define some input files. Input from files is not strictly necessary, but it is probably the most usual way since the model has to have some input and these typically come from measurements. The files are divided in three categories, ENV and MCM input and BG particles. We will soon see what the distinction between the first two is, but the last, BG particles, is used to send in aerosol measurements which can be used to initialize or restrict the aerosol module.

For this exercise, you can define the three files that are found in the ModelLib/Examples directory, which are measurements from Hyytiälä's SMEAR II station :

For demonstration, the BG particles shows how the wildcards work. Since the Date is still 2000-01-01 (as in the first figure), it means that the filename will be wrong, as shown by the tooltip, which shows the parsed path. Changing the Date to 2018-04-11 will point to the correct file.

Time options

Next we define the simulated time and output intervals. These are straightforward.

Runtime

Give the runtime as hours or seconds. Let's start with 20 hours.

Print every / Save every

"Prints" are the terminal outputs, that helps you to check that everything is going the way you want with the simulation. The prints can be viewed in the tab Run ARCA→Monitor. here we set the interval (in seconds) for the output.
"Saves" are the file saving events, so this is analogous to "Print every".

Here we set print to 600 seconds and save to 300 seconds.

X Samples

This option will calculate a nice interval for the Save every which will produce X saved points between the first and the last time. To get a number that coincides with the timestep, the exact amount of saves could still vary by a some amount.

Model timestep

We leave this to 10 seconds.

Modules in use

We will take the Full Monty and check all boxes!

Time dependent input

This is the important part - what time dependent input we send to the model and whether to modify the default values somehow. You should know what variables your simulation needs, and it would largely depend on the chemistry scheme. The default chemistry scheme is very narrow, but sufficient to give a demonstration how to use the model.

Variables

Here we need to pick all the time dependent variables that are sent in or manipulated in the model. First things you should pick are the compounds that were selected as precursors in MCM - or any other mechanism you are using. You can select these compounds by hand from the right hand panel, or you can use the two links below the left hand table. If you save the mcm_subset_mass.txt file when creating a new chemistry scheme, you can load the precursors form there. Another way is to press the select FIXED from chemistry (if there are any fixed variables).

We will pick a few variables. Some of the variables are read in from the ENV and MCM file, some are set to constant and some are parametrized. First, use the Tools→Print input headers with column numbers. In the terminal window that started ARCA gui, you will see the concents of the file - provided that the headers have been correctly written. Next, pick the following compounds, selecting one or more of them and clicking the button with left arrow to move them to our selected variables. Note that the color coding is used to mark which variables are loaded from the ENV_INPUT and which from the MCM_INPUT. Also note that one file can be used for both of them.

TEMPK and PRESSURE are always in use, just set their columns to 2 and 3, respectively. The input unit for TEMPK is C° in this file, and PRESSURE is hPa.
REL_HUMIDITY - we need some humidity in the air to produce OH. Set it to 65% by using the Shift and typing 65
O3 - also needed to OH production but also VOC oxidation reactions. Also constant and let's use Shift 20 and unit ppb. Be sure to check the units!
APINENE - alpha-pinene, a monoterpene compound that gets emitted from coniferous trees. These will come from the measurements file, so set the Column to 15, as can be seen from the terminal window. The thing is, Alpha-pinene can not be identified from other monoterpenes, so the measurement consists of all C10H16 compounds. From previous studies we know that roughly 50% of the monoterpenes emitted in Hyytiälä consists of alpha-pinene, so we correct the measuremens with setting Multi to 0.5. The unit of the measurements is ppb.
SO2 - let's throw in some sulfur to produce sulfuric acid. Also measured, these will be read from file ENV, and are on column 14 and are in ppb.
NH3 - gotta love the smell of ammonia in the morning. Let's use a constant 1e9 molecules/cm³. Use Shift again.
SW_RADIATION - Last but not least, let's add some sunlight. Here we will use the parametric input.

Parametric input

First load the current values for SW_RADIATION by selecting the variable name and pressing the button "Load selected". Next we alter the "minimum" and "maximum". Put 1 [W/m²] to minimum and 600 [W/m²] to maximum, let it be a sunny day. "Peaktime" is naturally somewhere around noon, and the dawn is already at four to five in the morning. Fiddling around with the sliders and lin/log scale should give you many enough options for a rough estimate or reasonable value for most variables. When you are happy with the shape, press the button "Save to selected" (and make sure SW_RADIATION is now selected). Were done. Check out the Variables tab, you will notice that SW_RADIATION has been highlighted to mark that parametric input is in use. You can always switch off this by selecting "No" from the corresponding Use PM drop down menu. The parameters for the function are still kept for the variable, they are just not used, instead the Column/Multi/Shift is used.

Chemistry

Phew, we are almost done. On this tab we will use the default settings.

Cluster formation

Likewise on this tab we will leave everything as they are. Skipping ACDC in low concentrations is ok, unless you explicitly want to study really small numbers. Make sure Run ACDC to Steady State is uncheked.

Aerosol

Condensing vapours

Here we need to make sure that the Vapour file and Elemental composition is pointing to the correct file. The default file "Vapours.dat" has only compounds whose saturation vapour pressure is less than 10^-6 ATM, and is good for this purpose.

PSD intialization

Here we choose "Use measured" since we have the BG particles file available. Initialize for 1 hour.

Aerosol structure

Here we pick the Fully stationary PSD scheme, set the number of bins to 100 and particle size range from 1.07e-9 to 2e-6 (1.4 nm to 2 µm) in diameter.

Losses

We will keep things simple and skip the losses in this example, so both tickboxes should be left unchecked.

Run ARCA

And that's it! Now you can Run the model with current settings! If you have aerosol module on, you can observe the particle size distribution evolving in real in View output → Surface plots by clicking on the tickbox Live update (only available when the model is running).