Problem Simulation - Forum - iMOD
Although the new site is live, work is still in progress to improve it.
- Forums: The organization of (sub-)categories and threads is disrupted. Please use the search box inside the "Message boards" to navigate.
I want to do a transient simulation with the MNW Package.
As layer type I took convertible (HNEW-BOT). Then I switched to modflow 2005 and pressed simulate. But each time come two errors. But I don't found a mistake in the ipf file.
best regards Franzi
In your issue you attach the fbr1.txt and fbr3.txt files. I presume you also have the 30 ILA*.txt files with a similar file format in the same directory.
At what date do you start your model?
I just guess: perhaps, if the first given pumping date in your TXT file is after the start of the model, perhaps this error can occur.
For sure, check if adding an early start date will help you out. So your txt files look like
Thanks for the answer. I changed the times as suggested by you. But it still does not work.
I've increased the time steps to days and changed the time format in the TxT file to year month day (no more year / month / day / hour / minute / second)
Thats how it works. As soon as I change it, it will not work anymore.
best regards Franzi
OK. Thanks for sharing the experience.
I will check it in the iMOD source code and the manual.
I tried to reproduce your issue in a run file with the well module. My experience is that iMOD version 4.2.1 can handle the hhmmss notation.
Can you report me if this is the solution?
Recently the time notation in iMOD is changed. You can read about it in the release notes. (https://oss.deltares.nl/web/imod/release-notes#iMOD4.1_Ad_1)
Thank you for the answer.
I have instalation the Version iMod_v2_1. The Simulation work with this timesteps (yyyymmddhhmmss) without problems.
But the next Problem is that when I will start from the results of this transient Simulation the Partikel Tracking the Program iMod crashed and closed without any message. Where could the problem be?
You write about the installation of iMod version 2.1 but I think you made a typing error and you mean iMOD 4.2.1.
Now about your question on Particle Tracking in the Graphical User Inter Face.
--> Can you show me where in the process iMOD crashes?
In general, these are the main steps to run a Particle Tracking in iMOD
1. run an iMOD model in order to get all the BDGFxF.IDF files describing the 3D fluxfield
2. Define your Starting points in the 'Define Starting points' toolbox to get a startpoing file *.ISD
3. Fill in the other required initialization data in the Pathlines Simulation window.
If you like to look at an example, please have a look at the steps at the end of tutorial 4 (https://content.oss.deltares.nl/imod/imod421/imod_um_html/imod-um-Start-Pathline-Simulation.html#autosec-329).
During the Partikel Tracking iMod creates the files IMODPATH.RUN and IMODPATH.LOG in the same directory as the iMODFLOW results.
The LOG file gives information on the process
--> so please check this log file to get an idea on the crash.
The RUN file is the iMODPATH run file containing all information needed to (re)run an iMODPATH run outside iMOD. So you can reuse this file to re-run iMODPATH (in a DOS window) without clicking around in iMOD every time you need to make a small change to the run. Therefore use the Batch option explained in the manual (8.6.5 RUNFILE-Function).
(please send those 2 files if created)
Kind regards, Frans
I have this results (see attachments 1 & 2) of the Model simulation.
Then I have create the startpoints.
Then I opened the Pathline Simulation window (see attachments 3). And when I click of the Models Folder of the transient Simulation than the Program iMod closed without error message.
Good choice to go for the iMOD batch option.
One suggestion on the Function iModpath: this function has a different set-up compared to the other batch functions. (I think because iMODPATH is a iMOD version of the standard software Modpath).
This iMODPATH functions need 2 files, an INI file and in your case a RUN file containing most input data (see also the description in the Manual on keyword RUNFILE and the example with this function).
The first 3 lines of your INI file are correct. All the other input data (from NLAY and further) should be in the RUN (without the keywords!) .
The RUN file will look like:
2 !## NLAY
48,1 !## NPER,ISTO=
2 !## NSDF
IMOD_USER\STARTPOINTS\Alle ILA.isd !## ISDFILE
IMOD_USER\MODELS\INSTATIONAR_2CAPTURE_LAYER2.IFF !## OUTFILE
Let me know the result.
I changed the file
one INI file and one run file
Unfortunately, the problem reappears.
we both are surprised because we see the keyword FUNCTION in the INI file.
So I tried to reproduce this error by using one of my own iMODPATH runs. That one works OK. So I looked deeper into the file and found out the reason: your INI file (and RUN file) are written in the UTF-8 format. Just like the Unix format, this format can give an error when read by iMOD. The file format for text files in an iMOD project should be ANSI. See the attachments for the ANSI format of your files.
One additional remark on you files: when referring to other files in an INI for RUN file 1) use the complete path, so including a drive letter and 2) for a file name with spaces in it (always try to avoid spaces in names of maps and files) always put them in brackets.
Thanks for the quick help and the great advice. Everything works after the changes.
One more question:
Could you please explain what the difference between the keywords STARTDATE and SWINDOW in runfile?
This figure will be included into the next version of the manual showing the relation between STARTDATE, SWINDOW and the dates of the IDF files in the RUN file.
In the not so successful sketch, I have tried to represent the temporal window. I would like to use the particle tracking method to show the influence of different pumps on the particles. With which stopcrit could I best represent this?
Generally, the runfile will only work with year / month / day input. As soon as I append the time indication comes an error message.
To start with, you could start the particle tracking from infiltration well X the same time well X starts infiltrating and than continue particle tracking till the end, the moment the particle leaves the model (boundary, river, well). In the resulting flowpaths, you will see the influence of the X+n pumping wells.
In case the whole process is a continuous cycle of starting and stopping of infiltration wells, use STOPCRIT=2.
In case all infiltration/abstraction activities run one time and stop after the last stress period use STOPCRIT=3. That will continue particle tracking till the pathline ends based on the flux field in the last stressperiod (should represent a stationary situation without pumping).
In addition: To see the dynamic in your system, you could additionally start particles from infiltration well X just before well X is shutting down, so at the moment well X+1 starts. Both flowline sets together will show the bandwidth of the flow from well x.
You are right, at the moment the runfile only works with year / month / day input.
The option to append the time indication (hhmmss) is added to the list of desired future developments.
As soon as budget is found for this upgrade, it will be implemented.
I am thinking of a workaround for the moment to avoid time notation:
- your model is set up in steps of 7 hours
- you might convert hours to days; 7 hours become 7 days.
- this means instead of 2016122700000 - 20170109170000 make the period 20161227 - 20190129
- with the iMODbatch function IDFCALC you should divide the flowfield by 24.0 and rename the FxF files for the new period
- also convert the time dependant keywords like STARTDATE, SWINDOW, MAXT etc.
- if you analyse the results (flowlines, start- and end points) the path will be OK but you should divide the age of a pathline by 24 (or multiply the velocity by 24)
Maybe I miss something zo please think this over yourself and see (or test) if it will work