forrtl: error (72): floating overflow

[爾文超]

Hi everyone, just met this error when running SWAT CUP "forrtl: error (72): floating overflow", have anybody found this error as well?. I can run my original project just fine, but as soon as I increase the number of subbasins and HRUs it doesnt work anymore!!! does anybody know the causes?. Id appreciate your help very much.

[Muhammad Shahid Iqbal]

HI

I am also facing the same problem in my project. My SWAT project runs fine but as soon as I start running my project in SWAT_CUP it gives me this error. Can any one tells whats the issue

[West Cotter]

I have the same problem with my computer. I have run SWAT CUP for a while with no problem, but suddenly in yesterday, SWAT CUP did not work. I checked SWAT 2012, reset up SWAT CUP, but still the same error: error (72): floating overflow.

 

I plan to reinstall SWAT CUP see if the problem can be solved. We need someone here to say something about this.

 

Kangsheng

[Abbaspour, Karim]

This error is from swat. Either a swat file is corrupted, or you have some strange parameter combination in the swat files.

Run debug swat to see which file is giving the problem. You can also replace all swat files in the main swatcup directory with the files in backup directory.

At any event make sure swat is working before running swatcup.

Karim

--
You received this message because you are subscribed to the Google Groups "SWAT-CUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jim Almendinger]

If the run failed right away, it could be an input data-formatting error.  But if the run failed after a few years of simulation, then (as Karim pointed out), you may have some odd parameter combination that is causing some instability in the output.  This could be almost anything in SWAT where a quantity goes to zero and is later used as a divisor, giving you a floating overflow (infinite value) -- for example, when a volume of a pond or reservoir that has gone dry is divided into a mass of nutrients or sediment, to calculate a concentration.  You might just have some ranges beyond reasonable values in your SWAT-CUP runs that wouldn't be apparent when just running SWAT alone with a singe parameter set. 

-- Jim

---

From: "West Cotter" <westcot...@gmail.com>
To: "SWAT-CUP" <swat...@googlegroups.com>
Sent: Friday, January 22, 2016 8:26:53 AM
Subject: Re: forrtl: error (72): floating overflow

--
You received this message because you are subscribed to the Google Groups "SWAT-CUP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

---

Spam
Not spam
Forget previous vote

--

Dr. James E. Almendinger
St. Croix Watershed Research Station
Science Museum of Minnesota
16910 152nd St N
Marine on St. Croix, MN  55047
tel: 651-433-5953 ext 19

[Steven Jepsen]

I had this problem yesterday, and found through trial-and-error (varying one parameter at a time in SWAT-CUP par_val.txt file) that it was being caused by parameter SNO50CV becoming too close to 0.95, where it causes the natural log function to blow up. The cause of your problem may be different. This just illustrates how excessive parameter limits can cause math functions to blow up, resulting in floating overflow errors.

Steve

[Pallav]

Hi Steve,

Right on spot! Your suggestion helped in my case. Changed limit of SNO50COV from 0-1 to 0.05 to 0.8 and it worked. I used to use the limits of value as prescribed in the SWAT Cup. However, it seems its wiser to shimmy a little within the both limits, just to be safe.

Thanks!

Pallav

[Tesfalem Abraham]

Hello Dears,

I frequently checked swat.exe run and it works properly, but SWAT-CUP showes error  forrtl: error (73): floating divide by zero and sometimes forrtl: error (72): floating overflow after few simulation like 4 or 5, i see some solutions for the problem but regarding parameter combination and how can i guess it? your help please

regards!

[Karim Abbaspour]

probably a different version of swat. Copy your version into swatcup project folder.

Karim

To unsubscribe from this group and stop receiving emails from it, send an email to swat-cup+unsubscribe@googlegroups.com.

[shara...@gmail.com]

I also got the same error. What is meant by Run debug swat and how to do it?

[Tesfalem Abraham]

I have encountered the same forrtl error <73>. As suggested by Karim some strange parameters are the cause. I repeatedly changed those parameters range but not running.

Any body can suggest me how to find suitable par ranges.

[Bendik Hansen]

Please paste your parameters (including the r__, v__, a__ at the start) so we can see what might be wrong. 

Best regards,

Bendik

On Thu, May 2, 2019 at 10:50 AM Tesfalem Abraham <ates...@gmail.com> wrote:

I have encountered the same forrtl error <73>. As suggested by Karim some strange parameters are the cause. I repeatedly changed those parameters range but not running.

Any body can suggest me how to find suitable par ranges.

[selam haftuu]

. Have you found a way to solve it?

[chriss lewis]

shara...@gmail.com how is the run debug process done? did you get how to solve it? i am encountering the same problem, just at the start of running simulation after executing year 1 then the floating overflow error appears, but in arc swat the program runs simulation successful and the txtinout file is obtained.

[2w2e]

where you installed ArcSWAT, there is a file called swat_64debug.exe 

Copy that into the swatcup project folder and run it in command line. 

[Jillian Wangari]

I have done this and there is no error in my execution it was completed successfully. 

Swat cup still gives the error 72 floating overflow

To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/2b02aa34-f765-4aa6-bab3-320c3e5b3e4an%40googlegroups.com.

[Filipe Pinhati]

In my case the run was failing right away, as soon as it started the first year of simulation.

But I could not identify any input data-formatting error.

So I created another SWAT-CUP project and pasted the content of the failing project in it, and now it works.

[abiy Gebremichael]

Copy swat2012.exe from main SWAT directory or from c drive or from your original SWAT instalstion c/Swat/arcswat/ directory and paste on SWATcup.sufi2 project and edit to swat.exe. there is other swat.exe in swatcup.sufi2 project and you can rename it to swatdefault.exe. finaly swat.exe will be that copied from SWAT instsllation. In evry new project preparation of swatcup  you need ti do this.

Cheers!

To view this discussion on the web visit https://groups.google.com/d/msgid/swat-cup/9ccaa5e0-457a-4a80-aadd-d72f5f6d27e0n%40googlegroups.com.

[Filipe Pinhati]

At first I was getting the error (72) floating overflow error.

Then I run Swat-Cup in the command line using the Swat64_debug.exe application and now I got the error (63): output conversion error

So the error (72) floating overflow apparently became error (63) output conversion error.

And it says it is occurring on the output.std file.

Can anyone shed some light in this problem?

               Error below:

forrtl: error (63): output conversion error, unit 26, file E:\PhD_2019_backup\SWAT-CUP\Period_A_2nd_Calibration.Sufi2.SwatCup\output.std

Image              PC                Routine            Line        Source

swat_64debug.exe   00007FF716B264F4  Unknown               Unknown  Unknown

swat_64debug.exe   00007FF716B3D453  Unknown               Unknown  Unknown

swat_64debug.exe   00007FF7163054ED  STDAA                     493  stdaa.f

swat_64debug.exe   00007FF7165BC5C2  WRITEAA                   557  writeaa.f

swat_64debug.exe   00007FF715B94457  MAIN__                    113  main.f

swat_64debug.exe   00007FF716BF56BE  Unknown               Unknown  Unknown

swat_64debug.exe   00007FF716BF6389  Unknown               Unknown  Unknown

swat_64debug.exe   00007FF716BF62AE  Unknown               Unknown  Unknown

swat_64debug.exe   00007FF716BF616E  Unknown               Unknown  Unknown

swat_64debug.exe   00007FF716BF63F9  Unknown               Unknown  Unknown

KERNEL32.DLL       00007FFA74C17C24  Unknown               Unknown  Unknown

ntdll.dll          00007FFA750ED721  Unknown               Unknown  Unknown

[Filipe Pinhati]

Just an update to my question, I have checked the output.std file and I found this strange data full of asterisk in the pathogen output.

It is the only portion in the entire output.std file with asterisks instead of numbers. I guess this is what is causing the problem

                   DIE-GRO P Q =     2349592.2 (No/M2)

                   DIE-GRO LP Q =  ************ (No/M2)

                   DIE-GRO P SED =       7553.8 (No/M2)

                   DIE-GRO LP SED =    -548094.3 (No/M2)

                   BACT P RUNOFF =       5490.7 (No/M2)

                   BACT LP RUNOFF = ************ (No/M2)

                   BACT P SEDIMENT =          1.5 (No/M2)

                   BACT LP SEDIMENT =          4.5 (No/M2)

                   BACT P INCORP =     472185.6 (No/M2)

                   BACT LP INCORP = ************ (No/M2)

[Mamush Tekle]

I face the same error what i should do

[Mamush Tekle]

Dear Dr. Abbabsour, I encountered forrtl: error (72) floating overflow while simulating a SWAT-CUP climate change scenario for 60 years, from 2031 to 2090, to assess the impact of climate change on streamflow. The calibration and validation stage was successful with good performance; however, I couldn't solve the trouble, so please help me locate the error and solve it.

[Sefa Nur Yeşilyurt]

Thank you Steven Jepsen, You helped me solve my problem. I was really struggling for days. And the problem was SNO50CV. It works fine between 0-0.8. Thanks

26 Aralık 2022 Pazartesi tarihinde saat 02:11:02 UTC+3 itibarıyla Mamush Tekle şunları yazdı: