There are significant differences in the channel_sd_mon*.txt files simulated by SWAT+ 61.0.1 and SWAT+ 61.0.2.

[FangPlus7904]

When I ran simulations using SWAT+ versions 61.0.1 and 61.0.2 separately, I noticed a significant difference in the simulated streamflow between the two versions. The results from each version are as follows:

Upon carefully examining the output files, I found that the difference in simulated streamflow between the two versions is approximately equal to the flo_cha values in the aquifer_*.txt files.

I have tried to identify the cause of this discrepancy but have been unsuccessful. Does anyone know what might be causing this issue?

[Atreyagaurav]

I think this is related to why the tests are failing in the repo. Could you try running ctest like suggested on the instructions.

I think going back and testing a few commits between 61.0.1 and 61.0.2 could help find the commit where this issue starts by running the cmake and ctest repeatedly. Once you find an issue where the test fails, but it passes on the previous commit, we can see the changes.

Sorry, I cannot help with this because I ran into trouble compiling several commits (I use gfortran).