HELP HELP reading mf6 budget file for a transient simulation returning error of the invalid imeth code 1085209134

[StellaLiu-gogogo]

Recently I'm encountering a really weird situation.
I'm working in a linux based HPC envinronment with python 3.9, flopy 3.5version.

The model runs fine in steady state and the .cbb file can be read and analyzed without any problem.
When I turned the simulation as transient. the model terminates normally and returns reasonable groundwater level results.
in the list file, I could also see the water budget values for each flow component. it also looks reasonable in my case.

HOWEVER, when I try to use the flopy.util.binaryfile to read my .cbb file, It returns me the following error.
To my level of understanding, it means the header of the binary file returns a imeth value of 1085209134, but it should be a value of 1-6, representing 6 different methods of storing the water budget records.

I have no clue why the cbb file I have is with such a weird long number of imeth code. Especially all the rest of the model results just looks perfectly fine and it only happens in this specific case. Since I also used this same set of script for other case studies and it never returns me this weird error.

gonna be really appreciate if anyone could give some thoughts on how can I fix it.

`---------------------------------------------------------------------------
Exception Traceback (most recent call last)
Cell In[28], line 3
1 import flopy.utils.binaryfile as bf
2 path = 'MYPATH/workspace_2/mf6_2l_ss_rch2.cbb'
----> 3 bud = bf.CellBudgetFile(path)
4 hds = bf.HeadFile('MYPATH/workspace_2/mf6_2l_ss_rch2.hds')
5 print(type(hds))

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/flopy/utils/binaryfile.py:1037, in CellBudgetFile.init(self, filename, precision, verbose, **kwargs)
1034 raise Exception(f"LayerFile error: unrecognized kwargs: {args}")
1036 if precision == "auto":
-> 1037 success = self._set_precision("single")
1038 if not success:
1039 success = self._set_precision("double")

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/flopy/utils/binaryfile.py:1133, in CellBudgetFile._set_precision(self, precision)
1130 self.header_dtype = np.dtype(hdt)
1132 try:
-> 1133 self._build_index()
1134 except BudgetIndexError:
1135 success = False

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/flopy/utils/binaryfile.py:1274, in CellBudgetFile._build_index(self)
1269 self.iposarray.append(
1270 ipos
1271 ) # store the position right after header2
1273 # skip over the data to the next record and set ipos
-> 1274 self._skip_record(header)
1275 ipos = self.file.tell()
1277 # convert to numpy arrays

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/flopy/utils/binaryfile.py:1338, in CellBudgetFile._skip_record(self, header)
1332 nbytes = nlist * (
1333 np.int32(1).nbytes * 2
1334 + self.realtype(1).nbytes
1335 + naux * self.realtype(1).nbytes
1336 )
1337 else:
-> 1338 raise Exception(f"invalid method code {imeth}")
1339 if nbytes != 0:
1340 self.file.seek(nbytes, 1)

Exception: invalid method code 1085209134`

[langevin-usgs]

Maybe try this:

hds = bf.HeadFile('MYPATH/workspace_2/mf6_2l_ss_rch2.hds', precision='double')

[StellaLiu-gogogo]

Thanks for the reply. But the head file can be read good. only the .cbb file can't be read.
any thought?
thx.

[jlarsen-usgs]

Similar to the previous suggestion, maybe try this:

bud = bf.CellBudgetFile(path, precision="double")

[StellaLiu-gogogo]

thx it works!

[jlarsen-usgs]

No problem. The "auto" precision setting has a couple of gotcha's we haven't been able to solve yet for CellBudgetFile.

Also, of note. You can access the budget object from your model object after you run a simulation. Here's some pseudocode to make it even easier.

sim = <my flopy simulation object>
m = sim.get_model(<my model name>)
bud = m.output.budget()

This method enforces the double precision setting natively and as an additional plus, you don't have to deal with file paths. In my personal opinion, it's the easiest and most robust way to get output from MF6 simulations.

[StellaLiu-gogogo]

THX! that is a super helpful suggestion. However I am indeed curious to know what is the reason it results different precision of my hds file and cbb file.
How come the hds file was with single precision while cbb file was in double precision?
Anything related to the datatype of the input?

I'd like to know how to avoid this issue and always keep the result files in either double of single precision.