Initial conditions file question

[Musselman]

Hi All! Happy to be the first to post a question to the Summa Git page.

I'm using the docker version uwhydro/summa:master pushed on May 28th. I'm running a simple test with 9 HRUs, 1 GRU. I'm getting a runtime error that I think is associated with my initial conditions file. I've exhausted my troubleshooting ideas.

At line 277 of file /code/build/source/engine/check_icond.f90
Fortran runtime error: Expected INTEGER for item 4 in formatted transfer, got REAL
(a,1x,i0)
      ^

Error termination. Backtrace:
 WARNING: routingRunoffFuture is not in the initial conditions file ... using zeros`

I don't know where the code is expecting an integer where there isn't already an integer. The header of my initial conditions file is:

dimensions:
	midToto = 8 ;
	hru = 9 ;
	scalarv = 1 ;
	ifcToto = 9 ;
	midSoil = 8 ;
variables:
	double mLayerVolFracIce(midToto, hru) ;
	double scalarCanairTemp(scalarv, hru) ;
	int64 nSnow(scalarv, hru) ;
	double iLayerHeight(ifcToto, hru) ;
	double mLayerMatricHead(midSoil, hru) ;
	double scalarSnowAlbedo(scalarv, hru) ;
	double dt_init(scalarv, hru) ;
	double mLayerTemp(midToto, hru) ;
	double scalarSfcMeltPond(scalarv, hru) ;
	double scalarCanopyTemp(scalarv, hru) ;
	double scalarSnowDepth(scalarv, hru) ;
	int64 nSoil(scalarv, hru) ;
	double scalarSWE(scalarv, hru) ;
	double scalarCanopyLiq(scalarv, hru) ;
	double mLayerVolFracLiq(midToto, hru) ;
	double mLayerDepth(midToto, hru) ;
	double scalarCanopyIce(scalarv, hru) ;
	double scalarAquiferStorage(scalarv, hru) ;
	double routingRunoffFuture(scalarv, hru) ; 

Likely a separate issue - I don't think I'm correctly defining routingRunoffFuture b/c it's provided in initial conditions but produces a warning that it is missing.

Thanks in advance for any thoughts on how I could troubleshoot this!
-Keith

[arbennett]

Hey Keith - I think there's a couple of things that might be going on off the top of my head (aka I haven't investigated at all). Is there any chance you could share your initial conditions file somehow? Or even better a setup including maybe a month/year of forcing data?

For completeness, here's my initial thoughts on the issues that might be cropping up:

• I'm not sure how the docker builds are updated, there might be something there that's off. We also have a conda version that you can install with conda install -c conda-forge summa that might work? Probably not though is my guess.
• I'm wondering if the routingRunoffFuture variable should be a function of GRU rather than HRU and maybe somehow that's throwing a weird error
• Is routingRunoffFuture actually required? You might just be able to omit it.
• Was this initial conditions file created as a restart file from an earlier version? I think we tried to keep backwards compatibility, but maybe something got overlooked?

[andywood]

Hi Keith, I'm not sure about the int/real problem -- perhaps try making your nSnow/nSoil variables 'int' (eg 4byte) -- I forget how the code looks for that. As Andy suggested, the warning on routingRunoffFuture can be ignored, and that variable can be deleted for a cold start. I added that variable to the state files to enable exact restarts when someone is using basin routing variables, but I also kept the code backward compatible, allowing you to skip it (with the warning). The variable will be created in a state file written by SUMMA -- leading to a state file like the following. Note the dimensions include 'tdh'. Cheers, -Andy netcdf wbout_restart_2011080900 { dimensions: scalarv = 1 ; hru = 43 ; spectral = 2 ; midToto = 8 ; midSoil = 3 ; ifcToto = 9 ; tdh = 2000 ; gru = 43 ; variables: double dt_init(scalarv, hru) ; dt_init:_FillValue = NaN ; dt_init:long_name = "length of initial time step at start of next data interval" ; dt_init:units = "s" ; double scalarCanopyIce(scalarv, hru) ; scalarCanopyIce:_FillValue = NaN ; scalarCanopyIce:long_name = "mass of ice on the vegetation canopy" ; scalarCanopyIce:units = "kg m-2" ; double scalarCanopyLiq(scalarv, hru) ; scalarCanopyLiq:_FillValue = NaN ; scalarCanopyLiq:long_name = "mass of liquid water on the vegetation canopy" ; scalarCanopyLiq:units = "kg m-2" ; double scalarCanopyWat(scalarv, hru) ; scalarCanopyWat:_FillValue = NaN ; scalarCanopyWat:long_name = "mass of total water on the vegetation canopy" ; scalarCanopyWat:units = "kg m-2" ; double scalarCanairTemp(scalarv, hru) ; scalarCanairTemp:_FillValue = NaN ; scalarCanairTemp:long_name = "temperature of the canopy air space" ; scalarCanairTemp:units = "K" ; double scalarCanopyTemp(scalarv, hru) ; scalarCanopyTemp:_FillValue = NaN ; scalarCanopyTemp:long_name = "temperature of the vegetation canopy" ; scalarCanopyTemp:units = "K" ; double spectralSnowAlbedoDiffuse(spectral, hru) ; spectralSnowAlbedoDiffuse:_FillValue = NaN ; spectralSnowAlbedoDiffuse:long_name = "diffuse snow albedo for individual spectral bands" ; spectralSnowAlbedoDiffuse:units = "-" ; double scalarSnowAlbedo(scalarv, hru) ; scalarSnowAlbedo:_FillValue = NaN ; scalarSnowAlbedo:long_name = "snow albedo for the entire spectral band" ; scalarSnowAlbedo:units = "-" ; double scalarSnowDepth(scalarv, hru) ; scalarSnowDepth:_FillValue = NaN ; scalarSnowDepth:long_name = "total snow depth" ; scalarSnowDepth:units = "m" ; double scalarSWE(scalarv, hru) ; scalarSWE:_FillValue = NaN ; scalarSWE:long_name = "snow water equivalent" ; scalarSWE:units = "kg m-2" ; double scalarSfcMeltPond(scalarv, hru) ; scalarSfcMeltPond:_FillValue = NaN ; scalarSfcMeltPond:long_name = "ponded water caused by melt of the \"snow without a layer\"" ; scalarSfcMeltPond:units = "kg m-2" ; double mLayerTemp(midToto, hru) ; mLayerTemp:_FillValue = NaN ; mLayerTemp:long_name = "temperature of each layer" ; mLayerTemp:units = "K" ; double mLayerVolFracIce(midToto, hru) ; mLayerVolFracIce:_FillValue = NaN ; mLayerVolFracIce:long_name = "volumetric fraction of ice in each layer" ; mLayerVolFracIce:units = "-" ; double mLayerVolFracLiq(midToto, hru) ; mLayerVolFracLiq:_FillValue = NaN ; mLayerVolFracLiq:long_name = "volumetric fraction of liquid water in each layer" ; mLayerVolFracLiq:units = "-" ; double mLayerVolFracWat(midToto, hru) ; mLayerVolFracWat:_FillValue = NaN ; mLayerVolFracWat:long_name = "volumetric fraction of total water in each layer" ; mLayerVolFracWat:units = "-" ; double mLayerMatricHead(midSoil, hru) ; mLayerMatricHead:_FillValue = NaN ; mLayerMatricHead:long_name = "matric head of water in the soil" ; mLayerMatricHead:units = "m" ; double scalarAquiferStorage(scalarv, hru) ; scalarAquiferStorage:_FillValue = NaN ; scalarAquiferStorage:long_name = "water required to bring aquifer to the bottom of the soil profile" ; scalarAquiferStorage:units = "m" ; double scalarSurfaceTemp(scalarv, hru) ; scalarSurfaceTemp:_FillValue = NaN ; scalarSurfaceTemp:long_name = "surface temperature (just a copy of the upper-layer temperature)" ; scalarSurfaceTemp:units = "K" ; double mLayerDepth(midToto, hru) ; mLayerDepth:_FillValue = NaN ; mLayerDepth:long_name = "depth of each layer" ; mLayerDepth:units = "m" ; double mLayerHeight(midToto, hru) ; mLayerHeight:_FillValue = NaN ; mLayerHeight:long_name = "height of the layer mid-point (top of soil = 0)" ; mLayerHeight:units = "m" ; double iLayerHeight(ifcToto, hru) ; iLayerHeight:_FillValue = NaN ; iLayerHeight:long_name = "height of the layer interface (top of soil = 0)" ; iLayerHeight:units = "m" ; int nSnow(hru) ; nSnow:long_name = "number of snow layers" ; nSnow:units = "-" ; int nSoil(hru) ; nSoil:long_name = "number of soil layers" ; nSoil:units = "-" ; double routingRunoffFuture(tdh, gru) ; routingRunoffFuture:_FillValue = NaN ; routingRunoffFuture:long_name = "runoff in future time steps" ; routingRunoffFuture:units = "m s-1" ;

…

On Tue, Jun 29, 2021 at 11:19 PM Andrew Bennett ***@***.***> wrote: Hey Keith - I think there's a couple of things that might be going on off the top of my head (aka I haven't investigated at all). Is there any chance you could share your initial conditions file somehow? Or even better a setup including maybe a month/year of forcing data? For completeness, here's my initial thoughts on the issues that might be cropping up: - I'm not sure how the docker builds are updated, there might be something there that's off. We also have a conda version that you can install with conda install -c conda-forge summa that might work? Probably not though is my guess. - I'm wondering if the routingRunoffFuture variable should be a function of GRU rather than HRU and maybe somehow that's throwing a weird error - Is routingRunoffFuture actually required? You might just be able to omit it. - Was this initial conditions file created as a restart file from an earlier version? I think we tried to keep backwards compatibility, but maybe something got overlooked? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#473 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABIKARP2YJIP3DHKWJTMHXDTVKSM7ANCNFSM47GSWETA> .