Shiny workflow plots not working with FATES outputs

[serbinsh]

See logs. Effectively, Shiny crashes when attempting to summarize data for plotting. I think it may be because it is expecting outputs in the netCDF files that aren't there. However, I am not sure why it is looking for the extra output vars as the only variables we define in the PEcAn outputs are:

AutoResp AutoResp
GPP GPP
NPP NPP
Qle Qle
TVeg TVeg
AbvGrndWood AbvGrndWood
CarbPools CarbPools
SoilTemp SoilTemp

Also it looks like the code is also trying to open up the original FATES output netCDF files as well, which it should not do.

Should the code use the *nc.var files to define which variables to find and extract? Or does it use the PEcAn standard output nc file metadata to define plotting vars?

Its possible I need to update the outputs or maybe we just need to modify shiny to work better with fates? I dont really have any familiarity with the Shiny code with PEcAn so I am not really sure where to begin.

[serbinsh]

Whoops, here are the logs

workflowPlot-sserbin-20180202-080757-33486.log workflowPlot-sserbin-20180201-091930-40362--LOG.log

[serbinsh]

Here is one of the outputs:

[sserbin@modex 2000066769]$ ncdump -h 1903.nc
netcdf \1903 {
dimensions:
    lon = 1 ;
    lat = 1 ;
    time = UNLIMITED ; // (8760 currently)
variables:
    double lon(lon) ;
        lon:units = "degrees_east" ;
        lon:long_name = "coordinate_longitude" ;
    double lat(lat) ;
        lat:units = "degrees_north" ;
        lat:long_name = "coordinate_latitude" ;
    double time(time) ;
        time:units = "days since 1903-01-01 00:00:00" ;
        time:long_name = "time" ;
        time:calendar = "standard" ;
    float AutoResp(time, lat, lon) ;
        AutoResp:units = "kgC m-2 s-1" ;
    float GPP(time, lat, lon) ;
        GPP:units = "kgC m-2 s-1" ;
    float NPP(time, lat, lon) ;
        NPP:units = "kgC m-2 s-1" ;
    float Qle(time, lat, lon) ;
        Qle:units = "W m-2" ;
    float TVeg(time, lat, lon) ;
        TVeg:units = "mm s-1" ;
    float AbvGrndWood(time, lat, lon) ;
        AbvGrndWood:units = "kgC m-2" ;
    float CarbPools(time, lat, lon) ;
        CarbPools:units = "kgC m-2" ;
    float SoilTemp(time, lat, lon) ;
        SoilTemp:units = "K" ;
}

[serbinsh]

It looks to me like Shiny wants to open any .nc file but it should only be trying to use PEcAn output netCDF correct?

2018-02-12 07:37:49 WARN   [read.output] :
   EFLX_LH_TOT missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/2005.nc
2018-02-12 07:37:49 WARN   [read.output] :
   GROWTH_RESP missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/2005.nc
2018-02-12 07:37:49 WARN   [read.output] :
   MAINT_RESP missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/2005.nc
2018-02-12 07:37:49 WARN   [read.output] :
   QVEGT missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/2005.nc
2018-02-12 07:37:49 WARN   [read.output] :
   TSOI_10CM missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/2005.nc
2018-02-12 07:37:49 WARN   [read.output] :
   AutoResp missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/case.clm2.h0.1902-01-01-00000.nc
2018-02-12 07:37:49 WARN   [read.output] :
   Qle missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/case.clm2.h0.1902-01-01-00000.nc
2018-02-12 07:37:49 WARN   [read.output] :
   TVeg missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/case.clm2.h0.1902-01-01-00000.nc
2018-02-12 07:37:50 WARN   [read.output] :
   AbvGrndWood missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/case.clm2.h0.1902-01-01-00000.nc
2018-02-12 07:37:50 WARN   [read.output] :
   SoilTemp missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/case.clm2.h0.1902-01-01-00000.nc
2018-02-12 07:37:50 WARN   [read.output] :
   AutoResp missing in
   /data/Model_Output/pecan.output/PEcAn_2000000842/out/2000066769/case.clm2.h0.1903-01-01-00000.nc

[serbinsh]

Potential work around: Store CLM/CLM-FATES raw outputs in a sub directory, controlled by the job template, and keep putting the PEcAn outputs in the standard ../out/ID/ folder

[github-actions[bot]]

This issue is stale because it has been open 365 days with no activity.