Running with NetCDF forcing provider requires forcing data and start_time to be the same

[mattw-nws]

NetCDF provider is returning wrong values for time values when the simulation start time is not the same as the forcing file start time.

Current behavior

When using "forcing":{"provider": "NetCDF"... and a start_time that does NOT equal the T0 time in the forcing file, the wrong data is returned for the wrong timesteps, effectively using the timestep index instead of the correct time value. CsvPerFeature does not exhibit this behavior.

Expected behavior

In the example below, the precip starting at 1700 (timestep 38) should start at 1400 (timestep 35) instead.

Steps to replicate behavior (include URLs)

1. Use the following realization config, with the test data in data:

{
    "global": {
        "formulations": [
            {
                "name": "bmi_multi",
                "params": {
                    "model_type_name": "bmi_multi_noahowp_cfe",
                    "forcing_file": "",
                    "init_config": "",
                    "allow_exceed_end_time": true,
                    "main_output_variable": "Q_OUT",
                    "output_variables": ["APCP_surface","RAIN_RATE","Q_OUT"],
                    "modules": [
                        {
                            "name": "bmi_fortran",
                            "params": {
                                "model_type_name": "bmi_fortran_noahowp",
                                "library_file": "./extern/noah-owp-modular/cmake_build/libsurfacebmi.so",
                                "forcing_file": "",
                                "init_config": "./data/bmi/fortran/noah-owp-modular-init-{{id}}.namelist.input",
                                "allow_exceed_end_time": true,
                                "main_output_variable": "QINSUR",
                                "variables_names_map": {
                                    "PRCPNONC": "atmosphere_water__liquid_equivalent_precipitation_rate",
                                    "Q2": "atmosphere_air_water~vapor__relative_saturation",
                                    "SFCTMP": "land_surface_air__temperature",
                                    "UU": "land_surface_wind__x_component_of_velocity",
                                    "VV": "land_surface_wind__y_component_of_velocity",
                                    "LWDN": "land_surface_radiation~incoming~longwave__energy_flux",
                                    "SOLDN": "land_surface_radiation~incoming~shortwave__energy_flux",
                                    "SFCPRS": "land_surface_air__pressure"
                                },
                                "uses_forcing_file": false
                            }
                        },
                        {
                            "name": "bmi_c",
                            "params": {
                                "model_type_name": "bmi_c_cfe",
                                "library_file": "./extern/cfe/cmake_build/libcfebmi.so",
                                "forcing_file": "",
                                "init_config": "./data/bmi/c/cfe/{{id}}_bmi_config.ini",
                                "allow_exceed_end_time": true,
                                "main_output_variable": "Q_OUT",
                                "registration_function": "register_bmi_cfe",
                                "variables_names_map": {
                                    "water_potential_evaporation_flux": "ETRAN",
                                    "atmosphere_air_water~vapor__relative_saturation": "SPFH_2maboveground",
                                    "land_surface_air__temperature": "TMP_2maboveground",
                                    "land_surface_wind__x_component_of_velocity": "UGRD_10maboveground",
                                    "land_surface_wind__y_component_of_velocity": "VGRD_10maboveground",
                                    "land_surface_radiation~incoming~longwave__energy_flux": "DLWRF_surface",
                                    "land_surface_radiation~incoming~shortwave__energy_flux": "DSWRF_surface",
                                    "land_surface_air__pressure": "PRES_surface"
                                },
                                "uses_forcing_file": false
                            }
                        }
                    ],
                    "uses_forcing_file": false
                }
            }
        ],
        "forcing": {
            "path": "./data/forcing/cats-27_52_67-2015_12_01-2015_12_30.nc",
            "provider": "NetCDF"
        }
    },
    "time": {
        "start_time": "2015-12-01 03:00:00",
        "end_time": "2015-12-30 22:00:00",
        "output_interval": 3600
    }
}

and you will get incorrect data for the timesteps, e.g. cat-27 timestep 38:

38,2015-12-02 17:00:00,0.100000001,0.000100115,0.059517931

This should start at 1400, not 1700, which is timestep 38 in the forcing data...i.e. if you change the start_time above to "2015-12-01 00:00:00", it will produce:

38,2015-12-02 14:00:00,0.100000001,0.000100115,0.059517931

instead.

Screenshots

[stcui007]

Does running bmi_c_cfe model reproduce this behavior? I just want to test on a simpler model at the start.

[stcui007]

My output:
38,2015-12-02 17:00:00,0.100000001,0.000100115,0.046988432
Note the last number is different, maybe we are using different model?

[mattw-nws]

The last column is Q_OUT, and if that output was produced with CFE only the result can definitely vary, as Noah will change the input that CFE receives.

Either way, the first two columns are relevant to the issue and it looks like you are reproducing it successfully.