Substantial wind and hydro fuel (mmbtu) in gf_eia923?

[zaneselvans]

For some reason, a substantial amount of "fuel" is listed as being consumed by wind and hydro facilities in the gf_eia923 table, in terms of mmbtu. However, the fuel type listed in the table is... wind, and no number of units of fuel consumed is reported. Does this mean anything? Is it correct? The amount of fuel by heat content being consumed is much smaller than gas, coal, or nuclear, but larger than waste, solar, and oil. See below.

import pudl
import sqlalchemy as sa
pudl_settings = pudl.workspace.setup.get_defaults()
pudl_engine = sa.create_engine(pudl_settings['pudl_db'])
pudl_out_orig = pudl.output.pudltabl.PudlTabl(pudl_engine)
gf_eia923_orig = pudl_out_orig.gf_eia923()
for fuel in ['wind', 'coal', 'gas', 'solar', 'hydro', 'oil', 'waste', 'nuclear']:
    byfuel = (
        gf_eia923_orig.
        query(f"fuel_type_code_pudl=='{fuel}'").
        query("fuel_consumed_for_electricity_mmbtu > 0.0").
        groupby(['report_date'])["fuel_consumed_for_electricity_mmbtu"].agg(sum).reset_index()
    )
    plt.plot(byfuel.report_date, byfuel.fuel_consumed_for_electricity_mmbtu, label=fuel)
plt.ylabel("Fuel Consumed [mmbtu]")
plt.title("Fuel Consumption by Fuel Type (gf_eia923)")
plt.legend()
plt.save;

[zaneselvans]

Wrote to Laura Martin at EIA about this issue on 2019-10-15.

[zaneselvans]

The response from EIA is that the heat content (in MMBTU) reported for renewable energy facilities which consume no fuel is calculated based on the weighted heat rate of all fossil fuel generation (?!) -- it's the amount of fuel heat content that the wind/solar/hydro/etc. power plant would have consumed to generate the electricity that it did, had it been operating with the efficiency of a typical fossil plant.

Which, honestly, seems bizarre to me. But that's why it's like this in the original data.