dapper.met.adapters.fluxnet

dapper module: met.adapters.fluxnet.

Functions

infer_fluxnet_dt_hours(df)

Infer native FLUXNET timestep from timestamp columns in hours.

Classes

FluxnetAdapter()

AmeriFlux FLUXNET → ELM adapter.
class dapper.met.adapters.fluxnet.FluxnetAdapter[source]

Bases: BaseAdapter

AmeriFlux FLUXNET → ELM adapter.

Assumptions

  • User provides a single FLUXNET CSV (FULLSET or SUBSET) per run.

  • CSV contains TIMESTAMP_START/TIMESTAMP_END (or TIMESTAMP) columns.

  • Missing values are coded as -9999.

  • Exporter supplies df_merged with [‘gid’,’lat’,’lon’,’zone’, …] already merged in from df_loc.

DRIVER_TAG = 'FLUXNET'
SOURCE_NAME = 'FLUXNET (AmeriFlux ONEFlux) tower data'
discover_files(csv_directory, calendar)[source]

Return (csv_files, start_year, end_year).

Parameters:

calendar (str)

native_dt_hours: Optional[float]
preprocess_shard(df_merged, start_year, end_year, calendar, dformat)[source]

Return a DataFrame with at least: [‘gid’,’time’,’LATIXY’,’LONGXY’,’zone’, <ELM vars>]

Return type:

DataFrame

Parameters:

  • df_merged (DataFrame)

  • start_year (int)

  • end_year (int)

  • calendar (str)

  • dformat (str)

required_vars(dformat)[source]

Optional: return a list of ELM var short names that this adapter will produce for the given dformat (‘BYPASS’ or ‘DATM_MODE’). Exporter doesn’t require it.

Parameters:

dformat (str)

resolution: Optional[str]
dapper.met.adapters.fluxnet.infer_fluxnet_dt_hours(df)[source]

Infer native FLUXNET timestep from timestamp columns in hours.

Handles:

  • half-hourly/hourly/weekly: TIMESTAMP_START, TIMESTAMP_END (YYYYMMDDHHMM)

  • daily/monthly/yearly: TIMESTAMP (YYYYMMDD or YYYYMM, etc.)

Returns:

Approximate timestep in hours.

Return type:

float

Raises:

ValueError – If no suitable timestamp columns are found or dt cannot be inferred.

Parameters:

df (DataFrame)