sMRFluxPartition: sEddyProc$sMRFluxPartition - Flux partitioning after...

In REddyProc: Data processing and plotting utilities of (half-)hourly eddy-covariance measurements

Description

Nighttime-based partitioning of measured net ecosystem fluxes into gross primary production (GPP) and ecosystem respiration (Reco)

Usage

sMRFluxPartition(FluxVar.s = "NEE_f", QFFluxVar.s = "NEE_fqc", 
    QFFluxValue.n = 0, TempVar.s = "Tair_f", QFTempVar.s = "Tair_fqc", 
    QFTempValue.n = 0, RadVar.s = "Rg", Lat_deg.n, 
    Long_deg.n, TimeZone_h.n, T_ref.n = 273.15 + 15, 
    Suffix.s = "", debug.l = list(useLocaltime.b = FALSE), 
    parsE0Regression = list())

Arguments

FluxVar.s	Variable name of column with original and filled net ecosystem fluxes (NEE)
QFFluxVar.s	Quality flag of NEE variable
QFFluxValue.n	Value of quality flag for _good_ (original) data
TempVar.s	Filled air- or soil temperature variable (degC)
QFTempVar.s	Quality flag of filled temperature variable
QFTempValue.n	Value of temperature quality flag for _good_ (original) data
RadVar.s	Unfilled (original) radiation variable
Lat_deg.n	Latitude in (decimal) degrees
Long_deg.n	Longitude in (decimal) degrees
TimeZone_h.n	Time zone (in hours)
T_ref.n	Reference temperature in Kelvin (degK) used in fLloydTaylor for regressing Flux and Temperature
Suffix.s	String suffix needed for different processing setups on the same dataset (for explanations see below)
debug.l	List with debugging control (passed also to sRegrE0fromShortTerm. useLocaltime.b see details on solar vs local time
parsE0Regression	list with further parameters passed down to sRegrE0fromShortTerm and fRegrE0fromShortTerm, such as TempRange.n

Details

Description of newly generated variables with partitioning results:

• PotRad - Potential radiation
  

• FP_NEEnight - Good (original) NEE nighttime fluxes used for flux partitioning
  

• FP_Temp - Good (original) temperature measurements used for flux partitioning
  

• E_0 - Estimated temperature sensitivity
  

• R_ref - Estimated reference respiration
  

• Reco - Estimated ecosystem respiration
  

• GPP_f - Estimated gross primary production
  

Background

This partitioning is based on the regression of nighttime respiration with temperature using the Lloyd-Taylor-Function fLloydTaylor. First the temperature sensitivity E_0 is estimated from short term data, see sRegrE0fromShortTerm. Next the reference temperature R_ref is estimated for successive periods throughout the whole dataset (see sRegrRref. These estimates are then used to calculate the respiration during daytime and nighttime and with this GPP. Attention: Gap filling of the net ecosystem fluxes (NEE) and temperature measurements (Tair or Tsoil) is required prior to the partitioning!

Selection of daytime data based on solar time

The respiration-temperature regression is very sensitive to the selection of night- and daytime data. Nighttime is selected by a combined threshold of current solar radiation and potential radiation. The current implementation calculates potential radiation based on exact solar time, based on latitude and longitude. (see fCalcPotRadiation ) Therefore it might differ from implementations that use local winter clock time instead.

Different processing setups on the same dataset

Attention: When processing the same site data set with different setups for the gap filling or flux partitioning (e.g. due to different ustar filters), a string suffix is needed! This suffix is added to the result column names to distinguish the results of the different setups. If a Suffix.s is provided and if the defaults for FluxVar.s and QFFluxVar.s are used, the Suffix.s will be added to their variable names (e.g. 'NEE_f' will be renamed to 'NEE_WithUstar_f' and 'NEE_fqc' to 'NEE_WithUstar_fqc' for the Suffix.s='WithUstar'). Currently, this works only with defaults of FluxVar.s='NEE_f' and QFFluxVar.s='NEE_fqc'.

Value

Flux partitioning results (see variables in details) in sTEMP data frame (with renamed columns). On success, return value is NULL. On failure an integer scalar error code is returned: -111 if regression of E_0 failed due to insufficient relationship in the data.

Author(s)

AMM,TW (Department for Biogeochemical Integration at MPI-BGC, Jena, Germany)

References

Reichstein M, Falge E, Baldocchi D et al. (2005) On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm. Global Change Biology, 11, 1424-1439.

REddyProc documentation built on May 2, 2019, 5:19 p.m.