construct_data_FN15: Format Fluxnet2015 data.frame
In bgctw/etpart: Partitioning evapotranspiration into transpiration and evaporation

Construct data.frame formatted for partitioning from Fluxnet2015 format https://fluxnet.org/data/fluxnet2015-dataset/

construct_data_FN15(
  data,
  GPP = "GPP_NT_VUT_USTAR50",
  GPP_sd = "NEE_VUT_USTAR50_RANDUNC",
  Ca = "CO2_F_MDS",
  H = "H_F_MDS",
  LE = "LE_F_MDS",
  LE_QC = "LE_F_MDS_QC",
  LE_sd = "LE_RANDUNC",
  NEE_QC = "NEE_VUT_USTAR50_QC",
  NIGHT = "NIGHT",
  precip = "P",
  Pair = "PA",
  Q = "PPFD_IN",
  Rg = "SW_IN_F_MDS",
  Rg_pot = "SW_IN_POT",
  Tair = "TA_F_MDS",
  ustar = "USTAR",
  VPD = "VPD_F_MDS",
  u = "WS_F"
)

`data`	data frame with column names given by the following arguments.
`GPP`	photosynthesis data (umol CO2 m-2 s-1).
`GPP_sd`	photosynthesis uncertainties (umol CO2 m-2 s-1).
`Ca`	atmospheric CO2 concentration (umol Co2 mol air-1).
`H`	sensible heat flux (W m-2).
`LE`	Latent heat flux (W m-2)
`LE_QC`	quality flag for LE (0 measured, 1 good quality, ...)
`LE_sd`	uncertainty of LE
`NEE_QC`	quality flag for NEE (0 measured, 1 good quality, ...)
`NIGHT`	Flag indicating nighttime interval (1 for night, zero otherwise)
`precip`	P used if measured (mm per dataset resolution: either hour or half-hour)
`Pair`	atmospheric pressure (kPa).
`Q`	photosynthetic active radiation (umol m-2 s-1).
`Rg`	Shortwave radiation, incoming (W/m2)
`Rg_pot`	potential Shortwave radiation, incoming (W/m2)
`Tair`	air temperature (deg C).
`ustar`	wind friction velocity (m s-1).
`VPD`	vapor pressure deficit (hPa).
`u`	wind velocity (m s-1).

data.frame with colums as given by argument names and

`timestamp`	POSIXct time: end of averaging period
`ET`	evapotranspiration computed from LE in mm/hour
`ET_sd`	uncertainty of ET
`RH`	relative humidity (1/1) computed from VPD and Tair
`isnight`	boolean TRUE if its nighttime computed from NIGHT
`qualtiy_flag`	(NEE_QC == 0) & (LE_QC == 0) & is.finite(ET) & is.finite(GPP)