rfluxnet: Process fluxnet data

## Known issues
# 2015 patch is not applied. LE, LH will lost some good values.

timename <- c("TIMESTAMP_START", "TIMESTAMP_END")

# except date, all variables were double
# Update 20191216 (FULLSET used, rename variabls)
# SW_OUT_F: SW_OUT
# LW_OUT_F: LW_OUT
# non_QC variables
# "SW_OUT_QC", "LW_OUT_QC", "NETRAD_QC", "USTAR_QC"
vars_val <- c(
  "SW_IN_F", # incoming Shortwave Radiation  | W/m2
  "SW_OUT", # outgoing Shortwave Radiation  | W/m2
  "LW_IN_F", # incoming Longwave Radiation   | W/m2
  "LW_OUT", # outgoing Longwave Radiation   | W/m2
  "NETRAD", # Rn, NET Radiation             | W/m2
  "LE_F_MDS", # Latent heat flux              | W/m2
  "LE_CORR", # LE_F_MDS_QC quality flag for LE_F_MDS, LE_CORR
  "H_F_MDS", # Sensible heat flux            | W/m2
  "H_CORR", # corrected H_F_MDS by energy balance closure correction factor, QC as H_F_MDC_QC
  "G_F_MDS", # Soil heat flux                | W/m2, could be negative
  "PA_F", # Atmospheric pressure          | kPa
  "P_F", # Precipitation                 | mm
  "VPD_F", # Vapor Pressure Deficit        | hPa
  "WS_F", # Wind Speed                    | m/s
  "WD", # Wind Direction                | Decimal degrees
  "USTAR", # Friction velocity             | m/s
  "RH", # Relative Humidity             | %
  "PPFD_IN", # incoming Photosynthetic photon flux density | W/m2
  "CO2_F_MDS", # CO2 mole fraction             | umol CO2/mol
  "TA_F", # air temperature               | degC
  "TS_F_MDS_1", # Soil temperature, '#' increases with the depth, 1 is shallowest | degC
  "SWC_F_MDS_1", # Soil water content            | %
  "NEE_VUT_REF", # Net Ecosystem Exchange, using Variable Ustar Threshold (VUT) | umol CO2/mol
  # for each year, reference selected on the basis of the model
  # efficiency (MEF). The MEF analysis is repeated for each time aggregation
  "RECO_DT_VUT_REF", # daytime partitioning Ecosystem Respiration          | umol CO2/mol
  "RECO_NT_VUT_REF", # nighttime partitioning Ecosystem Respiration        | umol CO2/mol
  "GPP_DT_VUT_REF", # daytime partitioning Gross Primary Production,      | umol CO2/mol
  "GPP_NT_VUT_REF" # nighttime partitioning Gross Primary Production,    | umol CO2/mol
)

################################################################################
vars_QC <- paste0(vars_val, "_QC")
vars_QC[match(c("LE_CORR_QC", "H_CORR_QC"), vars_QC)] <- c("LE_F_MDS_QC", "H_F_MDS_QC")
vars_QC[24:27] <- vars_QC[23] # RECO, GPP share the same QC with NEE

vars_all <- c(vars_val, vars_QC) # unique

# NA VALUES ---------------------------------------------------------------
#
# variables x < 0 set to be NA, values can't be negative
var_no_negative <- c(
  "SW_IN_F", "SW_OUT_F", "LW_IN_F", "LW_OUT_F",
  "PA_F", "PPFD_IN",
  "P_F", "VPD_F", "WS_F", "WD", "USTAR", "CO2_F_MDS", "SWC_F_MDS_1",
  "RECO_DT_VUT_REF", "RECO_NT_VUT_REF"
)

# vars_0na <- paste(c('SW_', 'LW_', 'PPFD_', 'PA_'),
#                   collapse = "|") %>% {vars_val[grep(., vars_val)]} #, 'GPP_'
# variables x < 0 set to be NA, except vars_0na variables
# fixed 17 Dec' 2017
# vars_noNeg <- vars_val[-grep("TA_F|TS_F|NEE|G_F_MDS|GPP|LE|H|NETRAD", vars_val)] %>% setdiff(vars_0na)
# all variables: vars_noNeg + vars_0na + "TA_F|TS_F|NEE|G_F_MDS"
#
# else variables (TA_F, TS_F, NEE) x < -9000 set to be NA
################################################################################