R/calc_lwnet.R

In aemon-j/rHeatFluxAnalyzer: Calculate heat fluxes from standart buoy data

#' Calculate net long wave radiation
#'
#' @name calc_lwnet
#' @param time Vector of times as POSIXct
#' @param lat latitude for which to calculate clear sky radiation
#' @param press Air pressure in (hPa) if available
#' @param ta Air temperature in (degrees C) if available
#' @param rh relative humidity in percentage if available
#' @param sw observed showrt wave radiation
#' @param ts Water surface temperature in degrees C
#' @return A data.frame with lw, LWo, and lwnet
#' @export
#' @references Read et al. 2012


calc_lwnet <- function(time,lat,press,ta,rh,sw,ts){



  clrSW <- clearSkySW(time,lat,press,ta,rh*.01)

  clf <-  1-sw/clrSW



  # impose limits

  ltI <- clf<0

  clf[ltI] <- 0

  gtI <- clf>1

  clf[gtI] <- 1



  # Create daily averages of cloud fraction for each day
  b <- format(time,"%Y-%m-%d")

  daily_av <- aggregate(clf,by=list(b),mean, na.rm=TRUE)

  # Replace clf on times where sw == 0, by the daily average
  c2=merge(list("Group.1"=b),daily_av,by="Group.1")

  clf[sw == 0] <- c2[sw == 0,2]



  # if still NaN, assume cloud cover is zero (at night, same as Read et al. 2012)

  clf[is.nan(clf)]  <-  0



  #month <- datevec(Jday)

  month <- as.numeric(format(time,"%m"))

  lw <- Mdl_LW(ta,rh*0.01,clf,month)

  Tk <- ts + 273.13     # sample Ts to match LW

  emiss <- 0.972

  S_B <- 5.67E-8

  LWo <- S_B*emiss*Tk^4

  lwnet <- lw-LWo



  return(data.frame(lw = lw,LWo = LWo,lwnet = lwnet))



}