R/plot_ESS.R

In MariekeDirk/WINS50_Lidar: Lidar data analysis for the WINS50 project

#'Plot similarity's
#'@description Plots S3 vs S2 and S2 vs S4.
#'@param SR similarities as output from \link{get_mean_S}.
#'@param df data input from \link{get_mean_S} to add statistics to the plot.
#'@param main header of the plot
#'@author Marieke Dirksen
#'@importFrom magrittr %>%
#'@importFrom rlang .data
#'@export
plot_ESS<-function(SR,df,main){
  #subset only the days which are used in the ESS analysis
  df$day<-as.Date(df$time)
  obs_per_day<-df %>% dplyr::group_by(.data$day) %>% dplyr::count()
  days_in<-obs_per_day[which(obs_per_day$n==144),]$day
  df<-df[df$day %in% days_in,]

  #Some numbers for the plotting routine
  n.obs<-length(unique(df$day))
  t.min<-min(as.Date(df$time),na.rm=TRUE)
  t.max<-max(as.Date(df$time),na.rm=TRUE)

  #Label for the plotting routine
  lab <- SR %>%
    dplyr::summarise(
      S3 = 0.1,
      S2 = mean(.data$S2),
      lab = paste0("nr.= ",n.obs,"\nmin= ",t.min,"\nmax= ",t.max)
    )

  p1<-ggplot2::ggplot(SR,ggplot2::aes(.data$S3,.data$S2))+  #ggtitle(main)+
    ggplot2::geom_point()+
    ggplot2::theme_bw()+
    ggplot2::theme(aspect.ratio = 1)+
    ggplot2::scale_y_log10(breaks = scales::trans_breaks("log10", function(x) 10^.data$x),
                  labels = scales::trans_format("log10", scales::math_format(10^.data$x)))+
    ggplot2::scale_x_log10(breaks = scales::trans_breaks("log10", function(x) 10^.data$x),
                  labels = scales::trans_format("log10", scales::math_format(10^.data$x)))+
    ggplot2::geom_text(data = lab,ggplot2::aes(label=lab), vjust = "top", hjust = "left")+
    ggplot2::coord_fixed()
  p2<-ggplot2::ggplot(SR,ggplot2::aes(.data$S2,.data$S4))+  #ggtitle(main)+
    ggplot2::geom_point()+
    ggplot2::theme_bw()+
    ggplot2::theme(aspect.ratio = 1)+
    ggplot2::scale_y_log10(breaks = scales::trans_breaks("log10", function(x) 10^.data$x),
                  labels = scales::trans_format("log10", scales::math_format(10^.data$x)))+
    ggplot2::scale_x_log10(breaks = scales::trans_breaks("log10", function(x) 10^.data$x),
                  labels = scales::trans_format("log10", scales::math_format(10^.data$x)))


  P<-gridExtra::grid.arrange(p1,p2,top=main,ncol=2)

  return(P)
}

#'Plot similarity's
#'@description Plots S3 vs S2 and S2 vs S4.
#'@param SR similarities as output from \link{get_mean_S}.
#'@param df data input from \link{get_mean_S} to add statistics to the plot.
#'@param main header of the plot
#'@param h_levels number of height levels
#'@author Marieke Dirksen
#'@importFrom magrittr %>%
#'@importFrom rlang .data
#'@export
plot_ESSh<-function(SR,df,h_levels=11,main){
  height_colors<-viridis::viridis(n=h_levels)
  df_hc<-data.frame(h_levels,height_colors)
  names(df_hc)<-c("h","clr")
  #subset only the days which are used in the ESS analysis
  df$day<-as.Date(df$time)
  obs_per_day<-df %>% dplyr::group_by(.data$day) %>% dplyr::count()
  days_in<-obs_per_day[which(obs_per_day$n==(144*h_levels)),]$day
  df<-df[df$day %in% days_in,]

  #Some numbers for the plotting routine
  n.obs<-length(unique(df$day))
  t.min<-min(as.Date(df$time),na.rm=TRUE)
  t.max<-max(as.Date(df$time),na.rm=TRUE)

  #Label for the plotting routine
  lab <- SR %>%
    dplyr::summarise(
      S3 = 0.1,
      S2 = mean(.data$S2),
      lab = paste0("nr.= ",n.obs,"\nmin= ",t.min,"\nmax= ",t.max)
    )
  #order of legend with setkey
  SR<-data.table::data.table(SR)
  SR$h<-as.numeric(SR$h)
  data.table::setkey(SR,"h")

  p1<-ggplot2::ggplot(SR,ggplot2::aes(.data$S3,.data$S2,color=as.factor(h)))+  #ggtitle(main)+
    ggplot2::geom_point()+
    ggplot2::theme_bw()+
    ggplot2::scale_color_manual(name="h [m]",values=height_colors)+
    ggplot2::theme(aspect.ratio = 1)+
    ggplot2::scale_y_log10(breaks = scales::trans_breaks("log10", function(x) 10^x),
                           labels = scales::trans_format("log10", scales::math_format(10^.x)))+
    ggplot2::scale_x_log10(breaks = scales::trans_breaks("log10", function(x) 10^x),
                           labels = scales::trans_format("log10", scales::math_format(10^.x)))+
    #ggplot2::geom_text(data = lab,ggplot2::aes(label=lab), vjust = "top", hjust = "left")+
    ggplot2::coord_fixed()

  p2<-ggplot2::ggplot(SR,ggplot2::aes(.data$S2,.data$S4,color=as.factor(h)))+  #ggtitle(main)+
    ggplot2::geom_point()+
    ggplot2::theme_bw()+
    ggplot2::scale_color_manual(name="h [m]",values=height_colors)+
    ggplot2::theme(aspect.ratio = 1)+
    ggplot2::scale_y_log10(breaks = scales::trans_breaks("log10", function(x) 10^x),
                           labels = scales::trans_format("log10", scales::math_format(10^.x)))+
    ggplot2::scale_x_log10(breaks = scales::trans_breaks("log10", function(x) 10^x),
                           labels = scales::trans_format("log10", scales::math_format(10^.x)))+
    #ggplot2::geom_text(data = lab,ggplot2::aes(label=lab), vjust = "top", hjust = "left")+
    ggplot2::coord_fixed()


  P<-gridExtra::grid.arrange(p1,p2,top=main,ncol=2)

  return(P)
}