mv subregion postprocessing to seperate script

Author: andybell

data/data_postprocessing.Rmd

@@ -134,94 +134,94 @@ write.csv(df_dsa_legal, 'data/dsa_legal/dsa_legal.csv', row.names = FALSE)
 
 ```{r subregions}
 
-####################################################################
-# data loading functions for processsing EE geojson for subregions
-# note this data does not include any crop information
-# All non-ag area should be masked out in EE
-
-# parse model and year from filename. Should be similiar to disalexi-DSAsubregions-2015.geojson
-get_filename_info <- function(file){
-  # get the basename from the file path
-  b <- basename(file)
-  
-  # split filename into parts
-  b_ex <- strsplit(b, "\\.") 
-  n <- b_ex[[1]][1] # name without extension
-  
-  n_split <- strsplit(n, "-")
-  m <- n_split[[1]][1]
-  yr <- n_split[[1]][3]
-  return(c(m, yr))
-}
-
-# calculate acre-feet per month
-acre_feet_per_month <- function(df, wy, month, reducer_size){
-  #num_days_in_month
-  numdays <- num_days_in_month_csv(wy, month)
-  
-  # name of the field with the month's mean
-  mean_field <- paste(month, '_', 'mean', sep='')
-  
-  # name of the field with the month's count
-  count_field <- paste(month, '_', 'count', sep='')
-  
-  # output name for the field with the month's acre-feet total
-  output_fieldname <- paste(month, '_', 'ACREFT', sep='')
-  
-  # add field that calculates acre-feet for month using mean, count, number of days in month and reducer
-  d <- df %>% mutate_(xyz = interp(~acre_feet(m, c, numdays, reducer_size), m=as.name(mean_field), c=as.name(count_field)))
-  names(d)[names(d) == "xyz"] <- output_fieldname
-  return(d)
-} 
-
-# read in geojson export and add fields parsed from filename and calculate monthly and wy acre-feet
-subregions_add_fields <- function(geojson, reducer_size){
-  model_year <- get_filename_info(geojson)
-  d <- st_read(geojson) # read geojson file to dataframe
-  d$model <- model_year[1] # add name of model parsed from filename
-  d$wateryear <- model_year[2] # add water year parsed from filename
-  d$source <- geojson # add the filepath as the source
-  
-  # calculate island/region area from geojson (units in square meters)
-  d$AREA_m <- st_area(d)
-  
-  #add fields for each month's acre-feet
-  for(m in month.abb){
-    d<-acre_feet_per_month(d, model_year[2], toupper(m), reducer_size)
-  }
-  
-  # calculate water year total acre-feet by adding all the monthly acre-feet columns
-  d$WY_ACREFT <- d$OCT_ACREFT + d$NOV_ACREFT + d$DEC_ACREFT + d$JAN_ACREFT + 
-    d$FEB_ACREFT + d$MAR_ACREFT + d$APR_ACREFT + d$MAY_ACREFT + d$JUN_ACREFT + 
-    d$JUL_ACREFT + d$AUG_ACREFT + d$SEP_ACREFT
-  return(d)
-}
-
-num_days_in_month_csv <- function(wy, month){
-  # load the months
-  month <- toupper(month)
-  months <- read.csv('lookups/months.csv', stringsAsFactors=FALSE)
-  wateryr <- mutate(months, cmb=paste(months$WaterYear, months$Month))
-  days <- wateryr$NumberDays[match(paste(wy, month), wateryr$cmb)]
-  return(days)
-}
-
-####################################################################################
-####################################################################################
-
-list_of_files <- list.files(path="data/subregions", pattern=".geojson", full.names=TRUE)
-listofdfs <- list() #Create a list in which you intend to save your df's.
-
-# for loop to load in all the raw json files to single data frame
-for (i in 1:length(list_of_files)) {
-  print(list_of_files[i])
-  g <- subregions_add_fields(list_of_files[i], 30)
-  listofdfs[[i]]<-g
-}
-df_subregions <- ldply(listofdfs, data.frame) # make into one dataframe 
-
-saveRDS(df_subregions, file="data/subregions/subregions.rds")
-write.csv(df_subregions, file="data/subregions/subregions.csv", row.names = FALSE)
+# ####################################################################
+# # data loading functions for processsing EE geojson for subregions
+# # note this data does not include any crop information
+# # All non-ag area should be masked out in EE
+# 
+# # parse model and year from filename. Should be similiar to disalexi-DSAsubregions-2015.geojson
+# get_filename_info <- function(file){
+#   # get the basename from the file path
+#   b <- basename(file)
+#   
+#   # split filename into parts
+#   b_ex <- strsplit(b, "\\.") 
+#   n <- b_ex[[1]][1] # name without extension
+#   
+#   n_split <- strsplit(n, "-")
+#   m <- n_split[[1]][1]
+#   yr <- n_split[[1]][3]
+#   return(c(m, yr))
+# }
+# 
+# # calculate acre-feet per month
+# acre_feet_per_month <- function(df, wy, month, reducer_size){
+#   #num_days_in_month
+#   numdays <- num_days_in_month_csv(wy, month)
+#   
+#   # name of the field with the month's mean
+#   mean_field <- paste(month, '_', 'mean', sep='')
+#   
+#   # name of the field with the month's count
+#   count_field <- paste(month, '_', 'count', sep='')
+#   
+#   # output name for the field with the month's acre-feet total
+#   output_fieldname <- paste(month, '_', 'ACREFT', sep='')
+#   
+#   # add field that calculates acre-feet for month using mean, count, number of days in month and reducer
+#   d <- df %>% mutate_(xyz = interp(~acre_feet(m, c, numdays, reducer_size), m=as.name(mean_field), c=as.name(count_field)))
+#   names(d)[names(d) == "xyz"] <- output_fieldname
+#   return(d)
+# } 
+# 
+# # read in geojson export and add fields parsed from filename and calculate monthly and wy acre-feet
+# subregions_add_fields <- function(geojson, reducer_size){
+#   model_year <- get_filename_info(geojson)
+#   d <- st_read(geojson) # read geojson file to dataframe
+#   d$model <- model_year[1] # add name of model parsed from filename
+#   d$wateryear <- model_year[2] # add water year parsed from filename
+#   d$source <- geojson # add the filepath as the source
+#   
+#   # calculate island/region area from geojson (units in square meters)
+#   d$AREA_m <- st_area(d)
+#   
+#   #add fields for each month's acre-feet
+#   for(m in month.abb){
+#     d<-acre_feet_per_month(d, model_year[2], toupper(m), reducer_size)
+#   }
+#   
+#   # calculate water year total acre-feet by adding all the monthly acre-feet columns
+#   d$WY_ACREFT <- d$OCT_ACREFT + d$NOV_ACREFT + d$DEC_ACREFT + d$JAN_ACREFT + 
+#     d$FEB_ACREFT + d$MAR_ACREFT + d$APR_ACREFT + d$MAY_ACREFT + d$JUN_ACREFT + 
+#     d$JUL_ACREFT + d$AUG_ACREFT + d$SEP_ACREFT
+#   return(d)
+# }
+# 
+# num_days_in_month_csv <- function(wy, month){
+#   # load the months
+#   month <- toupper(month)
+#   months <- read.csv('lookups/months.csv', stringsAsFactors=FALSE)
+#   wateryr <- mutate(months, cmb=paste(months$WaterYear, months$Month))
+#   days <- wateryr$NumberDays[match(paste(wy, month), wateryr$cmb)]
+#   return(days)
+# }
+# 
+# ####################################################################################
+# ####################################################################################
+# 
+# list_of_files <- list.files(path="data/subregions", pattern=".geojson", full.names=TRUE)
+# listofdfs <- list() #Create a list in which you intend to save your df's.
+# 
+# # for loop to load in all the raw json files to single data frame
+# for (i in 1:length(list_of_files)) {
+#   print(list_of_files[i])
+#   g <- subregions_add_fields(list_of_files[i], 30)
+#   listofdfs[[i]]<-g
+# }
+# df_subregions <- ldply(listofdfs, data.frame) # make into one dataframe 
+# 
+# saveRDS(df_subregions, file="data/subregions/subregions.rds")
+# write.csv(df_subregions, file="data/subregions/subregions.csv", row.names = FALSE)
 
 ```
 

data/subregions/DSAsubregions_postprocessing.R

@@ -1,4 +1,7 @@
-
+library(sf)
+library(plyr)
+library(tidyverse)
+library(lazyeval)
 
 ####################################################################
 # data loading functions for processsing EE geojson for subregions
@@ -40,6 +43,15 @@ acre_feet_per_month <- function(df, wy, month, reducer_size){
   return(d)
 } 
 
+# calculate acre-feet from monthly avg daily ET
+acre_feet <- function(mean_et, cell_count, number_days, reducer_size){
+  # Crop_acre_feet = (count) * (reducer pixel size) ^2 * (sq m to acres) * (mean daily ET) /10* (mm to feet) * (num days in month)
+  mm2ft <- 0.00328084
+  sqm2acres <- 0.000247105
+  crop_acft <- cell_count * reducer_size^2 * sqm2acres * mean_et / 10 * mm2ft * number_days 
+  return(crop_acft)
+}
+
 # read in geojson export and add fields parsed from filename and calculate monthly and wy acre-feet
 subregions_add_fields <- function(geojson, reducer_size){
   model_year <- get_filename_info(geojson)
@@ -84,7 +96,10 @@ for (i in 1:length(list_of_files)) {
   g <- subregions_add_fields(list_of_files[i], 30)
   listofdfs[[i]]<-g
 }
-df_subregions <- ldply(listofdfs, data.frame) # make into one dataframe 
+subregions <- ldply(listofdfs, data.frame) # make into one dataframe 
+
+# drop geo and ID columns
+df_subregions <- subregions %>% select(-c(id, geometry))
 
 saveRDS(df_subregions, file="data/subregions/subregions.rds")
 write.csv(df_subregions, file="data/subregions/subregions.csv", row.names = FALSE)