R/01_weather_gee.R

In energyandcleanair/creadeweather: Deweathering of air pollution

gee.get_weather <- function(){
  
  
  #-------------------------
  # Preparing weather data
  #-------------------------
  csv_pattern = Sys.glob("data/01_weather/output/glas_gadm1/*.csv")
  csv_files = lapply(csv_pattern, read.csv)
  
  csv_tbls <- lapply(csv_files, function(x){
    tryCatch({
      new_tibble(x) %>% dplyr::select(id, GID_1, NAME_1, Psurf_f_inst, Qair_f_inst, Rainf_tavg, Tair_f_inst, Wind_f_inst)
    },error=function(cond){
      print(cond)
      NA
    })
  })
  
  print(paste("Wrong tibbles: [",length(csv_tbls[is.na(csv_tbls)]),"/", length(csv_tbls),"]"))
  csv_tbls <- csv_tbls[!is.na(csv_tbls)]
  
  weath_3h_gadm <- do.call(bind_rows, csv_tbls)
  weath_3h_gadm <- weath_3h_gadm %>% mutate(date=as.POSIXct(id, format="A%Y%m%d_%H%M", tz="UTC"))
  
  # Average per day
  weath_1d_gadm <- weath_3h_gadm %>% mutate(date=lubridate::date(date)) %>%
    group_by(GID_1, NAME_1, date) %>%
    summarise(
      Psurf_f_inst=mean(Psurf_f_inst, na.rm=T),
      Qair_f_inst=mean(Qair_f_inst, na.rm=T),
      Rainf_tavg=mean(Psurf_f_inst, na.rm=T),
      Tair_f_inst=mean(Tair_f_inst, na.rm=T),
      Wind_f_inst=mean(Wind_f_inst, na.rm=T)) %>%
    rename(gadm1_id=GID_1, gadm1_name=NAME_1) %>% 
    group_by(gadm1_id, gadm1_name) %>%
    tidyr::nest() %>%
    rename(weather=data)
  
  
  result_path <- file.path('data','01_weather', 'output','gadm1_weather_gee.RDS')
  saveRDS(weath_1d_gadm, result_path)
}