1  Gather STRI data

1.1 Libraries

The packages we utilize are dplyr(Wickham et al. 2023) and tidyr(Wickham, Vaughan, and Girlich 2024) for data wrangling, ggplot2(Wickham 2016) for data visualization and readxl(Wickham and Bryan 2023) for reading excel files.

setwd("C:/Users/Vicente/repo/ClimatePanama/ClimatePanamaBook")
library(dplyr)
library(ggplot2)
library(readxl)
library(tidyr)

1.2 Data

The stri_met_stations.csv contains links and metadata for all sites listed in https://biogeodb.stri.si.edu/physical_monitoring/

DIRTOMET<-"../tables/stri_met_stations.csv"
met_stations<- read.csv(DIRTOMET)
all_stations<-unique(met_stations$alias)
print(all_stations)

 [1] "CELESTINO"  "AVA"        "BCICLEAR"   "BCIELECT"   "BOCAS"     
 [6] "PCULEBRA"   "FORTUNA"    "GALETASTRI" "PNM"        "SANBLAS"   
[11] "SHERMAN"   

The following function calculates monthly precipitation given the standard format of the STRI data, by default it only includes values label as “good”

monthly_precip<- function(file, include=("good"),site_name){
    sheets<-read.csv(file) %>% filter(chk_note %in% include) 
    if (nrow(sheets)<30){
        stop("Error: not enough days with good measurments")}
    daily_precip<-sheets%>%group_by(date)%>%summarize(daily_precip= sum(ra, na.rm = T))
    monthly_precip <- daily_precip %>% mutate(month_year = as.Date(paste0(format(as.Date(date, "%d/%m/%Y"), "%Y-%m"), "-01")))%>%
    group_by(month_year) %>% summarize(monthly_precip = sum(daily_precip, na.rm = T))
    monthly_precip$site<-site_name
    return(monthly_precip)
}

The download_met function download the file , unzips it and then erases the zip file.

download_met<- function(site,csv){
    link<-csv[csv$alias==site,'link']
    destfile<-paste0("../data_ground/met_data/STRI_data/",csv[csv$alias==site,'alias'],".zip")
    download.file(link,destfile =destfile  , mode = "wb")
    unzip(destfile, exdir = paste0("../data_ground/met_data/STRI_data/",csv[csv$alias==site,'alias']))
    file.remove(destfile)
}

1.3 Download data from sites with regular format

This sites only have a single csv file. Data sets provided by the Physical Monitoring Program of the Smithsonian Tropical Research Institute(Smithsonian Tropical Research Institute 2023).

sites <- c("PCULEBRA", "FORTUNA", "PNM", "SHERMAN", "AVA", "BCICLEAR", "BCIELECT", "BOCAS")
for (site in sites) {
  site_dir <- paste0("../data_ground/met_data/STRI_data/", met_stations[met_stations$alias == site, 'alias'])
  
  if (!dir.exists(site_dir)) {
    download_met(site, met_stations)
  }
  
  files_in_site <- list.files(site_dir, recursive = TRUE, pattern = ".csv", full.names = TRUE)
 
  This_site <- tryCatch(monthly_precip(files_in_site[1], include = c("good", "adjusted"), site_name = met_stations[met_stations$alias == site, 'alias']),
                        error = function(e) NULL)
  
  if (!is.null(This_site)) {
    write.csv(This_site, paste0(site_dir, '.csv'))
  }
  rm(This_site)
}

1.4 Download data from sites with non-regular format

1.4.1 Galeta station special case

GALETASTRI and GALETALAB share the same values in the first years, then in 2002 the values are different GALETALAB stops in 2007.

We accept the adjusted values for GALETASTRI which is the tower station, the metadata inform us that the ACP close-by station Limon bay was used for the gap filling.

site= "GALETASTRI"

download_met(site, met_stations)

[1] TRUE

files_in_site<-list.files(paste0("../data_ground/met_data/STRI_data/",met_stations[met_stations$alias==site,'alias']),recursive = T, pattern = ".csv", full.names = T)

This_site<-monthly_precip(files_in_site[2],include=c("good","adjusted"), site_name = met_stations[met_stations$alias==site,'alias'])

if (exists('This_site')) {
    write.csv(This_site, paste0('../data_ground/met_data/STRI_data/', met_stations[met_stations$alias == site, 'alias'], '.csv'))
}
remove(This_site)

GALETALAB has a second sensor in the laboratory. It has no notes, and does not follow the format of the other stations

GALETALAB<- read.csv(files_in_site[1])%>%mutate(month_year = as.Date(paste0(format(as.Date(date.time..yyyy.mm.dd.hh.mm.ss., "%Y-%m-%d"), "%Y-%m"), "-01")))%>%
group_by(month_year)%>% summarise(monthly_precip = sum(ra.mm., na.rm = T))%>% mutate(site="GALETALAB")

if (exists('GALETALAB')) {
    write.csv(GALETALAB, paste0('../data_ground/met_data/STRI_data/', "GALETALAB", '.csv'))
}

remove(GALETALAB)

1.4.2 San Blas station special case

The san blas station does not follow the standard format and only provides with the summaries.

site= "SANBLAS"
download.file(met_stations[met_stations$alias==site,'link'],destfile =paste0("../data_ground/met_data/STRI_data/",met_stations[met_stations$alias==site,'alias'],".xlsx")  , mode = "wb")

file<- read_excel(paste0("../data_ground/met_data/STRI_data/",met_stations[met_stations$alias==site,'alias'],".xlsx"), sheet = 4)
file<-file[2:9,1:13]
SANBLAS<- file%>%rename('01'='...2','02'= '...3','03'= '...4','04'= '...5','05'= '...6','06'= '...7','07'= '...8','08'= '...9','09'= '...10','10'= '...11','11'= '...12','12'= '...13')%>%
pivot_longer(cols = c('01','02','03','04','05','06','07','08','09','10','11','12'), names_to = 'month', values_to = 'monthly_precip')%>%mutate(month_year = as.Date(paste0(format(as.Date(paste0('01/',month,'/2000'), "%d/%m/%Y"), "%Y-%m"), "-01")))%>%
select(month_year, monthly_precip)%>%mutate(site="SANBLAS")
if (exists('SANBLAS')) {
    write.csv(SANBLAS, paste0('../data_ground/met_data/STRI_data/', "SANBLAS", '.csv'))
}
remove(SANBLAS)
file.remove(paste0("../data_ground/met_data/STRI_data/",met_stations[met_stations$alias==site,'alias'],".xlsx"))

[1] TRUE

1.4.3 Agua salud station special case

This site has 2 sensors within meters of each other

site= "CELESTINO"

download_met(site, met_stations)

[1] TRUE

files_in_site<-list.files(paste0("../data_ground/met_data/STRI_data/",met_stations[met_stations$alias==site,'alias']),
                                 recursive = T, pattern = ".csv", full.names = T)

This_siteN=monthly_precip(files_in_site[1],include=c("good","adjusted"), site_name = met_stations[met_stations$alias==site,'alias'])

This_siteS=monthly_precip(files_in_site[2],include=c("good","adjusted"), site_name = met_stations[met_stations$alias==site,'alias'])

plot(This_siteN$monthly_precip, This_siteS$monthly_precip)

Merge both sensors and average them.

CELESTINO= merge(This_siteN, This_siteS, by = "month_year", all = TRUE)%>%mutate(monthly_precip = rowMeans(.[, c("monthly_precip.x", "monthly_precip.y")], na.rm = TRUE))%>%
select(month_year, monthly_precip)%>%mutate(site="CELESTINO")
if (exists('CELESTINO')) {
    write.csv(CELESTINO, paste0('../data_ground/met_data/STRI_data/', met_stations[met_stations$alias == site, 'alias'], '.csv'))
}
remove(CELESTINO)

1.5 Combine all the downloaded sites

all <- data.frame()
for (site in all_stations) {
  thecsv <- file.path('../data_ground/met_data/STRI_data', paste0(site, '.csv'))
  if (file.exists(thecsv)) {
    append_stations <- read.csv(thecsv)
    all <- rbind(all, append_stations)
  }
}
all<- all%>%select(month_year, monthly_precip, site)%>%mutate(month_year = as.Date(month_year))

Write the STRI_monthltPrecip file

write.csv(all, paste0('../data_ground/met_data/STRI_data/', "STRI_monthlyPrecip", '.csv'))

Smithsonian Tropical Research Institute, Steve Paton. 2023. Data Sets Provided by the Physical Monitoring Program of the Smithsonian Tropical Research Institute.

Wickham, Hadley. 2016. Ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. https://ggplot2.tidyverse.org.

Wickham, Hadley, and Jennifer Bryan. 2023. Readxl: Read Excel Files.

Wickham, Hadley, Romain François, Lionel Henry, Kirill Müller, and Davis Vaughan. 2023. Dplyr: A Grammar of Data Manipulation. https://dplyr.tidyverse.org.

Wickham, Hadley, Davis Vaughan, and Maximilian Girlich. 2024. Tidyr: Tidy Messy Data. https://tidyr.tidyverse.org.