batch-StreamCat-download-by-state.R

#FTP location
ftpdir  <- 'ftp://newftp.epa.gov/EPADataCommons/ORD/NHDPlusLandscapeAttributes/StreamCat/States/'
#Your output directory
out_dir <- 'D:/Lixo/'
#Desired table (change to name of desired table)
tables <- c('NLCD2001RipBuf100_CA','Kffact_CA','ImperviousSurfaces2001_CA')
#Get URL, split returned list, select out only desired tables by name ('table' above)
library(RCurl)
url_list <- getURL(ftpdir, dirlistonly = TRUE)
url_list <- strsplit(url_list, split = '\r\n')[[1]]
url_list <- url_list[grep(paste(tables,collapse="|"), url_list)]

#Loop through files on FTP, download, and write out as .csv
for(i in 1:length(url_list)){
  print(url_list[i])
  temp <- tempfile()
  download.file(paste0(ftpdir, url_list[i]), temp)
  #replace .zip with .csv in file name
  csv_file <- gsub('.zip', '.csv', url_list[i])
  #read in csv
  tmp_metric <- read.csv(unz(temp, csv_file))
  print(head(tmp_metric))
  write.csv(tmp_metric, paste0(out_dir, csv_file), row.names = F)
}