I have a list of dataframes for years 2025 to 2050. These are for 3 variables: temperature, precpitation and RH. I want to combine (cbind) all 3 variables for each year, rename their columns and then save as csv. E.g.
> combined_2025<-cbind(tasmean_2025,prmean_2025,rhmean_2025)
> colnames(combined_2025) <- c("tasmean","pr","rh")
> write.csv(combined_2025,"D:/...../combined_annual_mean_245_2025.csv")
Now, I can easily do this for one year, but I have 25 years to work on. I was thinking of a for loop, but that isn't working for cbind. Please suggest a more efficient way.
1) We need a reproducible example to really know what you have and to test the code but here is some code that you can fix/modify accordingly. I tested it by replacing the first two lines (defining years and fmt) with years <- 2025; tasmean_2025 <- pr_2025 <- rh_2025 <- 1:3; fmt <- "combined_%d.csv" .
years <- 2025:2050
fmt <- "D:/combined_annual_mean_245_%d.csv" # change this
Names <- c("tasmean", "pr", "rh")
for (y in years) {
Names_ <- paste0(Names, "_", y)
# DF <- do.call("data.frame", L[Names_]) # if data is in list L
DF <- do.call("cbind", mget(Names_)) # if data is loose in workspace
names(DF) <- Names
file <- sprintf(fmt, y)
cat("writing", file, "\n")
write.csv(DF, file)
}
2) Another possibility if you would like to consider other ways of storing the result is to combine everything into a single data frame with an added year column and then write that out as a single csv.
library(dplyr)
library(purrr)
years <- 2025:2050
Names <- c("tasmean", "pr", "rh")
make_df <- function(year) {
Names_ <- paste0(Names, "_", y)
# DF <- do.call("data.frame", L[Names_]) # if data is in list L
DF <- do.call("data.frame", mget(Names_, .GlobalEnv)) # if data is loose in workspace
names(DF) <- Names
DF
}
DF <- map(years, make_df) %>% set_names(years) %>% bind_rows(.id = "year")
write.csv(DF, "combined.csv")
If at some later time we wanted to split it up we could still do that using R or we could use the command line miller utility which is available on all platforms that support R.
mlr --csv --from combined.csv split -g year