Create new column populated with existing column name(s) conditional on values in existing column(s)

Question

I have the following dataset:
crab_traps <- tibble(trap_id = 1:5, logger_1 = c(2, 3, 4, 5, 6), logger_2 = c(9, 7, 5, 3, 1), logger_3 = c(6, 5, 1, 7, 8)
trap_id identifies a specific crab trap and each logger_* column is the distance to a water quality logger.
I want to create a new column (preferably using dplyr::mutate) to figure out which logger is closest to each trap.
Hopefully the new column would be something like closest_logger <- c("logger_1", "logger_1", "logger_3", "logger_2", "logger_2") (using this example data)
I have tried crab_traps <- crab_traps %>% rowwise() %>% mutate(closest_logger = min(logger_1, logger_2, logger_3)) which gives me the smallest measurement, but I want to get the name of the column that the smallest measurement came from.
Any help would be wonderful!

Answer 1

The most efficient solution would be using max.col :

crab_traps$closest_logger <- names(crab_traps)[max.col(-crab_traps[-1]) + 1]

---

If you want a dplyr solution you can try c_across with rowwise.

library(dplyr)

crab_traps %>%
  rowwise() %>%
  mutate(closest_logger = names(.)[which.min(c_across(starts_with('logger')))+1])

#  trap_id logger_1 logger_2 logger_3 closest_logger
#    <int>    <dbl>    <dbl>    <dbl> <chr>         
#1       1        2        9        6 logger_1      
#2       2        3        7        5 logger_1      
#3       3        4        5        1 logger_3      
#4       4        5        3        7 logger_2      
#5       5        6        1        8 logger_2   

In both the cases, we add +1 because we are ignoring the first column while calculating minimum.