New next steps! about daynight-q10 HOT 2 OPEN

# we start with `dat` which holds timestamp, location, soil respiration
# e.g. one site, one week, hourly data = 24 * 7 = 168 rows
dat %>%
  # figure out what sm file each observation needs to draw from
  mutate(filename = compute_filename(timestamp)) %>%  # takes 168 unique timestamps -> 3 unique filenames
  # we don't want to duplicate effort, so `distinct()`
  distinct(lat, lon, filename) %>%   # after this we have only three rows!
  # for each filename, extract all the lat/lon s.m. data we need
  group_by(filename) %>%
  mutate(sm = get_sm_data(lat, lon, filename)) %>%  # still have three rows
  # after this right_join, we're going to end up with EXACTLY the same number of rows we started with!
  right_join(dat, by = c("filename", "lat", "lon")) ->
  dat