Adjust datetime x-axis to shown year as a sec axis below de main axis

Question

I have the following data example:

structure(list(cycle_rounded = structure(c(1604188800, 1606780800, 
1609459200, 1612137600, 1614556800, 1617235200, 1619827200, 1622505600, 
1625097600, 1627776000, 1630454400, 1633046400, 1635724800, 1638316800, 
1640995200, 1643673600, 1646092800, 1648771200, 1651363200, 1654041600
), tzone = "UTC", class = c("POSIXct", "POSIXt")), media = c(27.0163533756641, 
27.3523628886565, 27.6874774847651, 27.9632663969036, 28.2324797447653, 
28.5946591329209, 28.0774171223198, 27.0218334461407, 26.3074215031224, 
25.762714516129, 25.9331010513628, 26.1019612833841, 27.4678015082705, 
27.8551394156665, 27.61177313685, 28.5468447629364, 28.6978991786281, 
29.0753629090163, NaN, NaN), desvio = c(0.349566729323312, 0.354049553514593, 
0.316576405105695, 0.28004282459359, 0.339372318521672, 0.287591989484519, 
0.312261182269958, 0.246786673478273, 0.358857114435788, 0.119453586990502, 
0.256559466294029, 0.300030737021837, 0.348529074834953, 0.383358876272664, 
0.465191392255828, 0.596852747140639, 0.324808751418437, 0.222745848411023, 
NA, NA), year = c(2020, 2020, 2021, 2021, 2021, 2021, 2021, 2021, 
2021, 2021, 2021, 2021, 2021, 2021, 2022, 2022, 2022, 2022, 2022, 
2022)), row.names = c(NA, -20L), class = "data.frame")

And plot at this way:

ggplot(df, aes(x = cycle_rounded, y = media)) +
  geom_ribbon(aes(y = media, ymin = media - desvio, ymax = media + desvio), 
              alpha = .2, fill = "red") +
  geom_line(col = "grey10", linewidth = .5) +
  geom_point(col = "grey10", size = 1) +
  scale_x_datetime(date_labels = "%m", breaks = "1 month",
                   sec.axis = sec_axis(~., sec_axis(~ ., labels = year))) +
  labs(x = "data", y = "temperatura") +
  theme_classic() +
  theme(axis.text.x = element_text(size = 12, angle = 45, 
                                   vjust = 1, hjust = 1, color = "black"),
        axis.text.y = element_text(size = 12, color = "black"),
        axis.title.x = element_text(size = 14, color = "black"),
        axis.title.y = element_text(size = 14, color = "black"))

however the sec_axis won't works as I expected. I would like to shown years below the months as one value of year for all correspondent months.

Any helps are appreciated

Thanks, Wilson

I think it is not clear what you exactly mean with secondary axis? — TarJae, Jun 16 '23 at 18:26
I would like the month on first row of x-axis and year in the row below. I called secondary axis to refer the row of year. — Wilson Souza, Jun 16 '23 at 18:33
something like this? https://stackoverflow.com/questions/56339970/ggplot2-add-secondary-x-label-year-below-months — AdroMine, Jun 16 '23 at 18:37
Included ggplot code seems to have some issues and at least form me it fails at `sec.axis`. Perhaps include an image of what you have now and manually annotate it to illustrate what you'd like to achieve? — margusl, Jun 16 '23 at 18:48
@AdroMine exactly this, but it wont works for my data. The last month of the previous year plot above the first month of next year and the lines not connect. — Wilson Souza, Jun 16 '23 at 19:22

score 1 · Accepted Answer · answered Jun 16 '23 at 18:39

sec_axis(~., sec_axis(~ ., labels = year)) does not make sense as a nested double-call.
Further, labels=year is either a parsing error ('year' not found) or it uses year found in the current or parent environments (whether values or a function).
sec_axis is going to put another axis on the opposite side, in this case the top, not as a second row.

I think we can do a single axis and put the year after a newline.

ggplot(df, aes(x = cycle_rounded, y = media)) +
  geom_ribbon(aes(y = media, ymin = media - desvio, ymax = media + desvio), 
              alpha = .2, fill = "red") +
  geom_line(col = "grey10", linewidth = .5) +
  geom_point(col = "grey10", size = 1) +
  scale_x_datetime(labels = ~ { mon <- format(., format="%m"); if_else(!is.na(mon) & mon == "01", paste0(mon, format(., format = "\n%Y")), mon) },
                   breaks = "1 month") +
  labs(x = "data", y = "temperatura") +
  theme_classic() +
  theme(axis.text.x = element_text(size = 12, angle = 45, 
                                   vjust = 1, hjust = 1, color = "black"),
        axis.text.y = element_text(size = 12, color = "black"),
        axis.title.x = element_text(size = 14, color = "black"),
        axis.title.y = element_text(size = 14, color = "black"))

You can easily clean up that code a little by assigning the {...} block to a function.

I assumed putting the year once per 12 months, over to you do adjust this for your expectations.

Is possible to add all years that have? For example, 2020 not appear. And adding this second row, is possible format they different, for example, use the angle as 0 and not 45°? Sorry for this aestethics questions. — Wilson Souza, Jun 16 '23 at 18:56
It's all in the `labels=` function. Perhaps (untested) `if_else(!is.na(mon) & (mon == "01" | seq_along(mon) == which(!is.na(mon))[1]), paste0(mon, format(., format = "\n%Y")), mon)` or similar will force the first non-`NA` month to add the year to it — r2evans, Jun 16 '23 at 20:17

Adjust datetime x-axis to shown year as a sec axis below de main axis

1 Answers1