Adjusting tick marks

I want to modify the tick marks on my ggplot in order to create a package. However, no matter how diligently I search, I can't seem to find a solution that meets my expectations (attached is a screenshot depicting how I would like it to appear).

The closest I've come to achieving this was by using ggprism::guide_prism_bracket, but I'm encountering an issue where the axis line doesn't fill in completely and contains gaps:

library(ggplot2)

year <- factor(c(2000,2005,2010,2015,2020,2025,2030,2035,2040,2045,2050))
Population <- c(6.1,6.5,6.9,7.3,7.7,8.0,8.3,8.6,8.9,9.1,9.3)
labels <- as.character(Population)
labels <- gsub("." , "," , labels, fixed=T)
df <- data.frame(year,Population,labels)

graf <- ggplot(df, aes(x= as.character(year), y=Population)) +
  geom_bar(width = 0.5, stat = "identity", position=position_dodge(), color= NA) +
  scale_y_continuous(expand = expansion(mult = c(0, 0)))  +
  ggprism::theme_prism(
    base_fontface = "plain", 
    base_line_size = 0.7, 
    base_family = "Arial"
  ) + 
  scale_x_discrete(guide = ggprism::guide_prism_bracket(width = 0.15), 
                   labels = scales::wrap_format(5))
graf

Any recommendation?

Thank you

Probably the neatest way is to fake the tick lines using geom_rug

ggplot(df, aes(as.numeric(as.character(year)), Population)) +
  geom_col(width = 3) +
  scale_x_continuous("Year", breaks = seq(1995, 2050, 5)) +
  geom_rug(aes(x = as.numeric(as.character(year)) - 2.5),
           outside = TRUE, sides = "b", length = unit(5, "mm")) +
  coord_cartesian(expand = FALSE, clip = "off") +
  theme_classic(base_size = 14) +
  theme(axis.ticks.x = element_blank())

Another option that doesn't require annotations is to use existing tick marks and move the columns and the axis text

ggplot(df, aes(as.numeric(as.character(year)), Population)) +
  geom_col(width = 3) +
  scale_x_continuous("Year", breaks = seq(1995, 2050, 5)) +
  geom_rug(aes(x = as.numeric(as.character(year)) - 2.5),
           outside = TRUE, sides = "b", length = unit(5, "mm")) +
  coord_cartesian(expand = FALSE, clip = "off") +
  theme_classic(base_size = 14) +
  theme(axis.ticks.x = element_blank())

I think I get pretty close, I started with theme_void and added a few of the wanted theme items back. You might need to tweak some of the settings. I also kept year as numeric instead of a factor.

The idea is to make your x-axis continuous

• then adjust the size of your bar width = 2.8 but feel free to tweak to your taste.
• you showed a pretty squeezed down graph, you can adjust at aspect.ratio = 0.076
• the trick is that you want ticks between the bars and the labels below the bars. I decided to set the ticks half of the year gap skewed and show no labels
• then add the "labels" as annotation outside the range in combination with coord_cartesian with clip = "off"
• in annotate in combination with this aspect_ratio the y offset of y = -max(df$Population) / 5 worked out fine, but you might want to tweak that.

code

library(ggplot2)

ggplot(df, aes(x = year, y = Population)) +
  geom_bar(width = 2.8, stat = "identity", fill = "#006096") +
  theme_void() + # get rid of all, add a few below
  theme(
    axis.line = element_line(),
    axis.ticks.x = element_line(color = "black"),
    axis.line.y.right = element_line(color = "white"),
    axis.ticks.length = unit(0.2, "cm"),
    aspect.ratio = 0.076 # just tweak to get your "height"
  ) +
  scale_y_continuous(expand = c(0, 0), sec.axis = sec_axis(~ .)) +
  scale_x_continuous(breaks = seq(min(df$year) - 2.5, max(df$year) + 2.5, 5)) +
  annotate("text", x = year, y = -max(df$Population) / 5, label = year) +
  coord_cartesian(xlim = c(min(df$year), max(df$year)), ylim = c(0, max(df$Population)), clip = "off")

data used

df <- data.frame(
  year = c(2000,2005,2010,2015,2020,2025,2030,2035,2040,2045,2050),
  Population = c(6.1,6.5,6.9,7.3,7.7,8.0,8.3,8.6,8.9,9.1,9.3)
)