Ridgeline plots to study subnational rainfall patterns in Vietnam
Using {ggridges} and {ggplot2} to create ridgeline plots in R
Overview
Ridgeline plots are a type of chart that displays the distribution of a numeric variable for different categories. These plots are commonly used to visualize changes in distributions over time. To showcase how to create ridgeline plots with {ggplot2} and {ggridges} in R, I will use data on monthly rainfall at some stations by provincies and month in Vietnam, from 2002 to 2024.
Set-up
First, we install and load the necessary R packages.
About the data
Data come from the National Statistics Office of Vietnam and is available at their website. I downloaded the data and saved it in an Excel file called rainfall-vietnam.xlsx.
rainfall_vietnam <- read_excel("rainfall-vietnam.xlsx")Data is not in a tidy format, so I will need to wrangle it.
rainfall_vietnam <- rainfall_vietnam |>
rename(year = `Monthly rainfall at some stations by Year, Cities, provincies and Month`,
city = `...2`) |>
fill(year) |> # `fill()` replaces missing data from top to bottom
rename_at(vars(`...3`:`...14`),
~ month.name
) |> # row 2 contains the name of the months
slice(-c(1:2)) |> # remove first two rows
mutate_at(vars(January:December),
~ as.numeric(.)) |>
pivot_longer(cols = January:December,
names_to = "month",
values_to = "rainfall") After reshaping, data looks like this:
rainfall_vietnam |>
head(15) |>
kbl(
caption = "Monthly rainfall by month in Vietnam, 2002-2024"
) |>
kable_paper("hover", full_width = F)| year | city | month | rainfall |
|---|---|---|---|
| 2024 | Lai Chau | January | 185.2 |
| 2024 | Lai Chau | February | 44.5 |
| 2024 | Lai Chau | March | 69.1 |
| 2024 | Lai Chau | April | 27.7 |
| 2024 | Lai Chau | May | 383.9 |
| 2024 | Lai Chau | June | 541.8 |
| 2024 | Lai Chau | July | 395.4 |
| 2024 | Lai Chau | August | 362.6 |
| 2024 | Lai Chau | September | 164.7 |
| 2024 | Lai Chau | October | 148.1 |
| 2024 | Lai Chau | November | 4.4 |
| 2024 | Lai Chau | December | 21.9 |
| 2024 | Son La | January | 55.8 |
| 2024 | Son La | February | NA |
| 2024 | Son La | March | 14.3 |
Creating the ridgeline plots
I will create two versions of the ridgeline plot. First, a ridgeline plot to show the interquantile range of rainfall by month in Ha Noi, capital of Vietnam. Here, the colors represent different quantiles of the distribution of rainfall per month.
rainfall_vietnam <- rainfall_vietnam |>
mutate(month = factor(month, levels = month.name)) # Important to order months in the correct order
# More font alternatives can be found here https://fonts.google.com/
font_add_google("Oxygen", "Oxygen")
showtext_auto()
rainfall_vietnam |>
filter(city == "Ha Noi") |>
ggplot(aes(x = rainfall,
y = reorder(month, desc(month)),
fill = after_stat(quantile)
)) +
stat_density_ridges(quantile_lines = FALSE,
geom = "density_ridges_gradient",
calc_ecdf = TRUE,
color = "black",
scale = 1.5) + # Height of ridges
scale_fill_manual(
name = "Quantiles",
values = c("#CCFFFF", "#99EDFF", "#4CC3FF", "#007FFF"),
labels = c("(0, 0.25]", "(0.25, 0.5]", "(0.5, 0.75]", "(0.75, 1]")
) +
ggtitle("Mean monthly rainfall in Ha Noi, Vietnam | 2002-2024.",
subtitle = str_wrap("The rainy season in Ha Noi extends from May to early November, peaking in June, July, and August when precipitation is at its highest.", width = 110)
) +
xlab("Monthly rainfall in millimeters") +
ylab("") +
labs(caption = "Source: National Statistics Office of Vietnam") +
theme_minimal(
base_family = "Oxygen",
paper = "#EEEEEE"
) +
theme(
plot.margin = margin(0, 40, 0, 40),
plot.title = element_text(
color = "#0D47A1",
face = "bold",
size = 100,
hjust = 0,
margin = margin(15, 0, 0, 0)
),
plot.title.position = "plot",
plot.subtitle = element_text(
color = "#1976D2",
face = "bold",
size = 85,
hjust = 0,
margin = margin(0, 0, 20, 0),
lineheight = 0.3
),
axis.title = element_text(
color = "#0D47A1",
face = "bold",
size = 70
),
axis.text = element_text(
color = "#1976D2",
face = "bold",
size = 60
),
plot.caption = element_text(
color = "#0D47A1",
face = "bold",
size = 60,
hjust = 1,
margin = margin(25, 0, 5, 0)
),
legend.position = "top",
legend.title = element_text(
face = "bold",
color = "#0D47A1",
size = 60,
hjust = 1
),
legend.text = element_text(
face = "bold",
color = "#0D47A1",
size = 60,
hjust = 1
),
)# showtext_opts(dpi = 320)
# ggsave(
# "rainfall-vietnam-ridgeline.png",
# dpi = 320,
# width = 15,
# height = 11,
# units = "in"
# )
# showtext_auto(FALSE)
The first version of the ridgeline plot allows us to see the distribution of rainfall in Ha Noi by month. Particularly, it shows that the wettest months are June, July, and August, i.e. the summer months. The rainy season is easily identified as these months have their median values more to the right of the x-axis. On the other hand, the driest months are from November to April, i.e. the winter months. The interquantile ranges also show that there is a lot of variability in precipitation during the rainy season, while the dry season has less variablity.
In the second version of the ridgeline plot, I will use a gradient with a continuous fill color scale to represent the amount of rainfall. I will also use other font and color choices. To show this, I will plot the data from Nha Trang, a coastal city and the capital of Khánh Hòa Province.
# More font alternatives can be found here https://fonts.google.com/
font_add_google("Bree Serif", "Bree Serif")
showtext_auto()
rainfall_vietnam |>
filter(city == "Nha Trang") |>
ggplot(aes(x = rainfall,
y = reorder(month, desc(month)),
fill = stat(x)
)) +
stat_density_ridges(quantile_lines = FALSE,
geom = "density_ridges_gradient",
color = "white",
scale = 2.8) + # Height of ridges
scale_fill_distiller(
direction = 1,
name = "",
palette = "RdPu"
) +
ggtitle("Mean monthly rainfall in Nha Trang, Vietnam | 2002-2024.",
subtitle = str_wrap("The wet season in Nha Trang lasts from September to December.", width = 110)
) +
xlab("Monthly rainfall in millimeters") +
ylab("") +
labs(caption = "Source: National Statistics Office of Vietnam") +
theme_minimal(
base_family = "Bree Serif",
paper = "#767F8B"
) +
theme(
plot.margin = margin(0, 40, 0, 40),
plot.title = element_text(
color = "white",
face = "bold",
size = 100,
hjust = 0,
margin = margin(15, 0, 0, 0)
),
plot.title.position = "plot",
plot.subtitle = element_text(
color = "white",
face = "bold",
size = 85,
hjust = 0,
margin = margin(0, 0, 20, 0),
lineheight = 0.3
),
axis.title = element_text(
color = "white",
face = "bold",
size = 70
),
axis.text = element_text(
color = "white",
face = "bold",
size = 60
),
plot.caption = element_text(
color = "white",
face = "bold",
size = 60,
hjust = 1,
margin = margin(25, 0, 5, 0)
),
legend.position = "none"
)# showtext_opts(dpi = 320)
# ggsave(
# "rainfall-vietnam-ridgeline_gradient.png",
# dpi = 320,
# width = 15,
# height = 11,
# units = "in"
# )
# showtext_auto(FALSE)
The second version of the ridgeline plot allows us to see the distribution of rainfall in Nha Trang by month. As can be seen, the wettest months, i.e. those with the highest median values and the most spread out distributions, are from September to December. The dry season is from January to August, in which the values are more concentrated to the left of the x-axis.
Citation
@online{torres munguía2025,
author = {Torres Munguía, Juan Armando},
title = {Ridgeline Plots to Study Subnational Rainfall Patterns in
{Vietnam}},
date = {2025-09-15},
url = {https://juan-torresmunguia.netlify.app/blog/posts/vietnam-rainfall-ridgeline-plot},
langid = {en}
}