ggplot2 geom_area() in R: Area Charts With Examples
The geom_area() function in ggplot2 fills the area between a line and the x axis. It is the area version of geom_line() and the building block for stacked area charts and streamgraphs.
ggplot(df, aes(x, y)) + geom_area() # basic area ggplot(df, aes(x, y, fill = group)) + geom_area() # stacked ggplot(df, aes(x, y, fill = group)) + geom_area(position = "fill") # 100% stacked ggplot(df, aes(x, y, fill = group)) + geom_area(alpha = 0.5) # transparent ggplot(df, aes(x, y)) + geom_area(fill = "steelblue") + geom_line()# area + line ggplot(df, aes(x, ymin = lo, ymax = hi)) + geom_ribbon() # band (alternative)
Need explanation? Read on for examples and pitfalls.
What geom_area() does in one sentence
geom_area() fills the region between a line at y values and the x axis (or another baseline). It is geom_line() plus a fill below. With grouped data, it stacks areas by default.
Use area charts for time series where the magnitude (and how it composes by group) matters. They are common for sales over time, traffic source breakdown, and population composition.
Syntax
geom_area() requires aes(x, y). With grouping, also map fill.
The full signature:
geom_area(mapping = NULL, data = NULL, stat = "identity", position = "stack",
..., na.rm = FALSE, orientation = NA, show.legend = NA, inherit.aes = TRUE)position = "stack" automatically sums groups vertically. With aes(fill = group) and stack position, areas at each x add together. Set position = "identity" if you want them OVERLAID (and use alpha).Five common patterns
1. Single filled area
The simplest case: the area below the line is filled. Adding geom_line() on top emphasizes the upper boundary.
2. Stacked area by group
fill = group automatically stacks. The TOTAL height at each year is the sum across all groups. Each color shows that group's contribution.
3. 100% stacked (proportions)
position = "fill" rescales each year so the total reaches 1.0. Useful for showing how composition changes over time when total is not the focus.
4. Transparent overlapping areas
position = "identity" plots each group's area independently (not stacked). Combined with alpha = 0.4, overlapping areas are visible. Note: identity-positioned areas obscure each other; usually overlay is more readable as geom_line().
5. Confidence band with geom_ribbon
geom_ribbon() is a sibling: it fills the area between TWO y values (ymin and ymax), not between y and zero. Use ribbons for confidence intervals or forecast bands; use areas for compositions.
aes(fill = forcats::fct_reorder(group, value)). The bottom group has a stable baseline; higher groups fluctuate visually because their baseline is the sum below them.geom_area() vs geom_line() vs geom_ribbon()
Three line-related geoms, each filling a different region.
| Geom | Fills | Use for |
|---|---|---|
geom_line() |
Nothing | Trend or time series, no fill |
geom_area() |
Below line to baseline (0 or stacked group) | Time series with magnitude focus |
geom_ribbon() |
Between two y values (ymin, ymax) | Confidence intervals, forecast bands |
When to use which:
- Use
geom_line()for clean trend visualization without emphasis on magnitude. - Use
geom_area()for emphasizing magnitude or composition over time. - Use
geom_ribbon()for uncertainty bands around a central line.
Common pitfalls
Pitfall 1: stacked area when overlay was intended. Default position = "stack" sums groups. If you want to compare independent series, set position = "identity" and use alpha (or use geom_line()).
Pitfall 2: missing data creates gaps. geom_area() does not interpolate over NA. If a group has no value at year 2022, the area drops to zero there, distorting the visual. Either fill missing values explicitly or use tidyr::complete() to ensure regular structure.
Pitfall 3: order of fill groups matters. Default fill order is alphabetical, which may not be the most informative ordering. Reorder by recent value with forcats::fct_reorder(group, value) for cleaner visual hierarchy.
Try it yourself
Try it: Use the sample df to make a 100% stacked area chart showing the proportional composition of A, B, C over the years 2020 to 2024. Save to ex_plot.
Click to reveal solution
Explanation: position = "fill" rescales each year so the total reaches 1.0. scale_y_continuous(labels = scales::percent) formats the y axis as percentages instead of decimals. The viridis palette ensures colorblind-friendly fills.
Related ggplot2 functions
After mastering geom_area(), look at:
geom_line(): line without fillgeom_ribbon(): fill between two y valuesgeom_density(): filled density curve (special case)position_stack(),position_fill(): explicit position controlsforcats::fct_reorder(): reorder fill groups for cleaner visual hierarchyggstream::geom_stream(): streamgraph (centered around midline, smoothed)
For complex multi-series area charts, the ggstream package provides streamgraphs with smoothing and centering.
FAQ
What is the difference between geom_area and geom_ribbon in ggplot2?
geom_area() fills between a line and a BASELINE (default 0, or the stacked group below). geom_ribbon() fills between TWO y values (ymin and ymax). Use area for compositions; ribbon for confidence intervals.
How do I make a stacked area chart in ggplot2?
Map a categorical variable to fill inside aes(): geom_area(aes(fill = group)). Default position = "stack" stacks them. The y at each x is the sum across groups.
How do I make a 100% stacked area chart?
Set position = "fill": geom_area(position = "fill"). Each x value's total is rescaled to 1.0 (or 100%). Format the y axis with scale_y_continuous(labels = scales::percent).
Can I overlay area charts instead of stacking?
Yes, with position = "identity" and alpha: geom_area(position = "identity", alpha = 0.4). Overlapping areas are visible. For many groups, line plots (geom_line(color = group)) are usually clearer.
How do I fill between two lines in ggplot2?
Use geom_ribbon() with aes(ymin, ymax): geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.3). Common for confidence bands or forecast intervals around a central line.