If the outline of a polygon is deterministic but the fill is random, you should use position subdivide rather than varying the alpha value. This subdivide position can be used with geom_polygon_sample (soon to be extended to others such as geom_sf, geom_map, etc).
Examples
library(ggplot2)
library(distributional)
library(dplyr)
# make data polygon with uncertain fill values
ids <- factor(c("1.1", "2.1", "1.2", "2.2", "1.3", "2.3"))
values <- data.frame(
id = ids,
value = c(1, 2, 3, 4, 5, 6)
)
positions <- data.frame(
id = rep(ids, each = 4),
x = c(2, 1, 1.1, 2.2, 1, 0, 0.3, 1.1, 2.2, 1.1, 1.2, 2.5, 1.1, 0.3,
0.5, 1.2, 2.5, 1.2, 1.3, 2.7, 1.2, 0.5, 0.6, 1.3),
y = c(-0.5, 0, 1, 0.5, 0, 0.5, 1.5, 1, 0.5, 1, 2.1, 1.7, 1, 1.5,
2.2, 2.1, 1.7, 2.1, 3.2, 2.8, 2.1, 2.2, 3.3, 3.2)
)
datapoly <- merge(values, positions, by = c("id"))
uncertain_datapoly <- datapoly |>
mutate(value = dist_uniform(value, value + 0.8))
# ggplot
ggplot(datapoly , aes(x = x, y = y)) +
geom_polygon(aes(fill = value, group = id))
# ggdibbler
ggplot(uncertain_datapoly , aes(x = x, y = y)) +
geom_polygon_sample(aes(fill = value, group = id), times=50,
position = "subdivide")
