Fun pictures with ggplot2 and scico packages

This wonderful post from Jonathan Caroll give me idea to play with ggplot2 code and scico color palettes:

Also see this toot from Stefan Siegert, inspiring!

I took the code from the toot mentioned above, modified it, and used it to have a little fun creating graphics that look like paintings with imperfections around the edges, among other things.

All in all, I think it’s a good way to test color palettes.

When you get higher dot per inch images, it can be really cool.

# https://mastodon.social/@safest_integer/114296256313964335library(tidyverse)crossing(x=0:10, y=x) |>  mutate(dx = rnorm(n(), 0, (y/20)^1.5),         dy = rnorm(n(), 0, (y/20)^1.5)) |>ggplot() +geom_tile(aes(x=x+dx, y=y+dy, fill=y), colour='black',          lwd=2, width=1, height=1, alpha=0.8, show.legend=FALSE) +scale_fill_gradient(high='#9f025e', low='#f9c929') +scale_y_reverse() + theme_void()

library(tidyverse)crossing(x=8:300, y=x[1:100]) |>  mutate(dx = rnorm(n(), 0, (y/200)^1.5),         dy = rnorm(n(), 0, (y/100)^1.5)) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y+dy, fill=y), colour=NA,            lwd=4, width=8, height=4, alpha=0.1,             show.legend=FALSE) +  scale_fill_gradient(high='#9f025e', low='#f9c929') +  scale_y_reverse() + theme_void() +   coord_fixed()

Here is the “Oleron” color palette from scico. What a beautiful palette for those who know this french island!

library(tidyverse)crossing(x=8:300, y=x[1:100]) |>  mutate(dx = rnorm(n(), 0, (y/200)^1.5),         dy = rnorm(n(), 0, (y/100)^1.5)) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y+dy, fill=y), colour=NA,            lwd=4, width=8, height=4, alpha=0.1,             show.legend=FALSE) +  scico::scale_fill_scico(palette = "oleron",   direction = 1) +  scale_y_reverse() + theme_void() +   coord_fixed()

Now drawing waves, view at the top of the dune:

crossing(x=8:300, y=x[1:100]) |>  mutate(dx = rnorm(n(), 0, (y/200)^1.5),         dy = ifelse(y < 52,                      rnorm(n(), 0, (y/100)^1.5)+                        ifelse(x %% 2 == 0,                               cos(x * 3 + y / 5)*4 +                                sin(x * 3 + y / 2)*2, 0) / 1.7,                     rnorm(n(), 0, (y/100)^1.5))) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y+dy, fill=y), colour=NA,            lwd=4, width=8, height=4, alpha=0.2,             show.legend=FALSE) +  scico::scale_fill_scico(palette = "oleron", #begin = 0.6,                          direction = 1) +  scale_y_reverse() + theme_void() +   coord_fixed()

The green might be vegetation from the dunes, or maybe seaweed.The blue might also be sky and wind, blue hour, there is definitely sand.

Back to bigger rectangles with a cyclic scico palette: romaO.

crossing(x=1:280, y=x) |>  mutate(dx = rnorm(n(), 0, (x/40)^1.5),         dy = rnorm(n(), 0, (y/40)^1.5)) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y-dy*3, fill=y), colour='white',            lwd=0.03, width=13, height=6, alpha=0.4,             show.legend=FALSE) +  scico::scale_fill_scico(palette = "romaO", #begin = 0.6,                          direction = 1) +  scale_y_reverse() + theme_void() +   theme(panel.background = element_rect(fill = "#2b2828")) +  coord_fixed()

Now just play with a bit more magic: coord_polar

crossing(x=1:280, y=x) |>  mutate(dx = rnorm(n(), 0, (x/40)^1.5),         dy = rnorm(n(), 0, (y/40)^1.5)) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y-dy*3, fill=y), colour='white',            lwd=0.03, width=13, height=6, alpha=0.4,             show.legend=FALSE) +  scico::scale_fill_scico(palette = "romaO", #begin = 0.6,  direction = 1) +  scale_y_reverse() + theme_void() +   theme(panel.background = element_rect(fill = "#2b2828")) +  coord_polar()

Another one with vikO palette:

crossing(x=8:70, y=x[1:30]) |>  mutate(dx = rnorm(n(), 0, (y/20)^1.5),         dy = rnorm(n(), 0, (y/10)^1.5)+ ifelse(x %%2 == 0, cos(x * 3 + y / 5)*10, 0)) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y+dy, fill=y), colour='grey70',            lwd=0.1, width=8, height=4, alpha=0.6,            show.legend=FALSE) +  scico::scale_fill_scico(palette = "vikO", #begin = 0.6,                          direction = -1) +  scale_y_reverse() +  theme_void() +  theme(panel.background = element_rect(fill = "#2b2828")) +  coord_polar(theta = "x")

Finally back to Oleron palette with coord_polar:

crossing(x=8:70, y=x[1:30]) |>  mutate(dx = rnorm(n(), 0, (y/20)^1.5),         dy = rnorm(n(), 0, (y/10)^1.5)+ ifelse(x %%2 == 0, cos(x * 3 + y / 5)*10, 0)) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y+dy, fill=y), colour=NA,            lwd=1, width=8, height=4, alpha=0.4,            show.legend=FALSE) +  scico::scale_fill_scico(palette = "oleron", #begin = 0.6,                          direction = 1) +  scale_y_reverse() +  theme_void() +  theme(panel.background = element_rect(fill = "grey5")) + ##2b2828  coord_polar(theta = "x")

Kind of like Earth, but not from the Moon. Just from R.

bukavu palette:

crossing(x=8:70, y=x[1:30]) |>  mutate(dx = rnorm(n(), 0, (y/20)^1.5),         dy = rnorm(n(), 0, (y/10)^1.5)+ ifelse(x %%2 == 0, cos(x * 3 + y / 5)*10, 0)) |>  ggplot() +  geom_tile(aes(x=x+dx, y=y+dy, fill=y), colour='grey5',            lwd=0.2, width=8, height=4, alpha=0.4,            show.legend=FALSE) +  scico::scale_fill_scico(palette = "bukavu", #begin = 0.6,                          direction = 1) +  scale_y_reverse() +  theme_void() +  theme(panel.background = element_rect(fill = "#2b2828")) +  coord_polar(theta = "x")

Many thanks to:

• Thomas Lin Pedersen for this package: thomasp85/scico and Fabio Crameri for the palettes.
• Stefan Siegert
• Jonathan Caroll
• ggplot2 author and maintainers.