removing missing values labels from ggplot

Question

what I want appears to be simple but I can't figure it out: I want to take the NA values from my labs out. Problem is, it's my first time using the "na.value" argument, so I’m not quite sure how to proceed.

(btw, I can’t drop the NAs before plotting because the shapes that are not from the tourist regions will also disappear, and I need the full map.)

I have this code:

mun_tur_shape %>% 
  filter(abbrev_state == "BA") %>% 
  ggplot() +
  geom_sf(aes(fill=TOURIST_REGION, colour=TOURIST_REGION)) +
  scale_fill_manual(
    na.value = "grey90"
    values = c(viridis::inferno(13)),
    aesthetics = c("fill", "colour")
    ) +
    labs(fill = "Região Turística",
         colour = "Região Turística"
           ) +

And this is how it looks: Plot with NA value

Does anyone know what I can do to omit them?

# here's an sf for reproducible example:
#install.packages(geobr)
df <- geobr::read_state()

df %>%
  # creating NA values like my real dataset has
  mutate(name_region=case_when(name_region=="Nordeste"~NA_character_,
                          TRUE~name_region)) %>% 
  ggplot() +
  geom_sf(aes(fill=name_region, colour=name_region)) +
  scale_fill_manual(
    na.value = "grey90",
    values = (viridis::inferno(4)),
    aesthetics = c("fill", "colour")
  ) +
  labs(colour = "Regions",
       fill = "Regions")

Answer 1

My solution for this (see also NA values ?in choropleth plot legend with ggplot2 in R) is to add first a base layer with all the polygons (no aes) using the colors that I would want to use for NA. After that, you can overlay the layer with aes(). In practice, just add one line on your code

library(dplyr)
library(ggplot2)
library(geobr)

df <- geobr::read_state()


df %>%
  # creating NA values like my real dataset has
  mutate(name_region = case_when(
    name_region == "Nordeste" ~ NA_character_,
    TRUE ~ name_region
  )) %>%
  # Create map
  ggplot() +
  # Add this line a base layer with no aes, but a NA fill color
  geom_sf(fill = "grey50", color = "grey50") +
  # end
  geom_sf(aes(fill = name_region, colour = name_region)) +
  scale_fill_manual(
    na.value = "grey90",
    values = (viridis::inferno(4)),
    aesthetics = c("fill", "colour"),
    na.translate = FALSE
  ) +
  labs(
    colour = "Regions",
    fill = "Regions"
  )