Consider using tidyterra for plotting

[AndyBunn]

How could the content be improved?

The tidyterra library has a nice geom for SpatRaster objects. E.g.,

library(tidyterra)
DSM_HARV <-
  rast("/data/NEON-DS-Airborne-Remote-Sensing/HARV/DSM/HARV_dsmCrop.tif")
ggplot() + geom_spatraster(data = DSM_HARV , aes(fill = HARV_dsmCrop)) +
  scale_fill_viridis_c()

This would keep students from having to convert to data.frame or tibble for geom_raster

[jebyrnes]

Agree. I know it’s another dependency, but it takes away the need to continually turn rast objects into data frames, which confuses learners, and plays nice with geom_sf() for overlays.

[jonjab]

tidyterra is quite slow for me, I'm afraid it would overload Learners' laptops.

Does anyone have an opinion on this?

[jebyrnes]

Weird. It works fine for me? Use it all the time as part of my daily workflow.

[AndyBunn]

For me, using geom_raster and geom_spatraster are roughly equivalent, which makes sense given that geom_spatraster is based on geom_raster. They are however, both slower than using terra::plot but that is to be expected.

library(tidyverse)
library(terra)
library(tidyterra)
DSM_HARV <-
  rast("data/NEON-DS-Airborne-Remote-Sensing/HARV/DSM/HARV_dsmCrop.tif")

# use geom_raster
start_time <- Sys.time()
DSM_HARV_df <- as.data.frame(DSM_HARV, xy = TRUE)
ggplot() +
  geom_raster(data = DSM_HARV_df , aes(x = x, y = y, fill = HARV_dsmCrop)) +
  scale_fill_viridis_c() +
  coord_quickmap()
end_time <- Sys.time()
end_time - start_time

# use geom_spatraster
start_time <- Sys.time()
ggplot() + geom_spatraster(data = DSM_HARV , aes(fill = HARV_dsmCrop)) +
  scale_fill_viridis_c()
end_time <- Sys.time()
end_time - start_time