processing_datacube_raster.md

load libraries

library(tidyverse)      #for data munging
library(sf)             #for vector operations
library(raster)         #for raster operations

library(fasterize)      #for rasterizing vector data
library(exactextractr)  #for zonal statistics
library(tictoc)         #for time calculations

load input raster data

r <- 		raster("filepath here")
plot(r)

load H3 grid as polygon

grid <- 	st_read("filepath here")

convert raster data to H3 grid using zonal statistics

tic()
calc <- 	exactextractr::exact_extract(r, grid, c("majority", "minority", "max", "min", "mean"))
toc()

dropnames <- colnames(calc)

prepare lookup table to join zonal statistics with H3 grid

forbind <- 	calc %>%
		mutate(majority = majority, minority = minority, maximum = max, minimum = min, mean = mean) %>%
		dplyr::select(-all_of(dropnames))

combine lookup table with H3 grid

grid_join <-	bind_cols(grid, forbind)

drop geometries and convert to dataframe for export

grid_df <-	grid_join %>%
		st_drop_geometry() %>%
		as.data.frame() 

write dataframe to csv

write.csv(grid_df, row.names = FALSE, "filepath here")