R: reading 3-dim nc data (from gpw)

Question

I am trying to work with Columbia's 3-dimensional gridded population of the world data, v4 (Gridded Population of the World (GPW), v4). it has population estimates in a 1x1 grid for seq(2000,2020,5). https://sedac.ciesin.columbia.edu/data/collection/gpw-v4 . complicating the fact is that there are 5 years and 10 variables, vars <- c( "ccode", "datacode", "year", "admlevel", "sexlevel", "agelevel", "gyear0", "gyearT", "glevel", "ycensus" ) indicated in the textual description, but I can worry about these later. right now, I just want to get some variable(s) in one particular year.

library(ncdf4)
nc= nc_ncopen("gpw_v4_population_count_rev11_1_deg.nc")
print(nc)

tells me

     1 variables (excluding dimension variables):
        float Population Count, v4.11 (2000, 2005, 2010, 2015, 2020): 1 degree[longitude,latitude,raster]   
      ... more info ...
     3 dimensions:
        longitude  Size:360 
            units: degrees_east
            long_name: longitude
        latitude  Size:180 
            units: degrees_north
            long_name: latitude
        raster  Size:20   *** is unlimited *** 
            units: unknown
            long_name: raster

I am about to do a victory dance with mydf2000 <- as.data.frame( nc[,,1] ) but my hopes were cruelly dashed.

I got a little further with

library(raster)
library(ncdf4)

fnm <- "gpw_v4_population_count_rev11_1_deg.nc"
ncbrick <- brick(fnm)
dim(ncbrick) # 180, 360, 20
dim(ncbrick[,,1] ## 64800, 20 <-- huh?  I wanted 180, 360

could someone please let me know how to read the x-th dimensional 180x360 grid into a data frame?

Answer 1

Use theterra package:

library(terra)

Read using rast:

d = rast("./gpw_v4_population_count_rev11_1_deg.nc")

Get layers with [[index]]:

dim(d[[1]])
### [1] 180 360   1

Convert to a normal matrix with as.matrix and wide=TRUE:

dim(as.matrix(d[[1]],wide=TRUE))
### [1] 180 360