Vjylku

Here is some elevation data of some islands.

I wish to add depth info to the water. With the generous help of @Spacedman here I get:

To resolve the blockyness of the lower resolution water data I tried, resample(to_harbour, to_island, method = "bilinear") yielding:

I'm not sure if that is correct. Note that I have lost some of the water around the islands. I'm not sure why this is nor am I sure how I should identify the water in to_island. Currently I am just looking at the data and guessing that values less than 75.109m are water and should be looked up from the bathymetric raster layer.

Dropbox:

to_island

to_harbour

Code

library(sp)

library(raster)



library(tidyverse)

library(rayshader)



to_island <- readRDS(file = "to_island.RDS") # https://www.dropbox.com/s/hc86gzdt6nm4j4w/to_island.RDS?dl=0

to_harbour <- readRDS(file = "to_harbour.RDS") # https://www.dropbox.com/s/awl4clyb3254n14/to_harbour.RDS?dl=0



# Create matrix for 3D plotting

to_island_matrix = matrix(raster::extract(to_island, raster::extent(to_island), buffer = 10000), 

                          nrow = ncol(to_island), ncol = nrow(to_island))



# 3D plot land without water depth info ####################################################

to_island_matrix %>%

  sphere_shade(zscale = 0.1, texture = "imhof1") %>%  # Lower zscale maps points to wider range of colours

  plot_3d(to_island_matrix, zscale = 1, fov = 0, theta = 170, phi = 30, windowsize = c(1000, 800), zoom = 0.35,

          water = TRUE, waterdepth = 75.2, wateralpha = 0.5, watercolor = "#88DDFF",

          waterlinecolor = "white", waterlinealpha = 0.5)



render_snapshot()

rgl::clear3d()



###########################################################################################

# Add water depth values to land raster layer ##############################################



# Resample to_harbour to match higher resolution of to_island

to_harbour_res <- resample(to_harbour, to_island, method = "bilinear")





# Identify points on to_island that are water

w <- which(to_island[] < 75.109)

length(w)



# Get coordinates of those points (water)

xy <- SpatialPoints(to_island)[w]



# Add CRS

proj4string(xy) = projection(to_island)



# These locations aren't in the coordinate system of the water raster so we need to transform them (WGS84 v NAD83)

xyt = spTransform(xy, projection(to_harbour))



# Get the water raster values at those points:

ex = raster::extract(to_harbour_res, xyt)



# Replace the locations in the land raster with the values sampled from from the water raster:

to_island[w] = ex



# Create matrix for 3D plotting

to_island_matrix = matrix(raster::extract(to_island, raster::extent(to_island), buffer = 10000), 

                          nrow = ncol(to_island), ncol = nrow(to_island))



# # 3D plot land WITH water depth info

to_island_matrix %>%

  sphere_shade(zscale = 0.1, texture = "imhof1") %>%  # Lower zscale maps points to wider range of colours

  plot_3d(to_island_matrix, zscale = 1, fov = 0, theta = 170, phi = 30, windowsize = c(1000, 800), zoom = 0.35,

          water = TRUE, waterdepth = 75.2, wateralpha = 0.5, watercolor = "#88DDFF",

          waterlinecolor = "white", waterlinealpha = 0.5)



render_snapshot()

rgl::clear3d()