Modify an example in ch4 and add a solution

[iod-ine]

The last exercise in chapter 4 does not have a solution at the moment. In e1c41fd @Nowosad commented it out with a to-do of improving/replacing/modifying it. I think it's a good exercise, so I added a solution to the first part, which asks to calculate distance to the coast for Spain.

I removed the second part that asks to calculate weighted distance, at least for now. I could not figure out how to complete this from the text of the chapter and the documentation of terra::distance. I am more than willing to explore it further if you can guide me in the right direction :)

I also include a reproducible example of the code I added as a solution.

library(sf)
#> Linking to GEOS 3.8.1, GDAL 3.2.1, PROJ 7.2.1
library(terra)
#> terra version 1.4.20

# Fetch the DEM data for Spain
spain_dem = raster::getData("alt", country = "ESP", mask = FALSE)
#> Warning in showSRID(uprojargs, format = "PROJ", multiline = "NO", prefer_proj
#> = prefer_proj): Discarded datum Unknown based on WGS84 ellipsoid in Proj4
#> definition

# Convert the DEM to terra format
spain_dem = rast(spain_dem)
#> Warning: Unhandled datatype=INT2S (GDAL error 1)

# Assign the correct CRS to the DEM: WGS 84
crs(spain_dem) = "EPSG:4326"

# Reduce the resolution by a factor of 20 to speed up calculations
spain_dem = aggregate(spain_dem, fact = 20)

# According to the documentation, terra::distance() will calculate distance
# for all cells that are NA to the nearest cell that are not NA. To calculate
# distance to the coast, we need a raster that has NA values over land and any
# other value over water:
water_mask = is.na(spain_dem)
water_mask[water_mask == 0] = NA

# Use the distance() function on this mask to get distance to the coast
distance_to_coast = distance(water_mask)

# Plot the result
plot(distance_to_coast, main = "Distance to the coast")

^{Created on 2021-11-27 by the reprex package (v2.0.1)}

[Nowosad]

Awesome!

One thing: could you use geodata::elevation_30s() instead of raster::getData()?

[iod-ine]

@Nowosad Yes! Now it looks a lot cleaner, with no warnings and data conversions.

Attaching a new reprex.

library(sf)
#> Linking to GEOS 3.8.1, GDAL 3.2.1, PROJ 7.2.1
library(terra)
#> terra version 1.4.20

# Fetch the DEM data for Spain
spain_dem = geodata::elevation_30s(country = "Spain", path = ".", mask = FALSE)

# Reduce the resolution by a factor of 20 to speed up calculations
spain_dem = aggregate(spain_dem, fact = 20)

# According to the documentation, terra::distance() will calculate distance
# for all cells that are NA to the nearest cell that are not NA. To calculate
# distance to the coast, we need a raster that has NA values over land and any 
# other value over water
water_mask = is.na(spain_dem)
water_mask[water_mask == 0] = NA

# Use the distance() function on this mask to get distance to the coast
distance_to_coast = distance(water_mask)

# Plot the result
plot(distance_to_coast, main = "Distance to the coast")

^{Created on 2021-11-27 by the reprex package (v2.0.1)}

[iod-ine]

I also removed unnecessary spaces at the beginning of the lines.

[Nowosad]

Great!

[Nowosad]

@iod-ine Ivan, I also added a two new related exercises along with their solutions - see e2ee8c7. Do you have any suggestion on how I can improve the question in E8?

[iod-ine]

@Nowosad I like both the formulation and the solution for E8 :)
E9, however, seems redundant: we are simply restoring the scaled DEM that we added in E8.
I would suggest either ending with E7 and E8 or, if we feel more practice is needed for terra::distance(), do something different. Maybe add a use case where we compute distances to a vector, like distance to the capitals for a couple of countries in Europe or something similar?

[Nowosad]

Thanks for the suggestions!

[Robinlovelace]

Many thanks 🙏