Spatial interpolation (kriging and inverse distance weighting) for objects of class prevR.

These functions execute a spatial interpolation of a variable of the slot rings of an object of class prevR. The method krige() implements the ordinary kriging technique. The method idw() executes an inverse distance weighting interpolation.

# S4 method for ANY,prevR
krige(
  formula,
  locations,
  N = NULL,
  R = Inf,
  model = NULL,
  nb.cells = 100,
  cell.size = NULL,
  fit = "auto",
  keep.variance = FALSE,
  show.variogram = FALSE,
  ...
)

# S4 method for ANY,prevR
idw(
  formula,
  locations,
  N = NULL,
  R = Inf,
  nb.cells = 100,
  cell.size = NULL,
  idp = 2,
  ...
)

Arguments

formula: variable(s) to interpolate (see details).
locations: object of class prevR.
N: integer or list of integers corresponding to the rings to use.
R: integer or list of integers corresponding to the rings to use.
model: a variogram model returned by the function gstat::vgm().
nb.cells: number of cells on the longest side of the studied area (unused if cell.size is defined).
cell.size: size of each cell (in the unit of the projection).
fit: "auto" for using a variogram automatically fitted from the data, only if model is not defined (NULL).
keep.variance: return variance of estimates?
show.variogram: plot the variogram?
...: additional arguments transmitted to gstat::krige() or gstat::idw().
idp: inverse distance weighting power (see gstat::idw()).

Value

Object of class sf::sf. The name of estimated surfaces depends on the name of the interpolated variable, N and R (for example: r.radius.N300.RInf). If you ask the function to return variance (keep.variance=TRUE), corresponding surfaces names will have the suffix .var.

Details

formula specifies the variable(s) to interpolate. Only variables available in the slot rings of locations could be used. Possible values are "r.pos", "r.n", "r.prev", "r.radius", "r.clusters", "r.wpos", "r.wn" or "r.wprev". Variables could be specified with a character string or a formula (example: list(r.pos ~ 1, r.prev ~ 1}. Only formula like variable.name ~ 1 are accepted. For more complex interpolations, use directly functions gstat::krige() and gstat::idw() from gstat.

N and R determine the rings to use for the interpolation. If they are not defined, surfaces will be estimated for each available couples (N,R). Several interpolations could be simultaneously calculated if several variables and/or several values of N and R are defined.

A suggested value of N could be computed with Noptim().

In the case of an ordinary kriging, the method krige() from prevR will try to fit automatically a exponential variogram to the sample variogram (fit = "auto"). You can also specify directly the variogram to use with the parameter model.

Interpolations are calculated on a spatial grid obtained with make.grid.prevR().

Note

Results could be plotted with sf::plot() or with ggplot2 using ggplot2::geom_sf(). See examples.

prevR provides several continuous color palettes (see prevR.colors).

Results could be turned into a stars raster using stars::st_rasterize().

To export to ASCII grid, rasterize the results with stars::st_rasterize(), convert to SpatRast with terra::rast(), extract the desired layer with [[]] and then use terra::writeRaster(). See examples.

References

Larmarange Joseph, Vallo Roselyne, Yaro Seydou, Msellati Philippe and Meda Nicolas (2011) "Methods for mapping regional trends of HIV prevalence from Demographic and Health Surveys (DHS)", Cybergeo: European Journal of Geography, no 558, https://journals.openedition.org/cybergeo/24606, DOI: 10.4000/cybergeo.24606.

Examples

  if (FALSE) {
    dhs <- rings(fdhs, N = c(100,200,300,400,500))
    radius.N300 <- krige('r.radius', dhs, N = 300, nb.cells = 50)
    prev.krige <- krige(r.wprev ~ 1, dhs, N = c(100, 300, 500))

    plot(prev.krige, lty = 0)

    library(ggplot2)
    ggplot(prev.krige) +
      aes(fill = r.wprev.N300.RInf) +
      geom_sf(colour = "transparent") +
      scale_fill_gradientn(colors = prevR.colors.red()) +
      theme_prevR_light()

    # Export r.wprev.N300.RInf surface in ASCII Grid
    r <- terra::rast(stars::st_rasterize(prev.krige))
    # writeRaster(r[[2]], "wprev.N300.asc")
  }