CAST (version 0.9.0)

plot_geodist: Plot euclidean nearest neighbor distances in geographic space or feature space

Description

Density plot of nearest neighbor distances in geographic space or feature space between training data as well as between training data and prediction locations. Optional, the nearest neighbor distances between training data and test data or between training data and CV iterations is shown. The plot can be used to check the suitability of a chosen CV method to be representative to estimate map accuracy. Alternatively distances can also be calculated in the multivariate feature space.

Usage

plot_geodist(
  x,
  modeldomain,
  type = "geo",
  cvfolds = NULL,
  cvtrain = NULL,
  testdata = NULL,
  samplesize = 2000,
  sampling = "regular",
  variables = NULL,
  unit = "m",
  stat = "density",
  showPlot = TRUE
)

Value

A list including the plot and the corresponding data.frame containing the distances. Unit of returned geographic distances is meters.

Arguments

x

object of class sf, training data locations

modeldomain

SpatRaster, stars or sf object defining the prediction area (see Details)

type

"geo" or "feature". Should the distance be computed in geographic space or in the normalized multivariate predictor space (see Details)

cvfolds

optional. list or vector. Either a list where each element contains the data points used for testing during the cross validation iteration (i.e. held back data). Or a vector that contains the ID of the fold for each training point. See e.g. ?createFolds or ?CreateSpacetimeFolds or ?nndm

cvtrain

optional. List of row indices of x to fit the model to in each CV iteration. If cvtrain is null but cvfolds is not, all samples but those included in cvfolds are used as training data

testdata

optional. object of class sf: Data used for independent validation

samplesize

numeric. How many prediction samples should be used?

sampling

character. How to draw prediction samples? See spsample. Use sampling = "Fibonacci" for global applications.

variables

character vector defining the predictor variables used if type="feature. If not provided all variables included in modeldomain are used.

unit

character. Only if type=="geo" and only applied to the plot. Supported: "m" or "km".

stat

"density" for density plot or "ecdf" for empirical cumulative distribution function plot.

showPlot

logical

Author

Hanna Meyer, Edzer Pebesma, Marvin Ludwig

Details

The modeldomain is a sf polygon or a raster that defines the prediction area. The function takes a regular point sample (amount defined by samplesize) from the spatial extent. If type = "feature", the argument modeldomain (and if provided then also the testdata) has to include predictors. Predictor values for x are optional if modeldomain is a raster. If not provided they are extracted from the modeldomain rasterStack.

See Also

nndm

Examples

if (FALSE) {
library(sf)
library(terra)
library(caret)

########### prepare sample data:
dat <- readRDS(system.file("extdata","Cookfarm.RDS",package="CAST"))
dat <- aggregate(dat[,c("DEM","TWI", "NDRE.M", "Easting", "Northing")],
  by=list(as.character(dat$SOURCEID)),mean)
pts <- st_as_sf(dat,coords=c("Easting","Northing"))
st_crs(pts) <- 26911
pts_train <- pts[1:29,]
pts_test <- pts[30:42,]
studyArea <- terra::rast(system.file("extdata","predictors_2012-03-25.tif",package="CAST"))
studyArea <- studyArea[[c("DEM","TWI", "NDRE.M", "NDRE.Sd", "Bt")]]

########### Distance between training data and new data:
dist <- plot_geodist(pts_train,studyArea)

########### Distance between training data, new data and test data:
#mapview(pts_train,col.regions="blue")+mapview(pts_test,col.regions="red")
dist <- plot_geodist(pts_train,studyArea,testdata=pts_test)

########### Distance between training data, new data and CV folds:
folds <- createFolds(1:nrow(pts_train),k=3,returnTrain=FALSE)
dist <- plot_geodist(x=pts_train, modeldomain=studyArea, cvfolds=folds)

## or use nndm to define folds
AOI <- as.polygons(rast(studyArea), values = F) |>
  st_as_sf() |>
  st_union() |>
  st_transform(crs = st_crs(pts_train))
nndm_pred <- nndm(pts_train, AOI)
dist <- plot_geodist(x=pts_train, modeldomain=studyArea,
    cvfolds=nndm_pred$indx_test, cvtrain=nndm_pred$indx_train)

########### Distances in the feature space:
plot_geodist(x=pts_train, modeldomain=studyArea,
    type = "feature",variables=c("DEM","TWI", "NDRE.M"))

dist <- plot_geodist(x=pts_train, modeldomain=studyArea, cvfolds = folds, testdata = pts_test,
    type = "feature",variables=c("DEM","TWI", "NDRE.M"))

############ Example for a random global dataset
############ (refer to figure in Meyer and Pebesma 2022)
library(sf)
library(rnaturalearth)
library(ggplot2)

### Define prediction area (here: global):
ee <- st_crs("+proj=eqearth")
co <- ne_countries(returnclass = "sf")
co.ee <- st_transform(co, ee)

### Simulate a spatial random sample
### (alternatively replace pts_random by a real sampling dataset (see Meyer and Pebesma 2022):
sf_use_s2(FALSE)
pts_random <- st_sample(co.ee, 2000, exact=FALSE)

### See points on the map:
ggplot() + geom_sf(data = co.ee, fill="#00BFC4",col="#00BFC4") +
     geom_sf(data = pts_random, color = "#F8766D",size=0.5, shape=3) +
     guides(fill = FALSE, col = FALSE) +
     labs(x = NULL, y = NULL)

### plot distances:
dist <- plot_geodist(pts_random,co.ee,showPlot=FALSE)
dist$plot+scale_x_log10(labels=round)
}