Description Usage Arguments Details Value Examples
Sites don't need to be gridded or random; defines neighbors and calculates local spatial autocorrelation as Moran's I
1 | spatial_ac(lon, lat, value)
|
lon |
longitude |
lat |
latitude |
value |
value of interest for which to calculate AC |
The neighborhood is defined in a 2-step process. First, the nearest neighbor for each location is found, and the maximum distance among nearest neighbors recorded. Second, the neighborhood for a focal point is then defined as all points within that distance. So it's defined as all points within a raidus with a size that is the minimum distance required for all points to have at least 1 neighbor.
Local Moran's I is calculated for each site. P value is adjusted to number of sites in each "region" (not the same 'region' used in the trawlDiversity package); see spdep::localmoran
.
A named list of length 3. The first element of this list, max2NDist
is the minimum distance and is a length 1 numeric vector. The second element, nb
is the neighborhood object returned by spdep::knn2nb
. The third element, I
, is a data.table that contains columns as returned by spdep::localmoran
with a few adjustments: the column for the pvalue is changed to lI_pvalue
and columns for lon
and lat
of the sites are added.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | # calculate local I on random subset of volcano
data(volcano)
lon <- rep(seq_len(ncol(volcano)), each=nrow(volcano))
lat <- rep(seq_len(nrow(volcano)), ncol(volcano))
rind <- sample(seq_along(lon), 100) # random subset
value <- c(volcano)
spac <- spatial_ac(lon[rind], lat[rind], value[rind])
# plot neighborhood
plot(spac[["nb"]], cbind(lon[rind], lat[rind]))
# plot significant I's
sig_ind <- spac[["I"]][,lI_pvalue] < 0.05
plot(spac[["nb"]], cbind(lon[rind], lat[rind]))
spac[["I"]][sig_ind, points(lon, lat, col=rbLib::zCol(256, Ii), pch=19)]
rbLib::mapLegend(zlim=spac[["I"]][sig_ind,range(Ii)], cols=rbLib::zCol(6, 1:6))
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.