Learn R Programming

rvertnet (version 0.8.4)

vertmap: Make a simple map to visualize VertNet data.

Description

Plots record locations on a world or regional map using latitude/longitude data returned by a VertNet search.

Usage

vertmap(
  input = NULL,
  mapdatabase = "world",
  region = ".",
  geom = geom_point,
  jitter = NULL
)

Value

Map of record locations displayed on the selected base map

Arguments

input

Output from vertsearch, searchbyterm, or spatialsearch. Must include columns "decimallatitude" and "decimallongitude"

mapdatabase

The base map on which your data are displayed; what you choose here determines what you can choose in the region parameter; one of: county, state, usa, world, world2, france, italy, or nz

region

The region in which your data are displayed; to see region names for the "world" database layer, run sort(unique(map_data("world")$region)) after loading packages maps and ggplot2; to see region names for the US "state" layer, run sort(unique(map_data("state")$region))

geom

Specifies the type of object being plotted; one of: geom_point or geom_jitter (do not use quotes)

jitter

If geom = geom_jitter, the amount by which to jitter points in width, height, or both. Default

Details

vertmap uses decimal latitude and longitude data in records generated by an rvertnet search to display returned records on a specified base map. Taxa are color-coded by scientific name, if available. Adapt the vertmap code to construct maps according to your own specifications.

Examples

Run this code
if (FALSE) {
out <- vertsearch("Junco hyemalis") # get occurrence records
vertmap(out)                        # map occurrence records

# Records are color coded by dwc term "scientificname" - sometimes unavailble
out <- vertsearch("mustela nigripes")
vertmap(input = out, mapdatabase = "state")

# Use searchbyterm() to match records with mapped region
spec <- searchbyterm(genus = "ochotona", specificepithet = "princeps", state = "california",
limit = 200)
vertmap(input = spec, mapdatabase = "state", region = "california")

# Many species
splist <- c("Accipiter erythronemius", "Aix sponsa", "Haliaeetus leucocephalus",
		"Corvus corone", "Threskiornis molucca", "Merops malimbicus")
out <- lapply(splist, function(x) vertsearch(t=x, lim=100))
out <- dplyr::bind_rows(lapply(out, "[[", "data"))
vertmap(out)
## jitter points
library("ggplot2")
vertmap(out, geom = geom_jitter, jitter = position_jitter(1, 6))
}

Run the code above in your browser using DataLab