(CC-BY-NC-SA)

This document was produced on So Jun 12 2016 using mapview version 1.1.0

mapview is intended to provide a quick and easy way to visualize/plot spatial data in an interactive manner. As such, a one-liner is enough to produce an interactive map view of the data. Methods are defined for all objects from packages sp, raster and satellite:

Vector data

mapview includes 3 vector type data sets:

  • breweries91 - a SpatialPointsDataFrame of a selection of (micro-) breweries in Franconia, the region with the highest brewery density in the world.
  • gadmCHE - a SpatialPolygonsDataFrame of administrative boundaries fro Switzerland from gadm
  • atlStorms2005 - a SpatialLinesDataFrame of selected storm tracks in the Atlantic Ocean from 2005

For vector data a call to mapview() without any further arguments will produce a default map view including:

  • a layers control to switch between 5 different background maps - the default of which is CartoDB.Positron as it provides a color neutral environment for the visulizes features
  • zoom control buttons provided by the underlying leaflet map
  • attribution information of the active map layer in the bottom right corener of the map
  • information on mouse position (in latitude and longitude) and zoom level of the current view
  • in case you have the latest version of leaflet installed the map will also have a scale bar in the bottom left corner and provide labels of feature IDs if you hover over the features.
  • clicking on any feature will provide a standard table popup listing all attribute entries of the respective features.

Here’s an example for each vector type:

SpatialPoints(DataFrame)

library(mapview)
library(sp)

mapview(breweries91)

SpatialPolygons(DataFrame)

mapview(gadmCHE)
## Warning in spCheckObject(x): Columns CCN_1 in attribute table contain only
## NA values and are dropped.

SpatialLines(DataFrame)

mapview(atlStorms2005)

Controlling attribute columns

If we only want to plot certain columns of the attribute table we can use argument zcol. Each column will be rendered as a separate layer.

library(mapview)
library(sp)

mapview(breweries91, zcol = c("brewery", "village", "founded"))

Including legends

We can also include legends for the layers. Be careful, though, as it is currently not possible to link legends to layers so we end up with too many legends to fit the viewer window if we want to visualize many layers. This is also the reason why legends are not shown by default.

library(mapview)
library(sp)

mapview(breweries91, zcol = "founded", legend = TRUE)

Raster data

mapview includes 2 raster type data sets:

  • poppendorf - a RasterBrick including 5 bands of a landsat scene located in the same region as breweries91
  • kiliNDVI - a raw multiband raster data set of 23 16-day Aqua-MODIS NDVI layers of Mt. Kilimanajro in Tanzania for the year 2013. See Detsch et al. 2016 for details on how this data set was created

For raster data a call to mapview() without any further arguments will produce a default map view including:

  • a layers control to switch between 5 different background maps - the default of which is CartoDB.Positron as it provides a color neutral environment for the visulizes features
  • zoom control buttons provided by the underlying leaflet map
  • attribution information of the active map layer in the bottom right corener of the map
  • information on mouse position (in latitude and longitude) and zoom level of the current view
  • in case you have the latest version of leaflet installed the map will also have a scale bar in the bottom left corner and provide labels of feature IDs if you hover over the features
  • all cells containing NAs are rendered in light transparent grey (#BEBEBE80)
  • clicking on raster cells to get the respective cell values is currently not available, but we are working on a solution for this

Here’s an example for each raster type (including a sp::SpatialPixelsDataFrame):

RasterLayer

library(mapview)
library(raster)

raster_layer <- poppendorf[[5]]

mapview(raster_layer)

RasterStack/Brick

This will produce one map view layer for each layer in the stack/brick. Use the layers control to switch between the layers. By default only the first layer is shown.

library(mapview)
library(raster)

mapview(poppendorf)

SpatialGridDataFrame

Depending on the number of attribute columns, this is either rendered as a RasterLayer or RasterBrick

library(mapview)
library(raster)
library(sp)

# SpatialPixelsDataFrame
data(meuse.grid)
coordinates(meuse.grid) <- ~x+y
proj4string(meuse.grid) <- CRS("+init=epsg:28992")
gridded(meuse.grid) <- TRUE

mapView(meuse.grid)

Similar to the vector data functionality, if we want to render a certain column of the attribute table, we can use the zcol argument

library(mapview)
library(raster)
library(sp)

# SpatialPixelsDataFrame
data(meuse.grid)
coordinates(meuse.grid) <- ~x+y
proj4string(meuse.grid) <- CRS("+init=epsg:28992")
gridded(meuse.grid) <- TRUE

mapView(meuse.grid, zcol = "soil")

Identical to vector data plots use legend = TRUE to add legend(s) to raster visualizations.

library(mapview)
library(raster)

raster_layer <- poppendorf[[5]]

mapview(raster_layer, legend = TRUE)