C ONTRIBUTED A RTICLE

ggmap : Spatial Visualization with

ggplot2
30.5
30.0
29.5
29.0

Abstract In spatial statistics the ability to visualize data and models superimposed with their
basic social landmarks and geographic context
is invaluable. ggmap is a new tool which enables such visualization by combining the spatial
information of static maps from Google Maps,
OpenStreetMap, Stamen Maps or CloudMade
Maps with the layered grammar of graphics implementation of ggplot2. In addition, several
new utility functions are introduced which allow
the user to access the Google Geocoding, Distance Matrix, and Directions APIs. The result is
an easy, consistent and modular framework for
spatial graphics with several convenient tools for
spatial data analysis.

used to seeing areal plots in R which resemble Figure 1.

df$lat

by David Kahle and Hadley Wickham

-96.0

-95.5

-95.0

-94.5

df$lon

Introduction

30.0

30.5

While these kinds of plots are useful, they are not as

informative as we would like in many situations. For
instance, when plotting zip codes it is helpful to also
see major roads and other landmarks which form the
boundaries of areal units.
The situation for point referenced spatial data is
often much worse. Since we cant easily contextualize a scatterplot of points without any background
information at all, it is common to add points as an
overlay of some areal datawhatever areal data is
available. The resulting plot looks like Figure 2.

29.5
29.0

Plotting spatial data in R

Figure 1: Typical R areal plot zip codes in the

Greater Houston area

df$lat

Visualizing spatial data in R can be a challenging

task. Fortunately the task is made a good deal easier by the data structures and plot methods of sp,
RgoogleMaps, and related packages (Pebesma and
Bivand, 2006), (Bivand, Pebesma, and Gomez-Rubio,
2008), (Loecher and Sense Networks, 2010). Using
those methods, one can plot the basic geographic information of (for instance) a shape file containing
polygons for areal data or points for point referenced
data. However, compared to specialized geographic
information systems (GISs) such as ESRIs ArcGIS,
which can plot points, polygons, etc. on top of maps
and satellite imagery with drag-down menus, these
visualizations can be pretty disappointing. This article details some new methods for the visualization of spatial data in R using the layered grammar
of graphics implementation of ggplot2 in conjunction with the contextual information of static maps
from Google Maps, OpenStreetMap, Stamen Maps or
Cloudmade Maps (Wickham, 2009) (Wickham, 2010).
The result is an easy to use R package named ggmap.
After describing the nuts and bolts of ggmap, we
showcase some of its capabilities in a simple case
study concerning violent crimes in downtown Houston, Texas and present an overview of a few utility
functions.

-96.0

Areal data is data which corresponds to geographical

extents with polygonal boundaries. A typical example is the number of residents per zip code. Considering only the boundaries of the areal units, we are
The R Journal Vol. X/Y, Month, Year

-95.5

-95.0

-94.5

df$lon

Figure 2: Typical R spatial scatterplot murders in

Houston from January 2010 to August 2010
ISSN 2073-4859

C ONTRIBUTED A RTICLE

In most cases the plot is understandable to the researcher who has worked on the problem for some
time but is of hardly any use to his audience, who
must work to associate the data of interest with their
location. Moreover, it leaves out many practical
detailsare most of the events to the east or west of
landmark x? Are they clustered around more wellto-do parts of town, or do they tend to occur in disadvantaged areas? Questions like these cant really
be answered using these kinds of graphics because
we dont think in terms of small scale areal boundaries (e.g. zip codes or census tracts).
This is of course a (gross) exaggeration. With a
little effort better plots can be made, and tools such
as maps, maptools, sp, and RgoogleMaps make the
process much easier; in fact, RgoogleMaps was the
inspiration for ggmap. Moreover there has recently
been a deluge of interest in the subject of mapmaking
in RIan Fellows excellent interactive GUI-driven
DeducerSpatial package based on Bing Maps comes
to mind. ggmap takes another step in this direction
by situating the contextual information of various
kinds of static maps in the ggplot2 plotting framework. The result is an easy, consistent way of specifying plots which are readily interpretable by both
expert and audience and safeguarded from graphical inconsistencies by the layered grammar of graphics framework. The result is a spatial plot resembling
Figure 3.
murder <- subset(crime, offense == "murder")
qmplot(lon, lat, data = murder,
colour = I('red'), size = I(3), darken = .3)

One advantage of making the plots with ggplot2 is

the layered grammar of graphics on which ggplot2
is based (Wickham, 2010) (Wilkinson, 2005). By definition, the layered grammar demands that every plot
consist of five components :
a default dataset with aesthetic mappings,
one or more layers, each with a geometric
object ("geom"), a statistical transformation
("stat"), and a dataset with aesthetic mappings
(possibly defaulted),
a scale for each aesthetic mapping (which can
be automatically generated),
a coordinate system, and
a facet specification.
Since ggplot2 is an implementation of the layered
grammar of graphics, every plot made with ggplot2 has each of the above elements. Consequently,
ggmap plots also have these elements, but certain elements are fixed to map components : the x aesthetic
is fixed to longitude, the y aesthetic is fixed to latitude, and the coordinate system is fixed to the Mercator projection.1
The major theoretical advantage of using the layered grammar in plotting maps is that aesthetic
scales are kept consistent. In the typical situation
where the map covers the extent of the data, in
ggmap the latitude and longitude scales key off the
map (by default) and one scale is used for those
axes. The same is true of colors, fills, alpha blendings, and other aesthetics which are built on top of
the map when other layers are presentedeach is allotted one scale which is kept consistent across each
layer of the plot. This aspect of the grammar is particularly important for faceted plots in order to make
a proper comparison across several plots. Of course,
the scales can still be tricked if the user improperly
specifies the spatial data, e.g. using more than one
projection in the same map, but fixing such errors is
beyond any framework.
The practical advantage of using the grammar is
even better. Since the graphics are done in ggplot2
the user can draw from the full range of ggplot2s
capabilities to layer elegant visual contentgeoms,
stats, scales, etc.using the usual ggplot2 coding
conventions. This was already seen briefly in Figure
3 where the arguments of qmplot are identical to that
of ggplot2s qplot; much more will be seen shortly.

Figure 3: A spatial scatterplot based on Stamen

Maps terrain tile set made with the qmplot function
1 Note

The layered grammar of graphics

How ggmap works

The basic idea driving ggmap is to take a downloaded map image, plot it as a context layer using gg-

that because of the Mercator projection limitations in mapproject, anything above/below 80o cannot be plotted currently.

The R Journal Vol. X/Y, Month, Year

ISSN 2073-4859

C ONTRIBUTED A RTICLE

plot2, and then plot additional content layers of data,

statistics, or models on top of the map. In ggmap this
process is broken into two pieces (1) downloading
the images and formatting them for plotting, done
with get_map, and (2) making the plot, done with
ggmap. qmap marries these two functions for quick
map plotting (c.f. ggplot2s ggplot), and qmplot attempts to wrap up the entire plotting process into
one simple command (c.f. ggplot2s qplot).

The get_map function

In ggmap, downloading a map as an image and
formatting the image for plotting is done with the
get_map function. More specifically, get_map is
a wrapper function for the underlying functions
get_googlemap, get_openstreetmap, get_stamenmap,
and get_cloudmademap which accepts a wide array
of arguments and returns a classed raster object for
plotting with ggmap.
As the most important characteristic of any map
is location, the most important argument of get_map
is the location argument. Ideally, location is a longitude/latitude pair specifying the center of the map
and accompanied by a zoom argument, an integer
from 3 to 20 specifying how large the spatial extent
should be around the center, with 3 being the continent level and 20 being roughly the single building
level. location is defaulted to downtown Houston,
Texas, and zoom to 10, roughly a city-scale.
While longitude/latitude pairs are ideal for specifying a location, they are somewhat inconvenient on
a practical level. For this reason, location also accepts a character string. The string, whether containing an address, zip code, or proper name, is
then passed to the geocode function which then determines the appropriate longitude/latitude coordinate for the center. In other words, there is no need
to know the exact longitude/latitude coordinates of
the center of the mapget_map can determine them
from more colloquial ("lazy") specifications so that
they can be specified very loosely. For example, since

necessary) and passing them to each of the API specific get_* functions. To ensure that the resulting
maps are the same across the various sources for
the same location/zoom specification, get_map first
grabs the appropriate Google Map, determines its
bounding box, and then downloads the other map as
needed. In the case of Stamen Maps and Cloudmade
Maps, this involves a stitching process of combining
several tiles (small map images) and then cropping
the result to the appropriate bounding box. The result is a single, consistent specification syntax across
the four map sources as seen for Google Maps and
OpenStreetMap in Figure 4.
baylor <- "baylor university"
qmap(baylor, zoom = 14)

qmap(baylor, zoom = 14, source = "osm")

> geocode("the white house")

lon
lat
-77.03676 38.89784
works, "the white house" is a viable location argument. More details on geocode and other utility
functions are discussed at the end of this article.
In lieu of a center/zoom specification, some users
find a bounding box specification more convenient.
To accommodate this form of specification, location
also accepts numeric vectors of length four following
the left/bottom/right/top convention. This option
is not currently available for Google Maps.
While each map source has its own web
application programming interface (API), the
location/zoom specification in get_map works for
each by computing the appropriate parameters (if
The R Journal Vol. X/Y, Month, Year

Figure 4: get_map provides the same spatial extent

for Google Maps (top) and OpenStreetMaps (bottom)
with a single simple syntax, even though their APIs
are quite different
ISSN 2073-4859

C ONTRIBUTED A RTICLE

Before moving into the source and maptype arguments, it is important to note that the underlying API
specific get_* functions for which get_map is a wrapper provide more extensive mechanisms for downloading from their respective sources. For example,
get_googlemap can access almost the full range of the
Google Static Maps API as seen in Figure 5.

other hand, only provides the default style shown in

Figure 4.
Style is where Stamen Maps and Cloudmade
Maps really shine. Stamen Maps has three available
tile setsterrain (e.g. Figures 3 or 16), watercolor,
and toner.
qmap(baylor, zoom = 14, source = "stamen",
maptype = "watercolor")

set.seed(500)
df <- round(data.frame(
x = jitter(rep(-95.36, 50), amount = .3),
y = jitter(rep( 29.76, 50), amount = .3)
), digits = 2)
map <- get_googlemap('houston', markers = df,
path = df, scale = 2)
ggmap(map, extent = 'device')

qmap(baylor, zoom = 14, source = "stamen",

maptype = "toner")

Figure 5: Accessing Google Maps API features with

get_googlemap
Tile style - the source and maptype arguments of
get_map
The most attractive aspect of using different map
sources (Google Maps, OpenStreetMap, Stamen
Maps, and Cloudmade Maps) is the different map
styles provided by the producer. These are specified with the maptype argument of get_map and must
agree with the source argument. Some styles emphasize large roadways, others bodies of water, and still
others political boundaries. Some are better for plotting in a black-and-white medium; others are simply
nice to look at. This section gives a run down of the
various map styles available in ggmap.
Google provides four different familiar types
terrain (default), satellite (e.g. Figure 16), roadmap,
and hybrid (e.g. Figure 15). OpenStreetMap, on the
The R Journal Vol. X/Y, Month, Year

Figure 6: Stamen tile sets maptype = "watercolor"

and maptype = "toner"
Stamens terrain tile set is quite similar to
Googles, but obviously the watercolor and toner tile
sets are substantially different than any of the four
ISSN 2073-4859

C ONTRIBUTED A RTICLE

Google tile sets. The latter, for example, is ideal for

black-and-white plotting.

qmap("houston", zoom = 10, maptype = 58916,

api_key = api_key, source = "cloudmade")

Cloudmade Maps takes the tile styling even further by allowing the user to either (1) select among
thousands of user-made sets or (2) create an entirely
new style with a simple online editor where the
user can specify colors, lines, and so forth for various types of roads, waterways, landmarks, etc.,
all of which are generated by CloudMade and accessible in ggmap. ggmap, through get_map (or
get_cloudmademap) allows for both options. This is
a unique feature of CloudMade Maps which really
boosts their applicability and expands the possibilities with ggmap. The one minor drawback to using CloudMade Maps is that the user must register
with CloudMade to obtain an API key and then pass
the API key into get_map with the api_key argument.
API keys are free of charge and can be obtained in a
matter of minutes. Two low-light CloudMade map
styles are seen in Figure 7.

Both Stamen Maps and CloudMade Maps are

built using OpenStreetMap data. These data are
contributed by an open community of online users
in much the same way Wikipedia isboth are free,
both are user-contributed, and both are easily edited.
Moreover, OpenStreetMap has data not only on
roadways and bodies of water but also individual
buildings, fountains, stop signs and other apparent
minutiae. The draw back is that (like Google Maps)
not all locations are mapped with the same degree of
precision. For example, entire lakes might be missing
from the tile sets.
qmap(baylor, zoom = 14, maptype = 53428,
api_key = api_key, source = "cloudmade")

Figure 7: Two out of thousands of user made CloudMade Maps styles. The top is comparable to Figures
4 and 6, and the bottom is the bodies of water in Figure 5

The ggmap function

Once get_map has grabbed the map of interest, ggmap
is ready to plot it. The result of get_map is a specially
classed raster object (a matrix of colors as hexadecimal character strings)
> paris <- get_map(location = "paris")
> str(paris)
chr [1:1280, 1:1280] "#C6DAB6" "#C2D6B3" "#C2D6B3" ...
- attr(*, "class")= chr [1:2] "ggmap" "raster"
- attr(*, "bb")='data.frame': 1 obs. of 4 variables:
..$ ll.lat: num 48.6
..$ ll.lon: num 1.91
..$ ur.lat: num 49.1
..$ ur.lon: num 2.79

The purpose of ggmap is to take the map from the

raster object to the screen, and it fulfills this purpose
by creating a ggplot object which, when printed,
draws the desired map in the graphics device. This
is illustrated in Figure 8.
While ggmap requires a ggmap object, it accepts a handful of other arguments as wellextent,
base_layer, maprange, legend, padding, and darken.
With none of these additional arguments, ggmap effectively returns the following ggplot object
ggplot(aes(x=lon, y=lat), data=fourCorners) +
geom_blank() + coord_map("mercator") +
annotation_raster(ggmap,
xmin, xmax, ymin, ymax)
where fourCorners is the data frame resulting from
applying expand.grid to the longitude and latitude
The R Journal Vol. X/Y, Month, Year

ISSN 2073-4859

C ONTRIBUTED A RTICLE

ranges specified in the bb attribute of the ggmap object. Thus, the default base layer of the ggplot2 object
created by ggmap is ggplot(aes(x=lon,y=lat),data
= fourCorners), and the default x and y aesthetic
scales are calculated based on the longitude and latitude ranges of the map.

49.1

49.0

lat

48.9

48.8

48.7

48.6

2.0

2.2

2.4

2.6

2.8

lon

Figure 8: ggmap(paris, extent = "normal")

The extent argument dictates how much of the
graphics device is covered by the map. It accepts
three possible strings: "normal" shown in Figure 8,
"panel" shown in Figures 13 and 15, and "device"
shown in every other figure. "normal" situates the
map with the usual axis padding provided by ggplot2 and, consequently, one can see the panel behind it. "panel" eliminates this, setting the limits of
the plot panel to be the longitude and latitude extents
of the map with scale_[x,y]_continuous(expand =
c(0,0)). "device" takes this to the extreme by eliminating the axes themselves with the new exported
theme_nothing.
base_layer is a call which substitutes the
default base layer to the users specification.
Thus, in the above code the user can change
ggplot(aes(x=lon,y=lat),data=fourCorners) to a
different call. This is essential for faceting plots
since the referent of ggplot2 functions facet_wrap
and facet_grid is the base layer. Since changing the
base layer changes the base scales and therefore limits of the plot, it is possible that when the base layer
is changed only part of the map is visible. Setting
the maprange argument to TRUE (it defaults to FALSE)
ensures that the map determines the x and y axis limits (longitude and latitude) via the bb attribute of the
ggmap object itself, and not the base_layer argument.
The legend-related arguments of ggmap are
legend and padding, and they are only applicable
when extent = "device". The legend argument deThe R Journal Vol. X/Y, Month, Year

termines where a legend should be drawn on the

map if one should be drawn. Its options are "left",
"right" (default), "bottom", "top", "topleft",
"bottomleft", "topright", "bottomright" and
"none". The first four draw the legend according to
ggplot2s normal specifications (without any axes);
the next four draw the legend on top of the map itself
similar to ArcGIS; and the last eliminates the legend
altogether. padding governs how far from the corner
the legend should be drawn when on top of the map.
The darken argument, a suggestion by JeanOlivier Irisson, tints the map image. The default,
c(0,"black"), indicates a fully translucent black
layer, i.e. no tint at all. Generally, the first argument
corresponds to an alpha blending (0 = invisible, 1 =
opaque) and the second argument the color of the
tint. If only a number is supplied to the darken argument ggmap assumes a black tint. The tint itself is
made by adding a geom_rect layer on top of the map.
An example is provided by Figure 3, where a black
tint was added to the map to enhance the visibility
of the points.
Since ggmap returns a ggplot object, the product
of ggmap can itself act as a base layer in the ggplot2
framework. This is an incredibly important realization which allows for the full range of ggplot2 capabilities. We now illustrate many of the ways in which
this can be done effectively through a case study of
violent crime in downtown Houston, Texas.

Data
Crime data were compiled from the Houston Police Departments website over the period of January
2010-August 2010. The data were lightly cleaned and
aggregated using plyr and geocoded using Google
Maps (to the center of the block, e.g., 6150 Main St.);
the full data set is available in ggmap as the data set
crime.
> str(crime)
'data.frame': 86314 obs. of 17 variables:
$ time
: POSIXt, format: "2010-01-01 0...
$ date
: chr "1/1/2010" "1/1/2010" "1...
$ hour
: int 0 0 0 0 0 0 0 0 0 0 ...
$ premise : chr "18A" "13R" "20R" "20R" ...
$ offense : chr "murder" "robbery" "aggr...
$ beat
: chr "15E30" "13D10" "16E20" ...
$ block
: chr "9600-9699" "4700-4799" ...
$ street : chr "marlive" "telephone" "w...
$ type
: chr "ln" "rd" "ln" "st" ...
$ suffix : chr "-" "-" "-" "-" ...
$ number : int 1 1 1 1 1 1 1 1 1 1 ...
$ month
: Factor w/ 12 levels "january"...
$ day
: Factor w/ 7 levels "monday" ...
$ location: chr "apartment parking lot" ...
$ address : chr "9650 marlive ln" "4750 ...
$ lon
: num -95.4 -95.3 -95.5 -95.4 ...
$ lat
: num 29.7 29.7 29.6 29.8 29.7...
ISSN 2073-4859

C ONTRIBUTED A RTICLE

Since we are only interested in violent crimes which

take place downtown, we restrict the data set to those
qualifiers. To determine a bounding box, we first
use gglocator, a ggplot2 analogue of bases locator
function exported from ggmap. gglocator works
not only for ggmap plots, but ggplot2 graphics in
general.

Offense

# order violent crimes

violent_crimes$offense <- factor(
violent_crimes$offense, levels =
c("robbery", "aggravated assault",
"rape", "murder"))
# restrict to downtown
violent_crimes <- subset(violent_crimes,
-95.39681 <= lon & lon <= -95.34188 &
29.73631 <= lat & lat <= 29.78400)
The longitude and latitude in the last few lines
were obtained using the new exported function
gglocatora ggplot2 analogue of the graphics function locatoron a larger map.
The analysis performed only concerns data on the
violent crimes of aggravated assault, robbery, rape
and murder. Note that while some effort was made
to ensure the quality of the data, the data were only
leisurely cleaned and the data set may still contain
errors.

Murder

# only violent crimes

violent_crimes <- subset(crime,
offense != "auto theft" &
offense != "theft" & offense != "burglary")

Rape

Aggravated Assault

Robery

> # find a reasonable spatial extent

> qmap('houston', zoom = 13)
> gglocator(2)
lon
lat
1 -95.39681 29.78400
2 -95.34188 29.73631

Figure 9: Violent crime bubble chart of downtown

Houston
One of the problems with the bubble chart is
overplotting and point sizewe cant really get a feel
for what crimes are taking place and where. One way
around this is to bin the points and drop the bins
which dont have any samples in them. The result
(Figure 10) shows us where the crimes are happening at the expense of knowing their frequency.
HoustonMap +
stat_bin2d(
aes(x = lon, y = lat, colour = offense,
fill = offense),
size = .5, bins = 30, alpha = 1/2,
data = violent_crimes)
Offense
Robery
Aggravated Assault
Rape
Murder

Analysis ggmap in action

The first step we might want to take is to look at
where the individual crimes took place. Modulo
some simple ggplot2 styling changes (primarily in
the fonts and key-styles of the legends via ggplot2s
guide function), the following code produces the
spatial bubble chart in Figure 9.
theme_set(theme_bw(16))
HoustonMap <- qmap("houston", zoom = 14,
color = "bw", legend = "topleft")
HoustonMap +
geom_point(aes(x = lon, y = lat,
colour = offense, size = offense),
data = violent_crimes)
The R Journal Vol. X/Y, Month, Year

Figure 10: Contour plot of violent crimes by offense

type
ISSN 2073-4859

C ONTRIBUTED A RTICLE

Figure 10 is the first time we really begin to see

the power of having the maps in the ggplot2 framework. While it is actually not a very good plot,
it illustrates the practical advantage of the ggplot2
framework with the contextual information of the
mapsplitting the data frame violent_crimes into
chunks based on the offense variable, binning the
points of each, and aggregating back into one data set
to plot is all done entirely behind the scenes by ggplot2. ggplot2 has several such canned procedures.
What about violent crime in general? If we neglect the type of offense, we can get a good idea of the
spatial distribution of violent crimes by using a contour plot. Since the map image itself is based on ggplot2s new annotation_raster, which doesnt have
a fill aesthetic, we can access the fill aesthetic to make
a filled contour plot.
houston <- get_map('houston", zoom = 14)
HoustonMap <- ggmap("houston",
extent = "device", legend = "topleft")

overlay <- stat_density2d(

aes(x = lon, y = lat, fill = ..level..,
alpha = ..level..),
bins = 4, geom = "polygon",
data = violent_crimes)
HoustonMap + overlay + inset(
grob = ggplotGrob(ggplot() +
overlay + theme_inset()),
xmin = -95.35836, xmax = Inf,
ymin = -Inf, ymax = 29.75062
)
Violent
Crime
Density
1400
1200
1000
800
600
400

HoustonMap +
stat_density2d(
aes(x = lon, y = lat, fill = ..level..,
alpha = ..level..),
size = 2, bins = 4, data = violent_crimes,
geom = "polygon")
Violent
Crime
Density
1400
1200
1000

Figure 12: Filled contour plot with insert

800
600
400

Figure 11: Filled contour plot of violent crimes

These kinds of overlays can be incredibly effective; however, their ability to communicate information can be hindered by the fact that the map overlay
can be visually confused with the map itself. This is
particularly common when using colored maps. To
get around this problem the inset function can be
used to insert map insets containing the overlay on
a white background. The viewing is made easier by
the exported function theme_inset.
The R Journal Vol. X/Y, Month, Year

The image indicates that there are three main

hotspots of activity. Each of these three correspond to
locations commonly held by Houstonians to be particularly dangerous locations. From east to west, the
hotspots are caused by 1. a county jail which releases
its inmates twice daily, who tend to loiter in the area
indicated, 2. a commercial bus station in an area of
substantial homelessness and destitution, and 3. a
prostitution hotspot in a very diverse and pedestrian
part of town.
In addition to single plots, the base_layer argument to ggmap or qmap allows for faceted plots. This
is particularly useful for spatiotemporal data with
discrete temporal components (day, month, season,
year, etc.).
houston <- get_map(location = "houston",
zoom = 14, color = "bw", source = "osm")
HoustonMap <- ggmap(houston,
base_layer = ggplot(aes(x = lon, y = lat),
data = violent_crimes))
HoustonMap +
stat_density2d(aes(x = lon, y = lat,
ISSN 2073-4859

C ONTRIBUTED A RTICLE

fill = ..level.., alpha = ..level..),

bins = 5, geom = "polygon",
data = violent_crimes) +
scale_fill_gradient(low = "black",
high= "red") +
facet_wrap(~ day)
Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

29.78
29.77
29.76
29.75
29.74

Latitude

29.78

Violent
Crime
Density
1500

29.77
29.76

1000
29.75
29.74

500
Sunday

29.78
29.77
29.76
29.75
29.74
95.39 95.37 95.35

Longitude

Figure 13: Faceted filled contour plot by day

This last plot displays one of the known issues
with contour plots in ggplot2a "clipping" or "tearing" of the contours. Aside from that fact (which will
likely be fixed in subsequent ggplot2 versions), we
can see that in fact most violent crimes happen on
Monday, with a distant second being Friday. Fridays
pattern is easily recognizablea small elevated dot
in the downtown bar district and an expanded region to the southwest in the district known as midtown, which has an active nightlife. Mondays pattern is not as easily explainableit may be that more
jail inmates are released on Monday who had weekend stays.

ggmaps utility functions

ggmap has several utility functions which aid in spatial exploratory data analysis.
geocode
The ability to move from an address to a longitude/latitude coordinate is virtually a must for visualizing spatial data. Unfortunately however, the
process is almost always done outside R by using a
proper geographic information system (GIS), saving
the results, and importing them into R. The geocode
function simplifies this process to a single line in R.
geocode is a vectorized function which accepts
character strings and returns a data frame of geographic information. In the default case of output
The R Journal Vol. X/Y, Month, Year

= "simple", only longitudes and latitudes are returned. These are actually Mercator projections of
the ubiquitous unprojected 1984 world geodetic system (WGS84), a spheroidal earth model used by
Google Maps. When output is set to "more", a larger
data frame is returned which provides much more
Google Geocoding information on the query :
> geocode("baylor university", output = "more")
lon
lat
type
loctype
1 -97.11441 31.54872 university approximate
address
north
south
east
1 [long address] 31.55823 31.53921 -97.0984
west postal_code
country
1 -97.13042
76706 united states
administrative_area_level_2
1
mclennan
administrative_area_level_1 locality
street
1
texas
waco s 5th st
streetNo point_of_interest
1
1311
<NA>
In particular, administrative bodies at various levels
are reported. Going further, setting output = "all"
returns the entire JavaScript Object Notation (JSON)
tree given by the Google Geocoding API parsed by
rjson (Couture-Beil, 2011).
The Geocoding API has a number of request limitations in place to prevent abuse. An unspecified
short-term rate limit is in place (see mapdist below)
as well as a 24-hour limit of 2,500 requests. These are
monitored to some extent by the hidden global variable .GoogleGeocodeQueryCount and exported function geocodeQueryCheck. geocode uses these to monitor its own progress and will either (1) slow its rate
depending on usage or (2) error if the query limit is
exceeded. Note that revgeocode shares the same request pool and is monitored by the same variable and
function. To learn more about the Google Geocoding, Distance Matrix, and Directions API usage regulations, see the websites listed in the bibliography.
revgeocode
In some instances it is useful to convert longitude/latitude coordinates into a physical address.
This is made possible (to the extent to which it is possible) with the revgeocode function which also relies
on the Google Geocoding API.
> gc <- geocode("baylor university")
> (gc <- as.numeric(gc))
[1] -97.11441 31.54872
> revgeocode(gc)
[1] "S 1st St, Baylor University, Waco, TX
76706, USA"
Like geocode, more output can be provided as well
> revgeocode(gc, output = "more")
address
route
establishment
ISSN 2073-4859

C ONTRIBUTED A RTICLE

1 [long address] S 1st St Baylor University

neighborhood locality
1
Baylor
Waco
administrative_area_level_2
1
McLennan
administrative_area_level_1
country
1
Texas United States
postal_code
1
76706
Thus, in addition to the physical where of a reverse
geocode (i.e., the address), revgeocode can report
the what at various levels of granularity. Finally, an
output = "all" option is available which returns the
entire JSON list tree reported by Google.
mapdist
The ability to compute real distances in a spatial setting is another invaluable commodity which typically sends analysts to a GIS. Using the Google Distance Matrix API, ggmap is able to provided distances for Google-determined routes for driving, bicycling, or walking. In addition to the distances,
Google reports estimated travel durations as well.
The full output is placed in an easy-to-use data
frame. For example,
> from <- c("houston", "houston", "dallas")
> to <- c("waco, texas", "san antonio",
+
"houston")
> mapdist(from, to)
from
to
m
km
miles
1 houston waco, texas 298004 298.004 185.1797
2 houston san antonio 320764 320.764 199.3227
3 dallas
houston 387389 387.389 240.7235
seconds minutes
hours
1
11907 198.45 3.307500
2
11997 199.95 3.332500
3
14592 243.20 4.053333

10

time
url elements
1 2012-03-16 00:12:11 [url used]
1
2 2012-03-16 00:16:10 [url used]
2
If the user exceeds the limitations, mapdist either (1)
pauses until the short-term request limit has lapsed
or (2) errors if no queries are remaining. Thus, it
is almost identical to the mechanism in place for
geocoding. If the user believes this to be incorrect,
an override is available with the mapdist specification override_limit = TRUE.
The data frame output of mapdist is very convenient for use with ggplot2. An example is provided
by Figure 14, where travel times from one location
("My Office") to several nearby locations are (1) determined using mapdist, (2) binned into categories
using cut, and then (3) plotted using a combination
of qmap, geom_text, and geom_rect with the fill aesthetic set to the category of travel time. The full code
is in the examples section of the documentation of
ggmap.

Buzzard Billy's

Ninfa's Mexican
Cafe Cappuccino
Dr Pepper Museum

Mayborn Museum
Baseball Stadium

My Office

Salvation Army
HEB Grocery

Flea Market

The default mode of transportation is driving; however, the other modes are also available. The input
forms of from and to can be either physical addresses
(ideal), lazy ("the white house"), or geographic coordinates (which are reverse geocoded). While the
output defaults to the data frame format seen above,
setting output = "all" provides the full JSON tree
from Google.
The Distance Matrix API limits users to 100 requests per query, 100 requests per 10 seconds, and
2500 requests per 24 hours. To the extent to which
these can be easily monitored, the exported function
distQueryCheck helps the user keep track of their remaining balance of queries. It relies on the hidden
global variable .GoogleDistQueryCount
> distQueryCheck()
2495 distance queries remaining.
> .GoogleDistQueryCount
2 This

Basketball Arena

Administration

Minutes
Away
03 35 57 710 10+
by Bike

Figure 14: Distances by time provided by mapdist

route
The route function provides the map distances for
the sequence of "legs" which constitute a route between two locations. Each leg has a beginning and
ending longitude/latitude coordinate along with a
distance and duration in the same units as reported
by mapdist. The collection of legs in sequence constitutes a single route (path) most easily plotted with
geom_leg, a new exported ggplot2 geom which is
simply geom_segment with rounded ends.2
A nice illustration of plotting routes with
geom_leg can be seen in Figure 15 where three
routes are plotted between the same two locations.
These can be obtained using the alternatives =

was inspired by the beautiful work of J. Cheshire available at http://spatialanalysis.co.uk/2012/02/great-maps-ggplot2/.

The R Journal Vol. X/Y, Month, Year

ISSN 2073-4859

C ONTRIBUTED A RTICLE

TRUE specification in route. alternatives requests

more than one route from the origin to the destination; the returned value is again a data frame with an
additional variable which serves as a route identifier
(A, B, C, etc.).
Like map distances, there are restrictions on the
limit of routes which can be requested in a given period of time as well. The number of queries left is
monitored by the .GoogleRouteQueryCount variable
with the routeQueryCheck function.

Plotting shape files

As a final example which is perhaps too common to
ignore, plotting shape files with ggmap is a breeze
and can be done in several ways. The easiest way is
to convert the shape file to a data frame with fortify
(ggplot2) and then add a point/path/polygon layer
to the map, depending on what the contents of the
shape file are (note that fortify does not yet have
methods for all shape files). Additional layers can
be added by simply layering on more geom layers.
Figure 16 shows a basic example of plotting the U.S.
Census 2000 census tracts along with complete code
(U.S. Census Bureau, 2001).

Conclusion and future directions

Building on top of ggplot2, ggmap provides several
new useful tools for visualizing spatial data. Theoretically speaking, the layered grammar of graphics
attempts to enforce plotting consistency and therefore good plotting practice. Practically speaking,
building ggmap on ggplot2 makes the result even
better as the full range of ggplot2 capabilities can be
brought to bear.
There are a number of future directions in store
for ggmap. The new osmar package integrates R
and the OpenStreetMap data structures with which
OpenStreetMap maps, Stamen Maps, and CloudMade Maps are rendered, thereby opening a floodgate of possibilities for plotting geographic objects
on top of maps or satellite imagery all within R using ggmap (Schlesinger and Eugster, 2012). Alternatively, integration with other spatial packages in
R could provide several incredibly useful practical
tools for spatial data analysis. Finally, the Google Elevation API and Places API provide additional interesting frontiers which can be incorporated into the
ggmap framework just like the other Google APIs to
give users additional capabilities though freely available geographical data.
The authors would like to thank Jean-Olivier Irisson for his comments.

Bibliography
R.S. Bivand, E.J. Pebesma, and V. Gomez-Rubio. Applied spatial data analysis with R., Springer, New
The R Journal Vol. X/Y, Month, Year

11

York, 2008. URL http://www.asdar-book.org/.

A. Couture-Beil. rjson: JSON for R., R package
version 0.2.6. URL http://CRAN.R-project.org/
package=rjson.
Google, Inc.
The Google Geocoding API.
URL
http://code.google.com/apis/maps/
documentation/geocoding/.
Google, Inc.
The Google Distance Matrix API.
URL
http://code.google.com/apis/maps/
documentation/distancematrix/.
Google, Inc.
The Google Directions API.
https://developers.google.com/maps/
documentation/directions/.

URL

M. Loecher and Sense Networks. RgoogleMaps: Overlays on Google map tiles in R., R package version 1.1.9.3. URL http://CRAN.R-project.org/
package=RgoogleMaps.
E.J. Pebesma and R.S. Bivand. Classes and methods
for spatial data in R. R News, 5(2), 2005. URL
http://cran.r-project.org/doc/Rnews/.
T. Schlesinger and Manuel J. A. Eugster. osmar: OpenStreetMap and R., R package version 1.1-3. URL
http://CRAN.R-project.org/package=osmar.
U.S. Census Bureau, Geography Division, Cartographic Products Management Branch . Census 2000: Census Tract Cartographic Boundary
Files..
URL http://www.census.gov/geo/www/
cob/tr2000.html.
H. Wickham. ggplot2: elegant graphics for data analysis., Springer, New York, 2009.
H. Wickham. A layered grammar of graphics. Journal
of Computational and Graphical Statistics, 19(1):328,
2010.
H. Wickham. The Split-Apply-Combine Strategy
for Data Analysis. Journal of Statistical Software,
40(1):129, 2011. URL http://www.jstatsoft.
org/v40/i01/.
L. Wilkinson. The Grammar of Graphics., 2nd ed.,
Springer, New York, 2005.
David Kahle
Baylor University
Department of Statistical Science
One Bear Place #97140
Waco, TX 77005
[email protected]
Hadley Wickham
Rice University
Department of Statistics, MS 138
6100 Main St.
Houston, TX 77005
[email protected]
ISSN 2073-4859

C ONTRIBUTED A RTICLE

12

Route
A

31.558

31.556

Latitude

31.554

31.552

31.550

31.548

97.130

97.125

97.120

97.130

97.125

97.120

97.130

97.125

97.120

Longitude

Figure 15: Three routes returned by route plotted with geom_leg

download.file('http://www.census.gov/geo/cob/bdy/tr/tr00shp/tr48_d00_shp.zip', destfile = 'census.zip')
unzip('census.zip'); library(maptools); library(gpclib); library(sp); gpclibPermit()
shapefile <- readShapeSpatial('tr48_d00.shp', proj4string = CRS("+proj=longlat +datum=WGS84"))
data <- fortify(shapefile)
qmap('texas', zoom = 6, maptype = 'satellite') +
geom_polygon(aes(x = long, y = lat, group = group), data = data,
colour = 'white', fill = 'black', alpha = .4, size = .3)

Figure 16: Plotting shape files Census tracts in Texas from the 2000 U.S. Census
The R Journal Vol. X/Y, Month, Year

ISSN 2073-4859