[R-sig-Geo] Tearing of polygons using ggmap & readOGR

[Tyler Frazier]

Hi Roger, 

Thanks for your reply and pointers.  I was looking for a simplified shapefile of American state boundaries and downloaded this one from ESRI, states_basic.

https://www.arcgis.com/home/item.html?id=f7f805eb65eb4ab787a0a3e1116ca7e5 <https://www.arcgis.com/home/item.html?id=f7f805eb65eb4ab787a0a3e1116ca7e5>

The .shp is only about 222 KB which was OK for plotting speed in R as well as its appearance.  I tried 

states <- getData('GADM', country='USA', level=1)

as well as the census state subdivisions, but the resolution seemed to be too much for R to handle in a plot.

The states_basic file from ESRI plots just fine with 

plot(states)  — I think the summary() command returns what would be expected of a shapefile that was read into R as SPDF.

> summary(states)
Object of class SpatialPolygonsDataFrame
Coordinates:
         min       max
x -178.21760 -66.96927
y   18.92179  71.40624
Is projected: FALSE 
proj4string :
[+proj=longlat +datum=NAD83 +no_defs +ellps=GRS80 +towgs84=0,0,0]
Data attributes:
      STATE_NAME    DRAWSEQ       STATE_FIPS              SUB_REGION   STATE_ABBR
 Alabama   : 1   Min.   : 1.0   01     : 1   South Atlantic    : 9   AK     : 1  
 Alaska    : 1   1st Qu.:13.5   02     : 1   Mountain          : 8   AL     : 1  
 Arizona   : 1   Median :26.0   04     : 1   West North Central: 7   AR     : 1  
 Arkansas  : 1   Mean   :26.0   05     : 1   New England       : 6   AZ     : 1  
 California: 1   3rd Qu.:38.5   06     : 1   East North Central: 5   CA     : 1  
 Colorado  : 1   Max.   :51.0   08     : 1   Pacific           : 5   CO     : 1  
 (Other)   :45                  (Other):45   (Other)           :11   (Other):45  
> states at data
             STATE_NAME DRAWSEQ STATE_FIPS         SUB_REGION STATE_ABBR
0                Hawaii       1         15            Pacific         HI
1            Washington       2         53            Pacific         WA
2               Montana       3         30           Mountain         MT

etc…

sounds like I need to look into exactly what is happening as part of the fortify() command in order to achieve some consistency of integrating SpatialPolygonsDataFrames with the ggmap() function.  I may eventually go to tmap or mapview or perhaps leaflet, but have found ggplot:: / ggmap:: to be very useful packages (+geom_polygons, + geom_points etc…)and was hoping to keep working with their grammar of graphics.

Tyler


> On Jan 31, 2016, at 3:35 PM, Roger Bivand <Roger.Bivand at nhh.no> wrote:
> 
> On Sun, 31 Jan 2016, Tyler Frazier wrote:
> 
>> I'm trying to incorporate a simple shapefile of US states with ggmap and keep getting "tearing" of my polygons. I tried changing group=id to group=group which seemed to partially solve the problem, but not entirely.
>> 
>> library(rgdal)
>> 
>> states <- readOGR(dsn="shapefiles", layer="states")
>> 
>> proj4string(states)
>> 
>> [1] "+proj=longlat +datum=NAD83 +no_defs +ellps=GRS80 +towgs84=0,0,0"
>> 
>> states <- spTransform(states, CRS("+proj=longlat +datum=WGS84"))
> 
> Unnecessary, as you should know from the +towgs84= tag.
> 
>> 
>> states <- fortify(states)
> 
> The fortify methods turn data from marine animals with structure into fish soup (a data frame adding attribute values spuriously to each coordinate, I believe). Unless the objects are exactly as fortify wants them, it derails.
> 
> Do you know that they were polygons? Where did the shapefile come from? Did you plot it? Did you check that it contained what you thought it contained (summary())?
> 
> Have you considered alternatives (tmap, mapview, ...)?
> 
>> 
>> sstates <- get_map(location = c(-81, 35, -69, 45), zoom = 4, maptype = "watercolor")
>> 
>> sstates <- ggmap(sstates)
>> 
>> sstates <- sstates + geom_polygon(aes(x = long, y = lat, group=group), data = states, color ="white", fill ="orangered4", alpha = .4, size = .2)
>> 
>> Here is an image of the output.
>> 
>> http://pasteboard.co/1d7kiRfw.png <http://pasteboard.co/1d7kiRfw.png>
>> 
> 
> Image posted OK, but posting HTML (unnecessary).
> 
>> 
>> 	[[alternative HTML version deleted]]
>> 
>> _______________________________________________
>> R-sig-Geo mailing list
>> R-sig-Geo at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>> 
> 
> -- 
> Roger Bivand
> Department of Economics, Norwegian School of Economics,
> Helleveien 30, N-5045 Bergen, Norway.
> voice: +47 55 95 93 55; fax +47 55 95 91 00
> e-mail: Roger.Bivand at nhh.no
> http://orcid.org/0000-0003-2392-6140
> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
> http://depsy.org/person/434412


	[[alternative HTML version deleted]]