3

I only recently started playing with shapefiles and R packages for manipulating them. I am looking at creating a new shapefile based on merged polyons by following this example, however, I cannot seem to figure out what I am suppose to feed the as.data.frame("youData").

Using the Zambia shapefiles as an example, how would one go about creating a new shapefile composed of a superset district-level data?

require(maptools)
require(rgadl)

zambia <- readShapeSpatial("ZMB_adm2.shp")
zambiap <- unionSpatialPolygons(zambia, ID=zambia@data$ID_1)

zambiapd <-     SpatialPolygonsDataFrame(ob,data=as.data.frame("yourData"),proj4string=CRS("+proj=    aea > +ellps=GRS80 +datum=WGS84"))
writeOGR(zambiapd,"shapes","testShape",driver="ESRI Shapefile",)
Community
  • 1
lightonphiri
  • 131
  • 1
  • 5

1 Answers1

2

the str( ) will display the internal structure of your object

str(zambiap)

If you just want a data.frame to export:

dt<-data.frame(1:length(zambiap))
zambiapd <- SpatialPolygonsDataFrame(zambiap, data=dt)

If you want a meaningful data.frame, you need to get the data from the zambiap object.

dt.f<-NULL
for (i in 1:length(zambiap)){
dt<-zambiap@polygons[[i]]@area
dt.f<-rbind(dt.f, dt)
}

rownames(dt.f)<-1:length(zambiap)
dt.f<-data.frame(dt.f)
  • This can be done in a more elegant way, but I think that in this manner you can follow the idea based on the str( ) outputs above.

And then, you add the data, to your original object.

zambiapd <- SpatialPolygonsDataFrame(zambiap, data=dt)

To inspect the data.frame:

zambiapd@data

To manipulate the data.frame:

colnames(dt.f)<-("Area")
zambiapd@data<-dt.f

Be careful when manipulating the @data, adding more data, etc.

See this post: https://stackoverflow.com/questions/3650636/how-to-attach-a-simple-data-frame-to-a-spatialpolygondataframe-in-r

I am not sure if your proj4string parameters are correct.

For exporting your spdf:

writeOGR(zambiapd, dsn=getwd(), "Zambiapd", driver="ESRI Shapefile")

HTH,

Community
  • 1
alecamar
  • 106
  • 3
  • 1
    Thank you for this. How would I preserve some of the original descriptive information from the original shapefile though? The output shapefile only have one column---Area in this case. – lightonphiri Nov 08 '14 at 18:30