[R] ggmap, hexbin and length distortion in Lat/long

Mike Bock mbock at environcorp.com
Fri May 24 15:51:46 CEST 2013

I am working with spatial data in ggmap, generally with great success. I have a huge data set with the coordinates in NAD 83 UTM Zone 11 (meters).  To map the data the coordinates were converted to Lat Long in GIS prior to use in R and ggmap/ggplot. I am using hexagonal binning to aggregate the data :
#create bins and calculate stats

#create df for ggplot
hx_dat <- data.frame(hcell2xy(hb), count = hb at count,
    xo = hb at xcm, yo = hb at ycm, Mean=hb.avg,Max=hb.mx,

#Base Map
#BBox is the bounding box
m_hx<-ggmap(Base,legend = "bottom", base_layer=ggplot(aes(x=x,y=y),data=hx_dat))
#Map of means

m_hx+geom_hex(aes(x = x, y = y, fill = Mean),
  color = hc, ,alpha=a,stat = "identity") +

...and so on for other stats
I can also run statistical analyses on hx_dat.
By creating hexbins based on lat/long it seems there will be distortion due to the differences in length of a degree at different locations on the earth's surface. What is the most efficient way to eliminate this distortion? Should I run hexbin in NAD83 and convert the x/y coordinates to Lat Long? Can I get ggmap to convert the baselayer to NAD84 and just do everything in NAD(my preferred option)?
I have tried converting Lat Long to NAD84 and back but the coordinates are coming up in the eastern Pacific and not in California, so I am missing something and I am not sure that is the best way to solve the problem anyway. Thanks in advance, any help is greatly appreciated

Michael J. Bock, PhD | Senior Manager
mbock at environcorp.com

This message contains information that may be confidenti...{{dropped:8}}

More information about the R-help mailing list