Does this operation *have* to be done in R? Sounds an awful lot like a GIS would have no problem with this type of operation... Try QGIS, it's free, and has several overlay functions. Plus there are plugins available to connect it with R.
Carson On 13 Nov 2010, at 07:52 p.m., brwin...@aol.com wrote: > > > > Good Afternoon: > > I have spent a good part of today searching the discussion list and reading > "Applied Spatial Data Analysis with R" and can't seem to find how to do the > following: > > One of my colleague's grad students has two shapefiles. The first shapefile > contains local polygons with associated local attribute data. The second > shapefile consists of US county polygons and attribute data for each county. > The first shapefile's polygon boundaries do not correspond to county > boundaries i.e. the local polygon be a subset of one county or may overlay > more than one county. The student's immediate needs are to be able to > compute how many square miles each local polygon has in each county. For > example local polygon 1 might have 10, 5, and 2.5 square miles in counties > 12,15, and 24 respectively. > > This brings me to my larger question. Does anyone have R code or know of an > R package that would allow us to merge the polygons from the two shapefiles > into a new set of polygons (and eventually create a new shapefile) that > preserves all the polygon boundaries of both original shapefiles? We would > also like to have an associated attribute data file that preserves all the > original attribute data. > > Suppose that attribute data for the first and second shapefiles looks like: > > shapefile1 (plots of land): > polygon, name > 1, A > 2, B > 3, C > .... > > > shapefile2 (California counties): > polygon, sfips, cfips, name > 1, 6, 06051, Mono > 2, 6, 06019, Fresno > 3, 6, 06107, Tulare > 4, 6, 06027, Inyo > …. > > I want to merge or overlay shapefile 1 onto shapefile2 > > Assume that plot A's land is a subset of Mono county, plot B has land in > Mono, Fresno, and Inyo counties in California as well as land in Nevada that > does not overlay any of my California shapes. Plot C has land in Inyo and > Tulare counties. I would like to have a new shapefile with merged polygons > and with merged attribute data like: > > polygon, polygon.x, name.x, polygon.y, sfips.y, cfips.y, name.y > 1, 1, A, 1, 6, > 06051, Mono > 2, 2, B, 1, 6, > 06051, Mono > 3, 2, B, 2, 6, > 06019, Fresno > 4, 2, B, 4, 6, > 06027, Inyo > 5, 2, B, NA, NA, > NA, NA > 6, 3, C, 3, 6, > 06107, Tulare > 7, 3, C, 4, 6, > 06027, Inyo > …. > > If I can create and save the such a new shapefile, I know how to use the > geosphere package to calculate the areas of each of my new polygons and hence > can compute the area that each local polygon has in any given county. > > Does anyone know how I can merge the two shapefiles like this? > Thanks > Joe > > > [[alternative HTML version deleted]] > > _______________________________________________ > R-sig-Geo mailing list > R-sig-Geo@stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/r-sig-geo _______________________________________________ R-sig-Geo mailing list R-sig-Geo@stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-geo
