[R-sig-Geo] creating polygons

Thu Apr 23 19:55:09 CEST 2009

On Thu, 23 Apr 2009, Alina Sheyman wrote:

> I went back to the help page, but it still didn't clarify it for me.
> Yes, the two files are different length, but both have a zip variable that
> they can match on and then keep only those records that match. Or do I need
> to additionally create an index for the match to occur?
> I am understanding it that unionSpatialPolygons functions works very much in
> the same way as merge, except with geographical data. Am I wrong in this
> assumption?

Wrong, as you can see. That isn't how it works. As input, you have 
n Polygons objects in a SpatialPolygons object (with no variables at all, 
no data slot), as output m Polygons objects in a SpatialPolygons object. 
The IDs argument assigns the (possibly multiple) n input Polygons objects 
to the m output Polygons object, which means that IDs must be of length n, 
and express which j=1,m each i=1,n belongs to. So if you do length(IDs), 
it must equal n, and length(unique(IDs)) will equal m. Not complicated, 
not merge, but hard work if you don't have a list of zips broken down by 
districts that lets you create the IDs easily.

Hope this helps,

Roger

>
>
> On Wed, Apr 22, 2009 at 3:09 AM, Roger Bivand <Roger.Bivand at nhh.no> wrote:
>
>> On Tue, 21 Apr 2009, Alina Sheyman wrote:
>>
>>  Hi Roger,
>>>
>>
>> Alina,
>>
>> Don't take this offlist, summarise to the list when done so that others can
>> learn from your experiences.
>>
>>  when I follow the example in  maptools
>>>
>>> coordinates(mass) <- c("lat", "long")
>>> reg4 <- unionSpatialPolygons(zipmaps3, mass$center_id)
>>>
>>
>> Right, you have not read the unionSpatialPolygons help page carefully, have
>> you?
>>
>> Do:
>>
>> length(mass$center_id)
>> length(slot(zipmaps3, "polygons")
>>
>> They differ - your mass are the districts, zipmaps3 are the zips. zipmaps3
>> has a variable giving the district IDs, what is it called? If you do
>>
>> table(zipmaps3$district_IDs)
>>
>> you see the numbers of zips in each district. This is your second argument
>> to unionSpatialPolygons; it tells the functions which zips belong to each
>> district. If you don't have such a variable, you'll have to make it. This is
>> all explained on the help page, and in the example. It is very much like the
>> other tapply(), by() and aggregate() functions, but only returns the
>> geometries.
>>
>> Hope this helps,
>>
>> Roger
>>
>>
>>
>>> I get the following error message
>>> Error in unionSpatialPolygons(zipmaps3, mass$center_id) :
>>>  input lengths differ
>>>
>>> Do you know why that is? Do I in fact need to have two files of identical
>>> length in order for the
>>> function to work?
>>>
>>> thank you
>>>
>>> On Tue, Apr 21, 2009 at 3:17 PM, Roger Bivand <Roger.Bivand at nhh.no>
>>> wrote:
>>>
>>>  On Tue, 21 Apr 2009, Alina Sheyman wrote:
>>>>
>>>>  To rephrase my question and explain my problem further
>>>>
>>>>> I have already created a map for revenue by zip code, in the state of
>>>>> Massachusetts. To accomplish that I used a shapefile for Massachusetts
>>>>> and
>>>>> a
>>>>> dataframe with my revenue/zip codes.
>>>>>
>>>>> What I am trying to do now is to have the same map done not by zip code,
>>>>> but
>>>>> rather by districts, as defined by zip codes.
>>>>>
>>>>> Basically what I've done was to introduce a new variable "district" into
>>>>> my
>>>>> dataset along with a zip code. At the same time my basic state map (by
>>>>> zip)
>>>>> still comes from a shape file and I can't introduce a new variable into
>>>>> the
>>>>> shape file.
>>>>>
>>>>>  Does anyone  know if there's any way I can accomplish what I'm trying
>>>>> to
>>>>> do or do I basically need to create a new shapefile with districts
>>>>> already
>>>>> in the file?
>>>>>
>>>>>
>>>> OK, that wasn't too clear before. Look at unionSpatialPolygons() in
>>>> maptools - it will take a vector of districts matching all your zip
>>>> codes,
>>>> and join together the geometries belonging to each district, dropping
>>>> internal boundaries. The district IDs become the Polygons object IDs.
>>>> There
>>>> is no data frame in the object, because it isn't obvious how you might
>>>> want
>>>> to aggregate it.
>>>>
>>>> Hope this helps,
>>>>
>>>> Roger
>>>>
>>>>
>>>>
>>>>  On Mon, Apr 20, 2009 at 5:53 PM, Don MacQueen <macq at llnl.gov> wrote:
>>>>>
>>>>>  This is a pretty general question, and there are potentially many ways
>>>>> to
>>>>>
>>>>>> do it.
>>>>>>
>>>>>> One way to get started would be to follow the example(s) in the help
>>>>>> page
>>>>>> for the overlay() function in the sp package.
>>>>>>
>>>>>> -Don
>>>>>>
>>>>>>
>>>>>> At 1:26 PM -0400 4/20/09, Alina Sheyman wrote:
>>>>>>
>>>>>>  I have a list of made-up districts and corresponding zip codes. Based
>>>>>>
>>>>>>> on
>>>>>>> that I want to define my own polygons (for each district) and then use
>>>>>>> them
>>>>>>> to create a map. Does anyone know how I would go about defining these?
>>>>>>>
>>>>>>> thank you
>>>>>>>
>>>>>>>      [[alternative HTML version deleted]]
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> R-sig-Geo mailing list
>>>>>>> R-sig-Geo at stat.math.ethz.ch
>>>>>>> https:// stat.ethz.ch/mailman/listinfo/r-sig-geo
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>> --
>>>>>> --------------------------------------
>>>>>> Don MacQueen
>>>>>> Environmental Protection Department
>>>>>> Lawrence Livermore National Laboratory
>>>>>> Livermore, CA, USA
>>>>>> 925-423-1062
>>>>>> --------------------------------------
>>>>>>
>>>>>>
>>>>>>        [[alternative HTML version deleted]]
>>>>>
>>>>> _______________________________________________
>>>>> R-sig-Geo mailing list
>>>>> R-sig-Geo at stat.math.ethz.ch
>>>>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>>>>
>>>>>
>>>>>  --
>>>> Roger Bivand
>>>> Economic Geography Section, Department of Economics, Norwegian School of
>>>> Economics and Business Administration, Helleveien 30, N-5045 Bergen,
>>>> Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
>>>> e-mail: Roger.Bivand at nhh.no
>>>>
>>>>
>>>>
>>>
>> --
>> Roger Bivand
>> Economic Geography Section, Department of Economics, Norwegian School of
>> Economics and Business Administration, Helleveien 30, N-5045 Bergen,
>> Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
>> e-mail: Roger.Bivand at nhh.no
>>
>>
>

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no