[R-sig-eco] adehabitat conditional MCP creation and export to csv

Mathieu Basille basille at ase-research.org
Fri Jul 2 15:48:01 CEST 2010 

Hakim, you understood perfectly every step of the process (subset, 
factor, and mcp). In particular, have a look at ?mcp:

id: a factor giving the identity of the animal for each relocation

Which means that mcp will be computed for each level of 'id'. To make 
sure of this, run:

plot(bla)  ## where 'bla' is the object in which you stored the mcp
points(q[, c("X", "Y")]

If you just want to compute one single mcp for all individuals (which I 
believe is not your aim, otherwise you wouldn't ask to delete all 
animals with < 5 relocations), you could do something like that:

mcp(q[, c("X", "Y")], rep("AnyId", nrow(q)), percent = 95)

In any cases, ?mcp is your friend!

Hope this helps,
Mathieu.



Abdi, Abdulhakim a écrit :
> Thanks for your input Mathieu, your example seems to have worked. So, if I understand correctly the following code
> 
> subset(p, Name %in% names(which(table(Name) >= 5))) 
> 
> selects the rows in the data.frame which have 5 or more occurrences under the "Name" column and in order to eradicate the stored levels, the row in question has to be re-encoded as a factor:
> 
> q$Name <- factor(q$Name)
> 
> I would like to understand the MCP code further:
> 
> mcp(q[, c("X", "Y")], q$Name, percent = 95)
> 
> by including q$Name into the command you're opting to create separate MCPs for each factor in q$Name?
> 
> Again, thank you very much for your assistance, I really appreciate it. 
> 
> Best,
> 
> Hakim Abdi
> 
> 
> 
> _________________________________
> Abdulhakim Abdi, M.Sc.
> Research Intern
> 
> Conservation GIS/Remote Sensing Lab
> Smithsonian Conservation Biology Institute
> 1500 Remount Road
> Front Royal, VA 22630
> phone: +1 540 635 6578
> mobile: +1 747 224 7006
> fax: +1 540 635 6506 (Attn:GIS Lab)
> email: abdia at si.edu
> http://nationalzoo.si.edu/SCBI/ConservationGIS/ 
> 
> 
> 
> 
> -----Original Message-----
> From: Mathieu Basille [mailto:basille at ase-research.org] 
> Sent: Thursday, July 01, 2010 3:53 PM
> To: Abdi, Abdulhakim
> Cc: r-sig-ecology at r-project.org
> Subject: Re: [R-sig-eco] adehabitat conditional MCP creation and export to csv
> 
> Dear Hakim,
> 
> The problem comes from Name being a factor. Even if you subset a factor, 
> the levels are still stored, so that it gets 0 relocations for some 
> individuals. In addition, I'm a little bit perplexed about your 'for' 
> loop (especially the if(summary(p$Name)>=5) component).
> 
> I would do something like that:
> 
> (q <- subset(p, Name %in% names(which(table(Name) >= 5))))
> ## Check:
> table(q$Name)
> q$Name <- factor(q$Name)
> ## Check again:
> table(q$Name) # Now correct
> (bla <- mcp(q[, c("X", "Y")], q$Name, percent = 95))
> 
> Hope this helps.
> Mathieu
> 
> 
> Abdi, Abdulhakim a écrit :
>> Hi everyone,
>>
>> I've searched a few R lists for an answer to this dilemma but it's got me stumped. I have a dataset not unlike the Puechabon dataset that's included with the Adehabitat package:
>>
>>> data(puechabon)
>>> p = puechabon$locs
>>> p = p[-c(94:120,53:90,37:50,7:29),]
>>> p
>>     Name Age Sex      X       Y   Date
>> 1  Brock   2   1 699889 3161559 930701
>> 2  Brock   2   1 700046 3161541 930703
>> 3  Brock   2   1 698840 3161033 930706
>> 4  Brock   2   1 699809 3161496 930707
>> 5  Brock   2   1 698627 3160941 930708
>> 6  Brock   2   1 698719 3160989 930709
>> 30 Brock   2   1 699207 3161099 930831
>> 31 Calou   2   1 700098 3161648 930703
>> 32 Calou   2   1 700310 3161323 930704
>> 33 Calou   2   1 699877 3161471 930707
>> 34 Calou   2   1 700003 3161360 930708
>> 35 Calou   2   1 699820 3161387 930709
>> 36 Calou   2   1 699778 3161328 930711
>> 52  Chou   3   2 699640 3158702 920802
>> 53  Chou   3   2 699689 3158857 920803
>> 92  Jean   2   1 699448 3158689 930705
>> 93  Jean   2   1 699500 3158684 930706
>> 94  Jean   2   1 699409 3158794 930707
>>
>> I'm trying to create an MCP for each animal that has 5 or more relocation points (e.g. Brock & Calou) and export each MCP to a CSV file. I've tried the following, but I get an error message:
>>
>>> summary(p$Name)>=5
>> Brock Calou  Chou  Jean
>>  TRUE  TRUE FALSE FALSE
>>
>>> for (i in 1:nrow(p)){
>> + if(summary(p$Name)>=5){
>> + mcp.p = mcp(pxy, pid, percent=95)
>> + write.table(mcp.p, sep=",", file=paste(pid,".csv"), col.names=NA)
>> + }
>> + }
>> Error in mcp(pxy, pid, percent = 95) :
>>   At least 5 relocations are required to fit an home range
>> In addition: Warning message:
>> In if (summary(p$Name) >= 5) { :
>>   the condition has length > 1 and only the first element will be used
>>
>> I'm not an expert in R and would appreciate any help.
>>
>>
>> Regards,
>>
>> _________________________________
>> Hakim Abdi, M.Sc.
>> Research Intern
>>
>> Conservation GIS/Remote Sensing Lab
>> Smithsonian Conservation Biology Institute
>> 1500 Remount Road
>> Front Royal, VA 22630
>> phone: +1 540 635 6578
>> mobile: +1 747 224 7006
>> fax: +1 540 635 6506 (Attn:GIS Lab)
>> email: abdia at si.edu
>> http://nationalzoo.si.edu/SCBI/ConservationGIS/
>>
>> 	[[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-ecology mailing list
>> R-sig-ecology at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
> 

-- 

~$ whoami
Mathieu Basille, Post-Doc

~$ locate
Laboratoire d'Écologie Comportementale et de Conservation de la Faune
+ Centre d'Étude de la Forêt
Département de Biologie
Université Laval, Québec

~$ info
http://ase-research.org/basille

~$ fortune
``If you can't win by reason, go for volume.''
Calvin, by Bill Watterson.

More information about the R-sig-ecology mailing list