[R-sig-eco] diversity zones

[Marcelino de la Cruz]

Hi Katia,


Using some functions from spatstat you could get some maps.

For this, you should format your data as a ppp object of spatstat.


See my example below:


# data for my example
library(spatstat)
data(lansing)
plot(lansing)

# get richnes for each tree within a circle with R=0.05
# first tabulate number of neighbour trees for each species
lansing.r <- marktable(lansing, R=0.05)
# now summarize richness.
#This would be  equivalent to your vector of richness for each tree
lansing.rr <- apply(lansing.r, 1,function(x) sum(x>0))

# Add richness as a mark to the lansing ppp, maintaining the original 
species mark
lansing$marks=data.frame(sp=lansing$marks, richness=lansing.rr)
plot(lansing)

# get a ppp for each species  ("split" the ppp)
lansing.sp=split(lansing, f="sp")
lansing.sp

# now remove the sp marks and let only the richness one
lansing.sp<- lapply(lansing.sp, function(x){x=x; 
x$marks=x$marks$richness; return(x)})

# get "maps" of  richness around trees
    # with Smooth.ppp (Kernel smoothing)
    plot(Smooth(lansing.sp[["blackoak"]]))
    points(lansing.sp[["blackoak"]])

    # with idw (inverse distance weighting)
    dev.new()
    plot(idw(lansing.sp[["blackoak"]]))
    points(lansing.sp[["blackoak"]])

See the help page of Smooth.ppp and idw to see how to fine tuning 
"bandwith", weighting distance etc.



HTH,

MArcelino



El 04/05/2015 a las 16:19, Katia escribió:
>
>
>> De: Kátia Emidio
>> Dear all,
>>
>> I am trying to create a diversiy zones based on tree species distribution,  from neighboor analysis. I identify the 20th neighboor from each one of my tree species, and counted the richness associated to each one. I also have the associated distance to each kth tree... So, my study area is 100x100 m..As I want zone areas, how can I divide this area, so it can represent the richnees found? From 265 species, I am interested to do this for 30 one, generating maps for each one.
>>
>> I though to use the Dirichlet tessellation, but it is a little hard to me, to imagine how to do it...
>>
>> Any tips is welcome!!!
>>
>>
>> --
>> Kátia Emídio da Silva DSc
>> Eng. Florestal
>> Manaus/AM
>>
>>
>>
>> Forestry Engineer
>> Manaus/AM-Brazil
>
> 	[[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
>