[R-sig-Geo] underlying covariate data using R package 'spatstat'

[Raya A. Pruner]

I have another question for everyone related to the spatstat 
package and model fitting.

A brief background on my question of interest: We are interested 
in the spatial aggregation of snowy plover nests on a barrier 
island system in response to nest fates.  We hypothesize a greater 
level of aggregation in response to a positive nest fate.  Given 
this, we are also interested in underlying covariates.  So we have 
data for human activity, predator activity, and prey availability. 
 We would like to see how these covariates influence the 
aggregation of nests as well.  Currently I have been trying to 
create a list of images, but was unsuccessful.  So I used based on 
what I have been able to figure out and did each separately.  This 
was the only way I could figure out in associating a continuous 
covariate with a pixel inmage:

nests.extra <- read.csv("CIE 2009 covariates.csv", header = T)
extra.chop <- subset(nests.extra, select = c(Id, 
X,Y,Tracks,Predator_T, Invert_Tra))


covH.ppp <- ppp(extra.chop[,2], extra.chop[,3], marks 
=extra.chop[,4],window = area1)
human.act=density.ppp(covH.ppp)
human.act

covP.ppp <- ppp(extra.chop[,2], extra.chop[,3], marks 
=extra.chop[,5]_T,window = area1)
predator.act=density.ppp(covP.ppp)
predator.act

covI.ppp <- ppp(extra.chop[,2], extra.chop[,3], marks 
=extra.chop[,6],window = area1)
invert.act=density.ppp(covI.ppp)
invert.act

using this allows me to fit models such as this using multiple 
covariate:
cov2.cord.model=ppm(nestsF.ppp, ~x+y+Tracks+Pred, covariates = 
list(Tracks=human.act,Pred=predator.act ))

However, I don???t think it is working right.  I tried to plot 
human.act and it doesn???t seem to relate to the data provided in 
plotting covH.ppp.  The way that I collected the covariate data 
was systematic by counting human and predator tracks every 500m 
linearly along the beach.  So I want to be able to look at how the 
human intensity value effects nesting aggregation and not the 
density of covariate points themselves which is what I think it is 
doing currently.  I am interested in the value behind the point.  
If anyone has any insight on this, please let me know!

Cheers!!!
Raya


--
Raya A Pruner - Graduate Student
University of Florida
Department of Wildlife Ecology and Conservation
PO Box 110430
Gainesville, FL 32611-0430
352-214-3262