[R-sig-Geo] Simulation of inhomogeneous point process

[ASANTOS]

Dear Members,

- I have ant nests position in UTM in:

#Ant nests position-----------------------
a.x<-c(371349.2,371351.4,371277.1,371200.9,371139.9,371435.7,371344,371263.5,
371105.2,371023.2,371262.4,371273.2,371231.8,370959.5,370961.7,371020.3,371048.4,
371039.9,371086.8,371124.5,371034.7,371040,371025.8,370979.7,370954.3,370966.4,
370950.7,370914.1,370828.4,371294.4,371446.4,371468,371294.5,371276.1,371268.5,
371213.2,371055.2,371073.1,371333.6,371347.7,371242.4,371247,371081.9,371054.7,
370866.6,370885.9,370890,370916.2,370956.4,371125.5,371138.9,371094.8,371048.3,
371015.8,370990.2,370926.1,370915.8,370902.2,370882.6,371531.1)
a.y<-c(8247031,8247030,8247100,8247187,8247085,8246710,8246664,8246813,8247012,
8246970,8246750,8246696,8246644,8246803,8246802,8246748,8246711,8246696,8246624,
8246509,8246508,8246500,8246539,8246537,8246546,8246601,8246629,8246652,8246782,
8247238,8246805,8246742,8246811,8246829,8246836,8246903,8247033,8247014,8246730,
8246660,8246575,8246550,8246806,8246802,8246865,8246828,8246794,8246794,8246754,
8246474,8246452,8246460,8246533,8246615,8246615,8246661,8246692,8246710,8246738,
8246867)
ant.pos<-cbind(a.x,a.y)


- This nest are distributed in a stand area:

#Stand contour ----------------------------------
b.x<-c(371299.9,371266.4,371205.6,371111.8,371047.6,371018.2,371014,371009.3,
370983.1,370919.7,370853.6,370785.6,370748.8,370711.8,370687.8,370696.4,370785.9,
370885.5,371035.8,371148.1,371205.2,371231.7,371236.5,371240.3,371285.8,371326.5,
371397.2,371417.1,371432.9,371445,371455.7,371466.4,371476.6,371502.6,371536,371550,
371546.8,371528.3,371470,371393.3,371299.9)
b.y<-c(8246589,8246560,8246508,8246428,8246373,8246349,8246348,8246352,8246385,
8246465,8246551,8246638,8246685,8246732,8246764,8246771,8246846,8246932,8247062,
8247160,8247209,8247230,8247224,8247221,8247160,8247107,8247016,8246991,8246967,
8246939,8246914,8246892,8246875,8246846,8246821,8246809,8246802,8246785,8246735,
8246669,8246589)
bord<-cbind(b.x,b.y)


- I suspect that this ant nests have inhomogeneous Poisson distribution, 
than I make a CSR for inhomogeneous Poisson process using spastat package:

require(spatstat)
require(gpclib)
require(rgeos)
#Spatstat object 
---------------------------------------------------------------------
a <- as(bord, "gpc.poly")
w <- as.owin(a)
formi05.ppp<-ppp(x=coordinates(ant.pos)[,1],y=coordinates(ant.pos)[,2],window=w)
plot(formi05.ppp,main="")

## CSR inhomogeneous 
test-----------------------------------------------------
env.formi<-envelope(formi05.ppp,nsim=999,fun=Kinhom)
par(mfrow=c(1,2))
plot(env.formi,lwd=list(3,1,1,1),xlim=c(0,200), main="Stand 4")
#


- Ok, I have a  inhomogeneous Poisson process. If my process is 
homogeneous Poisson is very simple generate a random point pattern for 
ant nests using rpoispp() function, but now based in my results, what is 
the better approach for find in observed ant nests position the 
retention probability for lambda(x,y)/lmax, when my purposes is create a 
simulation of inhomogeneous point process?


Thanks,

Alexandre

-- 
======================================================================
Alexandre dos Santos
Proteção Florestal
IFMT - Instituto Federal de Educação, Ciência e Tecnologia de Mato Grosso
Campus Cáceres
Caixa Postal 244
Avenida dos Ramires, s/n
Bairro: Distrito Industrial
Cáceres - MT                      CEP: 78.200-000
Fone: (+55) 65 8132-8112 (TIM)   (+55) 65 9686-6970 (VIVO)

         alexandre.santos at cas.ifmt.edu.br
Lattes: http://lattes.cnpq.br/1360403201088680
======================================================================


---
Este email foi escaneado pelo Avast antivírus.
http://www.avast.com