BlockPredict: Block Predicton for Streams Data

In jayverhoef/SSN: Spatial Modeling on Stream Networks

Block Predicton for Streams Data

Description

Block prediction for objects of class glmssn-class

Usage

BlockPredict(object, predpointsID)

Arguments

object	an object of class glmssn
predpointsID	a valid prediction points ID

Details

This function operates on glmssn objects in much the same way as the predict function. BlockPredict uses the locations in the predpointsID data set to compute the average prediction value in the area defined by the prediction locations. These prediction locations are used to approximate the integral over that area, so they should be evenly spaced and dense in the area where block prediction is desired. The user needs to create these prediction locations and include them in the SSN object prior to fitting the model with glmssn.

Value

A data.frame with one row and two columns. The first column, BlockPredEst, is the average prediction value, and the second column, BlockPredSE, is the standard error of the block prediction.

Author(s)

Jay Ver Hoef [e-mail]

References

Ver Hoef, J. M.. Peterson, E. E. and Theobald, D. (2006) Spatial statistical models that use flow and stream distance. Environmental and Ecological Statistics 13, 449-464. DOI: 10.1007/s10651-006-0022-8.

Examples

## Not run: 
library(SSN)
#for examples, copy MiddleFork04.ssn directory to R's temporary directory
copyLSN2temp()
# NOT RUN
# Create a SpatialStreamNetork object that also contains prediction sites
#mf04p <- importSSN(paste0(tempdir(),'/MiddleFork04.ssn'), 
#  predpts = "pred1km", o.write = TRUE)
#use mf04p SpatialStreamNetwork object, already created
data(mf04p)
#for examples only, make sure mf04p has the correct path
#if you use importSSN(), path will be correct
mf04p <- updatePath(mf04p, paste0(tempdir(),'/MiddleFork04.ssn'))

# Not needed: already added,
# add densely gridded prediction points for two stream segments
# mf04p <- importPredpts(mf04p, "Knapp", "ssn")
# mf04p <- importPredpts(mf04p, "CapeHorn", "ssn")
names(mf04p)

# see densely gridded prediction points on stream
plot(mf04p, PredPointsID = "Knapp")

#  you could fit the model
#fitSpBk <- glmssn(Summer_mn ~ ELEV_DEM + netID,
#    ssn.object = mf04p, EstMeth = "REML", family = "Gaussian",
#    CorModels = c("Exponential.tailup","Exponential.taildown",
#    "Exponential.Euclid"), addfunccol = "afvArea")

# or load the pre-fit model
data(modelFits)
fitSpBk$ssn.object <- updatePath(fitSpBk$ssn.object, 
	paste0(tempdir(),'/MiddleFork04.ssn'))

# one-at-a-time predictions for CapeHorn stream
## NOTE: need the amongpreds distance matrices for block prediction
createDistMat(mf04p, predpts = "CapeHorn", o.write = TRUE, amongpreds = TRUE)
fitSpPredC <- predict(fitSpBk, "CapeHorn")
# plot densely gridded prediction points on stream
plot(fitSpPredC, "Summer_mn")
# block prediction for CapeHorn stream
BlockPredict(fitSpBk, "CapeHorn")

## Another example
# one-at-a-time predictions for Knapp stream
createDistMat(mf04p, predpts = "Knapp", o.write = TRUE, amongpreds = TRUE)
fitSpPredK <- predict(fitSpBk, "Knapp")
# plot densely gridded prediction points on stream
plot(fitSpPredK, "Summer_mn")
# block prediction for Knapp stream
BlockPredict(fitSpBk, "Knapp")

## End(Not run)

jayverhoef/SSN documentation built on May 1, 2023, 1:04 p.m.