Software to Implement Resource Utilization Function Estimation

The ruf package implements the statistical methods described in the paper Relating Resources to a Probabilistic Measure of Space Use: Forest Fragments and Steller's Jays by John M. Marzluff, J. J. Millspaugh, P. Hurvitz, and Mark S. Handcock, Ecology, 2004, 85:1411-1427 https://www.jstor.org/stable/3450181.

It is an R package to estimate resource utilization functions from point measurements of use and related functions. There are also functions to estimate summary measures and their uncertainties.

We recommend to install the current R version (R4.0) and the Rtools4.0.

The package can be installed from github via:

if (!requireNamespace("remotes")) install.packages("remotes")

remotes::install_github("handcock/ruf")

Once installed, it can be used by typing the call

library(ruf)

To see help on the functions type:

help(ruf.fit)

Case study: The resource utilization of Bird D412

As an example, below is an R program to reconstruct the analysis in the paper. It estimates the parameters of the model and related measures of uncertainty. This requires the above ruf package.

# load the "ruf" library

library(ruf)

# attach the small test data within the library

data(d412)

# Set initial estimates at the spatial range and smoothness

hval <- c(0.2, 1.5)

# Estimate the maximum likelihood values
# with unstandardized coefficients

d412.fit <- ruf.fit(ruf ~ CWED + IJI + NP + MSI,
         space= ~ x + y,
         data=d412, theta=hval,
         name="Bird 412",
         standardized=FALSE)
#> Fitting completed successfully!

Show the details of the results

summary(d412.fit)
#> 
#> Unstandardized Coefficients for name: Bird 412 
#> 
#> Matern Log-Lik = -145.6376 LS Log-Lik = -151.4624 
#> 
#> Change in Log-Lik 5.82484 p-value = 0.0029533 
#> 
#>                   MLE     s.e. LS estimate  LS s.e.
#> range        0.217498 0.041404          NA       NA
#> smoothness   1.500000       NA          NA       NA
#> (Intercept)  0.708582 1.653752    1.370407 1.650380
#> CWED         0.004491 0.046393    0.003097 0.046551
#> IJI         -0.014597 0.013885   -0.011364 0.013664
#> NP           0.368734 0.214525    0.463118 0.214353
#> MSI          0.211891 1.343349   -0.479440 1.348894

Estimate the maximum likelihood values with standardized coefficients

d412.fit <- ruf.fit(ruf ~ CWED + IJI + NP + MSI,
         space= ~ x + y,
         data=d412, theta=hval,
         name="Bird 412 standardized",
         standardized=T)
#> Fitting completed successfully!
summary(d412.fit)
#> 
#> Standardized Coefficients for name: Bird 412 standardized 
#> 
#> Matern Log-Lik = -145.6376 LS Log-Lik = -151.4624 
#> 
#> Change in Log-Lik 5.82484 p-value = 0.0029533 
#> 
#>                   MLE     s.e. LS estimate  LS s.e.
#> range        0.217498 0.041404          NA       NA
#> smoothness   1.500000       NA          NA       NA
#> (Intercept)  1.663544 0.106557    1.370407 1.650380
#> CWED         0.043437 0.448748    0.003097 0.046551
#> IJI         -0.234598 0.223152   -0.011364 0.013664
#> NP           0.364972 0.212336    0.463118 0.214353
#> MSI          0.035309 0.223854   -0.479440 1.348894

References

For further readings and references please consult:

• John M. Marzluff, J. J. Millspaugh, P. Hurvitz, and Mark S. Handcock, Relating Resources to a Probabilistic Measure of Space Use: Forest Fragments and Steller's Jays (2004) Ecology, 2004, 85:1411-1427 https://www.jstor.org/stable/3450181.