make_ghost {qfasar}

Make a ghost prey signature

Description

Bromaghin et al (2016) studied the performance of QFASA estimators when predators consumed a prey type that was not represented in the prey library, termed a ghost prey. make_ghost constructs a signature for a ghost prey type.

Usage

make_ghost(prey_sigs, loc, ghost_err = 0.25, dist_meas = 1, gamma = 1)

Arguments

prey_sigs

A matrix of prey signatures ready for analysis, intended to be the object sig_scale returned by a call to the function prep_sig with the prey data frame or the object sig_part returned by make_prey_part.

loc

A matrix giving the first and last locations of the signatures of each prey type within prey_sigs, intended to be the object loc returned by a call to the function prep_sig with the prey data frame or the object loc returned by make_prey_part.

ghost_err

A proportion strictly greater than 0 and less than 1 used to control the lower and upper bounds of ghost prey signature proportions. Default value 0.25.

dist_meas

An integer indicator of the distance measure to be used. Default value 1.

gamma

The power parameter of the chi-square distance measure. Default value 1.

Value

A list containing the following elements:

sig

A numeric vector containing the ghost prey signature.

dist

Summed distance between the ghost signature and the mean prey signatures.

err_code

An integer error code (0 if no error is detected).

err_message

A string contains a brief summary of the execution.

Details

One of the major assumptions of QFASA is that the prey library contains representatives of all prey types consumed by a predator. Bromaghin et al. (2016) investigated the robustness of diet estimators to violations of this assumption. The function make_ghost constructs a ghost prey signature using the methods of Bromaghin et al. (2016).

The ghost prey signature is constructed by maximizing the summed distance between the ghost prey signature and the mean prey signatures, while constraining the ghost signature proportions within reasonable bounds to ensure that the signature is somewhat realistic for the prey library. The definition of reasonable bounds is embodied in the argument ghost_err. ghost_err is a proportion greater than or equal to zero and less than 1 that is used to construct lower and upper bounds of the signature proportions. The lower bound is obtained by multiplying 1 - ghost_err by the minimum mean prey proportion for each fatty acid. Similarly, the upper bound is obtained by multiplying 1 + ghost_err by the maximum mean prey proportion for each fatty acid. The ghost prey signature is then obtained by maximizing the summed distance between the signature and the mean prey signatures, constraining the signature to lie within the bounds and sum to 1. See est_diet for information regarding distance measures.

This method ensures that the ghost prey signature is somewhat distinct from the other prey types, but not so wildly different that it represents a completely different pattern from the other prey types. Although research into suitable values for ghost_err has not been conducted, it is probably advisable to use small to moderate values. Bromaghin et al. (2016) used a value of 0.25. As the value of ghost_err is increased, the resulting signature will tend to become increasing different from any prey type in the library.

References

Bromaghin, J.F., S.M. Budge, G.W. Thiemann, and K.D. Rode. 2016. Assessing the robustness of quantitative fatty acid signature analysis to assumption violations. Methods in Ecology and Evolution 7:51-59.

Examples

make_ghost(prey_sigs = matrix(c(0.05, 0.10, 0.30, 0.55,
                                0.04, 0.11, 0.29, 0.56,
                                0.10, 0.05, 0.35, 0.50,
                                0.12, 0.03, 0.37, 0.48,
                                0.10, 0.06, 0.35, 0.49,
                                0.05, 0.15, 0.35, 0.45), ncol=6),
           loc = matrix(c(1, 3, 5, 2, 4, 6), ncol=2),
           ghost_err = 0.15,
           dist_meas = 1,
           gamma = NA)

make_ghost(prey_sigs = matrix(c(0.05, 0.10, 0.30, 0.55,
                                0.04, 0.11, 0.29, 0.56,
                                0.10, 0.05, 0.35, 0.50,
                                0.12, 0.03, 0.37, 0.48,
                                0.10, 0.06, 0.35, 0.49,
                                0.05, 0.15, 0.35, 0.45), ncol=6),
           loc = matrix(c(1, 3, 5, 2, 4, 6), ncol=2))