marknovak / FR_n-prey-at-a-time

Functional response when feeding on multiple prey at a time
0 stars 0 forks source link

A functional response for predators that can continue to search while handling up to $n$ prey at a time

This repository contains the code and data for the analyses of:

Novak, Coblentz & DeLong (2024) Feeding on multiple prey at a time: The frequency and population-dynamic effects of functional-response linearity. bioRxiv

Repository content

Mathematical analyses (isoclines and simulations) were performed in Mathematica v.13.3. Statistical analyses of the FoRAGE database were performed in R v.4.3.2.

code

Within the code/mathematica subfolder, the notebook FuncResp_n-prey-at-a-time.nb contains the primary analyses presented in the manuscript.

Within the code/R subfolder, Rpackages.md lists the required R packages, data_prep.R standardizes the format of all FoRAGE datasets, data_subset.R specifies which datasets are to be removed prior to model fitting, fit_datasets.R performs the model fitting, and analysis_Hn.R is used for the analysis of the resulting fits. Model fitting uses a library of functions located in the lib sub-folder originally developed with Daniel Stouffer (see General Functional Responses repository).

data

Contains the FoRAGE v4 database in .csv and .Rdata formats. Individuals interested in using the FoRAGE database for additional analyses should instead obtain it from KNB - The Knowledge Network for Biocomplexity repository where the FoRAGE database is maintained.

figs & tables

These contain all figures and statistical tables associated with the manuscript, as well as unused supplemental figures.

results & temp

The results folder contains each of the dataset-specific model fits produced by code/R/fit_datasets.R. The temp folder contains fitting error summaries for debugging purposes, as well as temporary Mathematica files generated during the mathematical analyses.

Warning: Parameter interpretation

As noted in the Supplementary Materials of the above-referenced manuscript, unlike in the original analyses of the FoRAGE database (e.g., Uiterwaal & DeLong 2022), the model-fitting analyses implemented here have not standardized prey and predator abundances by area (or volume) and have not standardized prey consumption values by time. Therefore, estimates of the attack rates and handling times are dataset-specific (i.e. are not in common units) and should not be compared across datasets. Estimates of the maximum number of prey that can be handled at a time (parameter n of the multi-prey functional response model) are unitless.

Warranty

All code is provided "as is" and without warranty.

Contact

Please email Mark Novak (mark.novak@oregonstate.edu) with any questions about the analyses and John DeLong (jpdelong@unl.edu) with any questions about the FoRAGE database.