This package contains code for creating Used-Habitat Calibration Plots as described in:
Fieberg, J.R., Forester, J.D., Street, G.M., Johnson, D.H., ArchMiller, A.A. and Matthiopoulos, J., 2018. Used‐habitat calibration plots: a new procedure for validating species distribution, resource selection, and step‐selection models. Ecography, 41(5), pp.737-752.
Programmer: John Fieberg | Secondary Programmer: Althea ArchMiller
Used-Habitat Calibration plots are a new method for model calibration that can be applied across a spectrum of species distribution models (SDMs) (Fieberg et al. In review for Ecography). To create a UHC plot, we first split our data into training and testing data sets. Then, we:
With UHC plots, we can validate models by comparing the distributions of observed and predicted values of the explanatory variables at the presence (or used) locations and compare them with the distributions of the absence (or background) locations. For instance, let's suppose that we have a species that is distributed on the landscape based on a quadratic relationship with temperature (x). The following figure demonstrates that when we fit a mis-specified model (i.e., linear temperature covariate only; left panel), our UHC simulation envelope (grey band) fails to contain the distribution of temperature at used/presence locations (solid black line). This indicates that the model is not well-calibrated. When we add the quadratic temperature term, the UHC simulation envelope contains the used location distribution (right panel), indicating a well-calibrated model. The data used in this example can be simulated with the uhcdatasimulator
function. The distribution of available or background points is given in this plot as a red dashed line.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.