Fitting a simple habitat suitability model

Overview

This is a short example vignette demonstrating how code, output, and figures can be embedded directly into a research page. Here, we simulate a simple presence/absence dataset and fit a logistic regression to estimate habitat suitability as a function of forest cover.

This is placeholder content to demonstrate the format — swap in real data and analysis for an actual project write-up.

Simulating example data

set.seed(42)
n <- 200
forest_cover <- runif(n, 0, 1)
linear_pred <- -2 + 5 * forest_cover
prob_presence <- 1 / (1 + exp(-linear_pred))
presence <- rbinom(n, 1, prob_presence)

data <- data.frame(forest_cover, presence)
head(data)

  forest_cover presence
1    0.9148060        1
2    0.9370754        1
3    0.2861395        0
4    0.8304476        1
5    0.6417455        1
6    0.5190959        1

Fitting the model

model <- glm(presence ~ forest_cover, data = data, family = binomial)
summary(model)

Call:
glm(formula = presence ~ forest_cover, family = binomial, data = data)

Coefficients:
             Estimate Std. Error z value Pr(>|z|)
(Intercept)   -2.0954     0.3403  -6.158 7.36e-10 ***
forest_cover   5.3781     0.6708   8.017 1.08e-15 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 273.93  on 199  degrees of freedom
Residual deviance: 173.92  on 198  degrees of freedom
AIC: 177.92

Visualising the relationship

Interpretation

The fitted model suggests presence probability increases with forest cover, consistent with expectations for forest-dependent species. In a real write-up, this section would discuss model diagnostics, validation, and ecological interpretation in more depth.

Related Publications

Forthcoming.