WorkspaceSpecies Distribution Modeling

Species Distribution Modeling

Fit explainable habitat occupancy models to presence–absence surveys and evaluate predictions on held-out spatial blocks.

Application guide

Verifying your session…

Using Species Distribution Modeling

Fit explainable species–habitat relationships and test out-of-block predictions and model sensitivity.

Presence–absence observations · Regularized logistic fit · Spatial cross-validation · Prediction spread

Inputs and working data
Spatially blocked, source-labelled presence/absence observations, moisture and canopy covariates, prediction grid and model parameters.
Working in the editor
Plot and grid import, penalty and classification settings, probability curves, fold diagnostics and coefficient report.
What can I produce?
Quadrat predictions, fold ranges, coefficients, training scales, source-linked held-out probabilities and model comparison.
What are the limits?
Two environmental covariates, observed presence/absence and bounded data. Sampling effort and imperfect detection are not inferred.
Example
Compare regularization on independent survey blocks, check reduced habitat contrast and restore the model.
Back to the editor