A habitat distribution modelling approach was used to model the spatial distribution of kelp forests within the Bay of Morlaix (France).

Lineage:

Data represents presence-absence prediction of kelp forest. Biological ground truth data were integrated with high resolution environmental datasets to develop statistical model that accurately predict the structure of Laminaria forests within the Bay of Morlaix. As a direct management output, high-resolution map (25 m2 grid) was produced.

Type of occurrence data used:

Forest occurrence (presence or absence),representative across the full range of environmental gradients, was sampled through a combination of underwater video surveys and direct diver observations.

Environmental covariates/explanatory variables:

The probability of kelp forest occurrence and its standard deviation was predicted using an additive multiple regression of water depth, light availability, significant wave height and sediment proximity.

Algorithm/modelling approach:

Kelp biological response (presence/absence) was estimated using Generalized Additive Models (GAM)