Bald eagle habitat suitability, Riding Mountain

A fuzzy-membership suitability surface, built because the binary version answered a question nobody asked

Academic coursework  /  February 2022

Discipline
Suitability modelling
Tools
ArcGIS
Methods
fuzzy membership, fuzzy overlay, distance decay, raster reclassification
Data
Manitoba Land Initiative

Notes on method

Binary suitability modelling answers a question with a hard edge: a cell either satisfies the criteria or it does not. For a nesting bird that is the wrong shape of answer. Bald eagles prefer to be near water and away from human activity, but preference is not a rule, and a map that draws a line at a fixed distance from a river is asserting something about eagles that no eagle agreed to.

So the criteria were expressed as fuzzy membership instead, each cell carrying a degree of belonging between zero and one rather than a verdict, and combined with fuzzy overlay. The output is a gradient, which is both less satisfying to look at and more honest about what is known.

The plate carries its own explanation in the margin, which is deliberate. A suitability surface without a stated method invites a reader to treat the colours as fact.

This is a separate study from the Sprague’s Pipit capstone, and the contrast between them is the point: the same park, two species, two modelling approaches chosen for different reasons.

Continuous suitability surface across Riding Mountain National Park, graded from pink for least suitable to pale green for most suitable, with suitability concentrating in narrow bands along rivers and around lakes rather than in sharp blocks.
Suitability as a continuous surface. The greener bands track water and fall away from roads, but they fade rather than stop.Base data: Manitoba Land Initiative. NAD 1983, Transverse Mercator, 1:400,000Open larger (2432px)