Fractal Dimension Estimation
One of the central geometric questions in the project is whether wildfire perimeters behave like smooth Euclidean curves or like rough fractal boundaries.
Candidate estimators
- box-counting methods
- perimeter-area scaling
- divider methods
Why multiple estimators matter
Fractal dimension can be sensitive to:
- spatial resolution
- smoothing procedures
- measurement scale
- event size
Using more than one estimator helps separate geometric signal from measurement artifact.
Why this matters for the proposal
The project is not only interested in whether perimeter growth follows a power law. It also asks whether the geometry of the boundary itself is consistent with a non-Euclidean growth process.