Naturalness Indicators of Forests in Southern Sweden derived from the Canopy Height Model

TL;DR

This study introduces a methodology for extracting a comprehensive and human-interpretable set of features from the Canopy Height Model (CHM) with a resolution of 1 meter to identify reliable indicators of the degree of naturalness of forests in Southern Sweden.

Abstract

Forest canopies embody a dynamic set of ecological factors, acting as a pivotal interface between the Earth and its atmosphere. They are not only the result of an ecosystem's ability to maintain its inherent ecological processes, structures, and functions but also a reflection of human disturbance. This study introduces a methodology for extracting a comprehensive and human-interpretable set of features from the Canopy Height Model (CHM), which are then analyzed to identify reliable indicators for the degree of naturalness of forests in Southern Sweden. Utilizing these features, machine learning models - specifically, the perceptron, logistic regression, and decision trees - are applied to predict forest naturalness with an accuracy spanning from 89% to 95%, depending on the area of the region of interest. The predictions of the proposed method are easy to interpret, something that various stakeholders may find valuable.

Figures (8)

• Figure 1: Overview of the proposed method. The classifier is trained in a supervised machine learning fashion.
• Figure 2: The study area in Southern Sweden: Continental view (left), overview (center), and zoom (right). Training, validation, and test regions are shown in (bright) green, yellow, and red, respectively. All the three images are centered in 58.4881 N, 15.1000 E; EPGS:3006 projection.
• Figure 3: Areas of the forests considered in training, validation, and test sets.
• Figure 4: Example of tree density ($\rm{TD}=0.574$). The image on the left shows the CHM (shades of green) and the bounds of the region of interest (in red). The image on the right shows the area covered by trees (dark blue) and the remaining area (yellow) inside the region of interest, as well as the area outside the region of interest (light pink).
• Figure 5: Example of treetop identification. The image on the left shows the CHM. The image on the right shows the treetops (in magenta) identified in the same area.