Using UAV Imagery to Detect and Map Woody Species Encroachment in a Subalpine Grassland: Advantages and Limits

Woody species encroachment on grassland ecosystems is occurring worldwide with both negative and positive consequences for biodiversity conservation ecosystem services. Remote sensing image analysis represent useful tools the monitoring of this process. In paper, we aimed at evaluating quantitatively potential using high-resolution UAV imagery to monitor process during its early development comparing performance manual semi-automatic classification methods. The RGB images an abandoned subalpine Western Italian Alps were acquired by drone then classified through photo-interpretation, pixel- object-based models, machine-learning algorithms. techniques applied different resolution levels tested their accuracy against reference data including measurements tree dimensions collected in field. Results showed that most accurate method was photo-interpretation (≈99%), followed pixel-based approach (≈86%) faster than technique more one (≈78%). dimensional threshold juvenile detection lower but comparable one. Therefore, mapping stages, proved be a promising pragmatic choice.