Forest disturbance dynamics in Bavaria: A novel approach for identifying fragmentation processes based on canopy cover loss and morphological image processing
- 1German Remote Sensing Data Center (DFD), German Aerospace Center (DLR), 82234 Wessling, Germany; Corresponding author.
- 2German Remote Sensing Data Center (DFD), German Aerospace Center (DLR), 82234 Wessling, Germany
- 3German Remote Sensing Data Center (DFD), German Aerospace Center (DLR), 82234 Wessling, Germany; Institute for Geography and Geology, University of Wuerzburg, 97074 Wuerzburg, Germany
Abstract
Unprecedented canopy loss has become a feature of central European temperate forests in recent years and is expected to continue. Bavaria, the largest and most forested state in Germany lost 146,320 ha from September 2017 – July 2025. To investigate the dynamics of forest disturbance, we classified interior, exterior, and convergent fragmentation processes (perimeter loss, shrinkage, perforation, and bisection) over an 8-year study period by intersecting the Forest Canopy Cover Loss (FCCL) dataset with a spatial classification using morphological image processing of 10 m forest mask rasters. We delineated 762,602 individual clusters of forest canopy loss, ranging in size from 0.01 to 1228 ha. The largest clusters were identified as bisection processes and were initiated early in the study period and contained multiple sub-clusters, suggesting expansion and merging of disturbance events. Fragmentation process labeling was sensitive to the edge depth, however the most robust processes were perforation, which covered 43% of total cluster area and perimeter loss (1.8%). Although nearly 34% of forest core areas transitioned into edge zones due to perforations, perimeters were the most relatively affected zone, whereby >7% of canopy was lost. While the number of small forests (‘Islets’) increased by 167.9%, core forest area declined along with a 12.6% decrease in the number of patches. This pattern suggests perimeter loss and bisection of core forests. Because fragmentation processes precipitate localized canopy restructuring and landscape reconfiguration, which have distinct ecological outcomes, this approach can be further applied both for spatially-explicit modeling of forest disturbance and as an indicator for monitoring habitats and biodiversity.
Related
- Pollen and non-pollen palynomorph signals of cultivation-related disturbance in a preservation-limited tropical soil profile, northern Thailand
- From fragmentation to fault zone architecture as a structural framework for fault weakening
- Tourism urbanization and landscape fragmentation in Taiwan’s Yilan Plain: an integrated analysis of intensity and landscape metrics
- Landscape fragmentation and carbon storage dynamics under co-evolving governance in the Shaanxi Yellow River wetland
- Trajectory tracking control of unmanned aerial vehicle with disturbance observer and robust adaptive neural dynamic surface
BibTeX
@article{coleman2026forest,
title = {{Forest disturbance dynamics in Bavaria: A novel approach for identifying fragmentation processes based on canopy cover loss and morphological image processing}},
author = {Kjirsten Coleman and Marco Wegler and Sarah Asam and Claudia Kuenzer},
journal = {Ecological Indicators},
year = {2026},
month = oct,
doi = {10.1016/j.ecolind.2026.115487},
url = {https://doi.org/10.1016/j.ecolind.2026.115487},
note = {CC BY},
}
RIS
TY - JOUR
TI - Forest disturbance dynamics in Bavaria: A novel approach for identifying fragmentation processes based on canopy cover loss and morphological image processing
AU - Kjirsten Coleman
AU - Marco Wegler
AU - Sarah Asam
AU - Claudia Kuenzer
PY - 2026
DA - 2026/10/23
JO - Ecological Indicators
DO - 10.1016/j.ecolind.2026.115487
UR - https://doi.org/10.1016/j.ecolind.2026.115487
AB - Unprecedented canopy loss has become a feature of central European temperate forests in recent years and is expected to continue. Bavaria, the largest and most forested state in Germany lost 146,320 ha from September 2017 – July 2025. To investigate the dynamics of forest disturbance, we classified interior, exterior, and convergent fragmentation processes (perimeter loss, shrinkage, perforation, and bisection) over an 8-year study period by intersecting the Forest Canopy Cover Loss (FCCL) dataset with a spatial classification using morphological image processing of 10 m forest mask rasters. We delineated 762,602 individual clusters of forest canopy loss, ranging in size from 0.01 to 1228 ha. The largest clusters were identified as bisection processes and were initiated early in the study period and contained multiple sub-clusters, suggesting expansion and merging of disturbance events. Fragmentation process labeling was sensitive to the edge depth, however the most robust processes were perforation, which covered 43% of total cluster area and perimeter loss (1.8%). Although nearly 34% of forest core areas transitioned into edge zones due to perforations, perimeters were the most relatively affected zone, whereby >7% of canopy was lost. While the number of small forests (‘Islets’) increased by 167.9%, core forest area declined along with a 12.6% decrease in the number of patches. This pattern suggests perimeter loss and bisection of core forests. Because fragmentation processes precipitate localized canopy restructuring and landscape reconfiguration, which have distinct ecological outcomes, this approach can be further applied both for spatially-explicit modeling of forest disturbance and as an indicator for monitoring habitats and biodiversity.
SN - 1470-160X
SN - 1872-7034
ER -