This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
Multi-Decadal Dynamics of Forest Canopy Water Stress and GIS-Based Risk Assessment of Drought-Induced Loss in a Mediterranean-Type Forest
by
Thai Son Le
Thai Son Le 1,2
,
Bernard Dell
Bernard Dell 1,3,*
and
Richard Harper
Richard Harper 1
1
Agriculture and Forest Sciences, Murdoch University, Murdoch, WA 6150, Australia
2
Department of Environmental Management, Vietnam National University of Forestry, Hanoi 13417, Vietnam
3
Forest Protection Research Centre, Vietnamese Academy of Forest Sciences, Hanoi 11910, Vietnam
*
Author to whom correspondence should be addressed.
Remote Sens. 2026, 18(12), 1975; https://doi.org/10.3390/rs18121975 (registering DOI)
Submission received: 4 May 2026
/
Revised: 4 June 2026
/
Accepted: 11 June 2026
/
Published: 13 June 2026
Abstract
Mediterranean-type forest ecosystems are becoming increasingly vulnerable to intensifying drought, threatening the resilience of even highly adapted ecosystems such as the Northern Jarrah Forest in south-western Australia. This study quantifies multi-decadal dynamics of canopy water stress using a 36-year multispectral satellite archive (1988–2024) and the newly developed Infrared Canopy Dryness Index (ICDI). We combined this spatiotemporal dataset with a MaxEnt-based risk assessment framework to identify the biophysical drivers of drought-induced canopy loss and to delineate high-risk zones under accelerating climate-forcing changes. Our results demonstrate a systematic spatial expansion of canopy dryness, paralleling a deteriorating regional climatic water balance. Hotspot analysis revealed a transition from localized, peripheral stress to widespread, chronic drought conditions across the landscape. The modelling achieved high diagnostic accuracy (AUC = 0.952), significantly outperforming conventional assessment methods. Regolith depth was identified as the primary determinant of drought-induced canopy collapse, followed by ICDI, NDVI, and slope. Crucially, high-biomass stands exhibited disproportionately higher risk of collapse, revealing a density-dependent vulnerability that suggests productive forests are approaching critical hydraulic thresholds. Conversely, lower-stature forests to the east of the study area demonstrated greater stability, likely due to reduced evapotranspirative demand. These findings provide robust spatial evidence for transitioning from reactive monitoring to proactive forest management. We conclude that targeted interventions, such as ecological thinning and prescribed burning in identified high-risk zones, are imperative to protect the forest and preserve the structural integrity of Mediterranean ecosystems in a drying climate.
Share and Cite
MDPI and ACS Style
Le, T.S.; Dell, B.; Harper, R.
Multi-Decadal Dynamics of Forest Canopy Water Stress and GIS-Based Risk Assessment of Drought-Induced Loss in a Mediterranean-Type Forest. Remote Sens. 2026, 18, 1975.
https://doi.org/10.3390/rs18121975
AMA Style
Le TS, Dell B, Harper R.
Multi-Decadal Dynamics of Forest Canopy Water Stress and GIS-Based Risk Assessment of Drought-Induced Loss in a Mediterranean-Type Forest. Remote Sensing. 2026; 18(12):1975.
https://doi.org/10.3390/rs18121975
Chicago/Turabian Style
Le, Thai Son, Bernard Dell, and Richard Harper.
2026. "Multi-Decadal Dynamics of Forest Canopy Water Stress and GIS-Based Risk Assessment of Drought-Induced Loss in a Mediterranean-Type Forest" Remote Sensing 18, no. 12: 1975.
https://doi.org/10.3390/rs18121975
APA Style
Le, T. S., Dell, B., & Harper, R.
(2026). Multi-Decadal Dynamics of Forest Canopy Water Stress and GIS-Based Risk Assessment of Drought-Induced Loss in a Mediterranean-Type Forest. Remote Sensing, 18(12), 1975.
https://doi.org/10.3390/rs18121975
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article Access Statistics
For more information on the journal statistics, click
here.
Multiple requests from the same IP address are counted as one view.