Phillips et al. (2025) Remotely sensed tree mortality rates in mesic forests of the US Southwest during an extended drought

Identification

• Journal: Forest Ecology and Management
• Year: 2025
• Date: 2025-10-09
• Authors: C. E. Lucas Phillips, Kyle C. Rodman
• DOI: 10.1016/j.foreco.2025.123198

Research Groups

• Department of Geography, Planning, and Recreation, Northern Arizona University
• Ecological Restoration Institute, Northern Arizona University

Short Summary

This study quantified recent tree mortality in isolated mesic forests of the US Southwest from 1997 to 2023 using high-resolution aerial imagery and Landsat time series, finding overall low mortality rates comparable to background levels but highlighting challenges in remotely sensing low-severity disturbances.

Objective

• To quantify recent tree mortality due to insect outbreaks and drought in isolated, mesic forests of the US Southwest.
• To create a 30-meter map of cumulative mortality (percentage of initial canopy cover that died) from 1997 to 2023.

Study Configuration

• Spatial Scale: Mesic forests of the US Southwest, across 1076 forest stands, mapped at 30-meter resolution.
• Temporal Scale: 1997 to 2023 (26 years).

Methodology and Data

• Models used: Random Forest (RF) modeling.
• Data sources: Manually interpreted high-resolution aerial imagery, Landsat time series.

Main Results

• Overall mortality rates averaged 0.4 % canopy cover per year.
• These rates were comparable to field-derived background mortality rates for large trees in the western US (0.5 % to 2 % per year).
• Low mortality (< 30 % cumulative; < 1 % per year) was observed in 94 % of the total area.
• Higher mortality (> 80 % cumulative; > 2.5 % per year) occurred in fragmented, high-elevation patches.
• Changes in shortwave infrared reflectance (e.g., 1.55–1.75 micrometers) were the strongest predictors of tree mortality.
• The RF model exhibited low-to-moderate accuracy (R² = 0.28, RMSE = 18 % mortality).
• Wildfire may have been a more significant driver of change in mesic forest systems of the US Southwest than drought-caused tree die-off and background mortality since approximately 2000.

Contributions

• Provides a quantitative, landscape-scale assessment of background tree mortality rates in vulnerable, isolated mesic forests of the US Southwest using remote sensing.
• Highlights the challenges and limitations of using Landsat imagery to capture low-severity disturbance effects.
• Offers insights for monitoring forest change in other isolated and vulnerable forest ecosystems globally.

Funding

• Not specified in the provided text.

Citation

@article{Phillips2025Remotely,
  author = {Phillips, C. E. Lucas and Rodman, Kyle C.},
  title = {Remotely sensed tree mortality rates in mesic forests of the US Southwest during an extended drought},
  journal = {Forest Ecology and Management},
  year = {2025},
  doi = {10.1016/j.foreco.2025.123198},
  url = {https://doi.org/10.1016/j.foreco.2025.123198}
}