AI Chat Paper
Note: Please note that the following content is generated by AMiner AI. SciOpen does not take any responsibility related to this content.
{{lang === 'zh_CN' ? '文章概述' : 'Summary'}}
{{lang === 'en_US' ? '中' : 'Eng'}}
Chat more with AI
PDF (1.6 MB)
Collect
Submit Manuscript AI Chat Paper
Show Outline
Outline
Show full outline
Hide outline
Outline
Show full outline
Hide outline
Research Article | Open Access

A quantitative framework for modeling the spatiotemporal dynamics of mountain pine bark beetle infestations utilizing Landsat time-series data and Random Forest

Hamza Taleb1( )Melinda Laituri2
Graduate Degree Program in Ecology (GDPE) at Colorado State University (CSU), Fort Collins, Colorado, USA
Department of Ecosystem Science and Sustainability (ESS) at Colorado State University, USA
Show Author Information

Abstract

Mountain pine bark beetles (MPBB, Dendroctonus ponderosae) are a primary driver of tree mortality in pine-dominant North American forests. To characterize the spatiotemporal dynamics of MPBB infestations in lodgepole pine forests of north central Colorado, we analyzed Landsat spectral trends for 2005 and 2009. Using a stratified random sampling design (N = 1,021), we classified land cover trajectories to identify stable forests, infestation, fire, clear-cutting, and regrowth sites. A Random Forest (RF) classifier was developed to detect infestation presence and predict mortality severity. The model achieved high classification accuracies of 96% (2005) and 97% (2009), while the regression for mortality severity yielded a strong fit (R2 = 0.878) with a low Root Mean Square Error (RMSE = 0.1425). A rigorous topographic analysis revealed that infestation risk is strongly non-random: South-facing slopes exhibited 9.0 times higher odds of infestation compared to north-facing slopes, likely driven by solar insolation and water stress. However, limitations remain in detecting low-severity mortality (<25% canopy loss) and distinguishing species-specific responses in mixed stands. A simple binary classification of "disturbed" versus "undisturbed, " as is the primary output of many earlier studies, is insufficient for prioritizing management actions. To address these challenges, we propose integrating multi-source data fusion (e.g., Landsat and Sentinel-2), leveraging UAV-based sub-pixel validation, and utilizing phenological metrics from Harmonized Landsat-Sentinel (HLS) data. These advanced approaches, combined with the RF modeling demonstrated, offer a pathway for more precise, early-warning monitoring of forest health in complex topographic landscapes. In addition, this approach can be used for future studies designed to track the location of trees that have developed self-immunity to the beetles with the hope of reforesting with seedlings of these resistant trees.

References

【1】
【1】
 
 
AIMS Geosciences
Pages 360-387

{{item.num}}

Comments on this article

Go to comment

< Back to all reports

Review Status: {{reviewData.commendedNum}} Commended , {{reviewData.revisionRequiredNum}} Revision Required , {{reviewData.notCommendedNum}} Not Commended Under Peer Review

Review Comment

Close
Close
Cite this article:
Taleb H, Laituri M. A quantitative framework for modeling the spatiotemporal dynamics of mountain pine bark beetle infestations utilizing Landsat time-series data and Random Forest. AIMS Geosciences, 2026, 12(2): 360-387. https://doi.org/10.3934/geosci.2026014

185

Views

20

Downloads

0

Crossref

0

Web of Science

Received: 18 August 2025
Revised: 04 January 2026
Accepted: 10 February 2026
Published: 15 June 2026
©2026 the Author(s), licensee AIMS Press.

This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)