Preprints
https://doi.org/10.5194/egusphere-2025-970
https://doi.org/10.5194/egusphere-2025-970
28 May 2025
 | 28 May 2025

Identifying alpine treeline species using high-resolution WorldView-3 multispectral imagery and convolutional neural networks

Laurel A. Sindewald, Ryan Lagerquist, Matthew D. Cross, Theodore A. Scambos, Peter J. Anthamatten, and Diana F. Tomback

Abstract. Alpine treeline systems are remote and difficult to access, making them natural candidates for remote sensing applications. Remote sensing applications are needed at multiple scales to connect landscape-scale responses to climate warming to finer-scale spatial patterns, and finally to community processes. Reliable, high-resolution tree species identification over broad geographic areas is important for connecting patterns to underlying processes, which are driven in part by species’ tolerances and interactions (e.g., facilitation). To our knowledge, we are the first to attempt tree species identification at treeline using satellite imagery. We used convolutional neural networks (CNNs) trained with high-resolution WorldView-3 multispectral and panchromatic imagery, to distinguish six tree and shrub species found at treeline in the southern Rocky Mountains: limber pine (Pinus flexilis), Engelmann spruce (Picea engelmannii), subalpine fir (Abies lasiocarpa), quaking aspen (Populus tremuloides), glandular birch (Betula glandulosa), and willow (Salix spp.). We delineated 615 polygons in the field with a Trimble geolocator, aiming to capture the high intra- and interspecies variation found at treeline. We adapted our CNN architecture to accommodate the higher-resolution panchromatic and lower-resolution multispectral imagery within the same architecture, using both datasets at their native spatial resolution. We trained four- and two-class models with aims to 1) discriminate conifers from each other and from deciduous species, and 2) to discriminate limber pine—a keystone species of conservation concern—from the other species. Our models performed moderately well, with overall accuracies of 44.1 %, 46.7 %, and 86.2 % for the six-, four-, and two-class models, respectively (as compared to random models, which could achieve 28.0 %, 35.1 %, and 80.3 %, respectively). In future work, our models may be easily adapted to perform object-based classification, which will improve these accuracies substantially and will lead to cost-effective, high-resolution tree species classification over a much wider geographic extent than can be achieved with uncrewed aerial systems (UAS), including regions that prohibit UAS, such as in National Parks in the U.S.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Share
Laurel A. Sindewald, Ryan Lagerquist, Matthew D. Cross, Theodore A. Scambos, Peter J. Anthamatten, and Diana F. Tomback

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2025-970', Anonymous Referee #1, 06 Jul 2025
    • AC1: 'Reply on RC1', Laurel Sindewald, 23 Aug 2025
    • AC2: 'Reply on RC1', Laurel Sindewald, 23 Aug 2025
  • RC2: 'Comment on egusphere-2025-970', Anonymous Referee #2, 22 Jul 2025
    • AC3: 'Reply on RC2', Laurel Sindewald, 23 Aug 2025
Laurel A. Sindewald, Ryan Lagerquist, Matthew D. Cross, Theodore A. Scambos, Peter J. Anthamatten, and Diana F. Tomback

Data sets

Sindewald et al - Identifying alpine treeline species using high-resolution WorldView-3 multispectral imagery and convolutional neural networks dataset Laurel A. Sindewald, Ryan Lagerquist, Matthew D. Cross, Theodore A. Scambos, Peter J. Anthamatten, and Diana F. Tomback https://zenodo.org/records/14942410

Model code and software

thunderhoser/wv3_species_classification Ryan Lagerquist, Laurel A. Sindewald, Matthew D. Cross, Theodore A. Scambos, Peter J. Anthamatten, and Diana F. Tomback https://doi.org/10.5281/zenodo.14946215

Laurel A. Sindewald, Ryan Lagerquist, Matthew D. Cross, Theodore A. Scambos, Peter J. Anthamatten, and Diana F. Tomback

Viewed

Total article views: 434 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
342 73 19 434 36 11 24
  • HTML: 342
  • PDF: 73
  • XML: 19
  • Total: 434
  • Supplement: 36
  • BibTeX: 11
  • EndNote: 24
Views and downloads (calculated since 28 May 2025)
Cumulative views and downloads (calculated since 28 May 2025)

Viewed (geographical distribution)

Total article views: 432 (including HTML, PDF, and XML) Thereof 432 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 09 Sep 2025
Download
Short summary
We used high-resolution satellite imagery and artificial intelligence models to identify six tree and shrub species commonly found at alpine treeline in the Rocky Mountains with accuracies from 44.1% to 86.2%. We are the first to attempt species identification using satellite imagery in treeline systems, where trees are small and difficult to identify remotely. Our work provides a method to identify species with satellite imagery over a broader geographic range than can be achieved with drones.
Share