Spatiotemporal Patterns of Intermittent Snow Cover From PlanetScope Imagery Using Deep Learning
Abstract Monitoring snow cover in regions with intermittent dynamics is a significant challenge due to the rapid changes occurring in snow accumulation and ablation over complex terrain. We trained a deep learning model with lidar‐derived labels and PlanetScope CubeSat imagery to map near‐daily snow...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-07-01
|
| Series: | Geophysical Research Letters |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2025GL116582 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Monitoring snow cover in regions with intermittent dynamics is a significant challenge due to the rapid changes occurring in snow accumulation and ablation over complex terrain. We trained a deep learning model with lidar‐derived labels and PlanetScope CubeSat imagery to map near‐daily snow cover dynamics at 3‐m resolution. The model demonstrated a high accuracy in the Salt and Verde River basins of Arizona and strong transferability to other sites in the western United States. Temporal analysis of snow line from 2021 to 2023 revealed distinct patterns of snowpack dynamics driven by seasonal and interannual climatic variability. The high‐resolution snow persistence maps also unveiled significant subgrid variability in snow cover at point and watershed scales, influenced by elevation, aspect, and vegetation cover. These findings illustrate the potential of integrating high‐resolution CubeSat imagery with deep learning models to enhance our understanding of intermittent snowpack spatiotemporal variability in complex terrain. |
|---|---|
| ISSN: | 0094-8276 1944-8007 |