The Rocky Mountain Snowpack dataset
Rocky Mountain Digerati publishes an open dataset of snowpack imagery collected in the Colorado backcountry: cross-sectional photos of extracted snow cores and magnified snowpack profile images, captured in chronological series so samples can be related to weather, geographic, and site conditions. Collection uses a novel mini-coring protocol — read about the field methods.
The dataset is public under a CC-BY-4.0 license and hosted on Hugging Face as rmdig/rocky_mountain_snowpack. You can pull it directly with the huggingface_hub Python library and use it for machine-learning research, snow-science coursework, or anything else the license permits.
The research it feeds
This corpus is the foundation for our avalanche-safety research line: the open-source snowGAN models over the two imaging modalities (magnified profile imagery and core samples) and, building on those, AvAI — Avalanche AI — a model in development that studies avalanche-risk signals in snowpack imagery alongside weather timeseries. This work is research in progress: it is not part of the AvAI app today, and when it first ships it will be clearly labeled as a research preview.
Contributing
The dataset grows through fieldwork. The AvAI app — our beta backcountry safety companion — includes field capture of snowpack profile and core photos; at this stage of the beta those photos stay on your device, and a future opt-in contribution flow will let users add their captures to this public dataset with explicit consent.
Citation & contact
Cite the dataset via its Hugging Face card, which carries the canonical citation text and revision history. Questions, collaboration, or corrections: support@rmdig.ai.