DOI · DOI-0148 · record R2614

Machine Learning for Avalanche Frequency Modeling

Department of the Interior · USGS

Overview

Development stage: DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
High-impact designation: Not high-impactSource: c) Not high-impact
Impact justification: Not reported
Start date: 2024-01-24 (day precision)Source: 1/24/2024
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Science
Operational functions: No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

The machine learning (Random Forest) was used to identify vegetation characteristics in avalanche paths. This helps determine avalanche return periods in specific avalanche paths.

Expected benefits

Benefits include results that inform avalanche forecasters, transportation departments, and infrastructure planners on estimating spatial extents of avalanche return periods.

System outputs

The Random Forest model outputs vegetation classification that is necessary for identifying avalanche return periods.

Technology

AI classification: Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…
System name(s): Not reported
Custom-developed code: No
Public source code: Not reported

Sourcing

How it was built: Developed in-houseSource: b) Developed in-house
Vendor (as reported): Not reported
Vendors (standardized): None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: No
Privacy Impact Assessment: Not reported
Authorization to Operate: No
Demographic features: k) None of the above
Training and evaluation data: We used high point density lidar data. Data collection and processing adhered to a maximum nominal post spacing of 0.35 m, with a mean point density of 16 points/m2. We also used four-band (red, green, blue, and near-infrared) NAIP imagery at 0.6-m horizontal spatial resolution.
Federal Data Catalog: Not reported