Skip to main content
ARS Home » Plains Area » Las Cruces, New Mexico » Range Management Research » Research » Research Project #440416

Research Project: Collection of Aeolian Landform Dataset for Inference Model Assessment

Location: Range Management Research

Project Number: 3050-11210-009-082-I
Project Type: Interagency Reimbursable Agreement

Start Date: Jun 5, 2021
End Date: Sep 30, 2022

Operational models used to generate authoritative weather forecasts are implementations of the Unified Model (UM), a state-of-the-art numerical weather prediction capability developed and managed by the United Kingdom Met Office. The current UM configuration struggles to simulate individual dust event hazards. A novel surface erodibility parameterization, called ERDC-Geo, uses geomorphic landform maps to generate a spatially varying dust emission flux-scaling factor. Preliminary results suggest ERDC-Geo enhances the UM dust emission scheme. To further improve UM simulations, spatial coverage of the landform dataset needs to be expanded to global coverage for all desert regions through machine learning and artificial intelligence (ML/AI) approaches. An independent aeolian landform dataset is therefore required to evaluate the representativeness of the resultant dataset in desert regions outside the ML/AI training data region.

1. Review selected field locations and provide feedback to the CRREL team. 2. Review the field methodology and provide feedback to the CRREL team. 3. Conduct fieldwork at locations specified by the CRREL team. 4. Provide both raw and summarized field data to the CRREL team. 5. Aggregate available rangeland monitoring datasets containing landform descriptions into a single database to support landform dataset validation. Public data from the BLM AIM program, NRCS LMF, and USDA Jornada MURV study will be harmonized and aggregated in an online Landscape Data Commons. 6. Crosswalk observed landforms in monitoring datasets to ERDC landform classes for validation. 7. Review landform dataset and provide feedback on landform dataset accuracy in northern Chihuahuan Desert. 8. Assist in final white paper documentation.