CONUS Crop Field Boundary Extraction

Feb 23

*The above images are a demonstration of Drs. Yan and Roy’s crop field extraction in Louisiana from commercial PlanetScope images with 3m resolution, more details at* *https://www.nasaacres.org/news/fieldboundaries*.

Project Summary

Spatially explicit data on crop field boundaries is lacking, yet this essential agricultural variable is an important proxy for other important indicators. Agricultural crop field sizes are indicative of the degree of agricultural capital investment, mechanization, and labor intensity. Information on field size, field-size change, and delineated field boundaries are needed to understand these factors, for land use planning, allocation of resources, and agricultural modeling. This project aims to help state and national policymakers make better-informed decisions by accessing data at field- or farm-level rather than pixel- or county-level. Drs. David Roy and Lin Yan are taking a computer-vision approach to extract crop field objects from Landsat satellite time series to extract wall-to-wall Conterminous United States (CONUS) field boundaries for 2008-2026. The core processing will be undertaken using the USGS Collection 2 Analysis Ready Data (ARD) stored in commercial cloud Amazon Web Services (AWS) and will be validated using Google-Earth and NASA high spatial resolution imagery.

Study Area:

Conterminous United States (CONUS)

Earth Observations Used

Collection 2 Landsat ARD

Related News

Featured