Related Experiment Video
Updated: Jun 20, 2026

Computer Vision-Based Biomass Estimation for Invasive Plants
Published on: February 9, 2024
Cropland mask dataset for the Canadian Prairies derived from Google satellite embedding imagery
Thuan Ha1,2, Kwabena Abrefa Nketia1,2, Shawn Neudorf1,2
1Department of Plant Sciences, College of Agriculture and Bioresources, University of Saskatchewan, Saskatoon, SK S7N 5A8, Canada.
A new cropland mask identifies persistent agricultural areas in the Canadian Prairies from 2017-2024. This stable mask supports land use assessment and agricultural monitoring using high-resolution geospatial data.
Area of Science:
- Geospatial Science
- Agricultural Science
- Remote Sensing
Background:
- Accurate cropland mapping is essential for agricultural monitoring and land-use assessments.
- Existing datasets may lack the spatial resolution or temporal consistency needed for detailed analysis.
Purpose of the Study:
- To develop a spatially explicit, persistent cropland mask for the Canadian Prairies (Alberta, Saskatchewan, Manitoba).
- To provide a stable, high-resolution (10 m) dataset for diverse agricultural and environmental applications.
Main Methods:
- Utilized 64-band AlphaEarth data embeddings for classification.
- Employed a Random Forest classifier trained on stratified-random points.
- Generated annual cropland classifications from 2017-2024 and aggregated them into a multi-year stable mask.
Main Results:
- Created a 10 m resolution binary raster mask identifying persistent cropland areas.
- The mask highlights areas with continuous cropping for over two years.
- Provided cloud-optimized GeoTIFFs for broad accessibility and use.
Conclusions:
- The persistent cropland mask offers a reliable tool for Canadian Prairie agricultural research.
- Enables consistent identification of long-term croplands for land-use change and monitoring.
- Supports advanced applications in precision agriculture, climate studies, and machine learning.
Related Concept Videos
Light Acquisition
GIS Software, Hardware, and Sources of GIS Data
Masking and Demasking Agents
There are many masking agents, such as cyanide, fluoride, triethanolamine, thiourea, and 2,3-bis(sulfanyl)propan-1-ol (formerly 2,3-dimercapto-1-propanol), with the masking agent chosen based on the metal...
Responses to Drought and Flooding
Levels of Use of a GIS
Topographic Surveying and Contours