370 Leveraging remote sensing products to estimate forage productivity in the Canadian Prairies
Bibliographic record
Abstract
Abstract Remote sensing is widely used for land cover classification applications. This technology supports the development of land management practices and policies to achieve conservation and economic goals in agricultural landscapes. In animal production, remote sensing aids in reducing grassland degradation and enhancing livestock and crop production. The objective of this work is to discuss recent satellite- and drone-based remote sensing applications for grassland mapping in the Canadian Prairies. Using a satellite-based approach, the agreement between remotely sensed land cover maps and non-spatial government records was assessed for public agricultural Crown lands in southern Manitoba. Two different land cover classification products derived from satellite imagery [i.e., the Manitoba Grassland Inventory (MGI) prepared by the Province of Manitoba and the Annual Crop Inventory (ACI) prepared by Agriculture and Agri-Food Canada] were merged into a single mosaic covering the entire Agro-Manitoba region. Non-spatial official records were georeferenced and summarized through parcel delineation using geographic information system (GIS) and R programming tools. Comparison of the two datasets revealed low agreement between related grassland classes (i.e., forests and shrublands; and native and tame grasslands) due to spectral similarities. However, grouping these related vegetation types into broader categories (i.e., forests and shrublands as woody vegetation; native and tame grasslands as grassy vegetation) significantly improved overall agreement, with a difference of less than 3% observed between remotely sensed datasets and official records. The derived grassland classification was then used to assess the grazing status of the agricultural Crown lands by estimating stocking rates in animal unit months (AUM) and utilizing historical yield data available from three separate field surveys conducted between 2004 and 2020 across the region. Average carrying capacities ranged from 0.71 AUM/ha to 1.99 AUM/ha. Overall, the analysis using remotely sensed data indicated that the forage resources in the Crown lands were underutilized by 45%. To further explore the potential of using high-resolution remote sensing to map grasslands, ongoing research involves the use of a drone-mounted hyperspectral sensor (HySpex Mjolnir VS-620) covering a spectral range from 400 to 2,500 nm across 490 spectral channels. This sensor enables grassland mapping at significantly greater spatial and spectral resolutions. Monthly surveys during the snow-free period will be conducted in 12 study sites across the Agro-Manitoba region in 2024 and 2025, with biomass samples collected to model grassland distribution and productivity. The outcomes of this study will offer critical insights into leveraging specific wavelengths for grassland mapping and strategies for extrapolating these insights to larger geographical areas using satellite imagery. Overall, the efforts described above underscore the significance of remote sensing approaches in managing agricultural landscapes and provide valuable insights for land managers, policymakers, and stakeholders to promote sustainable agricultural production in grasslands.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".