Mapping Shrub and Tree Encroachment in Canadian Prairies Using Stacking Ensemble and Sentinel-1/2 Imagery
Bibliographic record
Abstract
This repository contains the full set of data, code, and analytical outputs used in the development and evaluation of an ensemble learning framework for vegetation mapping in the Foam Lake and Aberdeen study areas (Canada). The materials are organized into three main folders—Codes, Data_all, and Results—to support full transparency, reproducibility, and reuse. 1. Codes This folder includes all R scripts used for data preprocessing, feature extraction, model training, ensemble classification, result evaluation, and spatial mapping. The codebase implements an ensemble learning framework designed to classify vegetation types using multi-temporal and multi-source remote sensing inputs. Scripts are organized by workflow steps, allowing users to reproduce the analysis from raw data to final maps. 2. Data_all This folder provides the complete dataset used in the study. It includes: Feature maps for both Foam Lake and Aberdeen regions derived from remote sensing imagery. These maps contain spectral indices, texture layers, and other predictor variables used for modeling. Extracted training dataset for Foam Lake (FoamLake_summer_new.txt), which was used to train and validate the ensemble model. This file contains labeled samples extracted from the feature maps, including vegetation classes and associated predictor variables. All datasets are provided in open formats to enable reuse in machine learning, remote sensing, and ecological modeling applications. 3. Results This folder contains the outputs generated by the ensemble learning analysis, including: Final trained model used for prediction. Predictions for the test dataset, representing classification outcomes across all evaluation samples. Summary tables of model performance, variable importance, and statistical evaluations. Final prediction maps for Foam Lake and Aberdeen, representing the spatial distribution of vegetation classes generated by the ensemble framework.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
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".