Rapid Post-Wildfire Burned Vegetation Assessment with Google Earth Engine (Case Study: 2023 Canada Wildfires)
Bibliographic record
Abstract
Abstract. Wildfires are significant environmental threats, requiring precise and prompt assessment to mitigate damage and guide recovery efforts. Remote sensing, mainly through satellite imagery multispectral data, provides practical tools for monitoring and evaluating wildfire impacts. Canada experiences significant wildfires each year, causing substantial damage to the country’s environment, particularly its vegetation. This study proposed a fast and efficient method using Google Earth Engine (GEE) cloud-based computing to rapidly assess burned vegetation following a wildfire in Canada in 2023, utilizing Sentinel-2 imagery data. This method computed NDVI, GNDVI, and EVI spectral indices for classifying pre-fire vegetation cover and NBR, dNBR, and MIRBI for classifying post-fire burn severity. These spectral indices served as input data for machine learning models, including K-Nearest Neighbors (KNN), Random Forest (RF), Gradient Boosting Decision Trees (GBDT), and Support Vector Machine (SVM). Ultimately, the results of vegetation cover and burn severity classifications, performed separately by these models, were combined using a decision-level fusion with a weighting approach based on an accuracy approach to produce integrated and final classifications. Subsequently, by overlapping the results of these fused classifications, the burned vegetation was assessed, and its area was estimated. According to the study's results, significant damage was observed in the vegetation after the wildfire. 4489 km2 of the study area, which was a Military Grid Reference System (MGRS) tile with an area of 12,000 km2, was burned due to the wildfire. 34.06% of this area was specifically burned vegetation, equating to approximately 4,088 km2.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 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".