An enhanced NHI algorithm configuration for fire detection and mapping
Bibliographic record
Abstract
The devastating fire events occurring during the intense fire season of 2023 have shown the importance of developing efficient fire detection methods capable of supporting the fire management activities. An enhanced configuration of the Normalized Hotspot Indices (NHI) algorithm has been developed in this direction to improve the fire mapping by satellite through near infrared (NIR) and short-wave infrared (SWIR) data (up to 20 m spatial resolution) from the Operational Land Imager (OLI/OLI2) and the Multispectral Instrument (MSI) aboard Landsat-8/9 (L8/9) and Sentinel-2 (S2) satellites, respectively. In this work, we show the results achieved by investigating the fire events occurring in California, Hawaii islands (USA), Yellowknife (Canada), Tenerife islands (Spain), Greece and Australia also through comparison with information from operational Landsat Fire and Thermal Anomaly (LFTA) product. Results of an extended validation analysis performed using information from well-established databases show that the enhanced NHI algorithm configuration enabled an accurate mapping of fire fronts with a very number of omission and commission errors. Moreover, the algorithm flagged up to 99% of fire pixels from the LFTA product over California and detected up to 70% of additional fire pixels, in night-time conditions, which better detailed the fire fronts and provided unique information about small-fire outbreaks. The effective integration of S2 (daytime) and L8/9 (daytime/night-time) observations, demonstrates that the enhanced NHI algorithm configuration may be used with success to analyse the dynamic evolution of flaming fronts by assessing/complementing information from satellite products at high-temporal/low-spatial resolution. The next implementation of the algorithm on from the Sea and Land Surface Temperature Radiometer (SLSTR) aboard Sentinel-3 satellite and the Flexible Combined Imager (FCI) of the Meteosat Third Generation (MTG) opens some interesting perspectives also regarding its usage for the near-real time monitoring of wildfires
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".