Wildfire detection employing an imaging spectrometer with a digital focal plane array
Bibliographic record
Abstract
Wildfires increasingly endanger people and property due to the growing population in the wildland urban interface, drought, and climate change. In the United States in 2023 over 1,000,000 acres burned in the western CONUS with no fire encompassing over 100,000 acres. Also, tragically the Lahaina Fire in Hawaii caused the deaths of over 100 people. In Canada, the extreme 2023 fire season resulted in almost 18,500,000 hectares burned, which was a factor of 2.6 larger than the previous high in 1995. The economic losses are enormous with resource expenditures running into the billions and insured losses running into the tens of billions of dollars in the United States. We propose the application of an imaging spectrometer for pre- and post-fire assessments and fire detection. MIT Lincoln Laboratory has developed three critical technologies that are applicable to the wildfire problem. The first is a compact spectrometer, the Chrisp Compact VNIR/SWIR Imaging Spectrometer (CCVIS), that can be modularly implemented for a wide-field imaging spectrometer. The second is the digital focal plane array (DFPA) technology with different detector materials, such as InGaAs or Mercury Cadmium Telluride (MCT), and extremely large well depths exceeding 108 electrons. The DFPA is critical for this application since traditional FPAs will saturate even for relatively cool fires with small spatial sample fill fractions. The DFPA also has sufficient signal to noise performance for pre- and post-fire products such as canopy cover, fuel quantification, and burnt area quantification and monitoring. The third is the TeraByte InfraRed Delivery (TBIRD) space-to-ground optical link that has a maximum data rate of 800 Gbps, which will not be addressed here. A small satellite implementation in a low Earth orbit (∼450 km) will have an entrance pupil on the order of 10 cm for a 50 m ground sample distance (GSD).
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| 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 | 0.000 |
| 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".