Evaluation of Fire Danger and Fire Potential Indices for South Africa : case studies in Mpumalanga and the Western Cape
Bibliographic record
Abstract
Wildfires are a common phenomenon on earth and can have disastrous effects on the environment, \ninfrastructure and surrounding communities. At the same time, many ecosystems are fire prone and \nrequire burning at regular intervals, in order to maintain the health of the ecosystems. It is necessary \nto minimise the negative effects of fires where possible. Information needs to be provided to fire \nmanagement officials to facilitate efficient planning and mitigation in order to minimise the negative \neffects. Wildfires are influenced by many variables including vegetation type, fuel load, fuel \nmoisture, proximity to roads, proximity to settlements, elevation, slope, aspect, temperature, \nprecipitation, wind and relative humidity. These variables can be used to build a fire potential index \nthat determines the probability of a fire occurrence and the possibility of the fire to become an out \nof control fire. Fire potential indices provide information on where fire potential is high so fire \nmanagement officials can plan resources accordingly and thus minimise negative impacts of \nwildfires. Many fire potential indices have been developed but their usefulness in South Africa has \nnot been verified. The aim of the research was to implement and evaluate different fire potential \nindices utilising geographic information, including remote sensing products, to predict fire potential \nin South Africa. The Mpumalanga and the Western Cape provinces were used as case studies. The \ntime periods included February to December 2015 for Mpumalanga and August 2014 to June 2015 \nfor the Western Cape. A number of candidate fire potential indices were implemented in the Python \nscripting language. A variety of data sources were used to implement the fire potential indices. The \nfire potential indices were evaluated along with a few fire danger indices. The performance \nevaluation compared satellite detected active fire events to the fire potential indices in the study \nareas based on statistical metrics including Pseudo R2, C-Index, Eastaugh’s Two-Part Parametric, \nBhattacharyya Coefficient and Percentile Shift. The evaluation was performed per pixel for the entire \ndate range. A performance ranking was then calculated for all the indices based on the pixel \nperformance and a final ranking was assigned to each index. The Fire Potential Index performed best \namongst the implemented candidate fire potential indices. The Canadian Fire Weather Index \nperformed well in Mpumalanga and the Fine Fuel Moisture Code performed well in the Western \nCape. The overall performance of the indices was not very high. This is due to the fact that even \nthough fire potential is high in an area, an ignition source might not be present to cause an actual \nfire event. The performance of fire potential indices and fire danger indices were different in the two \nprovinces. Future work can be done to develop an index based on South African conditions or \ncalibrate the indices implemented in this research for an area.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".