Bibliographic record
Abstract
The thesis discusses forest fire occurrence in the Karst forest management region in the period from 1995 to 2009. Data analysis has shown that fire occurrence has two season peaks which are highly associated with human activities in the natural environment (land cultivation, railway). Therefore fire mostly starts in unwooded areas, from where the wind spreads it into the woods. Most frequently it occurs in deciduous forests, and the largest burnt areas occur in coniferous forests. The kernel density showed that lightning fires most frequently occur in the northern part of the region, where hills rise above the surrounding landscape. Forest fire occurrence is impacted by people, tree composition and terrain, which coincides with the findings of authors who explore forest fires in different natural environments. Fire is a natural process with its own rules that does not change regardless of where it occurs. To predict the incidence of forest fire on the basis of weather variables we used the Canadian Forest Fire Weather Index System (CFFWIS), which is applied in Europe within the European Fire Forest Index System (EFFIS). Its performance has not been thoroughly assessed, especially in environments less prone to fire, but this has been done herein. Five fire danger classes (very high, high, moderate low and very low) were derived from percentile analysis of the CFFWIS Fire Weather Index. We established that these classes are related to former forest fire occurrence. Predicting the day when the forest fire might occur was done more precisely with the classification tree than with the logistic regression. Fire activity is more related to current weather conditions than to drought. Our findings stress the applicability of CFFWIS in forest fire protection, and simultaneously suggest limitations related to a small number of fires and a small study area. On the basis of data on forest stands, we created the forest fire occurrence probability model with the ordinary least squares method (OLS) and geographically weighted regression (GWR). Higher quality model was obtained with geographically weighted regression which showed which characteristics of forest stands influence forest fire occurrence and in what way. The GWR model has also shown that in addition to characteristics of forest stands we also need other data (railway). A combination of spatial statistical methods on the characteristics of forest stands has allowed detailed insight into forest fire occurrence and its characteristics. One of the duties of predicting forest fire danger is also predicting forest fire behaviour, which, in addition to weather data and topography, also requires a fuel model. We examined forest databases and established that they do not contain all data needed for calculating fuel model parameters. However, they do contain data on forest stands (forest stand maps), which can be applied to the spatial display of the fuel model (fuelbed). This serves as the basis for further studies in \nthe area of predicting forest fire behaviour. We established that usefulness of forest stand maps overcome their original purpose of collecting and preparing for forest management purposes. Their real significance will become clear only in the coming years as crucial information on forests and forest area, which provides a starting point for monitoring various habitat types and species, and environmental protection.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.000 | 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) | 0.002 | 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".