Forest fires play an important role in forest degradation in different parts of the world. The timely prevention of fires should be based on the use of forecast maps built on the basis of assessment of the characteristics of forest vegetation and stands in individual forest plots. According to this a scale for assessing the natural fire hazard in forest takes into account mainly type of the stands (coniferous or deciduous), hygrotope index and stand age for evaluation the fire hazard class. This parameter in the forest inventory databases does not take into account the influence of neighbor plots on fire spread, and therefore can be changed only in the case of forest age increase over 40 years.
Use of GIS technology gives the possibility to take into account the neighboring plots influence on fire hazard class (FHC), to build respective thematic maps, to plan appropriate measures as well as to develop algorithms for optimal vehicle mobilization in the event of forest fires.
In Ukraine, 55 % of cases and 74 % of forest fires occur in the eastern and southern regions. Therefore the new methodical approach was implemented for the forest fund of the State Enterprise “Kreminske Forest & Hunting Economy” (SE “Kreminske FHE”) of Luhansk region, where the large fire of 1996 killed over 7,000 hectares of forest. In 2004–2015, the fires in the forest fund of this enterprise happen annually on 140–350 hectares.
The aim of the research was to assess the shift of fire hazard level for SE “Kreminske FHE” forest stands using GIS technology.
Materials and Methods
Database of forest management in respect to “Kreminske FHE” for 2001 and 2011 was used. In the base scenario, fire hazard classes for each forest plots were evaluated without taking into account the land category of neighboring plots. They were the following: “1” – high, “2” – above average, “3” – average, “4” – below average, “5” – low.
In the second scenario, the land category of neighboring plots was taking into account for each forest plot using QGIS 2.18 as the query with the fulfillment of the predicates of neighborhood, affiliation and adjoining.
Fire hazard class decreased by one (fire hazard increases) if the plot borders with railways, buildings, mansions, swampy reeds. Fire hazard class for clear-cuts after coniferous forest assumes “1”; fire hazard class for clear-cuts after deciduous forest, glades and wastelands assumes “1” if the neighboring plot is coniferous forest, and it assumes “4” if the neighboring plot is a deciduous forest.
Forest stands area with each fire hazard class for all forestries of the SE “Kreminske FHE”, as well as weighted average fire hazard class, was calculated.
Thematic maps of fire hazard have been built on the example of Serebryanske forestry using QGIS 2.18 without taking into account and taking into account the land category of neighboring plots in 2001 and 2011.
Results and Conclusions
An increase of fire hazard (that is fire hazard class decreases) was revealed in the most of forestries. Spatial analysis shows that it was connected with increase the area and number of plots bordering with clear-cuts, unclosed plantations etc. Fire hazard class has decreased for 2001–2011 the most of all in the stands of Serebryanske forestry (from 3.57 to 3.38 points), where the area of plots bordering with clear-cuts and unclosed plantations in 2011 was 12.2 and 15.1 % (645.6 and 799.4 ha) respectively. A considerable part of plots simultaneously borders on different sides with clear-cuts, unclosed plantations, and other land categories, which neighboring increases fire hazard.
1 Fig., 5 Tables, 4 Refs.