• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2006.04a
• /
• pp.231-237
• /
• 2006

• Analytical Science and Technology
• /
• v.34 no.5
• /
• pp.231-239
• /
• 2021

Detection of fire accelerants from fire residues is critical to determine whether the case was arson or accidental fire. However, to develop a standardized model for determining the presence or absence of fire accelerants was not easy because of high temperature which cause disappearance or combustion of components of fire accelerants. In this study, logistic regression, random forest, and support vector machine models were trained and evaluated from a total of 728 GC-MS analysis data obtained from actual fire residues. Mean classification accuracies of the three models were 63 %, 81 %, and 84 %, respectively, and in particular, mean AU-PR values of the three models were evaluated as 0.68, 0.86, and 0.86, respectively, showing fine performances of random forest and support vector machine models.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.10 no.2
• /
• pp.122-128
• /
• 2007

GIS is used much research for efficient forest fire management and forecasting and slope has been known as high-leverage thing in spread of forest fire specially. Various algorithms are used usually to calculate slope angle of topography from DEM(Digital elevation model). However, because spread speed of forest fire is different according to uphill slope and downhill slope, it need new slope calculation algorithm. Therefore, developed slope calculation algorithm can reflect uphill slope and forest fire spread speed of looking downhill slope.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3D
• /
• pp.527-534
• /
• 2006

This study estimated the forest damage of Kangwon-Do fire disaster occurred April 2005. For the estimation, the delineation of fire damaged area was performed using SPOT-4 satellite image and DSM (Digital surface model)/DTM (Digital Terrain Model) was generated by airborne and ground LiDAR data to calculate forests height. The damaged amount of money was calculated in forest area using stand volume formula, combining the canopy height from forest height model and digital stock map. The total forest damage amounted to 3.9 billion won.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.15 no.3
• /
• pp.178-185
• /
• 2013

Forest fires are expected to increase in severity and frequency under global climate change and thus better understanding of fire dynamics is critical for mitigation and adaptation. Researchers with different background, such as ecologists, physicists, and mathematical biologists, have developed various simulation models to reproduce forest fire spread dynamics. However, these models have limitations in the fire spreading because of the complicated factors such as fuel types, wind, and moisture. In this study, we suggested a simple model considering the wind effect and two different fuel types. The two fuels correspond to susceptible tree and resistant tree with different probabilities of transferring fire. The trees were randomly distributed in simulation space with a density ranging from 0.0 (low) to 1.0 (high). The susceptible tree had higher value of the probability than the resistant tree. Based on the number of burnt trees, we then carried out the sensitivity analysis to quantify how the forest fire patterns are affected by wind and tree density. The statistical analysis showed that the total tree density had greatest effect on the forest fire spreading and wind had the next greatest effect. The density of the susceptible tree was relatively lower factor affecting the forest fire. We believe that our model can be a useful tool to explore forest fire spreading patterns.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.328-330
• /
• 2003

The land cover of burned area has changed dramatically since Daxinganling forest fire in Northeastern China during May 6 ? June 4, 1987. This research focused on determining the burn severity and assessment of forest recovery. Burned severity was classified into three levels from June 1987 Landsat TM data acquired just after the fire. A regression model was established between the forest canopy closure from 1999 forest stand map and the NDVI values from June 2000 Landsat ETM+ data. The map of canopy closure was got according to the regression model. And vegetation cover was classified into four types according to forest closure density. The change matrix was built using the classified map of burn severity and vegetation recovery. Then the change conversions of every forest type were analyzed. Results from this research indicate: forest recovery status is well in most of burned scars; and vegetation change detection can be accomplished using postclassification comparison method.

• Journal of agriculture & life science
• /
• v.43 no.6
• /
• pp.19-27
• /
• 2009

This study was carried out to obtain the basic model to estimate damage degree from the correlation analysis between forest fire and site environment factors and to clarify the restoration trends thorough multi-temporal survey by observing species diversity followed by various treatments at damaged forest area over time. From the derived model, the damage degree of forest fire was higher in the area of dense coniferous stands composed of simple story at the elevation of about 100m and 200m, and on steeper slope area over 30 degree. As results of this study, fire damaged trees are needed to cut down and a mixed stand with deciduous and coniferous species from the same area is desirable for the future species composition on fire damaged forest. Thus, site characteristics, local species, and mixed stands are the main consideration to enhance the vegetation recovery.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1273-1283
• /
• 2022

Recently, forest fires have frequently occurred due to climate change, leading to human and property damage every year. The forest fire monitoring technique using remote sensing can obtain quick and large-scale information of fire-damaged areas. In this study, the Gangneung and Donghae forest fires that occurred in March 2022 were analyzed using the spectral band of Sentinel-2, the normalized difference vegetation index (NDVI), and the normalized difference water index (NDWI) to classify the affected areas of forest fires. The U-net based convolutional neural networks (CNNs) model was simulated for the fire-damaged areas. The accuracy of forest fire classification in Donghae and Gangneung classification was high at 97.3% (f₁=0.486, IoU=0.946). The same model used in Donghae and Gangneung was applied to Uljin and Samcheok areas to get rid of the possibility of overfitting often happen in machine learning. As a result, the portion of overlap with the forest fire damage area reported by the National Institute of Forest Science (NIFoS) was 74.4%, confirming a high level of accuracy even considering the uncertainty of the model. This study suggests that it is possible to quantitatively evaluate the classification of forest fire-damaged area using a spectral band and indices similar to that of the Compact Advanced Satellite 500 (CAS500-4) in the Sentinel-2.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2008.11a
• /
• pp.434-437
• /
• 2008

In this study, we developed a fuel moisture variation prediction model on each day after precipitation during a spring forest fire exhibition period. For this research, we selected plots in pine forest on Sam-Chuck si and Dong-hae si in Kangwon do according to a forest density(low, mediate, high) and classified a surface woody fuel by a diameter.(below 0.6cm, $0.6{\sim}3cm$, $3{\sim}6cm$, and above 6cm). A validity of this model was verified by applying a fuel moisture variation after precipitation in this spring. In the result, $R^2$ was $0.76{\sim}0.92$. This model will be a useful for improvement of a forest fire danger rate forcast through a prediction a fule moisture in forest.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.10
• /
• pp.1966-1971
• /
• 2011

Generally, the characteristics of the conductor that was affected by forest fire can be analyzed only when the forest fire is accurately modeled and its effect is identified. Few studies have been conducted with a forest fire model for transmission lines, and no results of the examination of the actual test specimens that were exposed to forest fire have been reported. As the deterioration characteristics of a forest fire are difficult to analyze in the actual field, an environment that was similar to that in the field was used in this study. Deterioration was deposited on a wire using an artificial flame experiment device, to analysis the temperature, surface and component characteristics. It seems that this analysis data in this study can be used as the basic data for the database that can be utilized to analyze wires exposed to forest fire and deterioration and to predict the ACSR wire refurnishment life.