Machine-Learning Applications in Geosciences: Comparison of Different Algorithms and Vegetation Classes’ Importance Ranking in Wildfire Susceptibility

Trucchia, Andrea; Izadgoshasb, Hamed; Isnardi, Sara; Fiorucci, Paolo; Tonini, Marj

doi:10.3390/geosciences12110424

Cited by 4 publications

(1 citation statement)

References 59 publications

(82 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This technique involves constructing models or trees using random subsets of observations and a set of controlling predictors to capture patterns within the provided data and achieve an optimal prediction performance. The RF model includes two key parameters that require optimization: the number of variables considered (mtry) and the number of trees (ntree) [38]. For this study, we set these parameters to 3 for mtry and 700 for ntree.…”

Section: Random Forest (Rf)mentioning

confidence: 99%

Flood-Prone Zones of Meandering Rivers: Machine Learning Approach and Considering the Role of Morphology (Kashkan River, Western Iran)

Ghahraman,

Nagy,

Nooshin Nokhandan

2023

Geosciences

View full text Add to dashboard Cite

We utilized the random forest (RF) machine learning algorithm, along with nine topographical/morphological factors, namely aspect, slope, geomorphons, plan curvature, profile curvature, terrain roughness index, surface texture, topographic wetness index (TWI), and elevation. Our objective was to identify flood-prone areas along the meandering Kashkan River and investigate the role of topography in riverbank inundation. To validate the flood susceptibility map generated by the random forest algorithm, we employed Sentinel-1 GRDH SAR imagery from the March 2019 flooding event in the Kashkan river. The SNAP software and the OTSU thresholding method were utilized to extract the flooded/inundated areas from the SAR imagery. The results showed that the random forest model accurately pinpointed areas with a “very high” and “high” risk of flooding. Through analysis of the cross-sections and SAR-based flood maps, we discovered that the topographical confinement of the meander played a crucial role in the extent of inundation along the meandering path. Moreover, the findings indicated that the inner banks along the Kashkan river were more prone to flooding compared to the outer banks.

show abstract

Section: Random Forest (Rf)mentioning

confidence: 99%

Flood-Prone Zones of Meandering Rivers: Machine Learning Approach and Considering the Role of Morphology (Kashkan River, Western Iran)

Ghahraman,

Nagy,

Nooshin Nokhandan

2023

Geosciences

View full text Add to dashboard Cite

show abstract

Socioeconomic Impacts and Regional Drivers of Fire Management: The Case of Portugal

Parente,

Tonini,

Amraoui

et al. 2024

Fire Hazards: Socio-Economic and Regional Issues

View full text Add to dashboard Cite

Wildfires are uncontrolled and unwanted fires that usually occur in forested/rural areas and burn forests, agricultural areas, and wildlands. Land abandonment, with the consequent growth of the rural–urban interface, increases the exposure and vulnerability of fire-prone regions around the World. In the last two decades, Europe experienced a high number of wildfires causing large burnt areas mainly concentrated in the Mediterranean Basin. This high fire incidence seems to be the result of human activities including land use/land cover changes, but also of climate variability and change. In the present study, we analyse the current situation in Portugal, which is the European country with the highest total number of wildfires and the second-highest total burnt area. The spatial and temporal variability of the wildfires within the country is very heterogeneous, due to the human and biophysical drivers. In this regard, four main aspects are considered and discussed: (1) the spatial and temporal distribution of wildfires in mainland Portugal; (2) the main human and biophysical fire drivers; (3) socioeconomic impacts; and (4) the main strategies for fire risk mapping and management. The main results indicate high spatial heterogeneity of the fire incidence, with higher fire activity in the northern region than the southern region, mainly promoted by a higher irregular topography and significantly different types of climate and land use/land cover characteristics. We highlight how fire incidence is strongly dependent on many biophysical and human factors/drivers and the direct and indirect socioeconomic impacts of wildfires. Methodologies and indexes developed by Portuguese authorities to map fire risk and assess fire danger are described. The elements discussed in this chapter result from research and lessons learned in recent years on the fire regime in Portugal and Europe. These findings can contribute to improving forest, landscape, and fire management, in Mediterranean European countries which share similar characteristics.

show abstract

Developing novel machine-learning-based fire weather indices

Shmuel

Heifetz²

2023

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

Accurate wildfire risk estimation is an essential yet challenging task. As the frequency of extreme fire weather and wildfires is on the rise, forest managers and firefighters require accurate wildfire risk estimations to successfully implement forest management and firefighting strategies. Wildfire risk depends on non-linear interactions between multiple factors; therefore, the performance of linear models in its estimation is limited. To date, several traditional fire weather indices (FWIs) have been commonly used by weather services, such as the Canadian FWI. Traditional FWIs are primarily based on empirical and statistical analyses. In this paper, we propose a novel FWI that was developed using machine learning (ML) – the MLFWI. We present the performance of the MLFWI and compare it withvarious traditional FWIs. We find that the MLFWI significantly outperforms traditional indices in predicting wildfire occurrence, achieving an AUC score of 0.99 compared to 0.62-0.80. We recommend applying the MLFWI in wildfire warning systems.

show abstract

Machine-Learning Applications in Geosciences: Comparison of Different Algorithms and Vegetation Classes’ Importance Ranking in Wildfire Susceptibility

Cited by 4 publications

References 59 publications

Flood-Prone Zones of Meandering Rivers: Machine Learning Approach and Considering the Role of Morphology (Kashkan River, Western Iran)

Flood-Prone Zones of Meandering Rivers: Machine Learning Approach and Considering the Role of Morphology (Kashkan River, Western Iran)

Socioeconomic Impacts and Regional Drivers of Fire Management: The Case of Portugal

Developing novel machine-learning-based fire weather indices

Contact Info

Product

Resources

About