Trends and applications in wildfire burned area mapping: Remote sensing data, cloud geoprocessing platforms, and emerging algorithms

GEOMATICA, Год журнала: 2024, Номер 76(1), С. 100008 - 100008

Опубликована: Июль 1, 2024

Wildfires pose an increasing risk to expanding urban population centers, and critical habitats for plant animal species. Improving current wildland management strategies are vital mitigating loss of global biodiversity preventing the displacement residents. Accurate maps areas burned by wildfires is a primary source information required developing strategies. Advancements in underlying technologies mapping comes from three key areas: 1) remotely sensed data, 2) cloud geoprocessing platforms, 3) emerging image processing algorithms. Trends across these were explored this review, addition in-depth discussion comparison optimal usage scenarios. This review provides crucial insights researchers practitioners keen on exploring methods that hold potential improve wildfire area procedures.

Язык: Английский

---

Risk Factors and Prediction of the Probability of Wildfire Occurrence in the China–Mongolia–Russia Cross-Border Area

Remote Sensing, Год журнала: 2022, Номер 15(1), С. 42 - 42

Опубликована: Дек. 22, 2022

Wildfire is essential in altering land ecosystems’ structures, processes, and functions. As a critical disturbance the China–Mongolia–Russia cross-border area, it vital to understand potential drivers of wildfires predict where are more likely occur. This study assessed factors affecting wildfire using Random Forest (RF) model. No single factor played decisive role incidence wildfires. However, climatic variables were most critical, dominating occurrence The probability was simulated predicted Adaptive Network-based Fuzzy Inference System (ANFIS). particle swarm optimization (PSO) model genetic algorithm (GA) used optimize ANFIS hybrid models performed better than for training validation datasets. models, such as PSO-ANFIS GA-ANFIS, overcome over-fitting problem at learning stage pattern. high classification accuracy good performance suggest that can be occurrence. map illustrates high-risk areas mainly distributed northeast part especially grassland forest area Dornod Province Mongolia, Buryatia, Chita state Russia, Inner China. findings reliable estimates relative likelihood hazards management region covered or vicinity.

Язык: Английский

---

Spatial model of wildfire susceptibility using Machine Learning approaches on Rawa Aopa Watumohai National Park, Indonesia

GeoScape, Год журнала: 2024, Номер 18(1), С. 1 - 20

Опубликована: Июнь 1, 2024

Abstract Rawa Aopa National Park has experienced a severe wildfire. These fires are affected by several factors, including topography, meteorology, vegetation, and source of fire. This study uses Machine Learning approach based on re-sampling methods (e.g. crossvalidation, bootstrap, random subsampling) to evaluate, improve the performance twelve basic algorithms: Generalized Linear Model, Support Vector Machine, Random Forest, Boosted Regression Trees, Classification And Tree, Multivariate Adaptive Splines, Mixture Discriminate Analysis, Flexible Discriminant Maximum Entropy, Likelihood, Radial Basis Function, Multi-Layer Perceptron, analyze causes wildfires, correlation between variables. The model is evaluated Area Under Curve, Correlation, True Skill Statistics, Deviance. evaluation results show that Bt-RF good in predicting wildfire susceptibility TNRAW with AUC=0.98, COR=0.96, TSS=0.97, Deviance=0.15. An area 644.88 km 2 or equivalent 59.82% concentration occurring savanna ecosystem which around 245.12 88.95% jungle zone. Among 17 parameters cause fires, this strongly influenced Temperature, Land Use Cover, Distance from Road. There strong soil distance settlements = 0.96.

Язык: Английский

---

Monitoring Wildfires in Thailand: A Case Study of the ECSTAR-TeroSpace’s Earth Observation Project

Journal of Geoscience and Environment Protection, Год журнала: 2023, Номер 11(06), С. 23 - 36

Опубликована: Янв. 1, 2023

The primary objective of this paper is to present a comprehensive case study on monitoring wildfires in Nakhon Nayok, Thailand, utilizing Earth observation platforms. This initiative project has been undertaken by the Excellence Center Space Technology and Research (ECSTAR), partnership with its spin-off startup, TeroSpace. aims provide an in-depth analysis wildfire incidents region, advanced technologies such as satellite imagery data analytics, identify ways improve future management. In particular, focuses including thermal area comparison that ravaged land Nayok Province central Thailand from March April 18th, 2023. To conduct study, ECSTAR-TeroSpace analytic team utilized images platforms: MODIS Sentinel-2A. By presenting contributes broader understanding how monitor manage changing climate. findings underscore importance proactive collaborative efforts mitigating negative impacts other regions Thailand.

Язык: Английский

---

GeoAI for Disaster Mitigation: Fire Severity Prediction Models using Sentinel-2 and ANN Regression

Опубликована: Ноя. 24, 2022

Wildfire is a common disaster that hits Indonesia every dry season, especially on the islands of Kalimantan and Sumatra. In order to reduce impact fire hazards, preventive measures are needed before occurrence fires. One them by setting up an information system such as EWS. The aim this study create effective image- machine learning-based predictive model severity forest land fires based vegetation conditions prior burning. Three parameters prefire conditions, namely greenness indices, moisture, senescence, were selected independent variables predict postfire dependent variable, i.e., severity. There 25 index options tested, using either ANN regression or multiple linear regression. moisture represented Normalized Difference Moisture Index (NDMI). senescence extracted Plant Senescence Reflectance (PSRI). Meanwhile, wildfire measured Burned Area for Sentinel-2 (BAIS2). All from imageries. topology models configured one six hidden layers. More than 100,000 pixels used samples, which then separated into training samples validation samples. results development testing show with Inverted Red-Edge Chlorophyll (IRECI) parameter has highest accuracy in predicting

Язык: Английский

---