Integrating geospatial, remote sensing, and machine learning for climate-induced forest fire susceptibility mapping in Similipal Tiger Reserve, India

Forest Ecology and Management, Journal Year: 2024, Volume and Issue: 555, P. 121729 - 121729

Published: Jan. 31, 2024

Accurately assessing forest fire susceptibility (FFS) in the Similipal Tiger Reserve (STR) is essential for biodiversity conservation, climate change mitigation, and community safety. Most existing studies have primarily focused on climatic topographical factors, while this research expands scope by employing a synergistic approach that integrates geographical information systems (GIS), remote sensing (RS), machine learning (ML) methodologies identifying fire-prone areas STR their vulnerability to change. To achieve this, study employed comprehensive dataset of forty-four influencing including topographic, climate-hydrologic, health, vegetation indices, radar features, anthropogenic interference, into ten ML models: neural net (nnet), AdaBag, Extreme Gradient Boosting (XGBTree), Machine (GBM), Random Forest (RF), its hybrid variants with differential evolution algorithm (RF-DEA), Gravitational Based Search (RF-GBS), Grey Wolf Optimization (RF-GWO), Particle Swarm (RF-PSO), genetic (RF-GA). The revealed high FFS both northern southern portions area, nnet RF-PSO models demonstrating percentages 12.44% 12.89%, respectively. Conversely, very low zones consistently displayed scores approximately 23.41% 18.57% models. robust mapping methodology was validated impressive AUROC (>0.88) kappa coefficient (>0.62) across all validation metrics. Future (ssp245 ssp585, 2022–2100) indicated along edges STR, central zone categorized from susceptibility. Boruta analysis identified actual evapotranspiration (AET) relative humidity as key factors ignition. SHAP evaluation reinforced influence these FFS, also highlighting significant role distance road, settlement, dNBR, slope, prediction accuracy. These results emphasize critical importance proposed provide invaluable insights firefighting teams, management, planning, qualification strategies address future sustainability.

Language: Английский

---

Landslide susceptibility assessment using information quantity and machine learning integrated models: a case study of Sichuan province, southwestern China

Earth Science Informatics, Journal Year: 2025, Volume and Issue: 18(2)

Published: Jan. 18, 2025

Language: Английский

---

Forest fire probability zonation using dNBR and machine learning models: a case study at the Similipal Biosphere Reserve (SBR), Odisha, India

Environmental Science and Pollution Research, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 30, 2025

Language: Английский

---

Comparative Analysis of Tree-Based Ensemble Learning Algorithms for Landslide Susceptibility Mapping: A Case Study in Rize, Turkey

Water, Journal Year: 2023, Volume and Issue: 15(14), P. 2661 - 2661

Published: July 22, 2023

The Eastern Black Sea Region is regarded as the most prone to landslides in Turkey due its geological, geographical, and climatic characteristics. Landslides this region inflict both fatalities significant economic damage. main objective of study was create landslide susceptibility maps (LSMs) using tree-based ensemble learning algorithms for Ardeşen Fındıklı districts Rize Province, which second-most-prone province terms within Region, after Trabzon. In study, Random Forest (RF), Gradient Boosting Machine (GBM), CatBoost, Extreme (XGBoost) were used machine algorithms. Thus, comparing prediction performances these established second aim study. For purpose, 14 conditioning factors LMSs. are: lithology, altitude, land cover, aspect, slope, slope length steepness factor (LS-factor), plan profile curvatures, tree cover density, topographic position index, wetness distance drainage, roads, faults. total data set, includes non-landslide pixels, split into two parts: training set (70%) validation (30%). area under receiver operating characteristic curve (AUC-ROC) method evaluate models. AUC values showed that CatBoost (AUC = 0.988) had highest performance, followed by XGBoost 0.987), RF 0.985), GBM (ACU 0.975) Although models close each other, performed slightly better than other These results especially can be reduce damages area.

Language: Английский

---

Assessment of Wildfire Susceptibility and Wildfire Threats to Ecological Environment and Urban Development Based on GIS and Multi-Source Data: A Case Study of Guilin, China

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(10), P. 2659 - 2659

Published: May 19, 2023

The frequent occurrence and spread of wildfires pose a serious threat to the ecological environment urban development. Therefore, assessing regional wildfire susceptibility is crucial for early prevention formulation disaster management decisions. However, current research on primarily focuses improving accuracy models, while lacking in-depth study causes mechanisms wildfires, as well impact losses they cause This situation not only increases uncertainty model predictions but also greatly reduces specificity practical significance models. We propose comprehensive evaluation framework analyze spatial distribution effects influencing factors, risks damage local In this study, we used information from period 2013–2022 data 17 factors in city Guilin basis, utilized eight machine learning algorithms, namely logistic regression (LR), artificial neural network (ANN), K-nearest neighbor (KNN), support vector (SVR), random forest (RF), gradient boosting decision tree (GBDT), light (LGBM), eXtreme (XGBoost), assess susceptibility. By evaluating multiple indicators, obtained optimal Shapley Additive Explanations (SHAP) method explain decision-making mechanism model. addition, collected calculated corresponding with Remote Sensing Ecological Index (RSEI) representing vulnerability Night-Time Lights (NTLI) development vulnerability. coupling results two represent ecology city. Finally, by integrating information, assessed risk disasters reveal overall characteristics Guilin. show that AUC values models range 0.809 0.927, ranging 0.735 0.863 RMSE 0.327 0.423. Taking into account all performance XGBoost provides best results, AUC, accuracy, 0.863, 0.327, respectively. indicates has predictive performance. high-susceptibility areas are located central, northeast, south, southwest regions area. temperature, soil type, land use, distance roads, slope have most significant Based assessments, potential can be identified comprehensively reasonably. article improve prediction provide important reference response wildfires.

Language: Английский

---

Comparison of tree-based ensemble learning algorithms for landslide susceptibility mapping in Murgul (Artvin), Turkey

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 17(2), P. 1459 - 1481

Published: March 4, 2024

Abstract Turkey’s Artvin province is prone to landslides due its geological structure, rugged topography, and climatic characteristics with intense rainfall. In this study, landslide susceptibility maps (LSMs) of Murgul district in were produced. The study employed tree-based ensemble learning algorithms, namely Random Forest (RF), Light Gradient Boosting Machine (LightGBM), Categorical (CatBoost), eXtreme (XGBoost). LSM was performed using 13 factors, including altitude, aspect, distance drainage, faults, roads, land cover, lithology, plan curvature, profile slope, slope length, topographic position index (TPI), wetness (TWI). utilized a inventory consisting 54 polygons. Landslide dataset contained 92,446 pixels spatial resolution 10 m. Consistent the literature, majority (70% – 64,712 pixels) used for model training, remaining portion (30% 27,734 validation. Overall accuracy, precision, recall, F 1-score, root mean square error (RMSE), area under receiver operating characteristic curve (AUC-ROC) considered as validation metrics. LightGBM XGBoost found have better performance all metrics compared other algorithms. Additionally, SHapley Additive exPlanations (SHAP) explain interpret outputs. As per algorithm, most influential factors occurrence determined be whereas TWI, curvature identified least factors. Finally, it concluded that produced LSMs would provide significant contributions decision makers reducing damages caused by area.

Language: Английский

---

Comparison of diverse machine learning algorithms for forest fire susceptibility mapping in Antalya, Türkiye

Advances in Space Research, Journal Year: 2024, Volume and Issue: 74(2), P. 647 - 667

Published: April 16, 2024

Language: Английский

---

Assessing wildfire impact on Trigonella elliptica habitat using random forest modeling

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 353, P. 120209 - 120209

Published: Jan. 30, 2024

Language: Английский

---

SHAP-Driven Explainable Artificial Intelligence Framework for Wildfire Susceptibility Mapping Using MODIS Active Fire Pixels: An In-Depth Interpretation of Contributing Factors in Izmir, Türkiye

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(15), P. 2842 - 2842

Published: Aug. 2, 2024

Wildfire susceptibility maps play a crucial role in preemptively identifying regions at risk of future fires and informing decisions related to wildfire management, thereby aiding mitigating the risks potential damage posed by wildfires. This study employs eXplainable Artificial Intelligence (XAI) techniques, particularly SHapley Additive exPlanations (SHAP), map Izmir Province, Türkiye. Incorporating fifteen conditioning factors spanning topography, climate, anthropogenic influences, vegetation characteristics, machine learning (ML) models (Random Forest, XGBoost, LightGBM) were used predict wildfire-prone areas using freely available active fire pixel data (MODIS Active Fire Collection 6 MCD14ML product). The evaluation trained ML showed that Random Forest (RF) model outperformed XGBoost LightGBM, achieving highest test accuracy (95.6%). All classifiers demonstrated strong predictive performance, but RF excelled sensitivity, specificity, precision, F-1 score, making it preferred for generating conducting SHAP analysis. Unlike prevailing approaches focusing solely on global feature importance, this fills critical gap employing summary dependence plots comprehensively assess each factor’s contribution, enhancing explainability reliability results. analysis reveals clear associations between such as wind speed, temperature, NDVI, slope, distance villages with increased susceptibility, while rainfall streams exhibit nuanced effects. spatial distribution classes highlights areas, flat coastal near settlements agricultural lands, emphasizing need enhanced awareness preventive measures. These insights inform targeted management strategies, highlighting importance tailored interventions like firebreaks management. However, challenges remain, including ensuring selected factors’ adequacy across diverse regions, addressing biases from resampling spatially varied data, refining broader applicability.

Language: Английский

---

A Forest Fire Susceptibility Modeling Approach Based on Integration Machine Learning Algorithm

Forests, Journal Year: 2023, Volume and Issue: 14(7), P. 1506 - 1506

Published: July 24, 2023

The subjective and empirical setting of hyperparameters in the random forest (RF) model may lead to decreased performance. To address this, our study applies particle swarm optimization (PSO) algorithm select optimal parameters RF model, with goal enhancing We employ optimized ensemble (PSO-RF) create a fire risk map for Jiushan National Forest Park Anhui Province, China, thereby filling research gap this region’s studies. Based on collinearity tests previous results, we selected eight driving factors, including topography, climate, human activities, vegetation modeling. Additionally, compare logistic regression (LR), support vector machine (SVM), models. Lastly, evaluate feature importance generate map. Model evaluation results demonstrate that PSO-RF performs best (AUC = 0.908), followed by (0.877), SVM (0.876), LR (0.846). In created 70.73% area belongs normal management zone, while 15.23% is classified as alert zone. analysis reveals NDVI key factor area. Through utilizing PSO optimize have addressed problems hyperparameter setting, model’s accuracy generalization ability.

Language: Английский

---