Applications of Machine Learning and Remote Sensing in Soil and Water Conservation DOI Creative Commons
Kwang Jin Kim,

Woo Hyeon Park,

Yongchul Shin

и другие.

Hydrology, Год журнала: 2024, Номер 11(11), С. 183 - 183

Опубликована: Окт. 30, 2024

The application of machine learning (ML) and remote sensing (RS) in soil water conservation has become a powerful tool. As analytical tools continue to advance, the variety ML algorithms RS sources expanded, providing opportunities for more sophisticated analyses. At same time, researchers are required select appropriate technologies based on research objectives, topic, scope study area. In this paper, we present comprehensive review that been implemented advance conservation. key contribution paper is it provides an overview current areas within their effectiveness improving prediction accuracy resource management categorized subfields, including properties, hydrology resources, wildfire management. We also highlight challenges future directions limitations applications This aims serve as reference decision-makers by offering insights into fields

Язык: Английский

The potential of optical and SAR time-series data for the improvement of aboveground biomass carbon estimation in Southwestern China’s evergreen coniferous forests DOI Creative Commons
Yiru Zhang, Binbin He, Rui Chen

и другие.

GIScience & Remote Sensing, Год журнала: 2024, Номер 61(1)

Опубликована: Апрель 26, 2024

Accurate assessments of forest biomass carbon are invaluable for managing resources, evaluating effects on ecological protection, and achieving goals related to climate change sustainable development. Currently, the integration optical synthetic aperture radar (SAR) data has been extensively utilized in estimating aboveground (AGC), while it is limited by using single-phase remote sensing images. Time-series data, which capture interannual dynamic growth seasonal variations photosynthetic phenology forests, can sufficiently describe characteristics. However, there remains a gap research focusing utilizing satellite-based time-series AGC estimation, especially SAR sensors. This study investigated potential AGC. Here, we undertook nine quantitative experiments estimation from Landsat 8 Sentinel-1 tested several regression algorithms (including multiple linear (MLR), random forests (RF), artificial neural network (ANN), extreme gradient boosting (XGBoost)) explore contributions spatiotemporal features estimation. The results suggested that XGBoost algorithm was suitable with explanatory solid power stable performance. temporal representing trends periodic characteristics (such as coefficients continuous wavelet transform) were more valuable than spatial both sensor types, accounting around 40% ~50% variance compared 17% ~25%. combination produced best performance (R2 = 0.814, RMSE 18.789 Mg C/ha, rRMSE 26.235%), when or alone (optical: R2 0.657 35.317%; SAR: 0.672 34.701%). Feature importance analysis also verified vegetation indices, SWIR 1/2 bands, backscatter VV polarization most critical variables Furthermore, incorporating into modeling illustrated be effective reducing saturation within high-biomass forests. demonstrated superiority While applicability this methodology only evergreen coniferous may provide viable approach needed make full use increasingly better free satellite estimate high accuracy, supporting policy making management

Язык: Английский

Процитировано

4
PREDICTION OF FOREST FIRE SUSCEPTIBILITY USING MACHINE LEARNING TOOLS IN THE TRIUNFO DO XINGU ENVIRONMENTAL PROTECTION AREA, AMAZON, BRAZIL DOI
Kemuel Maciel Freitas, Ronie Silva Juvanhol, Christiano Jorge Gomes Pinheiro

и другие.

Journal of South American Earth Sciences, Год журнала: 2025, Номер unknown, С. 105366 - 105366

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
Study on power system resilience assessment considering cascading failures during wildfire disasters DOI
Baohong Li, Changle Liu, Yue Yin

и другие.

Energy Reports, Год журнала: 2025, Номер 13, С. 1819 - 1833

Опубликована: Янв. 24, 2025

Язык: Английский

Процитировано

0
A spatial weight sampling method integrating the spatiotemporal pattern enhances the understanding of the occurrence mechanism of wildfires in the southwestern mountains of China DOI
Wenlong Yang,

Mingshan Wu,

Lei Kong

и другие.

Forest Ecology and Management, Год журнала: 2025, Номер 585, С. 122619 - 122619

Опубликована: Март 10, 2025

Язык: Английский

Процитировано

0
Integrating DEM and Deep Learning for Forested Terrain Analysis: Enhancing Fire Risk Assessment Through Mountain Peak and Water System Extraction in Chongli District DOI Open Access

Yihui Wu,

Xueying Sun, Liang Qi

и другие.

Forests, Год журнала: 2025, Номер 16(4), С. 692 - 692

Опубликована: Апрель 16, 2025

Accurate fire risk assessment in forested terrain is crucial for effective disaster management and ecological conservation. This study innovatively proposes a novel framework that integrates Digital Elevation Models (DEMs) with deep learning techniques to enhance Chongli District. Our combines DEM data Faster Regions Convolutional Neural Networks (Faster R-CNN) CNN-based methods, breaking through the limitations of traditional approaches rely on manual feature extraction. It capable automatically identifying critical features, such as mountain peaks water systems, higher accuracy efficiency. DEMs provide high-resolution topographical information, which models leverage accurately identify delineate key geographical features. results show integration significantly improves by offering detailed precise analysis, thereby providing more reliable inputs behavior prediction. The extracted fundamental prediction, enable accurate predictions spread potential impact areas. not only highlights great combining geospatial advanced machine but also offers scalable efficient solution forest mountainous regions. Future work will focus expanding dataset include environmental variables validating model different areas further its robustness applicability.

Язык: Английский

Процитировано

0
An integrated framework for wildfire emergency response and post-fire debris flow prediction: a case study from the wildfire event on 20 April 2021 in Mianning, Sichuan, China DOI
Yao Tang,

Yuting Luo,

Wang Li-juan

и другие.

Natural Hazards, Год журнала: 2025, Номер unknown

Опубликована: Май 8, 2025

Язык: Английский

Процитировано

0
Incorporating fire spread simulation and machine learning algorithms to estimate crown fire potential for pine forests in Sichuan, China DOI Creative Commons
Rui Chen, Binbin He,

Yanxi Li

и другие.

International Journal of Applied Earth Observation and Geoinformation, Год журнала: 2024, Номер 132, С. 104080 - 104080

Опубликована: Авг. 1, 2024

Accurate estimation of crown fire potential (CFP) can improve guidance on control and management. However, robust simulations behavior are still challenging, limiting the accuracy regional-scale CFP mapping. This study aims to incorporate spread simulation machine learning algorithms mapping at a regional scale. First, we built dataset using from FARSITE model, as well multi-source data, including fuel, weather, topography variables. Fuel model parameters were optimized with four metaheuristic for simulations. Then, hybrid models (TBA-ML) established by coupling transfer AdaBoost (TrAdaBoost) algorithm three (ML) algorithms, i.e., Bayesian Network (BN), Random Forest (RF), Support Vector Machine (SVM), estimate danger assessment spatially. Results showed that TBA-BN performed best in estimating higher (AUC>0.9 F1 score > 0.8) than RF- SVM-based models. The variable importance causal analysis fuel variables have major contributions occurrence. Finally, mapped monthly average passive active scales qualitatively demonstrated our time-series products successfully captured dynamic change danger. above results suggest integrating accurately

Язык: Английский

Процитировано

2
Process-based and geostationary meteorological satellite-enhanced dead fuel moisture content estimation DOI Creative Commons
Chunquan Fan, Binbin He, Jianpeng Yin

и другие.

GIScience & Remote Sensing, Год журнала: 2024, Номер 61(1)

Опубликована: Март 5, 2024

Dead fuel moisture content (DFMC) is essential for assessing wildfire danger, fire behavior, and consumption. Several process-based models have been proposed to estimate DFMC. Previous studies employed DFMC, solely relying on meteorological data obtained from stations. Satellite can offer higher spatial resolution compared data, with the potential enhance DFMC estimates. Within this content, we aimed improve estimates by consideration of geostationary satellite-derived key variable (relative humility, RH) into Fuel Stick Moisture Model (FSMM). The RH was derived Himawari-8 satellite other variables required FSMM were Global Forecast System (GFS). As comparison, an equilibrium (EMC) model, Simard, random forest regression also used field measurement southwest China validate these three models. Results show that estimated reached a reasonable accuracy (R2 = 0.73, RMSE 3.60%, MAE 2.69%). comparison between two confirmed superior performance model. A case over region continuous decreasing trends until outbreak, highlighting applicability our approach in contributing risk assessment.

Язык: Английский

Процитировано

2
Applications of Machine Learning and Remote Sensing in Soil and Water Conservation DOI Creative Commons
Kwang Jin Kim,

Woo Hyeon Park,

Yongchul Shin

и другие.

Hydrology, Год журнала: 2024, Номер 11(11), С. 183 - 183

Опубликована: Окт. 30, 2024

The application of machine learning (ML) and remote sensing (RS) in soil water conservation has become a powerful tool. As analytical tools continue to advance, the variety ML algorithms RS sources expanded, providing opportunities for more sophisticated analyses. At same time, researchers are required select appropriate technologies based on research objectives, topic, scope study area. In this paper, we present comprehensive review that been implemented advance conservation. key contribution paper is it provides an overview current areas within their effectiveness improving prediction accuracy resource management categorized subfields, including properties, hydrology resources, wildfire management. We also highlight challenges future directions limitations applications This aims serve as reference decision-makers by offering insights into fields

Язык: Английский

Процитировано

1