Effects of spatial data resolution on the modelling and mapping of soil organic carbon content in hill country grassland landscapes DOI Creative Commons
Duy X. Tran, Estelle Dominati, John H. Lowry

et al.

Soil Use and Management, Journal Year: 2023, Volume and Issue: 40(1)

Published: Sept. 4, 2023

Abstract Limited use has been made of spatially explicit modelling soil organic carbon (SOC) in highly complex farmed landscapes to advance current mapping efforts. This study aimed address this gap knowledge by evaluating the spatial prediction SOC content 0–75 mm depth hill country New Zealand (NZ) using point‐based training data, along with topographic covariates and Sentinel 2 spectral band ratios an automated set machine learning (AutoML) tools ArcGIS. Subsequently, it also focused on quantifying effects data resolution (i.e., 1, 8, 15, 25 m) terms predicted map accuracy. Farmlets contrasting phosphorus fertilizer sheep grazing histories located at Ballantrae Hill Country Research Station, NZ were selected conduct research. Six candidate algorithms incorporated AutoML XGBoost, LightGBM, linear regression, decision trees, extra random forest) ensemble model utilized pattern content. The results show that combine predictions various applied for 1 m enables highest performance accuracy R = .76, RMSE 0.66%). Among predictive variables used model, slope, wetness, position indices found be most important topographical features explain patterns area. Inclusion derived from remote sensing, including surface moisture clay minerals ratio, further improvement prediction. reveals a decrease geospatial does not substantively affect mean estimation farm‐scale modelling. However, coarser reduces ability predict changes across grassland landscape.

Language: Английский

Evaluating landslide susceptibility: the impact of resolution and hybrid integration approaches DOI Creative Commons
Xia Zhao, Wei Chen,

Paraskevas Tsangaratos

et al.

Geomatics Natural Hazards and Risk, Journal Year: 2024, Volume and Issue: 15(1)

Published: Oct. 1, 2024

Language: Английский

Citations

9
FLOOD HAZARD ZONES PREDICTION USING MACHINE-LEARNING-BASED GEOSPATIAL APPROACH IN LOWER NIGER RIVER BASIN, NIGERIA DOI Creative Commons

Adedoyin Benson Adeyemi,

Akinola Adesuji Komolafe

Natural Hazards Research, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

Citations

1
Identification of soil erosion‑susceptible areas using fuzzy logic and hydrological indices aided by mineralogical-granulometric analysis in lower Subansiri basin, Assam, India DOI

Borneeta Dutta,

Pankaj Kumar Srivastava, Annapurna Boruah

et al.

Environmental Earth Sciences, Journal Year: 2025, Volume and Issue: 84(2)

Published: Jan. 1, 2025

Language: Английский

Citations

1
Use of spatial water database as an essential element of water management – a methodological overview DOI Creative Commons
Natalia Janczewska, Magdalena Matysik, Damian Absalon

et al.

Environmental & Socio-economic Studies, Journal Year: 2025, Volume and Issue: 13(1), P. 53 - 62

Published: March 1, 2025

Abstract In the current digital age, spatial management seems impossible without a set of data which maps real situation on computer screen. However, varying technologies (software, hardware) as well methodologies (vectorisation, automatic classification, deep learning, etc.), together with availability input materials, result in huge difference quality and timeliness collected infor example different countries. This statement also applies to hydrographic data, undeniably affects water efficiency. With increasing globalization, it necessary standardize transnational level. The main aim this article was review ways techniques collecting, updating sharing by selected countries or organizations. addition, use modern geo-information remote sensing tools reviewed, work towards interoperability inland surface databases. As review, authors identified strong need unify at both national continental levels, future, globally (considering dynamic change precision when changing mapping scale). addition,good practices were identified, methods that can be used create universal database waters identified.

Language: Английский

Citations

1
Application of GIS-based data-driven bivariate statistical models for landslide prediction: a case study of highly affected landslide prone areas of Teesta River basin DOI Creative Commons
Indrajit Poddar, Ranjan Roy

Quaternary Science Advances, Journal Year: 2023, Volume and Issue: 13, P. 100150 - 100150

Published: Dec. 6, 2023

Predicting landslides has become a critical global challenge for promoting sustainable development in mountainous regions. This study conducts comparative analysis of landslide susceptibility maps (L.S.M.s) generated using two GIS-based data-driven bivariate statistical models: (a) Frequency Ratio (F.R.) and (b) Evidential Belief Function (E.B.F). These models are applied evaluated the high landslide-prone upper middle Teesta basin Darjeeling-Sikkim Himalaya, leveraging geographic information system (GIS) remote sensing techniques. We compile comprehensive inventory map containing 2387 regional points. use approximately 70% this dataset model training reserve remaining 30% validation. In construction Landslide Susceptibility (LSMs), set twenty-one landslide-triggering parameters been considered.These encompass factors such as elevation, distance from drainage, lineament, roads, geology, geomorphology, lithology, land use, cover, normalized difference vegetation index, profile curvature, rainfall, relief amplitude, roughness, slope, slope aspect, classes, stream power sediment transport topographic position ruggedness wetness index. An examination multicollinearity statistics reveals no collinearity issues among causative utilized research. The final L.S.M.s demonstrate that combined application F.R. E.B.F. yields highest accuracy at 98.10%. insights derived hold significant promise valuable tools assessing environmental hazards planning.

Language: Английский

Citations

18
Flash Flood Susceptibility Mapping of North-east Depression of Bangladesh using Different GIS based Bivariate Statistical Models DOI Creative Commons
Md. Sharafat Chowdhury

Watershed Ecology and the Environment, Journal Year: 2024, Volume and Issue: 6, P. 26 - 40

Published: Jan. 1, 2024

Flash flood causes severe damage to the environment and human life across world, no exception is Bangladesh. Severe flash floods affect northeastern portion of Bangladesh in early monsoon pose a serious threat every aspect socioeconomic development environmental sustainability. To manage reduce loss, map susceptible zones plays key role. Thus, aim this research flood-susceptible areas haor utilizing GIS-based bivariate statistical models. The models utilized are frequency ratio (FR), weights evidence (WoE), certainty factor (CF), Shanon's entropy (SE) information value (IV). Among 250 identified locations, 80% data was used for training purposes 20% testing purposes. Eleven selected conditioning factors include elevation, slope, aspect, curvature, TWI, TRI, SPI, distance stream, stream density, rainfall physiography. calculated assigned using ArcGIS prepare final maps. Results AUC ROC indicate WoE (success rate = 0.833 prediction =0.925) best model susceptibility mapping followed by FR 0.828 =0.928) SE 0.827 =0.923). According models, topographic (flat area) hydrologic significantly control occurrence study area. prepared maps will be helpful disaster managers master planners

Language: Английский

Citations

7
Enhancing the Performance of Machine Learning and Deep Learning-Based Flood Susceptibility Models by Integrating Grey Wolf Optimizer (GWO) Algorithm DOI Creative Commons
Ali Nouh Mabdeh, R. S. Ajin, Seyed Vahid Razavi-Termeh

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(14), P. 2595 - 2595

Published: July 16, 2024

Flooding is a recurrent hazard occurring worldwide, resulting in severe losses. The preparation of flood susceptibility map non-structural approach to management before its occurrence. With recent advances artificial intelligence, achieving high-accuracy model for mapping (FSM) challenging. Therefore, this study, various intelligence approaches have been utilized achieve optimal accuracy modeling address challenge. By incorporating the grey wolf optimizer (GWO) metaheuristic algorithm into models—including neural networks (RNNs), support vector regression (SVR), and extreme gradient boosting (XGBoost)—the objective generate maps evaluate variation performance. tropical Manimala River Basin India, severely battered by flooding past, has selected as test site. This 15 conditioning factors such aspect, enhanced built-up bareness index (EBBI), slope, elevation, geomorphology, normalized difference water (NDWI), plan curvature, profile soil adjusted vegetation (SAVI), stream density, texture, power (SPI), terrain ruggedness (TRI), land use/land cover (LULC) topographic wetness (TWI). Thus, six are produced applying RNN, SVR, XGBoost, RNN-GWO, SVR-GWO, XGBoost-GWO models. All models exhibited outstanding (AUC above 0.90) performance, performance ranks following order: RNN-GWO (AUC: 0.968) > 0.961) SVR-GWO 0.960) RNN 0.956) XGBoost 0.953) SVR 0.948). It was discovered that hybrid GWO optimization improved three RNN-GWO-based shows 8.05% MRB very susceptible floods. found SPI, LULC, TWI top five influential factors.

Language: Английский

Citations

7
Evaluation of urban flood susceptibility through integrated Bivariate statistics and Geospatial technology DOI

Kalidhas Muthu,

R. Sivakumar

Environmental Monitoring and Assessment, Journal Year: 2024, Volume and Issue: 196(6)

Published: May 9, 2024

Language: Английский

Citations

6
A machine learning-based approach for flash flood susceptibility mapping considering rainfall extremes in the northeast region of Bangladesh DOI
Md. Enayet Chowdhury, A. K. M. Saiful Islam, Rashed Uz Zzaman

et al.

Advances in Space Research, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 1, 2024

Language: Английский

Citations

5
GDAL and PROJ Libraries Integrated with GRASS GIS for Terrain Modelling of the Georeferenced Raster Image DOI Creative Commons
Polina Lemenkova, Olivier Debeir

Technologies, Journal Year: 2023, Volume and Issue: 11(2), P. 46 - 46

Published: March 22, 2023

Libraries with pre-written codes optimize the workflow in cartography and reduce labour intensive data processing by iteratively applying scripts to implementing mapping tasks. Most existing Geographic Information System (GIS) approaches are based on traditional software a graphical user’s interface which significantly limits their performance. Although plugins proposed improve functionality of many GIS programs, they usually ad hoc finding specific solutions, e.g., cartographic projections conversion. We address this limitation principled approach Geospatial Data Abstraction Library (GDAL), library for conversions between (PROJ) Resources Analysis Support (GRASS) geospatial morphometric analysis. This research presents topographic analysis dataset using scripting methods include several tools: (1) GDAL, translator raster vector formats used converting Earth Global Relief Model (ETOPO1) GeoTIFF XY Cartesian coordinates into World Geodetic 1984 (WGS84) ‘gdalwarp’ utility; (2) PROJ projection transformation ETOPO1 WGS84 grid (Cassini–Soldner equirectangular, Equal Area Cylindrical, Two-Point Equidistant Azimuthal, Oblique Mercator); (3) GRASS sequential use following modules: r.info, d.mon, d.rast, r.colors, d.rast.leg, d.legend, d.northarrow, d.grid, d.text, g.region, r.contour. The depth frequency was analysed module ‘d.histogram’. provided systematic way measuring combine advantages PROJ, tools that informativeness, effectiveness, representativeness spatial processing. included computed slope, aspect, profile, tangential curvature study area. revealed distribution pattern data: 24% elevations below 400 m, 13% depths −5000 −6000 4% have values −3000 −4000 least frequent (−6000 7000 m) <1%, 2% −2000 3000 m basin, while other distributed proportionally. Further, incorporating generic coordinate transformed various demonstrate distortions shape Scripting techniques demonstrated applications modelling shows effectiveness visualization, compatibility libraries (GDAL, PROJ), technical flexibility combining Graphical User Interface (GUI), command-line contributes development.

Language: Английский

Citations

11