Assessing and projecting land use land cover changes using machine learning and artificial neural network models in Guder watershed, Ethiopia DOI Creative Commons
Sintayehu Fetene Demessie, Yihun T. Dile, Bobe Bedadi

et al.

Environmental Challenges, Journal Year: 2024, Volume and Issue: unknown, P. 101074 - 101074

Published: Dec. 1, 2024

Language: Английский

An integrated GEE and machine learning framework for detecting ecological stability under land use/land cover changes DOI Creative Commons

Atiyeh Amindin,

Narges Siamian,

Narges Kariminejad

et al.

Global Ecology and Conservation, Journal Year: 2024, Volume and Issue: 53, P. e03010 - e03010

Published: May 27, 2024

Ecological stability (ES) is recognized as a crucial factor for sustainable development at global and regional scales. However, the importance of this was not considered significant. Hence, main aim study to introduce new approach that focuses on detecting ES over Maharloo watershed in Iran. To achieve goal, we extracted land use cover (LULC) data from Google Earth Engine (GEE) platform by applying random forest (RF) machine learning method, which obtained Kappa statistics 0.85, 0.86, 0.87 years 2002, 2013, 2023, respectively. We identified both stable unstable regions based LULC changes employed them using forecast ES. The most important predictors ecological were elevation, soil organic carbon index, precipitation, salinity. results research revealed certain areas within have experienced instability recent years, with gardens showing highest percentage (60.65%) among all land-use categories. performance validation our model suggest are reliable (AUC = 0.86). This offers detailed maps trends, offering valuable insights decision makers support landscape conservation restoration efforts. Overall, findings contribute more comprehensive understanding dynamics provide efforts other regions.

Language: Английский

Citations

12

Random Forest Classifier Algorithm of Geographic Resources Analysis Support System Geographic Information System for Satellite Image Processing: Case Study of Bight of Sofala, Mozambique DOI Creative Commons
Polina Lemenkova

Coasts, Journal Year: 2024, Volume and Issue: 4(1), P. 127 - 149

Published: Feb. 26, 2024

Mapping coastal regions is important for environmental assessment and monitoring spatio-temporal changes. Although traditional cartographic methods using a geographic information system (GIS) are applicable in image classification, machine learning (ML) present more advantageous solutions pattern-finding tasks such as the automated detection of landscape patches heterogeneous landscapes. This study aimed to discriminate patterns along eastern coasts Mozambique ML modules Geographic Resources Analysis Support System (GRASS) GIS. The random forest (RF) algorithm module ‘r.learn.train’ was used map landscapes shoreline Bight Sofala, remote sensing (RS) data at multiple temporal scales. dataset included Landsat 8-9 OLI/TIRS imagery collected dry period during 2015, 2018, 2023, which enabled evaluation dynamics. supervised classification RS rasters supported by Scikit-Learn package Python embedded GRASS Sofala characterized diverse marine ecosystems dominated swamp wetlands mangrove forests located mixed saline–fresh waters coast Mozambique. paper demonstrates advantages areas. integration Earth Observation data, processed decision tree classifier land cover characteristics recent changes ecosystem Mozambique, East Africa.

Language: Английский

Citations

9

Identification of potential dam sites for severe water crisis management in semi-arid fluoride contaminated region, India DOI Creative Commons
Arijit Ghosh, Biswajit Bera

Cleaner Water, Journal Year: 2024, Volume and Issue: 1, P. 100011 - 100011

Published: Feb. 28, 2024

Pressure on freshwater resources is tremendously increasing due to large-scale global population explosion, socio-economic development, climate change and infrastructural development worldwide. The study area faces severe water crisis, groundwater fluoride contamination, high drought frequency. Thus, the principal objectives are i) assess recent surface subsurface dynamics in this plateau fringe using satellite datasets Google Earth Engine (GEE)and ii) demarcate suitable sites for dam construction manage crisis substitute drinking sources. Satellite such as Sentinel 2 Gravity Recovery Climate Experiment (GRACE) have been used access dynamics. Numerous criteria or influencing factors including geology, geomorphology, lineament, elevation, slope, rainfall, land use/land cover, soil, stream density, normalized vegetation index (NDVI), distance from river considered new site suitability. In study, four advanced machine learning models namely support vector (SVM), random forest (RF), logistic regression (LR) gradient boosting (XGBoost) applied recommend construction. Average changes 157.375 km2 (2012-2016) 156.185 km2(2017-2022). Estimated thickness (EWT) values vary 28.58 cm -27.07 (2002-2017). case of soil moisture (SM), lowest value (2.4 cm) was June 2009, highest (21.51 September 2003. After deduction SM EWS, it specifies that maximum storage (9.48 occurred July 2004 whereas a minimum (-30.21 March 2016. Dam suitability results denote 10% areas come under very curve (AUC) SVM, RF, LR, XGBoost 0.94, 0.95, 0.91, 0.92 respectively. Therefore, RF model has comparatively higher regarding performance. output research will be advantageous define places sustainable resource management semi-arid environment.

Language: Английский

Citations

4

Enhancing Hyperspectral Image Classification for Land Use Land Cover With Dilated Neighborhood Attention Transformer and Crow Search Optimization DOI Creative Commons

Ganji Tejasree,

A. Loganathan

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 59361 - 59385

Published: Jan. 1, 2024

The classification of Land Use Cover (LULC) can be accomplished with the help hyperspectral imaging, which is a cutting-edge technology. Nevertheless, despite its efficacy, utilization images for LULC continues to present difficulties and demands significant amount time. limited availability training samples poses challenge in achieving accurate LULC. through meticulous deliberation examination, this impediment surmounted. To tackle task classification, we have developed Dilated Neighbourhood Attention Transformer (DNAT). Firstly, employ LeNet-5 extract features from provided data. Subsequently, perform band selection using Crow Search Optimization (CSO). Following extraction bands, proceed classify In our study, used Salinas, Indian Pines (IP), Washington DC Mall datasets classification. performance proposed approach evaluated commonly metrics, namely, Average Accuracy (AA), Overall (OA), Kappa Coefficient (KC). We achieved 99.85% as OA, 99.83% AA, 99.73% KC Salinas Dataset. This highest accuracy DNAT classifier. experimental results proved beyond reasonable doubt that method possible performance, surpassing all prior methods.

Language: Английский

Citations

4

Perception study of urban green spaces in Singapore urban parks: Spatio-temporal evaluation and the relationship with land cover DOI
Wenting Zhang,

Yuxin Su

Urban forestry & urban greening, Journal Year: 2024, Volume and Issue: 99, P. 128455 - 128455

Published: July 24, 2024

Language: Английский

Citations

4

Tracking land use land cover changes in the twin cities of Odisha, India using a machine learning based Google Earth Engine approach DOI
Ajaya K. Nayak, Anil Kumar Kar

Urban Water Journal, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 22

Published: Jan. 30, 2025

The current study is based on analyzing the land use cover (LULC) changes and its corresponding effects water surface temperature (LST) twin cities of Odisha, i.e. Bhubaneswar Cuttack using a machine learning Google Earth Engine (GEE) platform. A random forest (RF) classification model was adopted due to simplicity high popularity for providing accurate results. For study, Landsat 8 (OLI/TRIS) Sentinel 2 were accessed via GEE. With an overall accuracy about 99% RF algorithm, results indicate alarming situation cities, especially where there has been reduction in by 59% response increments built-up area 90% LST 1.5%. expanding city radius, faced 28% increase 17% 3.4%. respectively.

Language: Английский

Citations

0

Research on the spatial temporary evolution of urban expansion in Xining city and its surrounding areas based on Landsat time series data DOI Creative Commons

Xiaomin Cao,

Xiaohong Gao, Runxiang Li

et al.

Heliyon, Journal Year: 2024, Volume and Issue: 10(3), P. e24846 - e24846

Published: Jan. 21, 2024

Quantitative analysis of the process urban expansion and evolution is great practical significance for future planning development potential valley cities. Based on GEE cloud platform Landsat satellite data, this paper analyzed spatio-temporal change characteristics transfer rules land cover in Xining City its surrounding areas past 33 years by using random forest algorithm, consistency test, use dynamic attitude, matrix hot spot methods. The results show that accuracy range preliminary classification construction improved 1.57%–3.53 % test algorithm. study area are mainly increase area, decrease cultivated grassland small water body unused land, type from to prominent. have gradually shifted central eastern districts city 1987 dominated Haihu New District West city, Biological Park higher education base North South Duoba Town Ganhe Industrial 2019.

Language: Английский

Citations

2

Macroinvertebrate Community in a Mediterranean Mountain River: Relationship with Environmental Factors Measured at Different Spatial and Temporal Scales DOI Open Access
Cristóbal García-García, Juan Diego Gilbert, Francisco Guerrero

et al.

Sustainability, Journal Year: 2024, Volume and Issue: 16(5), P. 1777 - 1777

Published: Feb. 21, 2024

The macroinvertebrate community, physical–chemical water variables and hydromorphological indices were studied in the Turón River (Málaga, Southern Spain). Our study aims to improve knowledge of most influential environmental factors at different spatial temporal scales Mediterranean rivers, order establish better management river ecosystems. To this end, work, seasonal sampling was carried out for one year evaluate effect characteristics drainage basin (i.e., geology, topography, land use) on community. catchment basins evaluated three scales: (i) watershed level, (ii) valley segment level (iii) local level. results showed that 13 variables, 3 scale, 5 scale influenced Land use is main explanatory variable while stream channel curvature common habitat assessment index with strongest influence scale. presented a variation. During spring, autumn winter, exhibited highest resolution (adjusted R2 = 0.20–0.29), summer, became significant explaining presence taxa 0.17). obtained emphasize significance rivers adequate ecosystem management.

Language: Английский

Citations

2

Systemic design of the very-high-resolution imaging payload of an optical remote sensing satellite for launch into the VLEO using an small launch vehicle DOI Creative Commons

Mojtaba Abolghasemi Najafabadi,

Iman Kazemi

Heliyon, Journal Year: 2024, Volume and Issue: 10(6), P. e27404 - e27404

Published: March 1, 2024

The very low Earth orbit (VLEO), which includes orbits with altitudes of 100-450 km, has distinct advantages for remote-sensing spacecraft. Lower allow payloads smaller dimensions, weight, and power to achieve performances similar or better than larger in higher remote sensing observation missions. most important advantage optical imaging systems is the resolution these can attain. present paper systemically designs high-resolution payload an satellite operations VLEO. aim this design a high spatial platform reach 1-m ground sampling distance (GSD) as reported. Therefore, considering orbital altitude 300 km system calculations, implementation method described. results obtained show that proposed (VHR4) diffraction-limited compared several examples operational platforms under approximately equal conditions. Also, by calculating estimating power, dimensions (VHR4), it shown meets requirements small launcher terms power.

Language: Английский

Citations

2

Study on erosion deformation of dry-red soil in Yuanmou dry-hot valley with different elevation gradients based on SBAS-InSAR technology DOI Creative Commons

Junqi Guo,

Wenfei Xi, Guangcai Huang

et al.

Frontiers in Earth Science, Journal Year: 2024, Volume and Issue: 12

Published: April 10, 2024

The Yuanmou dry-hot valley has been confirmed as a typical area subjected to severe soil erosion in southwestern China. research on the deformation exhibited by dry-red that is extensively distributed this region takes critical significance deepening investigation of and water loss control efforts valley. In study, time series was established at different altitudinal gradients from March 2018 October 2022 using Small Baseline Subset InSAR (SBAS-InSAR) technology explore patterns Next, fractional vegetation cover (FVC) monthly average rainfall identical period were analyzed comprehensively. result study are presented follows: 1) regions valley, which observed line sight (LOS) direction, attained rates ranging −101.683 mm/yr 30.57 (Ascending),-79.658 41.942 (Descending). general, areas with significant surface concentrated Longchuan River basin flowing through north south County well river confluence zones. Uplifted have more widely reported central northern (e.g., Wudongde hydroelectric power station reservoir area). 2) A gradient effect exerted valley-dam medium low mountain most erosion, maximum reached over 80 mm. Erosion mitigated around dam high areas, 60 mm 30 mm, respectively. At an altitude 1,350 m, significantly affected rainfall. Nevertheless, variations FVC become primary factor for soil. results can scientifically support

Language: Английский

Citations

2