Land Use and Land Cover Mapping Using Sentinel-2, Landsat-8 Satellite Images, and Google Earth Engine: A Comparison of Two Composition Methods

Remote Sensing, Journal Year: 2022, Volume and Issue: 14(9), P. 1977 - 1977

Published: April 20, 2022

Accurate and real-time land use/land cover (LULC) maps are important to provide precise information for dynamic monitoring, planning, management of the Earth. With advent cloud computing platforms, time series feature extraction techniques, machine learning classifiers, new opportunities arising in more accurate large-scale LULC mapping. In this study, we aimed at finding out how two composition methods spectral–temporal metrics extracted from satellite can affect ability a classifier produce maps. We used Google Earth Engine (GEE) platform create cloud-free Sentinel-2 (S-2) Landsat-8 (L-8) over Tehran Province (Iran) as 2020. Two methods, namely, seasonal composites percentiles metrics, were define four datasets based on series, vegetation indices, topographic layers. The random forest was classification identifying most variables. Accuracy assessment results showed that S-2 outperformed L-8 overall class level. Moreover, comparison indicated percentile both series. At level, improved performance related their better about phenological variation different classes. Finally, conclude methodology GEE an fast way be

Language: Английский

---

Integrating forest cover change and carbon storage dynamics: Leveraging Google Earth Engine and InVEST model to inform conservation in hilly regions

Ecological Indicators, Journal Year: 2023, Volume and Issue: 152, P. 110374 - 110374

Published: May 22, 2023

Forests are vital in combating climate change by storing and sequestrating CO2 from the atmosphere. Measuring influence of land use/land cover (LULC) changes on capacity carbon storage (CS) within forest ecosystems presents a significant challenge. This study employs remote sensing techniques to examine spatiotemporal patterns CS Chittagong Hill Tracts (CHT), resulting LULC alterations between 1996 2021. were identified for six different years utilizing Google Earth Engine (GEE). The Integrated Valuation Ecosystem Services Tradeoffs (InVEST) model was combined with GEE evaluate changing CS. discovered that CHT region experienced loss 21.65 × 106 Mg CS, owing 21% reduction vegetation (2862.85 km^2) during period. central city area (Chittagong) accounted most (7.99 Mg), while suburban areas Khagrachari (0.92 Mg) Rangamati (3.53 contributed least. multiple regression revealed elevation characteristics significantly influenced findings underscore importance developing policies strategies mitigate adverse effects advocate sustainable management practices strike balance ecological, social, economic concerns.

Language: Английский

---

A review of carbon monitoring in wet carbon systems using remote sensing

Environmental Research Letters, Journal Year: 2022, Volume and Issue: 17(2), P. 025009 - 025009

Published: Jan. 20, 2022

Abstract Carbon monitoring is critical for the reporting and verification of carbon stocks change. Remote sensing a tool increasingly used to estimate spatial heterogeneity, extent change within across various systems. We designate use term wet system interconnected wetlands, ocean, river streams, lakes ponds, permafrost, which are carbon-dense vital conduits throughout terrestrial aquatic sections cycle. reviewed studies that utilize earth observation improve our knowledge data gaps, methods, future research recommendations. To achieve this, we conducted systematic review collecting 1622 references screening them with combination text matching panel three experts. The search found 496 references, an additional 78 added by Our study considerable variability utilization remote global progress nine systems analyzed. highlighted routinely globally map in mangroves oceans, whereas seagrass, tidal marshes, rivers, permafrost would benefit from more accurate comprehensive maps extent. identified gaps twelve recommendations continue progressing increase cross scientific inquiry.

Language: Английский

---

Advances in Earth observation and machine learning for quantifying blue carbon

Earth-Science Reviews, Journal Year: 2023, Volume and Issue: 243, P. 104501 - 104501

Published: July 13, 2023

Blue carbon ecosystems (mangroves, seagrasses and saltmarshes) are highly productive coastal habitats, considered some of the most carbon-dense on earth. They an important nature-based solution for both climate change mitigation adaptation. Quantifying blue stocks assessing their dynamics at large scales through remote sensing remains challenging due to difficulties cloud coverage, spectral, spatial temporal limitations multispectral sensors speckle noise synthetic aperture radar (SAR). Recent advances in airborne space-borne SAR imagery Light Detection Ranging (LiDAR) data, sensor platforms such as unmanned aerial vehicles (UAVs), combined with novel machine learning techniques have offered different users a wide-range spatial, multi-temporal information quantifying from space. However, number challenges posed by various traits atmospheric correction, water penetration, column transparency issues environments, multi-dimensionality size LiDAR limitation training samples, backscattering mechanisms acquisition process. As result, existing methodologies face major accurately estimating using these datasets. In this context, emerging innovative artificial intelligence often required robustness reliability estimates, particularly those open-source software signal processing regression tasks. This review provides overview Earth Observation state-of-the-art deep that currently being used quantify above-ground carbon, below-ground soil mangroves, saltmarshes ecosystems. Some key future directions potential use data fusion advanced learning, metaheuristic optimisation also highlighted. summary, quantification approaches holds great contributing global efforts towards mitigating protecting

Language: Английский

---

Detection of Water Hyacinth (Eichhornia crassipes) in Lake Tana, Ethiopia, Using Machine Learning Algorithms

Water, Journal Year: 2023, Volume and Issue: 15(5), P. 880 - 880

Published: Feb. 24, 2023

Lake Tana is Ethiopia’s largest lake and infested with invasive water hyacinth (E. crassipes), which endangers the lake’s biodiversity habitat. Using appropriate remote sensing detection methods determining seasonal distribution of weed important for decision-making, resource management, environmental protection. As demand reliable estimation E. crassipes mapping from satellite data grows, comparing performance different machine learning algorithms could help in identifying most effective method lake. Therefore, this study aimed to examine ability random forest (RF), support vector (SVM), classification regression tree (CART) detect estimating spatial coverage on Google Earth Engine (GEE) platform using Landsat 8 Sentinel 2 images. Cloud-masked monthly median composite October 2021 2022, January 2022 2023, March June were used represent autumn, winter, spring, summer, respectively. Four spectral indices derived combination bands improve accuracy. All achieved greater than 95% 90% overall accuracy when images, both sets, all a 93% F1 score detection. Though difference between was small, RF accurate, while SVM CART had same The maximum area observed autumn (22.4 km2), minimum (2.2 km2) summer. Based data, decreased significantly by 62.5% winter spring increased 81.7% summer autumn. findings suggested that classifier accurate algorithm, an season Tana.

Language: Английский

---

Enhancing Carbon Sequestration through Afforestation: Evaluating the Impact of Land Use and Cover Changes on Carbon Storage Dynamics

Earth Systems and Environment, Journal Year: 2024, Volume and Issue: 8(4), P. 1563 - 1582

Published: June 15, 2024

Language: Английский

---

A New Hybrid Firefly–PSO Optimized Random Subspace Tree Intelligence for Torrential Rainfall-Induced Flash Flood Susceptible Mapping

Remote Sensing, Journal Year: 2020, Volume and Issue: 12(17), P. 2688 - 2688

Published: Aug. 20, 2020

Flash flood is one of the most dangerous natural phenomena because its high magnitudes and sudden occurrence, resulting in huge damages for people properties. Our work aims to propose a state-of-the-art model susceptibility mapping flash using decision tree random subspace ensemble optimized by hybrid firefly–particle swarm optimization (HFPS), namely HFPS-RSTree model. In this work, we used data from inventory map consisting 1866 polygons derived Sentinel-1 C-band synthetic aperture radar (SAR) field survey conducted northwest mountainous area Van Ban district, Lao Cai Province Vietnam. A total eleven flooding conditioning factors (soil type, geology, rainfall, river density, elevation, slope, aspect, topographic wetness index (TWI), normalized difference vegetation (NDVI), plant curvature, profile curvature) were as explanatory variables. These indicators compiled geological mineral resources map, soil type ALOS PALSAR DEM 30 m, Landsat-8 imagery. The was trained verified variables then compared with four machine learning algorithms, i.e., support vector (SVM), forests (RF), C4.5 trees (C4.5 DT), logistic (LMT) models. We employed range statistical standard metrics assess predictive performance proposed results show that had best achieved better than those other benchmarks ability predict flood, reaching an overall accuracy over 90%. It can be concluded approach provides new insights into prediction regions.

Language: Английский

---

Spatiotemporal Mapping and Monitoring of Mangrove Forests Changes From 1990 to 2019 in the Northern Emirates, UAE Using Random Forest, Kernel Logistic Regression and Naive Bayes Tree Models

Frontiers in Environmental Science, Journal Year: 2020, Volume and Issue: 8

Published: July 16, 2020

Mangrove forests are acting as a green lung for the coastal cities of United Arab Emirates, providing habitat wildlife, storing blue carbon in sediment and protecting shoreline. Thus, first step toward conservation better understanding ecological setting mangroves is mapping monitoring mangrove extent over multiple spatial scales. This study aims to develop novel low-cost remote sensing approach spatiotemporal forest northern part Emirates (NUAE). The was developed based on random (RF), Kernel logistic regression (KLR), Naive Bayes Tree (NBT) machine learning algorithms which use multitemporal Landsat images. Our results accuracy metrics include accuracy, precision, recall, F1 score revealed that RF outperformed KLR NB with an more than 0.90. Each pair produced maps (1990-2000, 2000-2010, 2010-2019 1990-2019) used image difference algorithm (ID) monitor by applying threshold ranges from +1 -1. great importance research community. new presented this will be good reference useful source management organization.

Language: Английский

---

Land Use/Land Cover Changes Impact on Groundwater Level and Quality in the Northern Part of the United Arab Emirates

Remote Sensing, Journal Year: 2020, Volume and Issue: 12(11), P. 1715 - 1715

Published: May 27, 2020

This study aims to develop an integrated approach for mapping and monitoring land use/land cover (LULC) changes investigate the impacts of LULC population growth on groundwater level quality using Landsat images hydrological information in a Geographic system (GIS) environment. All (1990, 2000, 2010, 2018) were classified support vector machine (SVM) spectral analysis mapper (SAM) classifiers. The result validation metrics, including precision, recall, F1, indicated that SVM classier has better performance than SAM. obtained maps have overall accuracy more 90%. Each pair enhanced (1990–2000, 2000–2010, 2010–2018, 1990–2018) used as input data image difference algorithm monitor changes. Maps change detection then imported into GIS environment spatially correlated against spatiotemporal quality. results also show approximate built-up area increased from 227.26 km2 (1.39%) 869.77 (7.41%), while vegetated areas (farmlands, parks gardens) about 76.70 (0.65%) 290.70 (2.47%). observed are highly linked depletion across Oman Mountains coastal areas.

Language: Английский

---

Semantic segmentation of seagrass habitat from drone imagery based on deep learning: A comparative study

Ecological Informatics, Journal Year: 2021, Volume and Issue: 66, P. 101430 - 101430

Published: Sept. 21, 2021

Language: Английский

---