Joining Application of Unmanned Aerial Vehicle Imagery with GIS for Monitoring of Soft Cliff Linear Habitats

Journal of Marine Science and Engineering, Journal Year: 2025, Volume and Issue: 13(1), P. 80 - 80

Published: Jan. 5, 2025

In the coastal zone, two types of habitats—linear and areal—are distinguished. The main differences between both are their shape structure hydro- litho-dynamic, salinity, ecological gradients. Studying linear littoral habitats is essential for interpreting ’coastal squeeze’ effect. study’s objective was to assess short-term behavior soft cliffs as during calm season storm events in example Olandų Kepurė cliff, located on a peri-urban protected seashore (Baltic Sea, Lithuania). approach combined surveillance cliff using unmanned aerial vehicles (UAVs) with data analysis an ArcGIS algorithm specially adjusted habitats. authors discerned forms—cliff base cavities scarp slumps. slumps more widely spread. It particularly noticeable at beginning spring–summer period when difference occurrence forms 3.5 times. contrast, proliferate spring. This phenomenon might be related seasonal Baltic Sea level rise. conclusion that 55 m long cells optimal analyzing UAV GIS.

Language: Английский

---

The spatiotemporal changes and influencing mechanisms of the coastline in the Yellow River Delta, China

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 11

Published: Jan. 6, 2025

Using remote sensing imagery of the Yellow River Delta (YRD) from 1984 to 2024, Digital Shoreline Analysis System (DSAS) model was employed analyze coastline position, migration rate, and characteristics four typical coastal sections. The response changes in study area global climate change human activities quantitatively assessed. Over past 40 years, modern YRD has generally advanced seaward at an average rate 109.64 m/a. This progression can be divided into three distinct phases: (i) rapid transition period 2000, during which total length reached its maximum nearly 440.65 km last years. In 1986, proportion artificial surpassed that natural for first time. (ii) A decreasing trend characterized slow 2000 2015. types continued previous period, with coastlines exceeding 90% time 2015, marking highest (iii) stable 2015 present, shown increasing trend. stabilized, while growth been concentrated around estuary. However, increase gradually slowed due water sediment regulation projects 2001. evolution shifted early control by river diversions a current primary influence human-driven land reclamation projects. Coastal present estuarine sections are mainly controlled inflows, abandoned northern channels experience pronounced effects extreme weather, such as cold wave-induced winds. Additionally, factors sea-level rise delta subsidence caused compaction have lowered relative elevation coastline, further accelerating erosion retreat. these had lesser impact on than activities.

Language: Английский

---

GCL_FCS30: a global coastline dataset with 30-m resolution and a fine classification system from 2010 to 2020

Scientific Data, Journal Year: 2025, Volume and Issue: 12(1)

Published: Jan. 22, 2025

The coastline reflects coastal environmental processes and dynamic changes, serving as a fundamental parameter for coast. Although several global datasets have been developed, they mainly focus on morphology, the typology of coastlines are still lacking. We produced Global CoastLine Dataset (GCL_FCS30) with detailed classification system. extraction employed combined algorithm incorporating Modified Normalized Difference Water Index an adaptive threshold segmentation method. was performed hybrid transect classifier that integrates random forest stable training samples derived from multi-source geophysical data. GCL_FCS30 offers significant advantages in capturing artificial coastlines, reflecting strong alignment location validation found to achieve overall accuracy Kappa coefficient over 85% 0.75. Each category accurately covered majority area represented third-party data exhibited high degree spatial relevance. Therefore, is first dataset covering latitudes continuous smooth line vector format.

Language: Английский

---

A remote monitoring approach for coastal engineering projects

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 23, 2025

High costs and project-based (short-term) financing mean that coastal engineering projects are often undertaken in the absence of appropriate post-construction monitoring programmes. Consequently, performance shoreline-stabilizing structures or beach nourishments cannot be properly quantified. Given high value beaches increase erosion problems responses, managers require as much accurate data possible to support efficient decision-making. This work presents a methodological approach characterise coastline position changes result actions. We describe new, low-cost method based on satellite remote sensing monitor shoreline evolution at temporal spatial resolution pre-, during post-implementation. Initially, satellite-derived waterlines identified extracted from publicly available imagery Landsat 5, 7, 8, 9, Sentinel-2 constellations using automatic extraction tool SHOREX. The waterline positions then compiled, differences over time quantified, matrix is constructed allows easy depiction interpretation patterns erosion/accretion. access comprehension morphological by non-expert. Two examples application Valencian coast Spain different scales demonstrate how response actions can characterised levels detail (from local regional) periods time. These applications evidence utility it analysis pre- post-intervention change offers means overcome widespread lack hence improve practice.

Language: Английский

---

Advances in Remote Sensing and Deep Learning in Coastal Boundary Extraction for Erosion Monitoring

Geomatics, Journal Year: 2025, Volume and Issue: 5(1), P. 9 - 9

Published: Feb. 6, 2025

Erosion is a critical geological process that degrades soil and poses significant risks to human settlements natural habitats. As climate change intensifies, effective coastal erosion management prevention have become essential for our society the health of planet. Given vast extent areas, efforts must prioritize most vulnerable regions. Identifying prioritizing these areas complex task requires accurate monitoring forecasting its potential impacts. Various tools techniques been proposed assess risks, impacts rates erosion. Specialized methods, such as Coastal Vulnerability Index, specifically designed evaluate susceptibility boundaries, factor in monitoring, are typically extracted from remote sensing images. Due extensive scale complexity data, manually extracting boundaries challenging. Recently, artificial intelligence, particularly deep learning, has emerged promising tool this task. This review provides an in-depth analysis learning assist monitoring. imaging modalities (optical, thermal, radar), platforms (satellites, drones) datasets first presented provide context field. Artificial intelligence associated metrics then discussed, followed by exploration algorithms boundaries. The range basic convolutional networks encoder–decoder architectures attention mechanisms. An overview how other can be utilized also provided. Finally, current gaps, limitations future directions field identified. aims offer insights into through learning-based boundary extraction.

Language: Английский

---

A comprehensive review of geomatics based coastal zone management in the Realm of Arabian Gulf, Saudi Arabia

Deleted Journal, Journal Year: 2025, Volume and Issue: 7(2)

Published: Feb. 14, 2025

Language: Английский

---

Land Use/Land Cover Changes Over the Quarter Century in the Paradip Port Region in the Mahanadi Delta of India: An Investigation Using Remote Sensing and Geographical Information System

Published: Jan. 1, 2025

Language: Английский

---

An assessment of the long-term change of the Mersin west coastline using digital shoreline analysis system and detection of pattern similarity using fuzzy C-means clustering

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 12

Published: May 1, 2025

The study focused on analyzing shoreline changes along the western beaches of Mersin Province, located Turkey’s Mediterranean coast. Landsat satellite imagery from 1985 to 2022 was used detect long-term coastal alterations. Google Earth Engine (GEE) platform facilitated data acquisition, classification, and edge detection. A Support Vector Machine (SVM) classification algorithm applied distinguish land water. To enhance accuracy, additional indices—Normalized Difference Water Index (NDWI), Modified NDWI (MNDWI), Normalized Moisture (NDMI)—were incorporated alongside spectral bands. Canny detection employed delineate shorelines classified images. Resulting positions were analyzed using DSAS, an open-source ArcGIS extension, quantify erosion accretion. Key change metrics— Net Shoreline Movement (NSM), Change Envelope (SCE), End Point Rate (EPR), Linear Regression (LRR) —were derived DSAS outputs. Over 38-year period, maximum advancement reached 588.59 meters, while retreat −130.63 meters. highest rates −3.53 m/year (EPR) −2.8 (LRR), whereas most pronounced accretion 15.91 15.47 (LRR). identify spatial patterns in change, Fuzzy C-Means (FCM) clustering NSM, SCE, EPR, LRR metrics. resulting clusters then interpreted relation cover provided by European Space Agency (ESA) WorldCover dataset.

Language: Английский

---

Research on coastline extraction and dynamic change from remote sensing images based on deep learning

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

Published: Oct. 3, 2024

Accurate coastline extraction is crucial for the scientific management and protection of coastal zones. Due to diversity ground object details complexity terrain in remote sensing images, segmentation sea land faces challenges such as unclear boundaries discontinuous contours. To address these issues, this study improve accuracy efficiency by improving DeepLabv3+ model. Specifically, constructs a sea-land network, DeepSA-Net, based on strip pooling coordinate attention mechanisms. By introducing dynamic feature connections pooling, connection between different branches enhanced, capturing broader context. The introduction allows model integrate information during extraction, thereby allowing capture longer-distance spatial dependencies. Experimental results has shown that can achieves land-sea mean intersection over union (mIoU) ration Recall 99% all datasets. Visual assessment show more complete edge segmentation, confirming model’s effectiveness complex environments. Finally, using data from area China an application instance, change analysis were implemented, providing new methods

Language: Английский

---

Granite Extraction Based on the SDGSAT-1 Satellite Thermal Infrared Spectrometer Imagery

Sensors, Journal Year: 2024, Volume and Issue: 24(6), P. 1750 - 1750

Published: March 8, 2024

Earth observation by remote sensing plays a crucial role in granite extraction, and many current studies use thermal infrared data from sensors such as ASTER. The challenge lies the low spatial resolution of these satellites, hindering precise rock type identification. A breakthrough emerges with Thermal Infrared Spectrometer (TIS) on Sustainable Development Science Satellite 1 (SDGSAT-1) launched Chinese Academy Sciences. With an exceptional 30 m resolution, SDGSAT-1 TIS opens avenues for accurate extraction using sensing. This study, exemplified Xinjiang’s Karamay region, introduces BR-ISauvola method, leveraging data. approach combines band ratio adaptive k-value selection local grayscale statistical features Sauvola thresholding. Focused large-scale results show F1 scores above 70% Otsu, Sauvola, BR-ISauvola. Notably, achieves highest accuracy at 82.11%, surpassing Otsu 9.62% 0.34%, respectively. underscores potential valuable resource extraction. proposed method efficiently utilizes spectral information, presenting novel rapid imagery, even scenarios single source.

Language: Английский

---