Search Results (136)

The number of CNN application areas is growing, which leads to the need for training data. The research conducted in this work aimed to obtain effective detection models trained only using simplified synthetic objects (SSOs). The research was conducted on inland shallow water areas, while images of bottom objects were obtained using a UAV platform. The work consisted in preparing SSOs, thanks to which composite images were created. On such training data, 120 models based on the YOLO (You Only Look Once) network were obtained. The study confirmed the effectiveness of models created using YOLOv3, YOLOv5, YOLOv8, YOLOv9, and YOLOv10. A comparison was made between versions of YOLO. The influence of the amount of training data, SSO type, and augmentation parameters used in the training process was analyzed. The main parameter of model performance was the F1-score. The calculated statistics of individual models indicate that the most effective networks use partial augmentation, trained on sets consisting of 2000 SSOs. On the other hand, the increased transparency of SSOs resulted in increasing the diversity of training data and improving the performance of models. This research is developmental, and further research should improve the processes of obtaining detection models using deep networks. Full article

Aquatic vegetation plays a crucial role in freshwater ecosystems by providing habitats, regulating water quality, and supporting biodiversity. This study aims to monitor and analyze the dynamics of aquatic vegetation in Chimaditida Shallow Lake, Greece, using Sentinel-2 satellite imagery, with validation from field measurements. Data processing was performed using Google Earth Engine and QGIS. The study focuses on discriminating and mapping two classes of aquatic surface conditions: areas covered with Floating and Emergent Aquatic Vegetation and open water, covering all seasons from 1 March 2024, to 28 February 2025. Spectral bands such as B04 (red), B08 (near infrared), B03 (green), and B11 (shortwave infrared) were used, along with indices like the Modified Normalized Difference Water Index and Normalized Difference Vegetation Index. The classification was enhanced using Otsu’s thresholding technique to distinguish accurately between Floating and Emergent Aquatic Vegetation and open water. Seasonal fluctuations were observed, with significant peaks in vegetation growth during the summer and autumn months, including a peak coverage of 2.08 km² on 9 September 2024 and a low of 0.00068 km² on 28 December 2024. These variations correspond to the seasonal growth patterns of Floating and Emergent Aquatic Vegetation, driven by temperature and nutrient availability. The study achieved a high overall classification accuracy of 89.31%, with producer accuracy for Floating and Emergent Aquatic Vegetation at 97.42% and user accuracy at 95.38%. Validation with Unmanned Aerial Vehicle-based aerial surveys showed a strong correlation (R² = 0.88) between satellite-derived and field data, underscoring the reliability of Sentinel-2 for aquatic vegetation monitoring. Findings highlight the potential of satellite-based remote sensing to monitor vegetation health and dynamics, offering valuable insights for the management and conservation of freshwater ecosystems. The results are particularly useful for governmental authorities and natural park administrations, enabling near-real-time monitoring to mitigate the impacts of overgrowth on water quality, biodiversity, and ecosystem services. This methodology provides a cost-effective alternative for long-term environmental monitoring, especially in regions where traditional methods are impractical or costly. Full article

Surface water is the primary freshwater supply for Earth. Small lakes and ponds provide important ecological and economic services to society but are often left undocumented, or their documentation is outdated, due to their small sizes and temporal dynamics. This study tested the feasibility of the new Surface Water and Ocean Topography (SWOT) mission regarding the 3D documentation of small waterbodies in a coastal area of South Carolina, USA. Via deep learning using a recent 15 cm aerial image, small waterbodies (>0.02 ha) were extracted at an average precision score of 0.81. The water surface elevation (WSE) of each waterbody was extracted using the SWOT Level-2 Water Mask Pixel Cloud (PIXC) product, with the data collected on 1 June 2023. Using a statistical noise-removal approach, the average WSE values of small waterbodies revealed a significant correlation (Pearson’s r = 0.64) with their bottom elevations. Via spatial interpolation, the water levels of small waterbodies across the study area were generally aligned with the state-reported Cone of Depression of ground water surfaces in underlying aquifers. While the WSE measurements of SWOT pixel points are noisy due to the land–water interactions in small waterbodies, this study indicates that the SWOT PIXC product could provide a valuable resource for assessing freshwater availability to assist in water-use decision-making. Full article

Water surface inspection is a crucial instrument for safeguarding the aquatic environment. UAVs enhance the efficiency of water area inspections due to their high mobility and extensive coverage. This paper introduces two UAV inspection methodologies for the characteristics of rivers and lakes, along with an efficient semantic segmentation algorithm, WaterSegLite (Water Segmentation Lightweight algorithm), for UAV visual navigation. The algorithm employs the UAV’s aerial perspective alongside a streamlined neural network architecture to facilitate rapid real-time segmentation of water bodies and to furnish positional data to the UAV for visual navigation. The experimental findings indicate that WaterSegLite achieves a segmentation accuracy (mIoU) of 93.81% and an F1 score of 95.44%, surpassing the baseline model by 2.7% and 2.23%, respectively. Simultaneously, the processing frame rate of this algorithm on the airborne device attains 28.27 frames per second, fully satisfying the requirements for real-time water surface inspection by UAVs. This paper offers technical assistance for UAV inspection techniques in aquatic environments and presents innovative concepts for the intelligent advancement of water surface inspection. Full article

Alpine wetland ecosystems, as important carbon sinks and water conservation areas, possess unique ecological functions. Driven by climate change and human activities, the spatial distribution changes in alpine wetlands directly affect the ecosystems and water resource management within a basin. To further refine the evolution processes of different types of alpine wetlands in different zones of a basin, this study combined multiple field surveys, unmanned aerial vehicle (UAV) flights, and high-resolution images. Based on the Google Earth Engine (GEE) cloud platform, we constructed a Random Forest model to identify and extract alpine wetlands in the Shule River Basin over a long-term period from 1987 to 2021. The results indicated that the accuracy of the extraction based on this method exceeded 90%; the main wetland types are marsh, swamp meadow, and river and lake water bodies; and the spatial–temporal distribution of each wetland type has obvious heterogeneity. In total, 90% of the swamp meadows areas were mainly scattered throughout the study area’s section 3700 to 4300 m above sea level (a.s.l.), and 80% of the marshes areas were concentrated in the Dang River source 3200 m above sea level. From 1987 to 2021, the alpine wetland in the study area showed an overall expansion trend. The total area of the wetland increased by 51,451.8 ha and the area increased by 53.5%. However, this expansion mainly occurred in the elevation zone below 4000 m after 2004, and low-altitude marsh wetland primarily dominated the expansion. The analysis of the spatial–temporal heterogeneity of alpine wetlands can provide a scientific basis for the attribution analysis of the change in alpine wetlands in inland water conservation areas, as well as for protection and rational development and utilization, and promote the healthy development of ecological environments in nature reserves. Full article

Tailings generated by mining account for the largest world-wide waste from industrial activities. As an element, copper is relatively uncommon, with low concentrations in sediments and waters, yet is very elevated around mining operations. On the Keweenaw Peninsula of Michigan, USA, jutting out into Lake Superior, 140 mines extracted native copper from the Portage Lake Volcanic Series, part of an intercontinental rift system. Between 1901 and 1932, two mills at Gay (Mohawk, Wolverine) sluiced 22.7 million metric tonnes (MMT) of copper-rich tailings (stamp sands) into Grand (Big) Traverse Bay. About 10 MMT formed a beach that has migrated 7 km from the original Gay pile to the Traverse River Seawall. Another 11 MMT are moving underwater along the coastal shelf, threatening Buffalo Reef, an important lake trout and whitefish breeding ground. Here we use remote sensing techniques to document geospatial environmental impacts and initial phases of remediation. Aerial photos, multiple ALS (crewed aeroplane) LiDAR/MSS surveys, and recent UAS (uncrewed aircraft system) overflights aid comprehensive mapping efforts. Because natural beach quartz and basalt stamp sands are silicates of similar size and density, percentage stamp sand determinations utilise microscopic procedures. Studies show that stamp sand beaches contrast greatly with natural sand beaches in physical, chemical, and biological characteristics. Dispersed stamp sand particles retain copper, and release toxic levels of dissolved concentrations. Moreover, copper leaching is elevated by exposure to high DOC and low pH waters, characteristic of riparian environments. Lab and field toxicity experiments, plus benthic sampling, all confirm serious impacts of tailings on aquatic organisms, supporting stamp sand removal. Not only should mining companies end coastal discharges, we advocate that they should adopt the UNEP “Global Tailings Management Standard for the Mining Industry”. Full article

This paper presents an assessment of potential losses and damage costs to rare coastal plant species of Lake Baikal (UNESCO World Heritage Site) as a result of inundation at high water levels. The lake’s ecosystem is characterized by an exceptional diversity of rare and endemic animal and plant species. The construction of a hydroelectric power plant caused an increase in the water level of Lake Baikal, resulting in the inundation of low-lying coastal areas, the destruction of the coastline, alterations to the hydrological regime, etc. However, there are practically no works devoted to water-level modeling and the assessment of its impact on riparian vegetation, including rare species. We conducted fieldwork to determine the abundance of four vulnerable species and identified inundation zones at different high water levels on the basis of digital elevation models based on aerial photography data. The analysis revealed that at the maximum level of inundation, the number of plant species affected would total 5164, amounting to a financial loss of biodiversity estimated at 3098.4 thousand rubles. To mitigate the projected losses, it is imperative to implement measures that restrict water-level fluctuations above the 457.00 m threshold. The absence of flora as an object of state environmental monitoring, which is not specified in the regulatory legal document, must be rectified in a timely manner. Full article

This study investigates the formation of cold air pools during calm, anticyclonic winter nights in a topographic basin bounded by a medium-sized mountain to the east and near-flat terrain elsewhere. The main objective is to understand how local topography drives unique topoclimatic conditions—specifically cold air lakes and an inversion layer at approximately 100/120 m altitude—in a peri-urban depression where a major cement factory and several residential areas are located. To achieve this, the research design combined surface measurements (collected at 10:00 p.m., 3:00 a.m., 7:00 a.m., and 3:00 p.m.) using a motorized vehicle, with vertical measurements (at 7:00 a.m.) collected via two unmanned aerial vehicles (UAVs), with the three vehicles equipped with Tinytag data loggers. The Empirical Bayesian Kriging tool in ArcGIS Pro was employed to generate the surface temperature cartograms. The results show that shortly after sunset, a cold air layer of approximately 100–120 m thickness forms, with nocturnal air temperature variations of up to 8 °C on the night measurements. An inversion layer was detected at around 120–130 m, while near-zero wind speeds in the basin’s core facilitate the retention of cold air. Surface spatialization confirms earlier findings of a cold air lake and thermal belts on the basin’s perimeter, forming in the early evening and dissipating by late morning. A 3D visualization underscores the influence of the mountain in directing cold air downslope, leading to stabilization and stratification within the lower atmospheric layers. These findings carry significant health implications: air pollutants released by the cement plant tend to accumulate within the cold air pool and beneath the inversion layer, posing potential risks to nearby populations. Full article

Turing’s instability has been widely introduced to explain the formation of several biological and ecological patterns, such as the skin patterning of fish or animals, wings of butterflies, pigmentation, and labyrinth patterns of the cerebral cortex of mammals. Such a mechanism may occur in the ecosystem due to the differential diffusion dispersal that happen if one of the constituent species results in the activator or the prey, showing a tendency to undergo autocatalytic growth. The diffusion of the constituent species activator is a random mobility function called passive diffusion. If the other species in the system (the predator/inhibitor) disperses sufficiently faster than the activator, then the spatially uniform distribution of species becomes unstable, and the system will settle into a stationary state. This paper introduced Turing’s mechanism in our reaction–taxis–diffusion model to simulate the harmful algal bloom (HAB) pattern. A numerical approach, the Runge–Kutta method, was used to deal with this system of reaction–taxis–diffusion equations, and the findings were qualitatively compared to the aerial patterns obtained by a drone flying over Torment Lake in Nova Scotia (Canada) during the bloom season of September 2023. Full article

Optical remote sensing images are easily affected by atmospheric absorption and scattering, and the low contrast and low signal-to-noise ratio (SNR) of aerial images as well as the different sensors of aerial and satellite images bring a great challenge to image matching. A tilted aerial image and satellite image matching algorithm based on edge curve direction angle features (ECDAF) is proposed, which accomplishes image matching by extracting the edge features of the images and establishing the curve direction angle feature descriptors. First, tilt and resolution transforms are performed on the satellite image, and edge detection and contour extraction are performed on the aerial image and transformed satellite image to make preparations for image matching. Then, corner points are detected and feature descriptors are constructed based on the edge curve direction angle. Finally, the integrated matching similarity is computed to realize aerial–satellite image matching. Experiments run on a variety of remote sensing datasets including forests, hills, farmland, and lake scenes demonstrate that the effectiveness of the proposed algorithm shows a great improvement over existing state-of-the-art algorithms. Full article

Small lakes play an essential role in maintaining regional ecosystem stability and water quality. However, turbidity in these lakes is increasingly influenced by anthropogenic activities, which presents a challenge for traditional monitoring methods. This study explores the feasibility of using consumer-grade UAVs equipped with RGB cameras to monitor water turbidity in small lakes within the Taihu Lake Basin of eastern China. By collecting RGB imagery and in situ turbidity measurements, we developed and validated models for turbidity prediction. RGB band indices were used in combination with three machine learning models, namely Interpretable Feature Transformation Regression (IFTR), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). Results showed that models utilizing combinations of the R, G, B, and ln(R) bands achieved the highest accuracy, with the IFTR model demonstrating the best performance (R² = 0.816, RMSE = 3.617, MAE = 2.997). The study confirms that consumer-grade UAVs can be an effective, low-cost tool for high-resolution turbidity monitoring in small lakes, providing valuable insights for sustainable water quality management. Future research should investigate advanced algorithms and additional spectral features to further enhance prediction accuracy and adaptability. Full article

Marine and freshwater plastic pollution is a worldwide problem affecting ecosystems and human health. Although remote sensing has been used to map large floating plastic rafts, there are research gaps in detecting submerged plastic due to the limited amount of in situ data. This study is the first to collect in situ data on submerged and floating plastics in a freshwater environment and analyse the effect of water submersion on the strength of the plastic signal. A large 10 × 10 m artificial polymer tarpaulin was deployed in a freshwater lake for a two-week period and was captured by a multi-sensor and multi-resolution unmanned aerial vehicle (UAV) and satellite. Spectral analysis was conducted to assess the attenuation of individual wavelengths of the submerged tarpaulin in UAV hyperspectral and Sentinel-2 multispectral data. A K-Means unsupervised clustering algorithm was used to classify the images into two clusters: plastic and water. Additionally, we estimated the optimal number of clusters present in the hyperspectral dataset and found that classifying the image into four classes (water, submerged plastic, near surface plastic and buoys) significantly improved the accuracy of the K-Means predictions. The submerged plastic tarpaulin was detectable to ~0.5 m below the water surface in near infrared (NIR) (~810 nm) and red edge (~730 nm) wavelengths. However, the red spectrum (~669 nm) performed the best with ~84% true plastic positives, classifying plastic pixels correctly even to ~1 m depth. These individual bands outperformed the dedicated Plastic Index (PI) derived from the UAV dataset. Additionally, this study showed that in neither Sentinel-2 bands, nor the derived indices (PI or Floating Debris Index (FDI), it is currently possible to determine if and how much of the tarpaulin was under the water surface, using a plastic tarpaulin object of 10 × 10 m. Overall, this paper showed that spatial resolution was more important than spectral resolution in detecting submerged tarpaulin. These findings directly contributed to Sustainable Development Goal 14.1 on mapping large marine plastic patches of 10 × 10 m and could be used to better define systems for monitoring submerged and floating plastic pollution. Full article

Fractional vegetation cover (FVC) plays a key role in ecological and environmental status assessment because it directly reflects the extent of vegetation cover and its status, yet vegetation is an important component of ecosystems. FVC estimation methods have evolved from traditional manual interpretation to advanced remote sensing technologies, such as satellite data analysis and unmanned aerial vehicle (UAV) image processing. Extraction methods based on high-resolution UAV data are being increasingly studied in the fields of ecology and remote sensing. However, research on UAV-based FVC extraction against the backdrop of the high soil reflectance in arid regions remains scarce. In this paper, based on 12 UAV visible light images in differentiated scenarios in the Ebinur Lake basin, Xinjiang, China, various methods are used for high-precision FVC estimation: Otsu’s thresholding method combined with 12 Visible Vegetation Indices (abbreviated as Otsu-VVIs) (excess green index, excess red index, excess red minus green index, normalized green–red difference index, normalized green–blue difference index, red–green ratio index, color index of vegetation extraction, visible-band-modified soil-adjusted vegetation index, excess green minus red index, modified green–red vegetation index, red–green–blue vegetation index, visible-band difference vegetation index), color space method (red, green, blue, hue, saturation, value, lightness, ‘a’ (Green–Red component), and ‘b’ (Blue–Yellow component)), linear mixing model (LMM), and two machine learning algorithms (a support vector machine and a neural network). The results show that the following methods exhibit high accuracy in FVC extraction across differentiated scenarios: Otsu–CIVE, color space method (‘a’: Green–Red component), LMM, and SVM (Accuracy > 0.75, Precision > 0.8, kappa coefficient > 0.6). Nonetheless, higher scene complexity and image entropy reduce the applicability of precise FVC extraction methods. This study facilitates accurate, efficient extraction of vegetation information in differentiated scenarios within arid and semiarid regions, providing key technical references for FVC estimation in similar arid areas. Full article

Shrinking shorelines and the exposed playa of saline lakes can pose public health and air quality risks for local communities. This study combines a community science method with models to forecast future shorelines and PM10 air quality impacts from the exposed playa of the Salton Sea, near the community of North Shore, CA, USA. The community science process assesses the rate of shoreline change from aerial images collected through a balloon mapping method. These images, captured from 2019 to 2021, are combined with additional satellite images of the shoreline dating back to 2002, and analyzed with the DSAS (Digital Shoreline Analysis System) in ArcGIS desktop. The observed rate of change was greatly increased during the period from 2017 to 2020. The average rate of change rose from 12.53 m/year between 2002 and 2017 to an average of 38.44 m/year of shoreline change from 2017 to 2020. The shoreline is projected to retreat 150 m from its current position by 2030 and an additional 172 m by 2041. To assess potential air quality impacts, we use WRF-Chem, a regional chemical transport model, to predict increases in emissive dust from the newly exposed playa land surface. The model output indicates that the forecasted 20-year increase in exposed playa will also lead to a rise in the amount of suspended dust, which can then be transported into the surrounding communities. The combination of these model projections suggests that, without mitigation, the expanding exposed playa around the Salton Sea is expected to worsen pollutant exposure in local communities. Full article

The acquisition of the three-dimensional (3D) morphology of the complete tufa dam system is of great significance for analyzing the formation and development of a pellucid tufa lake in a fluvial tufa valley. The dam system is usually composed of the dams partially exposed above-water and the ones totally submerged underwater. This situation makes it difficult to directly obtain the real 3D scene of the dam system solely using an existing measurement technique. In recent years, unmanned aerial vehicle (UAV) digital photogrammetry has been increasingly used to acquire high-precision 3D models of various earth surface scenes. In this study, taking Wolong Lake and its neighborhood in Jiuzhaigou Valley, China as the study site, we employed a fixed-wing UAV equipped with a consumer-level digital camera to capture the overlapping images, and produced the initial Digital Surface Model (DSM) of the dam system. The refraction correction was applied to retrieving the underwater Digital Elevation Model (DEM) of the submerged dam or dam part, and the ground interpolation was adopted to eliminate vegetation obstruction to obtain the DEM of the dam parts above-water. Based on the complete 3D model of the dam system, the elevation profiles along the centerlines of Wolong Lake were derived, and the dimension data of those tufa dams on the section lines were accurately measured. In combination of local hydrodynamics, the implication of the morphological characteristics for analyzing the formation and development of the tufa dam system was also explored. Full article