Surface sediment classification using a deep learning model and unmanned aerial vehicle data of tidal flats.

This study proposes a deep learning model, U-Net, to improve surface sediment classification using high-resolution unmanned aerial vehicle (UAV) images. We constructed training datasets with UAV images and corresponding labeling data acquired from three field surveys on the Hwangdo tidal flat. The labeling data indicated the distribution of surface sediment types. We compared the performance of the U-Net model trained in various implementation environments, such as surface sediment criteria, input datasets, and classification models. The U-Net trained with five class criteria-derived from previous classification criteria-yielded valid results (overall accuracy:65.6 %). The most accurate results were acquired from trained U-Net with all input datasets; in particular, the tidal channel density caused a significant increase in accuracy. The accuracy of the U-Net was approximately 20 % higher than that of other classification models. These results demonstrate that surface sediment classification using UAV images and the U-Net model is effective.

High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery.

Red tides caused by Margalefidinium polykrikoides occur continuously along the southern coast of Korea, where there are many aquaculture cages, and therefore, prompt monitoring of bloom water is required to prevent considerable damage. Satellite-based ocean-color sensors are widely used for detecting red tide blooms, but their low spatial resolution restricts coastal observations. Contrarily, terrestrial sensors with a high spatial resolution are good candidate sensors, despite the lack of spectral resolution and bands for red tide detection. In this study, we developed a U-Net deep learning model for detecting M. polykrikoides blooms along the southern coast of Korea from PlanetScope imagery with a high spatial resolution of 3 m. The U-Net model was trained with four different datasets that were constructed with randomly or non-randomly chosen patches consisting of different ratios of red tide and non-red tide pixels. The qualitative and quantitative assessments of the conventional red tide index (RTI) and four U-Net models suggest that the U-Net model, which was trained with a dataset of non-randomly chosen patches including non-red tide patches, outperformed RTI in terms of sensitivity, precision, and F-measure level, accounting for an increase of 19.84%, 44.84%, and 28.52%, respectively. The M. polykrikoides map derived from U-Net provides the most reasonable red tide patterns in all water areas. Combining high spatial resolution images and deep learning approaches represents a good solution for the monitoring of red tides over coastal regions.

Satellite-based observations of unexpected coastal changes due to the Saemangeum Dyke construction, Korea.

Spatial and temporal changes around an area of conventional coastal engineering can be easily observed from field surveys because of the clear cause-and-effect observable in the before and after stages of the project. However, it is more difficult to determine environmental changes in the vicinity of tidal flats and coastal areas that are a considerable distance from the project. To identify any unexpected environmental impacts of the construction of Saemangeum Dyke in the area, we examined morphological changes identified by satellite-based observations through a field survey on Gomso Bay tidal flats (15km from Saemangeum Dyke), and changes in the suspended sediment distribution identified by satellite-based observations through a hydrodynamic analysis in the Saemangeum and Gomso coastal area. We argue that hydrodynamic changes due to conventional coastal engineering can affect the sedimentation pattern in the vicinity of tidal flats. We suggest that the environmental impact assessment conducted before a conventional coastal engineering project should include a larger area than is currently considered.

Ocean color products from the Korean Geostationary Ocean Color Imager (GOCI).

The first geostationary ocean color satellite sensor, Geostationary Ocean Color Imager (GOCI), which is onboard South Korean Communication, Ocean, and Meteorological Satellite (COMS), was successfully launched in June of 2010. GOCI has a local area coverage of the western Pacific region centered at around 36°N and 130°E and covers ~2500 × 2500 km(2). GOCI has eight spectral bands from 412 to 865 nm with an hourly measurement during daytime from 9:00 to 16:00 local time, i.e., eight images per day. In a collaboration between NOAA Center for Satellite Applications and Research (STAR) and Korea Institute of Ocean Science and Technology (KIOST), we have been working on deriving and improving GOCI ocean color products, e.g., normalized water-leaving radiance spectra (nLw(λ)), chlorophyll-a concentration, diffuse attenuation coefficient at the wavelength of 490 nm (Kd(490)), etc. The GOCI-covered ocean region includes one of the world's most turbid and optically complex waters. To improve the GOCI-derived nLw(λ) spectra, a new atmospheric correction algorithm was developed and implemented in the GOCI ocean color data processing. The new algorithm was developed specifically for GOCI-like ocean color data processing for this highly turbid western Pacific region. In this paper, we show GOCI ocean color results from our collaboration effort. From in situ validation analyses, ocean color products derived from the new GOCI ocean color data processing have been significantly improved. Generally, the new GOCI ocean color products have a comparable data quality as those from the Moderate Resolution Imaging Spectroradiometer (MODIS) on the satellite Aqua. We show that GOCI-derived ocean color data can provide an effective tool to monitor ocean phenomenon in the region such as tide-induced re-suspension of sediments, diurnal variation of ocean optical and biogeochemical properties, and horizontal advection of river discharge. In particular, we show some examples of ocean diurnal variations in the region, which can be provided effectively from satellite geostationary measurements.

Macrobenthos habitat potential mapping using GIS-based artificial neural network models.

This paper proposes and tests a method of producing macrobenthos habitat potential maps in Hwangdo tidal flat, Korea based on an artificial neural network. Samples of macrobenthos were collected during field work, and eight control factors were compiled as a spatial database from remotely sensed data and GIS analysis. The macrobenthos habitat potential maps were produced using an artificial neural network model. Macrobenthos habitat potential maps were made for Macrophthalmus dilatatus, Cerithideopsilla cingulata, and Armandia lanceolata. The maps were validated by compared with the surveyed habitat locations. A strong correlation between the potential maps and species locations was revealed. The validation result showed average accuracies of 74.9%, 78.32%, and 73.27% for M. dilatatus, C. cingulata, and A. lanceolata, respectively. A GIS-based artificial neural network model combined with remote sensing techniques is an effective tool for mapping the areas of macrobenthos habitat potential in tidal flats.

Monitoring of suspended sediment variation using Landsat and MODIS in the Saemangeum coastal area of Korea.

Suspended sediment concentration (SS) is an important indicator of marine environmental changes due to natural causes such as tides, tidal currents, and river discharges, as well as human activities such as construction in coastal regions. In the Saemangeum area on the west coast of Korea, construction of a huge tidal dyke for land reclamation has strongly influenced the coastal environment. This study used remotely sensed data to analyze the SS changes in coastal waters caused by the dyke construction. Landsat and MODIS satellite images were used for the spatial analysis of finer patterns and for the detailed temporal analysis, respectively. Forty Landsat scenes and 105 monthly composite MODIS images observed during 1985-2010 were employed, and four field campaigns (from 2005 to 2006) were performed to verify the image-derived SS. The results of the satellite data analyses showed that the seawater was clear before the dyke construction, with SS values lower than 20 g/m(3). These values increased continuously as the dyke construction progressed. The maximum SS values appeared just before completion of the fourth dyke. Values decreased to below 5 g/m(3) after dyke construction. These changes indicated tidal current modification. Some eddies and plumes were observed in the images generated from Landsat data. Landsat and MODIS can reveal that coastal water turbidity was greatly reduced after completion of the construction.

Centroid method: reliable method to determine the coronal curvature of scoliosis: a case control study comparing with the Cobb method.

STUDY DESIGN: Observational study with three examiners. OBJECTIVE: To compare the reliability of the Cobb and centroid methods. SUMMARY OF BACKGROUND DATA: The Cobb method is considered to be the gold standard in scoliosis measurement despite its low reliability. In adolescent idiopathic scoliosis (AIS) patients, the centroid method can be a good method for measuring scoliosis. METHODS: Sixty whole spine postero-anterior radiographs were collected to compare the reliability of the Cobb and centroid methods in AIS patients. Sixty radiographs were measured twice by each of the three examiners using the two measurement methods. The data were analyzed statistically to determine the inter- and intraobserver reliability. RESULT: In comparisons of inter- and intraobserver reliability of all 60 radiographs, the inter- and intraclass coefficients (ICCs) were higher in the centroid (>0.969) than in the Cobb method (>0.832), although both were in the excellent range. The mean absolute difference (MAD) values were higher in the Cobb method (<7.15° vs. <3.75°), and >5° in five comparisons. Regarding measures of mismatched radiograms, the inter- and intraobserver MAD values were higher in the Cobb method (<9.81° vs. <3.82°), and >5° in six comparisons. And, the ICCs were higher in the centroid method (>0.972) than the Cobb method (>0.758). In immature radiograms, the ICCs were higher in the centroid (>0.973) than in the Cobb method (>0.764), even though it was in the excellent range. And, the inter- and intraobserver MAD values were higher in the Cobb method (<8.49° vs. <3.99°), and >5° in seven comparisons. CONCLUSION: The centroid method is more reliable for measuring scoliosis in AIS than the Cobb method, and it can substitute the Cobb method, which showed high variability.

Macrobenthos habitat mapping in a tidal flat using remotely sensed data and a GIS-based probabilistic model.

This paper proposes and tests a method of producing macrofauna habitat potential maps based on a weights-of-evidence model (a probabilistic approach) for the Hwangdo tidal flat, Korea. Samples of macrobenthos were collected during field work, and we considered five mollusca species for habitat mapping. A weights-of-evidence model was used to calculate the relative weights of 10 control factors that affect the macrobenthos habitat. The control factors were compiled as a spatial database from remotely sensed data combined with GIS analysis. The relative weight of each factor was integrated as a species potential index (SPI), which produced habitat potential maps. The maps were compared with the surveyed habitat locations, revealing a strong correlation between the potential maps and species locations. The combination of a GIS-based weights-of-evidence model and remote sensing techniques is an effective method in determining areas of macrobenthos habitat potential in a tidal flat setting.