SciELO - Scientific Electronic Library Online

Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Gayana (Concepción)

versión impresa ISSN 0717-652Xversión On-line ISSN 0717-6538

Gayana (Concepc.) v.68 n.2 supl.TIIProc Concepción  2004 

  Gayana 68(2): 586-589, 2004



Hong-Rhyong Yoo1, Joo-Hyung Ryu2, Il-Hwan Bae1 & Yu-Hwan Ahn2

1. Marine Geoenvironment and Resources Research Division,
2. Satellite Ocean Research Laboratory Korea Ocean Research and Development Institute, Ansan P.O. Box 29, Seoul, 425-600, Korea Tel: 031-400-7601, Fax: 031-400-6439, Email:


The southern tidal flat of Kanghwa Island with an area of approximately 90 km2 is one of the biggest flats on the western coast of Korea. Remote sensing has been identified as potential tool to provide synoptic information of the intertidal environments. To estimate quantitatively the morphological changes in a tidal flat, it is necessary to measure both the amount of sedimentation or erosion and the sediment facies involved. Thus, the objectives of this study are : (i) to generate an intertidal digital elevation model (DEM) using the waterline method around 1989; (ii) to quantitatively estimate the morphological changes from the generated DEM for the years 1989 and the extracted waterlines for the year 2002; (iii) to discriminate the surface sedimentary facies by texture analysis and classification method, using high-resolution satellite data such as IKONOS and Kompsat EOC; iv) to ultimately evaluate the morphological changes with respect to sedimentary facies. The morphological changes are dependent upon the sedimentary facies. The changes in sand or mixed flats are greater by at least three times than that in the mudflats. Although we are not able to assess the accuracy of the estimation of morphologic changes, the resulting trend of sedimentation or erosion is complied with the general rule of sedimentology. The results demonstrate that satellite remote sensing is a more efficient tool for estimating long-term morphologic changes in the tidal flats.


The west coast of the Korean Peninsula is famous for its large tidal range (up to 9 m) and vast tidal flats. Sedimentation and/or erosion on the Korean tidal flats are significant due to the high tidal energy. Land reclamation, which has occurred on a large scale such as New Airport, has also accelerated environmental changes in the tidal flat. Generally, tidal flats show dynamic morphologic changes that arise from high tidal energy and sediment transportation. The driving force of coastal changes results from sediment budget processes, tectonic processes, marine energy processes, relative sea-level movements, and human impacts (Fletcher, 2000). Among these, the sediment budget process is very important, especially to ecological systems. The sediment budget can be estimated only if one is able to measure the total morphological change that has occurred in a certain period, and if one can discriminate about the types of sediments. Sedimentologists have estimated morphologic changes by measuring the accumulation rate in the field, and by 210Pb analysis of sediment cores (Patchineelam et al, 1999; Lee et al., 1989). It is, however, very difficult to estimate morphological changes from the field observation alone, because of the limited accessibility, short exposure to air, and lack of suitable transportation. Thus, remote sensing combined with in situ surveying is an effective tool for monitoring tidal flats. With improving spectral resolution and the spatial resolution, the application of satellite remote sensing to tidal flat analysis is now increasing. In addition, it is possible to monitor rapid changes in the tidal flats, owing to the improvement in the temporal resolution of the available Earth observation satellites. Therefore, the field measurements and remote sensing techniques have become accepted as complementary tools in geomorphology (Kevin et al., 1999).


Kanghwa tidal flat is open type 7-9 km wide and 20 km long, and is located on the western coast of the Korea Peninsula, as shown in Figure 1(a). The sediment distribution in Kanghwa tidal flat can be classified into mud flat, mixed flat, and sand flat environments, according to the textural characteristics. The sand flats contain more than 75% sand, the mud flats less than 25% sand, and the mixed flats 25­75% sand (Folk, 1968). The sand content, and the mean grain size generally increases on moving towards the sea. The mud flats are located on the eastern part of the Kanghwa tidal flat, the sand flats near the western part , and the mixed flats in a broad transition zone between the mud flats and the sand flats. The grain size has an important implication to the geomorphological changes in the tidal flats: For instance, the coarser the grain size results when the tidal energy is high. Therefore, we can expect sand flats to be more vulnerable to morphologic changes than mud flats. To carry out grain size analysis, samples were collected from the surface sediment at 230 sites by van Veen Grab Sampler (Figure 1(b)). After removing carbonates and organic materials from the collected samples, the grain size was measured using a Gradex 2000 Particle Size Analyzer and Sedigraph 5100.

Figure 1 : (a) Landsat ETM+ image of Kanghwa tidal flat with a location map
(b) Sampling positions of grain size by grab
(c) A digital elevation model constructed using waterline method

For our study, we acquired a total 25 Landsat TM and ETM+ images (path/row: 116/34). Geometric rectification was conducted using 1:5,000 and 1:25,000 scale topographic maps. A horizontal accuracy of less than 0.3 pixels (corresponding to 10 m on the ground) was achieved after geometric rectification. We used Landsat TM data to extract the waterlines and high-resolution satellite data such as KOMPSAT-1, IKONOS to discriminate the surface sediment types.


The waterline method exploits the different tide conditions that are rendered in each image as a topographic contour line. The principal tactic in the waterline method is to collect as many images as possible, since the tidal conditions are different in almost all the images. The tidal flat DEM can be generated by stacking all the waterlines acquired over a given short period. The waterline method is based on three assumptions; (i) that the waterline represents an equal elevation at the moment of image acquisition; (ii) that topographic change is negligible during the period of data acquisition; and (iii) that the absolute elevation of each waterline is known. The accuracy of the resulting DEM from the waterline method largely depends on the accuracy of the waterline extracted from a given image, and of the absolute elevation assigned to the waterline (Lohani et al., 1999).

Waterline extraction

Although the core idea of the waterline method is straightforward, it is not such a simple task to the extract waterline from a tidal flat. Ryu et al. (2002) investigated the characteristics of the spectral reflectance from a tidal flat, and described in detail the waterline extraction process. The tidal flat was characterized by a low reflectance from the exposed surface, high seawater turbidity, and a variation in moisture content that was in turn governed by grain size, local slope, and the existence of tidal channels or creeks. Near infrared (NIR), short wavelength infrared (SWIR), and thermal infrared (TIR) images were equally effective for waterline extraction under flood tides, but extreme caution was required for the ebb images. Under ebb tide conditions, TIR and NIR were proven to be more effective than SWIR. The location of the waterline in a tidal flat (i.e., in the lower, middle, or upper tidal flat) is also an important factor that must be accounted for. When the waterline is in the middle tidal flat, SWIR often leads to determination of an incorrect waterline. We extracted waterlines by applying the procedures of Ryu et al. (2002).

Conversion to Absolute Elevation

The waterline method also requires a reference elevation of the extracted waterline. Tidal level data from a tide gauge or hydrodynamic tide model can be used for this purpose (Mason et al., 1997; Chen and Rau, 1998; Klocke et al., 1999). Because of the relatively large tidal range and high velocity, a general hydrodynamic tide model could not be used in this study. It may be more appropriate to use tide gauge data than data from the hydrodynamic tide model, but data from the local hydrodynamic tide model would still be required if the tide gauge was deployed far away from the study area.

Results and Discussion

We selected seven images that were suitable for generating an intertidal DEM, after reviewing 11 images, taken over a time interval of 16 months from February 2002 to May 2003. The intertidal DEM'2002 was then generated using the minimum curvature interpolation technique. Since neither the sedimentation nor the erosion in an upper tidal flat is usually significant, most scientists are interested in lower tidal flats. Mason et al. (1997) pointed out that the expected DEM accuracy from the waterline method would be 14 cm at best. Therefore, our result is extremely pleasing, and can be applied to the estimation of the topographic changes. The generated DEMs themselves provide useful information for intertidal morphologic analysis. In the inner tidal flat, intertidal channels and creeks are well developed on mud flats. In contrast, the surface topography is relatively smooth in the outer tidal flat, where the sand and/or mixed flats are dominant. A similar process was carried out on 6 TM images selected out of 12 available images, taken over a time interval of 46 months between January 1989 and December 1991 (Figure 1(c)).

Quantitative morphologic changes can be estimated from the differences of the two DEMs. In this estimation process, the seaward area below -200 cm in elevation was excluded, because there was no reference elevation or waterlines. The landward area above +150 cm in elevation was also excluded, because it was too close to the landmass, and so no waterline was extracted. The vicinities of the big stream and other tidal channels were also excluded in the estimation, to avoid the effect of inter-tidal channel migration. The sedimentation/erosion rate represents an average elevation change over the entire tidal flats over a thirteen-year period. The annual mean morphologic change was obtained by summing the volumetric change per year in each grid for all the regions. This result shows that there is a slow accumulation of sediment in the inner tidal flat area, while erosion dominates the outer tidal flat area. It implies that the high tidal energy in the outer tidal flat contributes to erosion. Meanwhile, the low tidal energy in the inner tidal flat results in the slow sedimentation of suspended fine-grained materials. This sediment budget process is generally accepted by sedimentologists, but no attempt has ever been made to estimate it three-dimensionally. The changes in the sand or mixed flats (i.e., the sand or silty sand) are larger by at least three times than the mud flats (sandy silt) values. The results strongly support sedimentologic common sense: that sand flats are more vulnerable to morphologic change than mud flats (Ryu et al., 2004). Contemporary analysis of morphologic changes in tidal flats has depended solely upon in situ sedimentation rate measurements taken at sporadic points. Because we do not have sedimentation rate data, it was impossible to quantitatively evaluate the morphologic changes using this method in our study. The results demonstrate that satellite remote sensing and the waterline method are very useful for evaluating the long-term morphologic changes in tidal flats.



Chen, L.C., Rau, J.Y., 1998. Detection of shoreline changes for tideland areas using multi-temporal satellite images. International Journal of Remote Sensing, 19(17), 3383-3397.         [ Links ] [1]

Fletcher C., 2000. National Science Foundation Geology and Paleontology Program Research in the Sedimentary Geology of the Coastal Zone and Inner Shelf. Coastal Change. URL: _Conf/PDF/NSF.PDF.         [ Links ] [2]

Folk R.L., 1968. A review of grain size parameters. Sedimentology, 6, 73-93.         [ Links ] [3]

Kevin, W., Hesham, M., 1999. Monitoring changing position of coastlines using Thematic Mapper imagery, an example from the Nile Delta. Geomorphology, 29, 93-105.         [ Links ] [4]

Klocke, B., Schmidt, Klocke, B., Schmidt, R., Muller-Navarra, Ellmer, W., 1999. Topographic mapping of the German tidal flat areas using ERS-SAR images. IEEE Proceedings of International Geoscience and Remote Sensing Symposium 1999, Hamburg, Germany, 28 June-2 July, CD-ROM version.         [ Links ] [5]

Lee, H.J., Chough, S.K., 1989. Sediment distribution, dispersal and budget in the yellow sea. Marine Geology, 87, 195-205.         [ Links ] [6]

Lohani, B. Mason, D.C., 1999. Construction of a Digital Elevation Model of the Holderness Coast using the waterline method and Airborne Thematic Mapper data. International Journal of Remote Sensing, 20(3), 593-607.         [ Links ] [7]

Mason, D., Hill, D., Davenport, I., Flather, R., Robinson, G., 1997. Improving inter-tidal digital elevation models constructed by the waterline technique. Third ERS Symposium, Florence, Italy, 14-21 March, European Space Agency Publications Division, 1079-1082.         [ Links ] [9]

Patchineelam, S.M., Kjerfve, B., Gardner, L.R., 1999. A preliminary sediment budget for the Winyah Bay estuary, South Carolina, USA. Marine Geology, 162, 133-144.         [ Links ] [10]

Ryu, J.H., Won, J.S., Min, K.D., 2002. Waterline extraction from Landsat TM data in a tidal flat: A case study in Gomso Bay, Korea. Remote Sensing of Environment, 83, 442-456.         [ Links ] [11]

Ryu, J.H., Na, Y.H., Won, J.S., Doerffer, R., 2004. A critical grain size for Landsat ETM+ investigations into intertidal sediments: a case study of the Gomso tidal flats, Korea. Estuarine, Coastal and Shelf Science 60, 491-502.         [ Links ] [12]


Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons