Joong-Sun Won

Determination and application of the weights for landslide susceptibility mapping using an artificial neural network

The purpose of this study is the development, application, and assessment of probability and artificial neural network methods for assessing landslide susceptibility in a chosen study area. As the basic analysis tool, a Geographic Information System (GIS) was used for spatial data management and manipulation. Landslide locations and landslide-related factors such as slope, curvature, soil texture, soil drainage, effective thickness, wood type, and wood diameter were used for analyzing landslide susceptibility. A probability method was used for calculating the rating of the relative importance of each factor class to landslide occurrence. For calculating the weight of the relative importance of each factor to landslide occurrence, an artificial neural network method was developed. Using these methods, the landslide susceptibility index (LSI) was calculated using the rating and weight, and a landslide susceptibility map was produced using the index. The results of the landslide susceptibility analysis, with and without weights, were confirmed from comparison with the landslide location data. The comparison result with weighting was better than the results without weighting. The calculated weight and rating can be used to landslide susceptibility mapping.

Waterline extraction from Landsat TM data in a tidal flat. A case study in Gomso Bay, Korea

Waterline extraction is potentially one of the most effective satellite remote sensing tools for studying changes in tidal flat environment and coastlines. However, its application to the study of tidal flats has not been investigated in detail. The waterline in a tidal flat in Gomso Bay, Korea is characterized and evaluated using 27 sets of Landsat thematic mapper (TM) and Earth Observing System-Terra (EOS-Terra) Advanced Spaceborne Thermal Emission and Radiometer (ASTER) data. Ground truth data including grain size, soil moisture content, ground levelling, and waterline tracking have been obtained. The choice of spectral bands in terms of tidal conditions turned out to play a key role in locating the waterline. While the waterlines extracted from near-infrared (NIR), short-wavelength infrared (SWIR), and thermal infrared (TIR) data are commonly reliable on the flood tide, the discrepancies between different bands become large on the ebb tide. The bands TIR, NIR, and SWIR are, in that order, decreasingly effective in extracting the waterline while ebb tides are in progress. The effect of turbid water on NIR can be reduced by a ratioing of NIR and the visible band. The location of the waterline is also an important factor to consider; the discrepancy is largest on the middle tidal flat. Selection of the proper band for waterline extraction under various conditions is suggested. However, no single band can accommodate all aspects of the tidal flat environment. A schematic model associated with the spectral reflectance of the tidal flat is proposed in which the remnant water scattered on the surface after exposure is emphasized as an additional factor.

Mapping Three-Dimensional Surface Deformation by Combining Multiple-Aperture Interferometry and Conventional Interferometry: Application to the June 2007 Eruption of Kilauea Volcano, Hawaii

Surface deformation caused by an intrusion and small eruption during June 17-19, 2007, along the East Rift Zone of Kilauea Volcano, Hawaii, was three-dimensionally reconstructed from radar interferograms acquired by the Advanced Land Observing Satellite (ALOS) phased-array type L-band synthetic aperture radar (SAR) (PALSAR) instrument. To retrieve the 3-D surface deformation, a method that combines multiple-aperture interferometry (MAI) and conventional interferometric SAR (InSAR) techniques was applied to one ascending and one descending ALOS PALSAR interferometric pair. The maximum displacements as a result of the intrusion and eruption are about 0.8, 2, and 0.7 m in the east, north, and up components, respectively. The radar-measured 3-D surface deformation agrees with GPS data from 24 sites on the volcano, and the root-mean-square errors in the east, north, and up components of the displacement are 1.6, 3.6, and 2.1 cm, respectively. Since a horizontal deformation of more than 1 m was dominantly in the north-northwest-south-southeast direction, a significant improvement of the north-south component measurement was achieved by the inclusion of MAI measurements that can reach a standard deviation of 3.6 cm. A 3-D deformation reconstruction through the combination of conventional InSAR and MAI will allow for better modeling, and hence, a more comprehensive understanding, of the source geometry associated with volcanic, seismic, and other processes that are manifested by surface deformation.

Ionospheric Correction of SAR Interferograms by Multiple-Aperture Interferometry

Interferometric synthetic aperture radar (InSAR) is a powerful technique that precisely measures surface deformations at a fine spatial resolution over a large area. However, the accuracy of this technique is sometimes compromised by ionospheric path delays on radar signals, particularly with L- and P-band SAR systems. To avoid ionospheric effects from being misinterpreted as ground displacement, it is necessary to detect and correct their contributions to interferograms. In this paper, we propose an efficient method for ionospheric measurement and correction and validate its theoretical and experimental performance. The proposed method exploits the linear relationship between the multiple-aperture interferometry phase and the azimuth derivative of the ionospheric phase. Theoretical analysis shows that a total electron content (TEC) accuracy of less than

Ensemble-Based Landslide Susceptibility Maps in Jinbu Area, Korea

1.0 \times 10^{-4}

Polarimetric Features of Oyster Farm Observed by AIRSAR and JERS-1

TEC units can be achieved when more than 100 neighboring samples can be averaged (multilooked), and the coherence is 0.5. The regression analysis between the interferometric phase and the topographic height shows that the root-mean-square error can be improved by a factor of two after ionospheric correction. A 2-D Fourier spectral analysis indicates that the ionospheric wave pattern in the uncorrected power spectrum has disappeared in the power spectrum of the corrected interferogram. These results demonstrate that the proposed method can effectively remove ionospheric artifacts from an ionosphere-distorted InSAR image. Note that the method assumes that there is no appreciable surface displacement in the along-track dimension of the interferogram.

An application of L-band synthetic aperture radar to tide height measurement

Ensemble techniques were developed, applied and validated for the analysis of landslide susceptibility in Jinbu area, Korea using the geographic information system (GIS). Landslide-occurrence areas were detected in the study by interpreting aerial photographs and field survey data. Landslide locations were randomly selected in a 70/30 ratio for training and validation of the models, respectively. Topography, geology, soil and forest databases were also constructed. Maps relevant to landslide occurrence were assembled in a spatial database. Using the constructed spatial database, 17 landslide-related factors were extracted. The relationships between the detected landslide locations and the factors were identified and quantified by frequency ratio, weight of evidence, logistic regression and artificial neural network models and their ensemble models. The relationships were used as factor ratings in the overlay analysis to create landslide susceptibility indexes and maps. Then, the four landslide susceptibility maps were used as new input factors and integrated using the frequency ratio, weight of evidence, logistic regression and artificial neural network models as ensemble methods to make better susceptibility maps. All of the susceptibility maps were validated by comparison with known landslide locations that were not used directly in the analysis. As the result, the ensemble-based landslide susceptibility map that used the new landslide-related input factor maps showed better accuracy (87.11% in frequency ratio, 83.14% in weight of evidence, 87.79% in logistic regression and 84.54% in artificial neural network) than the individual landslide susceptibility maps (84.94% in frequency ratio, 82.82% in weight of evidence, 87.72% in logistic regression and 81.44% in artificial neural network). All accuracy assessments showed overall satisfactory agreement of more than 80%. The ensemble model was found to be more effective in terms of prediction accuracy than the individual model.

Interferometric Coherence Analysis of the Everglades Wetlands, South Florida

The polarimetric features of an oyster farm in a coastal area are analyzed to verify the applicability of radar polarimetry and interferometry. L-band Airborne Synthetic Aperture Radar (AIRSAR) data and Japan Earth Resources Satellite (JERS-1) data are used to examine the unique structure of an oyster farm located in South Korea. A specific feature of the oyster farm is the presence of numerous arrays of structures of various orientations that consist of exercise-bar-shaped poles protruding above sea level. This paper demonstrates that tide level is strongly correlated with the double-bounce scattering power from the vertical pole structures. This phenomenon is also verified by laboratory measurements using a network analyzer. In the laboratory experiment, double-bounce scattering and total power showed increasing trends with increased height of the vertical poles. Single-bounce scattering is sensitive to the orientation of horizontal poles relative to antenna orientation. HH-polarization is the most effective technique for imaging oyster farms from L-band polarimetric AIRSAR data. The authors were able to use a three-component decomposition of the AIRSAR data to distinguish an exposed tidal flat from a submerged tidal flat. The characteristics of the exposed tidal flat are similar to those of the carbon sponge in the laboratory test, except that the double-bounce scattering power is slightly greater in the real-world example. The single-bounce scattering component in AIRSAR data is generally greater than that in laboratory measurements because of sea-surface conditions and oyster growth. When the horizontal pole was aligned normal to the radar look direction, single-bounce scattering was greater than the double-bounce scattering, even under water-covered conditions. While a difference in tide height of 10 cm contributed approximately 3.0 dB in the laboratory experiment, a difference in tide height of 20 cm contributed to only approximately 1.7 dB in the JERS-1 SAR image intensity. JERS-1 SAR image intensity for areas dominated by double- and single-bounce scattering was 0.78 and 0.56, respectively. Results confirm that polarimetric SAR data are useful in selecting areas dominated by double-bounce scattering in oyster farms

A Study of Decadal Sedimentation Trend Changes by Waterline Comparisons within the Ganghwa Tidal Flats Initiated by Human Activities

A method for measuring the tide height near the coast from L-band synthetic aperture radar (SAR) data is presented. Twenty-one coherent interferograms have been successfully constructed from Japanese Earth Resources Satellite 1 (JERS-1) SAR data obtained over oyster sea-farming structures. A coherence analysis of the 21 interferometric pairs showed that a perpendicular baseline of less than 3 km, with a temporal baseline within 500 days, are required to obtain a coherent pair, with a coherence higher than 0.25, in the study area. The coherent phases preserved in the interferograms showed a close relation with the sea level. The problem of phase unwrapping to restore an absolute tide height was overcome by introducing normalized image intensities. The radar measurements estimated by the proposed method were verified using tide gauge data, and comparison of the two datasets yielded a correlation coefficient R/sup 2/ of 0.91, with a root mean square error of 5.76 cm. The results demonstrate that radar interferometry can be applied for a tide height measurement near the coast given sufficient structures that return off-nadir radar pulses to the antenna. The multipolarized L-band SAR system will provide better results, using only double-bounced signals, in the future.

Detection and Restoration of Defective Lines in the SPOT 4 SWIR Band

Interferometric synthetic aperture radar (InSAR) observations of wetlands reveal spatially detailed measurements of water-level changes and quantitative images of flow dynamics. However, lateral variability of wetland vegetation results in a heterogeneous scattering medium, which can affect interferometric coherence levels and can even limit the applicability of the technique. Here, we analyze coherence variations in Southern Florida, which consist of various wetland vegetation types, including sawgrass, graminoid, cypress, mixed shrubs, and mangrove marsh. We use JERS-1, ERS-1/2, ENVISAT, and RADARSAT-1 data, to investigate the effect of acquisition parameters and temporal baseline (time span between acquisitions) on the coherence level in the various wetland vegetation environments. The main findings of our coherence analysis are as follows: (1) Woody wetlands, such as cypress and mixed shrubs swamps, have higher coherence levels than herbaceous wetlands of sawgrass and graminoid (cattail) in all SAR data types; (2) the coherence level of C-band data is strongly dependent on temporal baseline, whereas the coherence level of L-band data depends mainly on perpendicular baseline, but to some degree also on temporal baseline; (3) backscatter from JERS-1 and RADARSAT-1 is correlated with coherence in four wetland vegetation types (sawgrass, cypress, mixed shrubs, and mangrove), but ERS backscatter has no relation to coherence, except over sawgrass marsh. Finally, our study clearly indicates that high resolution, HH polarization, and small incidence angle observations are most suitable for wetland InSAR applications.