Katsuaki Koike

Improved Accuracy of Chlorophyll-a Concentration Estimates from MODIS Imagery Using a Two-Band Ratio Algorithm and Geostatistics: As Applied to the Monitoring of Eutrophication Processes over Tien Yen Bay (Northern Vietnam)

Sea eutrophication is a natural process of water enrichment caused by increased nutrient loading that severely affects coastal ecosystems by decreasing water quality. The degree of eutrophication can be assessed by chlorophyll-a concentration. This study aims to develop a remote sensing method suitable for estimating chlorophyll-a concentrations in tropical coastal waters with abundant phytoplankton using Moderate Resolution Imaging Spectroradiometer (MODIS)/Terra imagery and to improve the spatial resolution of MODIS/Terra-based estimation from 1 km to 100 m by geostatistics. A model based on the ratio of green and blue band reflectance (rGBr) is proposed considering the bio-optical property of chlorophyll-a. Tien Yen Bay in northern Vietnam, a typical phytoplankton-rich coastal area, was selected as a case study site. The superiority of rGBr over two existing representative models, based on the blue-green band ratio and the red-near infrared band ratio, was demonstrated by a high correlation of the estimated chlorophyll-a concentrations at 40 sites with values measured in situ. Ordinary kriging was then shown to be highly capable of predicting the concentration for regions of the image covered by clouds and, thus, without sea surface data. Resultant space-time maps of concentrations over a year clarified that Tien Yen Bay is characterized by natural eutrophic waters, because the average of chlorophyll-a concentrations exceeded 10 mg/m 3 in the summer. The temporal changes of chlorophyll-a concentrations were consistent with average monthly air

Applying Bayesian Decision Classification to Pi-SAR Polarimetric Data for Detailed Extraction of the Geomorphologic and Structural Features of an Active Volcano

An understanding of the geomorphology and distribution of surface materials on an active volcano is crucial to characterize eruptions and mitigate volcanic hazards. For volcanoes, synthetic aperture radar (SAR) remote sensing is the only useful observation and monitoring technology that can be undertaken in any weather condition. This letter uses the data from one type of airborne SAR system termed polarimetric and interferometric airborne SAR and L-band microwaves to classify SAR imagery into geomorphologic units, based on a scattering mechanism, using the example of Mt. Sakurajima, a representative active volcano situated in southern Japan. This is accomplished by adopting a Bayesian decision classification (BDC) scheme applied to two polarimetric parameters, namely, entropy and the type of scattering mechanism, which are derived from Cloude-Pottier decomposition of full polarimetry. In spite of the thick vegetation cover, BDC can divide SAR imagery from Mt. Sakurajima into three geomorphologic units: volcanic cone, terrace, and foot. The suitability of the BDC classification of microwave sensor imagery-and its superiority over a traditional classification scheme, the K -means unsupervised classification-is confirmed by polarimetric signature analysis and ground-truth surveying that directly quantifies surface scattering.

Incorporation of fracture directions into 3D geostatistical methods for a rock fracture system

Simulating a rock fracture distribution is an important problem common to various fields in geosciences. This paper presents GEOFRAC, a geostatistical method to simulate a fracture distribution by incorporating the directions (strikes and dips) of the sampled fracture data into the simulation. Fracture locations are generated randomly following fracture densities assigned by a sequential Gaussian simulation. Fracture directions are transformed into an indicator set consisting of several binary (0 and 1) variables and the variables are compressed using the principal component analysis. Ordinary kriging is then employed to estimate the distributions of these principal values and the results are back-transformed into the coordinate system of the original indicator set. Fracture directions are generated randomly using their histograms within the defined directional interval. Finally, facets (fracture elements) are determined from the simulated locations and directions, and the fractures within the angle and distance tolerances are connected to form a fracture plane. From a case study of applying GEOFRAC to the fracture data in Kikuma granite, southwest Japan, GEOFRAC was shown to be able to depict a plausible fracture system because the simulated directions corresponded well to those measured. Furthermore, the simulated fracture system was available to estimate the hydraulic conductivity of the study site, which was roughly in agreement with the average of hydraulic test results.

Mapping an uncertainty zone between interpolated types of a categorical variable

Categorical data cannot be interpolated directly because they are outcomes of discrete random variables. Thus, types of categorical variables are transformed into indicator functions that can be handled by interpolation methods. Interpolated indicator values are then backtransformed to the original types of categorical variables. However, aspects such as variability and uncertainty of interpolated values of categorical data have never been considered. In this paper we show that the interpolation variance can be used to map an uncertainty zone around boundaries between types of categorical variables. Moreover, it is shown that the interpolation variance is a component of the total variance of the categorical variables, as measured by the coefficient of unalikeability.

Identifying Surface Materials on an Active Volcano by Deriving Dielectric Permittivity From Polarimetric SAR Data

Dielectric permittivity εr measured on the Earths surface is an effective property for characterizing surface materials in terms of rock type and water content, particularly in highly changeable environments such as active volcanoes. We propose a technique termed dielectric permittivity from polarimetric synthetic aperture radar (dPSAR) to quantify εr using a single scene of polarimetric SAR data, based on the small perturbation model of backscattering (SPMB). For an optimal solution, the Nelder-Mead simplex method was combined with SPMB. The application of dPSAR to a scene of ALOS PALSAR data from the vicinity of Mt. Merapi, Indonesia, correctly identified the relative value ranges of εr for pyroclastic flow and tephra deposits accompanying large eruptions that occurred on November 5, 2010; their means were 2.55 and 3.07, respectively. Pore water within porous ashes is a plausible factor for increases in the εr of the tephra.

Increased radon-222 in soil gas because of cumulative seismicity at active faults

This study demonstrates how the radon-222 (222Rn) concentration of soil gas at an active fault is sensitive to cumulative recent seismicity by examining seven active faults in western Japan. The 222Rn concentration was found to correlate well with the total earthquake energy within a 100-km radius of each fault. This phenomenon can probably be ascribed to the increase of pore pressure around the source depth of 222Rn in shallow soil caused by frequently induced strain. This increase in pore pressure can enhance the ascent velocity of 222Rn carrier gas as governed by Darcys law. Anomalous 222Rn concentrations are likely to originate from high gas velocities, rather than increased accumulations of parent nuclides. The high velocities also can yield unusual young gas under the radioactive nonequilibrium condition of short elapsed time since 222Rn generation. The results suggest that ongoing seismicity in the vicinity of an active fault can cause accumulation of strain in shallow fault soils. Therefore, the 222Rn concentration is a possible gauge for the degree of strain accumulation.

Relationship between Remotely Sensed Vegetation Change and Fracture Zones Induced by the 2008 Wenchuan Earthquake, China

The Wenchuan (汶川) earthquake triggered cascading disasters of landslides and debris flows that caused severe vegetation damage. Fracture zones can affect geodynamics and spatial pattern of vegetation damage. A segment tracing algorithm method was applied for identifying the regional fracture system through lineament extractions from a shaded digital elevation model with 25 m mesh for southern Wenchuan. Remote sensing and geographic information system techniques were used to analyze the spatiotemporal vegetation pattern. The relationship between vegetation type identified from satellite images and lineament density was used to characterize the distribution patterns of each vegetation type according to fracture zones. Broad-leaved forest, mixed forest, and farmland persist in areas with moderate lineament density. Deciduous broad-leaved and coniferous forest persists in less fractured areas. Shrub and meadow seem to be relatively evenly distributed across all lineament densities. Meadow, farmland, and shrub persist in the fractured areas. Changes of spatial structure and correlation between vegetation patterns before and after the earthquake were examined using semivariogram analysis of normalized difference vegetation indices derived from Landsat enhanced thematic mapper images. The sill values of the semivariograms show that the spatial heterogeneity of vegetation covers increased after the earthquake. Moreover, the anisotropic behaviors of the semivariograms coincide with the vegetation changes due to the strikes of fracture zones.

Applicability of computer-aided comprehensive tool (LINDA: LINeament Detection and Analysis) and shaded digital elevation model for characterizing and interpreting morphotectonic features from lineaments

Abstract Detection and analysis of linear features related to surface and subsurface structures have been deemed necessary in natural resource exploration and earth surface instability assessment. Subjectivity in choosing control parameters required in conventional methods of lineament detection may cause unreliable results. To reduce this ambiguity, we developed LINDA (LINeament Detection and Analysis), an integrated tool with graphical user interface in Visual Basic. This tool automates processes of detection and analysis of linear features from grid data of topography (digital elevation model; DEM), gravity and magnetic surfaces, as well as data from remote sensing imagery. A simple interface with five display windows forms a user-friendly interactive environment. The interface facilitates grid data shading, detection and grouping of segments, lineament analyses for calculating strike and dip and estimating fault type, and interactive viewing of lineament geometry. Density maps of the center and intersection points of linear features (segments and lineaments) are also included. A systematic analysis of test DEMs and Landsat 7 ETM+ imagery datasets in the North and South Eastern Deserts of Egypt is implemented to demonstrate the capability of LINDA and correct use of its functions. Linear features from the DEM are superior to those from the imagery in terms of frequency, but both linear features agree with location and direction of V-shaped valleys and dykes and reference fault data. Through the case studies, LINDA applicability is demonstrated to highlight dominant structural trends, which can aid understanding of geodynamic frameworks in any region.

Co-kriging for modeling shallow groundwater level changes in consideration of land use/land cover pattern

This study aimed at clarifying the relationship between the dynamics of land use/land cover (LULC) changes and decline in the groundwater levels, and specifying an LULC category strongly affecting such decline in a Quaternary sedimentary basin. Groundwater level data recorded at 26 observation wells for a 14-year period in the Kumamoto Plain, central Kyushu, southwest Japan, were used for the analysis. The general trends of LULC were detected by a satellite image classification technique and surface spline method, which highlighted the decreases in groundwater-recharge materials. As the next step, those trends of groundwater levels that were closely correlated with rainfall were removed from the level data set, and the resultant residual component levels were applied to co-kriging analysis with LULC categories. Co-kriging provided a detailed map of groundwater level variability. Furthermore, we propose a method, prediction of residual of groundwater level (PWL), to infer future residual groundwater levels from the supposed LULC pattern by co-kriging-based modeling. PWL was demonstrated to be effective because it clearly represented the decrease and increase in negative residual level areas, depending on the extent of rice fields in the past and in predicted future distribution scenarios.

Selecting the Best Band Ratio to Estimate Chlorophyll-a Concentration in a Tropical Freshwater Lake Using Sentinel 2A Images from a Case Study of Lake Ba Be (Northern Vietnam)

This study aims to develop a method to estimate chlorophyll-a concentration (Chla) in tropical freshwater lake waters using in situ data of Chla, water reflectance, and concurrent Sentinel 2A MSI imagery (S2A) over Lake Ba Be, a Ramsar site and the largest natural freshwater lake in Vietnam. Data from 30 surveyed sampling sites over the lake water in June 2016 and May 2017 demonstrated the appropriateness of S2A green-red band ratio (band 3 versus band 4) for estimating Chla. This was shown through a strong correlation of corresponded field measured reflectance ratio with Chla by an exponential curve (r2 = 0.68; the mean standard error of the estimates corresponding to 5% of the mean value of in situ Chla). The small error between in situ Chla, and estimated Chla from S2A acquired concurrently, confirmed the S2A green-red band ratio as the most suitable option for monitoring Chla in Lake Ba Be water. Resultant Chla distribution maps over time described a partially-seasonal pattern and also displayed the spatial dynamic of Chla in the lake. This allows a better understanding of the lake’s limnological processes to be developed and provides an insight into the factors that affect lake water quality. The results also confirmed the potential of S2A to be used as a free tool for lake monitoring and research due to high spatial resolution data (10 m pixel size).