Soft Computing Approach for Liquefaction Identification Using LANDSAT-7 Temporal Indices Data (original) (raw)
Related papers
Liquefaction Identification Using IRS-1D Temporal Indices Data
Journal of the Indian Society of Remote Sensing, 2013
was the most deadly in India's recorded history. Widespread appearance of soil liquefaction in the Rann of Kachchh and the coastal areas of Kandla port covering an area of tens of thousands of kilometers. Remote sensing products allow us to explore the land surface parameters at different spatial scales. This work is an attempt to document and identify the impact of using conventional band ratio indices from IRS-1D temporal images for liquefaction extraction was empirically investigated and compared with Class Based Sensor Independent (CBSI) spectral NDVI band ratio while applying possibilistic fuzzy classification as soft computing approach via supervised classification. It is found that CBSI temporal indices data approach was good for extraction liquefaction as well as water bodies.
FINDING LIQUEFACTION FEATURES BY USING SATELLITE DATA FOR NORTH-EAST INDIA REGION
Pre and post-earthquake(EQ) satellite images of area near the source of the 2016 Manipur EQ are used for finding the probable liquefaction. Liquefaction-induced surface effects are identified by measuring the increase in moisture content of the area with respect to the low moisture content of the surrounding area. In this study, satellite image/ data, LANDSAT 7 is acquired for EQ affected area before and after the EQ. These images contain data of different bands information in terms of Digital Numbers (DN). The DN is first converted to Radiance. The value of the radiance is affected by the atmospheric condition of the area. So, atmospheric correction is carried out by using the FLAASH module in ENVI software. The thermal infrared spectrum of the satellite data is used for calculating the radiant temperature. With the help of Planck's radiation equation, spectral radiance is converted to the temperature. As soil surface temperature decreases with increasing soil water content, it can be correlated with soil moisture content. After finding the soil moisture content, the pre and/or post-earthquake(EQ) satellite are compared to find out the changes. On the basis of the changes and soil moisture contain difference between nearby area the probable post EQ liquefaction features positions are found out for the region.
Remote Sensing, 2020
Using automated supervised methods with satellite and aerial imageries for liquefaction mapping is a promising step in providing detailed and region-scale maps of liquefaction extent immediately after an earthquake. The accuracy of these methods depends on the quantity and quality of training samples and the number of available spectral bands. Digitizing a large number of high-quality training samples from an event may not be feasible in the desired timeframe for rapid response as the training pixels for each class should be typical and accurately represent the spectral diversity of that specific class. To perform automated classification for liquefaction detection, we need to understand how to build the optimal and accurate training dataset. Using multispectral optical imagery from the 22 February, 2011 Christchurch earthquake, we investigate the effects of quantity of high-quality training pixel samples as well as the number of spectral bands on the performance of a pixel-based parametric supervised maximum likelihood classifier for liquefaction detection. We find that the liquefaction surface effects are bimodal in terms of spectral signature and therefore, should be classified as either wet liquefaction or dry liquefaction. This is due to the difference in water content between these two modes. Using 5-fold cross-validation method, we evaluate performance of the classifier on datasets with different pixel sizes of 50, 100, 500, 2000, and 4000. Also, the effect of adding spectral information was investigated by adding once only the near infrared (NIR) band to the visible red, green, and blue (RGB) bands and the other time using all available 8 spectral bands of the World-View 2 satellite imagery. We find that the classifier has high accuracies (75%-95%) when using the 2000 pixels-size dataset that includes the RGB+NIR spectral bands and therefore, increasing to 4000 pixels-size dataset and/or eight spectral bands may not be worth the required time and cost. We also investigate accuracies of the classifier when using aerial imagery with same number of training pixels and either RGB or RGB+NIR bands and find that the classifier accuracies are higher when using satellite imagery with same number of training pixels and spectral information. The classifier identifies dry liquefaction with higher user accuracy than wet liquefaction across all evaluated scenarios. To improve classification performance for wet liquefaction detection, we also investigate adding geospatial information of building footprints to improve classification performance. We find that using a building footprint mask to remove them from the classification process, increases wet liquefaction user accuracy by roughly 10%.
Inventory of Liquefaction Area and Risk Assessment Region Using Remote Sensing
This proposed paper is focused on the identification of liquefaction areas for the communal protection and suggesting the suitable build up region to improve the inventory of areas .The waterlogged sediments get loose up from the strong vibration of the earthquake causing liquefaction, so identifying the more vulnerable areas which become the source for the earthquake-related secondary effects, such as landslides, mud flow, ground subsidence and effects on human infrastructure should be considered gravely. The conventional methods used in analysis of liquefaction factor may be time consuming and really expensive, but the wide range of modern satellite imagery can easily be adopted for communal to access the bare earth and features, in the same advance used in this project for spotting the liquefaction areas which may cause various disaster/Land transform in future. Geographic Information Systems (GIS) and Remote Sensing methods along with the associated geodatabases can be assisted by local and national authorities to be better prepared and organized in providing infrastructure to the public. The assessment of satellite imageries, digital topographic data and Geo-data contribute to the attainment of the exact geologic and geomorphologic situation influencing the local site circumstances in an area and estimate all the probable damages that could happen. The main goal of this research is delineating the region which mainly corresponds to high liquefaction potential through the various Images processing technique and GIS analysis, using satellite imagery such as Landsat 7 ETM+ sensor and advanced space borne Thermal Emission and Reflection Radiometer (ASTER), collectively with different indices calculation, ground water table, digital elevation model, geomorphology and geological studies.
Geotechnical and Geological Engineering, 2006
The Bhuj earthquake (Mw=7.9) occurred in the western part of India on 26th January 2001 and resulted in the loss of 20,000 lives and caused extensive damage to property. Soil liquefaction related ground failures such as lateral spreading caused significant damage to bridges, dams and other civil engineering structures in entire Kachchh peninsula. The Bhuj area is a part of large sedimentary basin filled with Jurassic, Tertiary and Quaternary deposits. This work pertains to mapping the areas that showed sudden increase in soil moisture after the seismic event, using remote sensing technique. Multi-spectral, spatial and temporal data sets from Indian Remote Sensing Satellite are used to derive the Liquefaction Sensitivity Index (LSeI). The basic concept behind LSeI is that the near infrared and shortwave infrared regions of electromagnetic spectrum are highly absorbed by soil moisture. Thus, the LSeI is herein used to identify the areas with increase in soil moisture after the seismic event. The LSeI map of Bhuj is then correlated with field-based observation on Cyclic Stress Ratio (CSR) and Cyclic Resistance Ratio (CRR), depth to water table, soil density and Liquefaction Severity Index (LSI). The derived LSeI values are in agreement with liquefaction susceptible criteria and observed LSI (R 2 =0.97). The results of the study indicate that the LSeI after calibration with LSI can be used as a quick tool to map the liquefied areas. On the basis of LSeI, LSI, CRR, CSR and saturation, the unconsolidated sediments of the Bhuj area are classified into three susceptibility classes.
Documenting Earthquake-Induced Liquefaction Using Satellite Remote Sensing Image Transformations
Environmental & Engineering Geoscience, 2013
Documenting earthquake-induced liquefaction effects is important to validate empirical liquefaction susceptibility models and to enhance our understanding of the liquefaction process. Currently, after an earthquake, field-based mapping of liquefaction can be sporadic and limited due to inaccessibility and lack of resources. Alternatively, researchers have used change detection with remotely sensed pre-and postearthquake satellite images to map earthquake-induced effects. We hypothesize that as liquefaction occurs in saturated granular soils due to an increase in pore pressure, liquefaction-induced surface changes should be associated with increased moisture, and spectral bands/transformations that are sensitive to soil moisture can be used to identify these areas. We verify our hypothesis using change detection with pre-and postearthquake thermal and tasseled cap wetness images derived from available Landsat 7 Enhanced Thematic Mapper Plus (ETM +) for the 2001 Bhuj earthquake in India. The tasseled cap wetness image is directly related to the soil moisture content, whereas the thermal image is inversely related to it. The change detection of the tasseled cap transform wetness image helped to delineate earthquake-induced liquefaction areas that corroborated well with previous studies. The extent of liquefaction varied within and between geomorphological units, which we believe can be attributed to differences in the soil moisture retention capacity within and between the geomorphological units.
ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2018
While extracting land cover from remote sensing images, each pixel in the image is allocated to one of the possible class. In reality different land covers within a pixel can be found due to continuum of variation in landscape and intrinsic mixed nature of most classes. Mixed pixels may not be appropriately processed by traditional image classifiers, which assume that pixels are pure. The existence of mixed pixels led to the development of several approaches for soft (often termed fuzzy in the remote sensing literature) classification in which each pixel is allocated to all classes in varying proportions. However, while the proportions of each land cover within each pixel may be predicted, the spatial location of each land cover within each pixel is not. Thus, it is important to develop and implement a classifier that can work as soft classifiers for landslide identification. This work is an attempt to document and identify landslide areas by five spectral indices using temporal multi-spectral images from IRS-P6 LISS-IV images. To improve the spectral properties of spectral indices for specific class identification (in this case landslide) a Class Based Sensor Independent (CBSI) technique proposed. The result indicates that CBSI based Transformed Normalized Difference Vegetation Index (TNDVI) temporal indices data gives better results for landslide identification with minimum entropy and membership range.
Computational Geosciences, 2008
This study pertains to prediction of liquefaction susceptibility of unconsolidated sediments using artificial neural network (ANN) as a prediction model. The backpropagation neural network was trained, tested, and validated with 23 datasets comprising parameters such as cyclic resistance ratio (CRR), cyclic stress ratio (CSR), liquefaction severity index (LSI), and liquefaction sensitivity index (LSeI). The network was also trained to predict the CRR values from LSI, LSeI, and CSR values. The predicted results were comparable with the field data on CRR and liquefaction severity. Thus, this study indicates the potentiality of the ANN technique in mapping the liquefaction susceptibility of the area.
Evaluation of Liquefaction Potential for Large Areas Based on Geomorphologic Classification
Earthquake Spectra, 2015
Ground motion maps and observation records of liquefaction sites from ten historical earthquakes are used to develop predictive equations for the regional occurrence of liquefaction. Liquefaction occurrence ratio is determined for different geomorphological conditions and intervals of causative shaking intensity obtained from the observation data. Probability regression analysis of these data, based on a cumulative normal distribution, is then used to develop equations for estimating probability of liquefaction for different geomorphological conditions given shaking intensity. Utility of the model is demonstrated for a hypothetical Tonankai-Nankai earthquake to create an estimated liquefaction potential map having 250-m grid-cells. The approach shows promise for rapid online generation of liquefaction maps following an earthquake.