The present state of the art technologies for flood mapping are typically tested on small geographical regions due to limitation of resources, which hinders the implementation of real-time flood management activities. We proposed a unified framework (GEE4FLOOD) for rapid flood mapping in Google Earth Engine (GEE) cloud platform. With the unexpected spells of extreme rainfall in August 2018, many parts of Kerala state in India experienced a major disastrous flood. Therefore, we tested the GEE4FLOOD processing chain on August 2018 Kerala flood event. GEE4FLOOD utilizes multitemporal Sentinel-1 synthetic aperture radar images available in GEE catalog and an automatic Otsu’s thresholding algorithm for flood mapping. It also utilizes other remote sensing datasets available in GEE catalog for permanent water body mask creation and result validation. The ground truth data collected during the Kerala flood indicates promising accuracy with 82% overall accuracy and 78.5% accuracy for flood class alone. In addition, the entire process from data fetching to flood map generation at a varying geographical extent (district to state level) took ∼2 to 4 min.
Retreat of glaciers is an important phenomenon to be monitored as they have a direct bearing on flow of water in rivers and rise of water level in sea. Use of Synthetic Aperture Radar (SAR) images in glacier dynamics studies has been gaining interest in the recent years. The present study discusses the use of SAR coherence images for demarking the snout position on a glacier and thus measure the retreat. Being sensitive for even the slightest changes over the terrain, SAR coherence images seems to be very useful in glacier retreat measurement.
Gangotri is one of the major Himalayan glaciers which has been subjected to dominant retreat since 1850 AD.16 The retreat of Gangotri glacier has a huge impact on the flow of Ganges, the largest perennial river in India. Coherence images were generated over Gangotri glacier from SAR images with different repeat periods from 1996 (ERS-1 & 2), 2004 (Envisat) & 2012 (TerraSAR-X) and are resampled to 50x50 m grid using SRTM DEM. Profiles near the snout position were precisely marked in a GIS environment and the distance between the profiles (1996, 2004, 2012) is reported as retreat. It has been observed that the Gangotri glacier has been retreating at the rate of 24+/- 1 m per year which is in good agreement with several other studies.
This paper discusses the methodology of Synthetic Aperture Radar (SAR) data analysis for studying various aspects of
snow characteristics viz snow dielectric constant, snow wetness and snow density. ENVISAT- Advanced Synthetic
Aperture Radar (ASAR), single look complex (SLC) data have been processed for backscattering coefficient image
generation. ASAR Backscattering coefficient images have been calibrated and processed into terrain corrected images.
Corrected backscattering images are despeckled using Frost filter technique. The estimation of snow pack characteristics
is optimal at different incidence angles. The relation between snow characteristics like wetness, and snow density and
radar backscatter has been studied and the importance of radar backscatter to infer various snow characteristics has been
emphasized. This investigation shows the backscattering coefficient is inversely correlated to snow wetness and density.
The correlation between the backscattering coefficients and snow wetness and snow density were observed as 0.8 and
0.92 respectively. 14.74 % and 13.31% part of the study area was found affected by layover and low or grazing local
incidence respectively in ENVISAT-ASAR IS6 image. In this study, the wetness range was found to vary from
0.05% to 10.28% by volume and mean absolute error was found to be 0.64% by volume and snow density range varies
from < 0.1 to 0.48 gm/cc and mean absolute error for density was found 0.032 gm/cc. At higher elevation to moderate
elevation estimated snow wetness was observed to be 0.05 - 4% by volume, increasing to 4-10.28 % by volume at
moderate to lower elevation.
The main objective of the study is to estimate snow wetness using ENVISAT ASAR data. Snow surface backscattering
can be expressed as a function of permittivity of snow. Coding has been done for backscattering coefficient image
generation using ENVISAT- Advanced Synthetic Aperture Radar (ASAR), single look complex (SLC) data with dual
(HH and VV) polarization as well as single (HH) polarization data. Incidence angle images were extracted from the
ASAR header data using interpolation method. These mages were multi-looked 5 times in azimuth and 1 time in range
direction. ASAR backscattering coefficient images have been calibrated and processed into terrain corrected images in
Universal Transverse Mercator (UTM), zone 43 north and WGS-84 datum map projection using ERDAS Imagine
software. Corrected backscattering images are despeckled using Frost filter technique. For this study Integral equation
method (IEM) for first order surface scattering based inversion model has been used. A Software has been developed
using integral equation method (IEM) based inversion model to estimate snow permittivity, which can be further related
to estimating snow wetness. A comparison was done between inversion model estimated snow wetness and field values
of snow wetness in the study region. Comparison with field measurement showed that the correlation coefficient for
snow wetness estimated from ASAR data was observed to be 0.94 at 95% confidence interval and standard error is
observed as 1.28% by volume at 95% confidence interval. The comparison of ASAR derived snow wetness with ground
measurements shows the average absolute error at 95% confidence interval as 2.8%. The snow wetness range varies
from 0-15% by volume.
The measurement of snow parameters is important for hydrological modeling. Spatial and temporal changes in snow
grain size can help us to characterize the thermal state of snow pack and to estimate the timing and spatial distribution of
snowmelt. This paper discusses the methodology of Advanced Synthetic Aperture Radar (ASAR) data analysis for
estimating snow grain size. In this investigation, we have used ENVISAT-ASAR image mode SLC data in HH-polarization
with incidence angle range 39.1 °- 42.8 ° of 31st January 2006. Survey of India (SOI) topographical sheet
(52H3) in 1:50,000 scale is used for preparation of digital elevation model (DEM) and for the registration of satellite
data. Field data were measured synchronous with satellite pass. Envisat-advanced synthetic aperture radar single
polarized, single look complex (SLC) data have been processed for backscattering coefficient image generation.
Incidence angle image was extracted from the ASAR header data using interpolation method. These images were multi-looked
5 times in azimuth and 1 time in range direction. ASAR Backscattering coefficient images have been calibrated.
The scattering and absorption efficiencies of an ice particle are only weakly dependent on the shape of the particle. A
Snowflake, although non-spherical in shape, may be treated using the Rayleigh expression for a spherical particle of the
same mass provided the Rayleigh condition applies. This study has been done using Rayleigh scattering condition based
model. The effect of snow grain size on backscattering coefficient is studied in detail. The comparison of ASAR C-band
estimated value with field grain size measurement shows an absolute error of 0.045 mm and relative error 9.6%.
Backscattering coefficient increases as the grain size increases with elevation.
This paper discusses the methodology of Synthetic Aperture Radar (SAR) data analysis for studying snow porosity and its effect on electric property of snow. ENVISAT- Advanced Synthetic Aperture Radar (ASAR) single polarized, single look complex (SLC) data have been processed for backscattering coefficient image generation. Incidence angle image has been extracted from the ASAR header data using interpolation method. These images were multi-looked 5 times in azimuth and 1 time in range direction. ASAR Backscattering coefficient images have been calibrated. The estimation of snow porosity is optimal at different incidence angles. The effect of snow porosity on snow wetness, snow density and backscattering coefficient is studied in detail. The correlation coefficient between estimated and measured porosity is observed to be 0.88 and absolute error was 0.045.
The snow wetness in the Himalayan snow covered region is an important parameter, for the snow melt runoff modeling
and forecasting. The main objective of the study is to estimate snow wetness in parts of Himalayan snow covered
regions. Snow surface backscattering can expressed as function of permittivity of snow. Reflectivity at the air snow
interface increases greatly with wetness and volume scattering decreases abruptly. ENIVISAT-ASAR dual polarization
(HH and VV) data have been used to investigate permittivity and snow wetness in sub Himalayan region. Raw data have
been processed for backscattering coefficient (BSC) image generation for HH and VV polarization. BSC image is georeferenced
and topographically corrected using high precision digital elevation model (DEM). The BSC images are
despeckled using adaptive filter technique. For this study Physical optics Model (POM) for surface scattering based
inversion model has been used. Physical Optics Model based inversion model gives the permittivity which can be further
related for estimating snow wetness. A comparison was done between inversion model estimated snow wetness and field
values of snow wetness in the study region. Comparison with field measurement showed that the correlation coefficient
for snow wetness estimated from ASAR data was 0.8 at 95% confidence interval. The snow wetness ranges from
0-15% by volume.
AQUA AMSR-E L3 soil moisture data of 5 years (2002-06) were analyzed over Gujarat, India for flood analysis. We
have used a threshold value (Gujarat 40% and Rajasthan desert 20%) for soil moisture to estimate flood affected area.
Time series daily soil moisture data during flood period in Gujarat shows that soil moisture above threshold has been
observed during 2004-06. The estimated flood area in Gujarat was 8418, 22618, 6313 sq. km in 2004, 05 and 06
respectively. Recent 2006 flood in Barmer, Rajasthan shows 7366 sq. km flood affected area with 20% as threshold soil
moisture. The flood-area correlates well with rainfall data and surface soil moisture distribution. As the resolution of
AMSR-E (60 km) is very poor, the area estimation is not so accurate. But the soil moisture trend clearly shows variation
in flood affected area with days. The results from this method are to be compared with methods available using other
data sets or techniques. Fixing a threshold value for soil moisture of a particular test site is also very important in the
estimation of flood affected area.
Snow cover area (SCA) mapping is very important parameter for snowmelt runoff modeling and forecasting. Snow
cover information is also useful for managing transportation and avalanche forecasting. For SCA mapping, we selected
Gangotri area and ENVISAT ASAR swath-2 data sets acquired during 2003-2004. The ASAR SLC data are converted
into backscattering coefficient. The backscattering coefficient images are co-registered, multi-looked and geocoded
using freely available DORIS InSAR software package. Out of available images during 2 years, Nov. 2003 image was
taken as a reference image to form a ratio image with respect to others. Snow covered and non-snow areas are classified
using threshold value (less than -3 dB is considered as snow) based on change detection method as given by Nagler and
Rott. This value is shifted to -2dB to match with the classified snow cover area from optical data of IRS-1D. With five
pairs of ratio images we could observe seasonal change of snow. The problem with the technique is that it can not be
used with different ASAR mode data. The threshold value also depends on the location of the area.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.