3.1 Materials Used
Below are the materials used the project.
- Quantum GIS (QGIS)
- Sentinel Application Toolbox (SNAP) and SNAPHU
- Google Earth pro
3.1.1 Data Set and Source
The following datasets were used during the course of this study.
- The Nigerian shape file was obtained from the GADM database (www.gadm.org), version 2.5, July 2015.
- Satellite image of the study area obtained from Google Earth Pro
- SAR data from the Copernicus ESA
- Precise orbit determination (POD)
Table 3.1: Data Set and Source
S/N
|
DATA TYPE
|
DATA NAME
|
DATA DATE
|
DATA SOURCE
|
1
|
Secondary data
|
Google earth imagery
|
2021
|
Google earth pro
|
2
3
|
Secondary data
Secondary data
|
Sentinel 1 (SLC)
Precise orbit data (POD)
|
2019
2019
|
ESA
ESA
|
Table 3.2: Data Set Characteristics
File name
|
Type
|
Acquisition
|
Track
|
Orbit
|
S1A_IW_SLC_1SDV_20190923T1745…
|
SLC
|
23Sep2019
|
30
|
29152
|
S1A_IW_SLC_1SDV_20190514T1745…
|
SLC
|
14May2019
|
30
|
27227
|
S1A_IW_SLC_1SDV_20190526T1745…
|
SLC
|
26May2019
|
30
|
27402
|
S1A_IW_SLC_1SDV_20190502T1745…
|
SLC
|
02May2019
|
30
|
27052
|
3.2 Methods
This section presents the methodology adopted in this study. Figure 3.1 depicts the flow of the methodology. It is subjected to the following stages:
3.2.3 Data analysis (processing steps)
I. Processing with the Sentinel Application Tool Box (SNAP)
The processing was carried out with a sentinel application tool box (SNAP), an open-source software package developed by the European Space Agency (ESA), in the following steps:
Co-registration
Co-registration of SAR images is a necessary and critical component of any interferometric processing chain for accurate estimation of phase differences through coarse fine resampling, which estimates the range and azimuth shift offset from multiple pairs and applications such as DEM generation and interferometric deformation analysis. The S-1 TOPS Split was applied to select the AOI for the analysis. Currently, each sub-swath is processed separately using image statistics to align both products at sub-pixel accuracy. To exploit the phase difference of the acquisitions, a stack containing both products was created. The test was performed individually and successively.
Orbit auxiliary data containing the satellite position information acquisition were applied. The precise orbit determination (POD) service for Sentinel-1 provides precise orbit ephemerides (POEs). The POE files cover 28 hours with orbit state vectors at intervals of 10 seconds. They are generated one per day and are delivered within 20 days after data acquisition (Braun, 2021b).
a) Interferogram Formation and Coherence Estimation
An interferogram is formed by cross multiplying the complex conjugate of the reference image with the complex conjugate of the secondary image. The amplitude of both images is multiplied, while the phase represents the phase difference between the two images. The interferometric phase of each SAR image pixel depends only on the difference in the travel paths from each of the two SARs to the considered resolution cell. Accordingly, the computed interferogram contains phase variation.
b) Phase filtering
This process was performed to reduce noise to aid phase unwrapping. Goldstein filtering is adopted, and the interferometric fringes become smoother after filtering the peak in the spectrum (caused by the fringes), given a higher relative weight. Phase filtering greatly reduces the residue, and the phase unwrapping step enhances the phase unwrapping accuracy.
c) Phase Unwrapping
The ambiguity of phase measured modulo 2π faced by interferometric techniques for every pixel P limits access to the main value which is denoted by To determine the relief or movement field, it will be necessary to return to the exact phase value denoted by Φ(P). Such a stage is called phase unwrapping and consists of finding the right multiple k of 2π (Ferretti et al., 2007, Richards et al., 2011 and Melvin & Scheer, 2013):
In two dimensions, the measure modulo 2π turns a continuous model into a network of fringes, the edges of which (brutal transitions from 0 to 2π) depend only on the origin of phases on the complex circle. In the range direction, each line in all sub-swaths with the same time tag merges adjacent sub-swaths; for the overlapping region in the range, merging is performed midway between sub-swaths, creating a wrapped interferogram stack, using the SNAPHU package to prepare the data for unwrapping. The unwrapping process was implemented outside of the SNAP using the command terminal.
d) Range Doppler Terrain Correction
Due to the topographic variation in the scene and the tilt of the satellite sensor, the distances are distorted in the SAR IMAGE dataset. Image data not directly at the sensor nadir location will have some distortion. Terrain corrections are intended to compensate for this distortion so that the geometric representation of the image will be as close as possible to the real world (Laur et al., 2004).
After terrain correction, the generated DEM was reprojected to WGS 1984 zone 32N and was saved in Geotiff for further analysis. The DEM was also exported to Google Earth Pro to validate the exact extent of coverage and area of interest (AOI).
3.2.4 Maps Production using QGIS
a) Contour map
Contours do not extend beyond the spatial extent of the raster, and they are not generated in areas with no data. The contour type is used to produce either contour lines or polygons. The DEMs generated in this study have raster with values between 65.3 and 524.69, and the contour interval is set to 100m. Value output feature classes were created and represented with different color regions.
The generated DEM was used to create a contour map. A raster dataset was generated, and the x,y coordinate sets were computed using a raster computer and then exported to Excel. Afterwards, the surface (contour) was generated using the elevation data as the input file.
b) Slope map
The slope range of values in the output depends on the type of measurement unit. In this study, the metric measurement is used for degrees, and the range of slope values is 0 to 90; for percent rise, the range is 0 to essentially infinity. A flat surface is 0 percent, a 45-degree surface is 100 percent, and as the surface becomes more vertical, the percent rise becomes increasingly larger. The input raster needs to be resampled, and the bilinear technique was adopted with the addition of DEM data.
c) Flow direction
The generation of flow direction, which is quite similar to that of the formal (slope) DEM file, was used as the row data. The above steps constitute the conventional way of doing so, but in this study, the dataset was resampled to reduce the resolution and increase the cell size for proper visualization of each pixel. The resampled dataset was reclassified to properly arrange or change the value in the raster, and then the point to raster operation was carried out to identify which observer points were visible from each raster surface location. The basic component needed in this subset is the high-resolution DEM generated.