Applying inSAR principles to detect surface deformation in Metro Manila

I have just attended an InSAR training  showing how to detect surface deformation through phase differences between SAR signals.

SAR satellite transmits microwave to earth and receives reflected microwave signals from earth surface. The intensity of received microwave indicates electromagnetic characteristics of objects, and phase indicates distance between aperture and object. Surface deformation is measured by the difference in two observed phases. The observed phase differences still contains errors that ought to be corrected along processing enumerated by the steps below.

  1. Getting SLC (Single-Look Complex)
  2. Co-registration (registration of Master and Slave images)
  3. Interferometry ( measuring phase differences)
  4. Corrections (Removal of Orbital and Topographic Fringe)
  5. Phase Unwrapping (integration of extracted to actual phase difference)
  6. Geocoding
  7. Visualization of Results in ENVI
  8. Analysis and Interpretation
With the use of ALOS Palsar images from two different instances in time, the InSAR processing flow is initiated in a LINUX environment with the use of SIGMA-SAR software. 

After a real image and an interferometric image is produced, these image files are then viewed, further processed and analysed in ENVI. 

Our study area is Metro Manila, Philippines. Two ALOS Palsar images were used (2011 and 2007 images).

The following images were produced. By linking displays, the suspected are of deformation is along the area enclosed by the ellipse.



This is a close-up view of the interferometric image and the deformation (subsidence) is approximately 9.2cm along Caloocan City, Metro Manila.

The issue on land subsidence within this area is beyond the scope of this post. Land subsidence , as suspected to be in Caloocan City, may be due to its soil type and its proximity to Manila Bay. The results should also be compared with a current surface deformation study such as the results of a differential GPS campaign.

Comments

Post a Comment

Popular posts from this blog

Free Open Street Map data (i.e.shapefiles etc.) for the Philippines

Free Road data extracted from Open Street Map made available by Cloudmade . philippines.ai.zip   (28.7M) philippines.img.zip   (8.3M) philippines.navit.bin.zip   (12M) philippines.osm.administrative.bz2   (627K) philippines.osm.bz2   (29M) philippines.osm.coastline.bz2   (3.5M) philippines.osm.highway.bz2   (16.2M) philippines.poi.gpx.zip   (521.7K) philippines.poi.tomtom.zip   (334.6K) philippines.poly   (2.3M) philippines.shapefiles.zip   (20.5M) Abra Agusan Del Norte Agusan Del Sur Aklan Albay Antique Apayao Aurora Basilan Bataan Batanes
Read more

Free Philippine Administrative Boundaries shapefile

Image
Free Administrative Boundary shapefile can be downloaded here . The coordinate reference system is latitude/longitude and the WGS84 datum. Other file formats such as ESRI geodatabase (.gdb), Google Earth (.kmz), and RData for RProject. Geospatial data from other countries are also available for download here. File Formats A "shapefile" consist of at least three actual files. This is a commonly used format that can be directly used in Arc-anything, DIVA-GIS , and many other programs. Unfortunately, non-latin characters are garbled. An "ESRI geodatabase" is an MS Access fole that can be opened in ArcGIS (an expensive but powerful program). One of its advantages, compared to a shapefile, is that it can store non-latin characters (e.g. Cyrillic and Chinese characters). A "Google Earth .kmz" file can be opened in Google Earth . A "RData" file can be used in R (with the sp package loaded) Some files have been compressed and grouped i
Read more

University of the Philippines Diliman Campus

Image
In our GIS class for Geomatics Engineering, we were asked to digitize a pan-sharpened Quickbird image of UP Diliman and specify the roads, buildings, and open spaces in the campus. Also this is my first time to import data from a handheld GPS into my GIS. Here's my output map: We we're also asked to georeference a "very old" aerial photo of UP Diliman campus. Here's the result:
Read more