Satgen225    ERS 1  Earthquake Pictures        by GM4IHJ        17 July 93
This is replacement for Satgen225 msg which was badly corrupted        

At about 0500 California time on 28th June 92, Dave GM4JJJ woke suddenly.
His hotel bed was shaking and even more alarming , the drawers in all the
cupboards were falling out. Clearly Southern California was having an
earthquake. A point which became more obvious when he went to the window
and saw the hotel desk clerk who had abandonned ship and run out into the
middle of the car park on the good old principal of sauve-qui-peut.
Fortunately not much more happened in Anaheim , but at Landers 120 kms
east in relatively open desert country there had been a movement of more
than 4 metres along an 85 km fault line.
Up to now earthquakes have been poorly defined, because pre quake and post
quake studies at a large number of sites was simply not possible. But now
with the advent of ERS1 the situation has changed. ERS1 orbits at a height
of 750 kms and views the world via a side scan synthetic aperture radar
using a frequency of 5.36 GHz.
ERS1 collected images of the Landers area 4 times in 1992. Once before the
quake, and 3 times after it. Each pixel is 4 x 20 metres and the satellite
receiver records both signal range and phase data of the echo returns from
the ground. Despite the fact that no two orbits follow exactly the same
track , it has proved possible to take the data from runs on 24 April pre
quake and, 7 August post quake and put them together with a high degree of
coherence. To do this the operators reconstruct the phase of each pixel by
minimizing the number of fringes at the corners of the picture, assuming
that far field data in the picture corners will be least disturbed by the
quake. They then eliminate the stereo path difference using a differential
elevation model and, then calculate the interferometric fringes of the
image at map coordinates , before resampling to improve overall signal to
noise ratio and finally plotting what is basically a ground map covered in
colour coded fringes where the quake has shifted the surface betwee the
prequake and post quake records.
The result is a map of the Landers area overlayed by the "diffraction
pattern" of coloured fringes. Each fringe can reveal a shift as little as
28mm , that is the half wave length of the radar. Naturally with such a
precise record the middle of the pattern over the fault itself is blurred
because even with the small pixel size at the ground of 4 x 20 metres ,
there has been a much larger ground shift than 28mm. Never the less the
area near the fault has obvious breaks in the surface so a record at that
point is hardly necessary , and ERS1 provides a beautiful record out to
far beyond the edges of the total affected area.
This ERS1 technique has a precision never seen before. Indeed geologists
are now calling for records to be taken by ERS1 of all the world danger
spots - volcanoes which swell just before erupting - glaciers having
sudden spurts of movement which could foretell large killer landslides as
have occured in South American , and of course other active earthquake
fault lines.
ERS1 scans the same patch of ground every 35 days . It does not require
the positioning of ground stations in remote and often dangerous places.
So we should hear more about its successes in the future. ERS1 was built
and launched by the European Space Agency/ArianeSpace. 73 GM4IHJ @ GB7SAN