×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Remote sensing for resources development and environmental management
Author
Damen, M. C. J.

157
Symposium on Remote Sensing for Resources Development and Environmental Management / Enschede / August 1986
Geological analysis of Seasat SAR and SIR-B data in Haiti
Ph.Rebillard
SEP/DTI, Puteaux, France
B.Mercier de l’Epinay
Université PM. Curie, Paris, France
ABSTRACT : Synthetic Aperture Radar data obtained by Seasat in 1978 and by SIR-B in 1984 were registered to
the corresponding Landsat MSS data over the Western part of Haiti. Two areas were studied : the island of La
Gonave and the Miragoane area. Seasat SAR data were registered to the Landsat data over the island of La
Gonave and both Seasat and SIR-B data were compared over the Miragoane area.
In both cases, difficulties occured in the interpretation due to the relief which distorted the radar data
particularly the Seasat data for which the radar incidence angle was 20 degrees ; futhermore, on the Miragoane
area difficulties occured using the SIR-B data due to low radar signal.
Structural lines and lithological boundaries were pointed out thanks to the compositional and textual
informations provided by the radar and MSS data. Geological interpretation maps of La Gonave and the Miragoane
area were drawn : boundaries between rocks Eocene of age and rough basalts were precised as well as 1) domes,
2) ancients faults oriented NE-SW, 3) faults oriented NW-SE linked to the La Gonave anticline and 4) faults
oriented E-W integrated in the tectonic model of the Caribbean plate.
1 INTRODUCTION
The Western part of Hispaniola has been
imaged by Seasat in 1978 and by SIR-B in 1984
(Fig. 1). Two areas were studied: the
island of La Gonave and the Miragoane area
(Southern branch of Haiti).
Seasat SAR data were registered to the
Landsat MSS data over the island of La Gonave
and a map was produced showing the interest
of compositional and roughness informations
produced by MSS and Seasat respectively.
Over the Miragoane area, both Seasat and
SIR-B data were contrasted to the existing
maps
Seasat SAR acquired data at 20° incidence
angle. In this case, surface slopes played a
dominant role because of the quasi-specular
scattering mechanism. Therefore topography
is strongly highlighted in the Seasat images.
The effect of foreshortening as well as
foldover, in some cases, even make topography
effects more dramatic.
SIR-B data were acquired with an incidence
angle of 60°. In this geometry, the
scattering is mostly dominated by the surface
roughness at the scale of the wavelength (L
band = 25 cm). The roughness characteristics
include both the soil roughness as well as
the vegetation surficial and volumetric
roughness. Surface slopes also impact the
scattered return by modulating the scattering
from the surface roughness. Surface
dielectric constant also plays a role in the
scattered intensity.
Therefore combined Seasat/SIR-B data contain
informations on surface topography as well as
on roughness and provide a better knowledge
of the surface physical properties. The
following paper presents our preliminary
results.
2 SPACEBORNE DATA
Seasat SAR data were acquired on August 17,
1978 during rev. 737. The swath width was
100 Km and the incidence angle was 23.5°.
The SIR-B data (data take KI-086.10) were
obtained in October 1984. The SIR-B swath
was 14.6 Km wide and the incidence angle was
60° (Ford et al. 1986). Both Seasat and
SIR-B were operating in L-band and HH
polarization. Coverage of the radar paths
are indicated in Fig. 1.
Landsat MSS data were obtained on June 29,
1979 (path 009, row 47). Bands 4, 5 and 7
were used during the study.
Figure 1. SAR data coverage, 1: Seasat rev.
737, 2: SIR-B data take KI-086.10.
3 DATA REGISTRATION
The registration of the Seasat and the
Landsat MSS data over the La Gonave area has
been made as a pixel to pixel registration
using the SEP's Vizir Image Processing System
procedures. The Landsat MSS data image was
taken as the reference image. Fifty nine
ground control points were chosen (Rebillard
et al, 1983 ) .
Several difficulties occured due to the
relief of the island and as a consequence,
the registration is not perfect in the center
of the study area.
In the Seasat images (acquired data at 20°
incidence angle), surface slopes played a
dominant role because of the quasi-specular
scattering mechanism. Therefore topography
is strongly highlighted and the effect of