International Archives of Photogrammetry and Remote Sensing. Vol. XXXII Part 7C2, UNISPACE III, Vienna. 1999
115
15PR5
UNISPACE III - ISPRS/EARSeL Workshop on
“Remote Sensing for the Detection, Monitoring
and Mitigation of Natural Disasters”
2:30-5:30 pm, 22 July 1999, VIC Room B
Vienna, Austria
The INCO Limited Copper Cliff tailings area is the largest
repository of acid generating tailings in North America, and
possibly the world. The site covers approximately 5500 acres
and contains more than 10 percent of all tailings in Canada.
Mill tailings have been deposited since the 1937. The
impoundments contain silt to sand size tailings including sulfide
minerals, such as pyrrhotite, pentlandite, calcopyrite and pyrite
(McGregor et al 1995). The mill waste is impounded in a basin
enclosed by bedrock ridges and dams made of till material. The
tailings dams were designed to seep in order to maintain a low
phreatic head for structural stability (Puro et al 1995).Tire
sulfide minerals contained in the tailings are oxidized by
atmospheric oxygen. A 1-3 m thick oxidation zone is present at the
old tailing surface (De Vos et al 1995). The water infiltrating the
tailings surface will ultimately seep as acid drainage. Some of the
acid drainage accumulates in depressions below the dam, and
gradually kills the nearby vegetation.
Restoration of the tailings area involves treating the acid soils with
a lime- rich fertilizer and revegetation with varieties of grasses and
legumes. The casi image was used to discriminate vegetation, lime
and oxidized tailings (Figure 2). In this image Levesque et al.,
1997 use a linear unmixing technique to separate three surfaces of
the tailings. Tire linear unmixing technique uses the spectrum of
each pixel which is decomposed into a linear combination of
"endmembers" spectra, representing the purest pixels.Manual
selection of endmember spectra from the extremities of various
combinations of first three principal components scatter plots were
used to separate the tailings surface. Field verification shows that
the resulting thematic images accurately corrispond to the tailing
surfaces. Images taken at subsequent periods were used to monitor
the revegetation progress at specific sites.
Monitoring Pollution of Mining Lakes environmental
restoration in Germany.
In Germany, today there is a need to provide low cost
techniques for the assessment and long term monitoring of
mining areas. Because of the size of the mining areas and the
changing landscape multi temporal and multisensor remote
sensing data offer the best possibilities to fulfil these demands
(Glaesser at 1999, Birger et.al 1998). In this section, we report
the results of Boine et. al. 1999 and Glaesser et al 1999. They
used satellite and airborne multispectral (casi) images, and field
spectroradiometric measurements to map sulphate content of
mining lakes and monitor restoration programs
The test area is part of the Eocene lignite deposit (bituminous
brown lignite) in Central Germany. It is characterised by
widespread lignite seams - mostly two seams of 5 to 15 m
thickness- that are overlaid by marine Tertiary and glacial
Pleistocene sediments. In these mining areas many acid lake
exists because of the oxidation of pyrite and marcasite minerals
of waste sediments, and the rising groundwater. To explore the
uses of airborne multispectral data to characterise the acid lakes,
casi data were flown with the enhanced spectral mode with 48
adjacent non-overlapping bands with a 10.6 nm band with
within the spectral range from 428 mn to 975 nm. The casi data
were corrected and classified, based on field spectral
information (Figure 3). Water sampling and spectral field
measurements were made on the acid lakes at the time of the
casi flights overpass. Figure 3 summarises some of the field
measurements, which included pH. secchi depth, total dissolved
iron content and sulphate content and reflectance spectra of the
acid lakes (Boine et al 1999)
This case study suggests that remote sensing methods can be
used to characterise acid lakes in mining areas. To achieve a
more reliable classification results Boine et.al suggests that
information related to the hydrochemical and hydrobiological
properties of the lakes, be considered. Also because of extensive
variability in optical and chemical properties of the lakes the
classification procedures should be done within individual lakes
and between single lakes. These new methods provide another
tool to monitor the temporal and spatial variability of acid lakes
in mining areas.
The classification of the sediments and vegetation uses three
different satellite and airborne sensors for comparison of their
specific potentials in mining areas. During each flight a spectral
field measurement of sediments were carried out with a field
spectrometer GER Mark V in the wavelength range from 400 to
2500 nm. Classification methods include the maximum-
likelihood, spectral angle mapping, and linear spectral
unmixing. The highest spectral resolution airborne data
(HYMAP and DIAS) and TM5 were the most useful for the
classification of the sediments For the vegetation, it was
possible to classify the airborne data to produce information at
the level of the vegetation units. Glaesser at al 1999 noted that
the airborne data was useful to classify single trees and quite
good for determination of differences in the density of the
vegetation even when then pixel size is much larger then the
size of the plants.
Image Maps for Groundwater Protection in Jordan
In Jordan, there is an urgent need to manage and utilize the
water resources in order to comply with the immediate and
projected water demand in the country. Hydrogeological
mapping is fundamental to the economic development of
Jordan. Pumping of groundwater is already in danger of
exceeding levels at which the resource cannot be renewed; and
water quality is declining (World Bank 1995). Groundwater is
the major water resources in many areas in the country. All the
known aquifers have been explored. Water harvesting is a
national priority as water supplies are being depleted by
intensive pumping. For this reason it is necessary to map in
detail the hydrogeology of all the aquifers. Since the distribution
of groundwater in Jordan is fairly well known, the main
problem is to optimize the utilization of groundwater resources,
and protect them against depletion and pollution. Although
several studies have shown that there has been considerable
work done for the search for potable water in Jordan (Bajjalli
1994, GTZ 1977, etc.), the use of remote sensing techniques for
groundwater protection is increasing. This case study reports on
the use of and SAR/TM data fusion technique integration for
providing information on the surface hydrogeological
characteristics of a sensitive aquifer in Azraq. Jordan. The
image map produced from the data fusion technique was used to
map wadis, rock units, fractures and land practices necessaiv for
environmental protection of the aquifers.
The SAR/TM synergy image was integrated using the IHS
integration method (Cliche et al., 1985). The three Eigen vectors
of the TM image were transferred to the IHS space then, the
