HYPERTEMPORAL ANALYSIS OF REMOTELY SENSED SEA ICE DATA
FOR CLIMATE CHANGE STUDIES
Joseph M. Piwowar
Ellsworth F. LeDrew
Department of Geography and Earth Observations Laboratory
Institute for Space and Terrestrial Science
University of Waterloo
Waterloo, ON N2L 3G1
Canada
tel.:
(519)885-1211 ext. 6563
fax:
(519)888-6768
internet:
[email protected]
ABSTRACT
Climatologists have been concerned for some time about the possible
climatic ramifications of increased atmospheric
concentrations
of
anthropogenically produced greenhouse gases. Many general circulation
model experiments show amplified warming in the polar regions as t h e
strongest
response
to
enhanced
atmospheric
greenhouse
gas
concentrations. In light of this, there has been speculation that a spatially
coherent pattern of high-latitude temperature trends could be an early
indicator of climatic change. The sensitivity of sea ice to the t e m p e r a t u r e
of the overlying air implies that observed trends in Arctic ice conditions
may also indicate general climatic changes. For example, changes in t h e
global average temperature might be detectable by observing variations i n
sea ice extent. Extent, then, is one common climatic indicator obtainable
from sea ice. Other indicators include ice type, concentration, thickness,
and dynamics. These are shown to have distinct regional and spatial
distributions within the Arctic.
The advent of satellite remote sensing has offered intriguing o p p o r t u n i t i e s
to investigate, in good spatial and temporal detail, even the most r e m o t e
or rarely visited areas of the earth's surface. The large extent of sea ice i n
polar regions, the difficulty and expense of access, and the normally
adverse weather conditions make satellite data almost indispensable f o r
studies requiring global or large-scale ice cover characteristics in t h e s e
regions. Remote sensing in the microwave portion of the electromagnetic
spectrum
is particularly
relevant for polar applications
because
Progress in Physical Geography, Vol. 19, No. 2, pp. 216-242, 1995.
microwaves are capable of penetrating the atmosphere under virtually all
conditions (of particular significance during the frequent and e x t e n d e d
periods of cloudiness in the Arctic) and microwave remote sensing
systems are not dependent on the sun as a source of illumination ( a n
important consideration during the long polar night).
In particular,
analyses of passive microwave imagery can provide us with daily
information on sea ice extent, type, concentration, dynamics, and m e l t
onset . An historical record of Arctic imagery from orbiting passive
microwave sensors starting from 1973 provides us with an excellent d a t a
source for climate change studies.
The development of analysis tools to support large area monitoring is
integral to advancing global change research.
Traditionally, maps h a v e
shown phenomena distributed over space at some point in time. Explicit
representation of time in cartographic displays has either taken the f o r m
of change maps or been expressed by a series of static maps. The critical
need is to create techniques which highlight the space-time relationships
in the data rather than simply displaying voluminous quantities of d a t a .
In particular, hypertemporal image analysis techniques are required t o
help find anticipated trends and to discover unexpected t e m p o r a l
relationships. Direct hypertemporal classification, principal c o m p o n e n t s
analysis, and spatial time series analysis are identified as three p r i m a r y
techniques for enhancing change in temporal image sequences. There is a
need for the development of new tools for spatial-temporal modelling,
however.
1. INTRODUCTION
The earth's polar regions play a critical role in global change scenarios.
Polar climates exist through a delicate balance between the limited
amount of radiant energy received from solar radiation and general
atmospheric and oceanic circulation, and the high proportion of that h e a t
which is reflected back into space or otherwise tied up with in changes o f
phase of surface moisture. The net result is a permanent presence of i c e
and snow. In areal extent, sea ice accounts for nearly two-thirds of t h e
earth's ice cover. In the north polar region, it spans an area ranging f r o m
a summer minimum of 8.5 x 10 6 k m 2 to almost twice that extent i n
winter.
Climatologists have been concerned for some time about the possible
climatic ramifications of increased atmospheric
concentrations
of
anthropogenically
produced
pollutants.
At the
1987
World
Meteorological Organization / United Nations Environmental Programme /
International Council of Scientific Unions workshop on climate c h a n g e
held in Villach, Austria a scenario was developed which suggests that d u e
to progressive accumulation of greenhouse gases (mainly carbon dioxide
and methane) in the atmosphere, the global average surface t e m p e r a t u r e
would rise between 1.5°C and 4.5°C above the mean 1980s d e c a d a l
temperature by about the year 2030.
In high northern latitudes, t h e
average rise in winter temperature is expected to be more than twice t h e
global average.
Thus, if the median Villach scenario of a global
temperature rise of 3°C by 2030 were taken, one could expect, on t h e
basis of current knowledge and modelling, that the polar regions would b e
about 6-7°C warmer than at present (Roots, 1989).
Although the predictions of the Villach scenario are substantiated b y
other working groups, such as the International Panel on Climate Change
(Houghton, 1990) and the Canadian Climate Centre (Daley, 1989), not all
scientists are in agreement (Mysak, 1990; Kukla, 1993). Washington a n d
Meehl (1989) report on a study where an atmospheric model was c o u p l e d
to a four layer ocean model to investigate the response of the climate
system to increases of greenhouse gases. Their results show the o c e a n
surface warming only near 30-50°N latitude and a cooling of the n o r t h e r n
North Atlantic resulting in greater ice cover in the Greenland Sea.
A common hypothesis is that the most direct effect of global warming o n
the climate of the Arctic would be manifested in changes of the ice c o v e r
(Hare, 1982). Recent studies on the climatic role of sea ice do show t h e
establishment of a bi-directional feedback mechanism between the i c e
and the adjacent atmospheric and oceanic systems (Walsh and J o h n s o n ,
1979a; Parkinson, 1989; LeDrew et al., 1991).
That is, sea ice has a
significant influence on the atmosphere and adjacent oceans and in turn is
significantly influenced by the atmosphere and oceans. There is now little
doubt about the intimate connection between sea ice and climate a n d
since the atmosphere, oceans, and ice cover are all components of a
single thermodynamic system, a change in any one part necessarily r e s u l t s
in compensating changes in the others (Maykut and Untersteiner, 1 9 7 1 ;
LeDrew, 1992a).
Our research in this area is built upon the assumption that, due to the bidirectional feedbacks operating in the polar basin, observed trends i n
Arctic sea ice conditions may be indicative of global climatic changes.
What is not clear, however, is what sort of changes we should b e
expecting to find in the sea ice cover and whether the ice is a leading o r
trailing indicator of global change.
Recent advances in two related technologies may hold the key to helping
us detect and monitor changes in global climates. First, we note t h a t
repetitive, continuous, and global imaging of the earth's surface b e g a n
early in the 1970s with the launch of the Nimbus, NOAA, and Landsat
series of remote sensing satellites.
Thus, the length of the historical
record of global remote sensing data is quickly approaching the thirtyyear mark usually deemed necessary for the establishment of climatic
norms. Secondly, new techniques are being developed to analyze t h e s e
data. It is not sufficient to simply apply existing time series and t r e n d
analyses to long sequences of remote sensing imagery because these h a v e
not been developed to capture the spatial relationships inherent in t h e
data. Research has begun into a new type of image analysis t e c h n i q u e ,
which we call hypertemporal image analysis, to work with these data t o
extract temporal signals in a spatially coherent manner. It is the coupling
of these new hypertemporal
image analysis techniques with t h e
increasingly long remote sensing record which holds the promise o f
significantly advancing our understanding of the changes now h a p p e n i n g
in the global climate system.
Although much of the discussion about hypertemporal image analysis o f
the historical remote sensing archive is generic to almost any global
location, we focus our attention on the north polar region because of o u r
interests in the sea ice-climate linkages. In the balance of this paper t h e n ,
we summarize the properties of the Arctic sea ice cover which a r e
significant for climate change studies in light of the data available in t h e
remote sensing record. We then explore the potential of h y p e r t e m p o r a l
image analysis techniques for extracting these parameters.
2. ARCTIC SEA ICE AND CLIMATE CHANGE
The climatic role which sea ice plays is generally expected to be based o n
changes in its surface albedo, insulative properties, thermal capacitance,
and brine capacitance (Walsh, 1983; Anderson, 1987; Parkinson, 1 9 8 9 ) .
Table 1 lists the physical properties of sea ice which can be used t o
estimate each of these climatic properties on a global (or at least
hemispheric) basis. Any analysis of climate change and variability using
sea ice as a proxy indicator must ultimately reference one or more o f
these physical properties.
In this section we assess the i n f o r m a t i o n
potential of sea ice extent, type, concentration, thickness, m o t i o n
dynamics, and seasonality for climate monitoring.
2.1. SEA ICE EXTENT
The polar regions tend to amplify climatic variations occurring elsewhere,
thus relatively minor climatic changes are expected to produce large
variations in the areal extent of the ice cover. Since the existence of a
large thermal gradient at the ice-ocean boundary significantly affects t h e
large-scale transfer of energy between the atmosphere and o c e a n s ,
continued monitoring of the extent of the floating ice pack is crucial t o
our understanding of the present climatic state.
Areal extent is the most easily, hence most frequently, mapped p a r a m e t e r
of the Arctic sea ice. Arctic sea ice has an average seasonal cycle ranging
from a summer minimum of 8.5 x 10 6 k m 2 in September to a w i n t e r
maximum of 15 x 106 km2 in March (Figure 1). The historical record as a
whole does not show any definitive indications of consistent growth o r
decay of the north polar ice extent, although there is considerable
interannual, regional, and seasonal variation (Gloersen and Campbell,
1988; Parkinson and Cavalieri, 1989). In a seasonal analysis, however,
Chapman and Walsh (1992) have found statistically significant d e c r e a s e s
of the summer ice extent and three new summer minima between 1 9 7 6
and 1990. This is in contrast to the 1960s where five of the largest
summer extent values are found.
2.2. SEA ICE CONCENTRATION/OPEN WATER AREA
Although it is generally very compact, sea ice never completely covers t h e
entire ocean surface because of atmospheric and oceanic forcing.
Expansion and deformation within the ice pack can cause the formation o f
open water areas known as "leads" (long, narrow cracks ranging f r o m
metres to kilometres in width) and "polynyas" (more expansive regions o f
open water up to several thousands of square kilometres in size). Sea i c e
concentration - the percentage of a given area of ocean covered by ice - is
an often used measure of the amount of open water within the i c e
margins. Even at high ice concentrations, an open lead 10-20 m a c r o s s
can have a dramatic impact on the air-ice-ocean climate system. Ice a c t s
as a thermal regulator, controlling the exchange of energy between t h e
ocean and the atmosphere.
Whenever the relatively warm ocean is
exposed directly to the cold winds blowing over the ice surface, the very
strong temperature contrasts establish a massive energy flux. The o p e n
water areas thus act as "thermal volcanoes" (LeDrew, 1992a).
Figure 2 shows that even at the time of maximum ice extent there a r e
large portions of the polar ice pack which have ice concentrations of less
than 100%.
An annual examination of similar images may p r o d u c e
evidence of climate change through significant differences in i c e
concentrations both regionally and/or seasonally.
2.3. SEA ICE TYPE
Sea ice is a complex material formed by the freezing of sea water. Many
of the geophysical properties of sea ice are largely determined by its age.
Of significance to climate research is the metamorphosis which occurs i n
sea ice as it survives the first summer melt season. Before this transition,
so called "first-year" ice tends to be relatively thin, attaining a m a x i m u m
thickness of about 2 m (Weeks, 1976). Ice which has survived at least
one summer melt period - termed "multiyear" ice - is substantially
altered from first-year ice, because water from melt ponds p e r c o l a t e s
through the ice causing desalination of the surface and underlying layers
(Parkinson et al., 1987b). The low brine content of multiyear ice i m p a r t s
a particularly high strength to an already thickening ice sheet.
Sea ice is, therefore, generally classed by age: thinner, weaker first-year
ice; and thicker, stronger multiyear ice. The maximum age of Arctic s e a
ice is estimated to be nineteen years old, with only about 2% of the a r e a
occupied by this group (Vowinckel and Orvig, 1970).
Changes in t h e
relative distributions of first-year and multiyear ice across the Arctic a r e
potential indicators of climatic variability although no significant findings
have been reported to date.
2.4. SEA ICE THICKNESS
Pack ice reaches its final equilibrium thickness of 2-4 m in about five
years when the amount of ice lost during summer melt is made up d u r i n g
the following winter (Vowinckel and Orvig, 1970; Maykut a n d
Untersteiner, 1971).
Although local bathymetry, tides, and c u r r e n t s
affect the growth of landfast ice, the relative thickness of sea ice in a
particular year reflects average winter and early spring temperatures a s
well as the magnitude of mid-summer solar radiation (Jacobs and Newell,
1979; Ledley, 1990). Thus, sustained decreases in the average t h i c k n e s s
of sea ice on a regional scale may well be indicative of a warmer climate.
2.5. SEA ICE DYNAMICS
Much of the Arctic sea ice exists in a geographically c o n s t r a i n e d
environment. The Arctic Basin holds a small ocean which is s u r r o u n d e d
by two large continents (Figure 2). The Arctic Ocean has r e s t r i c t e d
interaction with the world's ocean system, the principal access p o i n t
being through Fram Strait, a deep trench between Greenland and Svalbard.
The connection of the Arctic Basin with the Pacific Ocean through t h e
Bering Strait is of lesser consequence because its shallow bench inhibits
the passage of deep ocean currents (Clarke and Jones, 1989).
The general drift pattern of the ice in the Arctic Ocean follows t w o
dominant features:
the Beaufort Sea Gyre and the Transpolar Drift
Stream.
The oldest ice in the Arctic circulates in a generally c l o s e d
clockwise drift stream in the Beaufort Sea/Arctic Ocean and is c o n t r o l l e d
by the Arctic High atmospheric pressure system. The Transpolar Drift
Stream, on the other hand, stretches from the East Siberian Sea across t h e
North Pole to the northeast of Greenland where it exits through Fram
Strait. The Transpolar Drift Stream is the major carrier of ice out of t h e
Arctic Basin. Mean annual net drift rates vary from 0.4 to 4.8 k m / d a y
with monthly averages observed as high as 10.7 km/day (Weeks, 1976).
Recent investigations of ice motion using satellite data have revealed t h a t
the mean drift rate of the ice pack is quite variable and can, at times b e
reversed. In particular, the predominantly clockwise circulation of t h e
Beaufort Gyre has been seen to reverse for periods lasting at least 30 d a y s
in late summer (Serreze et al., 1989). LeDrew et al. (1991) show t h i s
phenomenon to be triggered and maintained by persistent cyclonic
activity during that time of year. It is clear a reversal of this nature m u s t
have a direct impact on the climate system. LeDrew et al. (1991) suggest
that there is some evidence of ice divergence during the reversal episodes
exposing large amounts of relatively warm ocean to the cold a t m o s p h e r e .
Hypertemporal remotely sensed imagery can play a key role in mo n i t o r i n g
the motion of the ice pack and perhaps reveal other anomalous d r i f t
phenomena. In addition, tracking of ice masses across image s e q u e n c e s
may help verify the relative ages of sea ice in different regions of t h e
Arctic.
2.6. SEASONALITY
In the Northern Hemisphere, the seasonal cycle of ice advance and r e t r e a t
is approximately symmetrical in all regions, although the periods of m o s t
rapid growth and decay do vary regionally (Comiso and Zwally, 1984). In
the late autumn, sea ice morphology continues to change until the s u r f a c e
air temperature drops below about -10°C in November or December, a t
which point the ice stabilizes and remains so until the air t e m p e r a t u r e
rises back above that level in the spring. Comiso (1986) and C h a p m a n
and Walsh (1992) did not find any significant trends of winter ice e x t e n t
in the Arctic between 1961 and 1990.
The variations in sea ice are particularly significant during the m e l t
season, when rapid changes in ice extent and surface conditions occur. In
an examination of the onset of melt across the Arctic Basin in 1979 a n d
1980, Anderson (1987) found the melt signature is observable first in t h e
Chukchi Sea and the Kara and Barents Seas. As melting progresses, t h e
melt signature moves westward from the Chukchi Sea and eastward f r o m
the Kara and Barents Seas to the Laptev Sea region. Anderson (1987) a l s o
notes that the initial location and date of the melt signal varies with y e a r .
This leads him to conclude that monitoring the occurrence of m e l t
signatures could be used as an indicator of climate variability in t h e
Arctic's seasonal ice zones.
The significant processes of the summer Arctic ocean are the loss of t h e
snow cover, the opening up of extensive areas of ocean, and t h e
formation of melt ponds on the existing ice surfaces. In the context o f
climatological monitoring, the effects these changes have on the s u r f a c e
albedo and the interannual variation of albedo are of prime i m p o r t a n c e
(Carsey, 1985). Global warming scenarios generally project a more r a p i d
retreat of Arctic sea ice in summer than in winter. Chapman and Walsh
(1992) report finding statistically significant decreases of the summer i c e
extent of Arctic ice and three new summer minima in the past fifteen
years. When they compare this decrease with the concomitant fact t h a t
the strongest atmospheric warming has occurred over those land a r e a s
(northwestern North America, northern Asia) and seasons (late winter,
spring) when the albedo-temperature feedback is potentially strongest,
they conclude that northern land areas as well as the summer sea i c e
regime should be closely monitored for greenhouse signals (Chapman a n d
Walsh, 1992).
Crane (1978) observed an "early" and "late" mode of ice retreat in t h e
spring seasons from 1964-1974. He also noted a similar pattern in t h e
autumn period of ice advance.
When he related this pattern to t h e
general synoptic circulation over the area he found that for the years o f
early ice advance, there is an increased frequency of northerly a n d
westerly air flow, bringing lower temperatures and an influx of multiyear
ice into the Davis Strait-Labrador Sea area (Crane, 1978). Ice-atmosphere
feedbacks are thought to operate in both directions during the ice-growth
regime (Barry, 1983). Not only do the cooler air temperatures precipitate
ice formation, but the strongest correlations between ice anomalies a n d
subsequent atmospheric fluctuations are found in the autumn m o n t h s
(Walsh and Johnson, 1979a). This bi-directional coupling has a t e n d e n c y
to increase the variability within the ice pack and to mask any potential
climatic signals.
Thus, the autumn is a less desirable season f o r
monitoring the sea ice than the spring or summer.
2.7. SEA ICE AS A CLIMATIC INDICATOR
Table 2 summarizes what we currently know about the possible climatic
influences of sea ice extent, type, concentration, thickness, m o t i o n
dynamics as well as an estimation of their ideal observational
requirements. Analyses of these parameters are compounded by the f a c t
that the oceanic, atmospheric, and geographic composition of the Arctic
combine to create distinct regional variations in sea ice. For example,
there is only a 20% seasonal difference in sea ice extent in the Greenland
and Norwegian Seas while Baffin and Hudson Bays experience s u m m e r winter variations of over 200% (Comiso and Zwally, 1984). This m e a n s
certain regions may exhibit climatic signals with more clarity; regions
exhibiting high interannual variability would not be good candidates f o r
climate change studies since a particularly large change from average s e a
ice conditions would have to occur before a climate change c o u l d
confidently be said to have taken place (Zwally et al., 1983b; Parkinson,
1991). High variability is the noise in which a climatic signal must b e
found.
3. REMOTE SENSING OF SEA ICE
Remote sensing has a particularly prominent role in the monitoring o f
climate change; it is the only source of data with which we can view t h e
entire planet and monitor the change in the nature of the surface of t h e
planet through time, in a consistent, integrated, synoptic and n u m e r i c a l
manner (Davis et al., 1991; LeDrew, 1992b). Global observations of t h e
earth by satellite sensors are required simply because no other a p p r o a c h
permits measurements to be made at the required spatial and t e m p o r a l
resolutions.
This is particularly relevant in Arctic research where t h e
large extent of the sea ice, the difficulty and expense of access, and t h e
normally adverse weather conditions make satellite data indispensable f o r
studies requiring global or large-scale ice cover characteristics in t h e s e
regions (Comiso and Sullivan, 1986; CGCP, 1989).
Many of the changes in the global environment may be monitored on a
worldwide basis with remote sensing satellite systems. This m o n i t o r i n g
can be direct, through such observations as surface temperature increase,
or indirect, through the observations of parameters thought to b r i n g
about changes in climate or to respond to the changing climate, such a s
changes in polar ice extent (CGCP, 1989). Little quantitative i n f o r m a t i o n
existed on seasonal or interannual changes of the polar oceans' sea i c e
cover until repetitive and synoptic satellite observations began.
The
ability of earth observing satellites to collect data from the polar regions
at a variety of scales for extended periods of time have been of significant
value in our understanding of ocean-ice-atmosphere climate i n t e r a c t i o n
processes. These data provide information on current climatic conditions
in polar regions as well as information useful in studies of the t i m e dependent behaviour of sea ice. In short, satellite data have h e l p e d
scientists see the ocean and ice systems as an integral part of a c o m p l e x
global climate system (Johnson, 1989).
Of particular relevance to remote sensing in polar environments are t h e
restrictions imposed by the limited times of direct sunlight and a c l e a r
atmosphere.
Further, for any variable to be useful in climate c h a n g e
studies, it must be routinely measured over a period of at least 1 0 - 2 0
years. In the following review of remote sensing of sea ice, optical,
thermal and microwave sensors are evaluated against these criteria i n
addition to assessing their potential to monitor the specific sea i c e
parameters highlighted above.
3.1. OPTICAL AND THERMAL SENSORS
Data from the AVHRR sensor carried by the NOAA satellite series h a v e
been the most widely used in climate change studies in general, and h a v e
particular relevance for polar remote sensing.
These data have high
temporal resolution (at least two images can be acquired daily for a n y
location north of 60° latitude), medium spatial resolution (1.1 k m :
detailed enough for monitoring individual ice floes while minimizing
unnecessary data volume), thermal band imaging capabilities (which allow
the collection of data through the long polar night), and inexpensive
acquisition costs. Results from ice mapping experiments show that i c e
concentrations determined from optical imagery differ from t h o s e
derived from passive microwave data by ±15-20%, probably due to t h e
presence of thin new ice not visible in the optical imagery and the ability
of the microwave sensors to measure partial concentrations within t h e
pack more accurately (Dey, 1981; Comiso and Zwally, 1982; Alfultis a n d
Martin, 1987; Burns and Viehoff, 1991; Emery et al., 1991).
Remote sensing at optical and thermal wavelengths is limited for long
term climatic monitoring of the northern cryosphere, however, b e c a u s e
of the inability to collect data during the long polar night of winter ( w i t h
the exception of bands located at thermal infrared wavelengths) a n d
through an optically opaque atmosphere (Arctic cloudiness averages 7 0 80% from June through August). In addition, since these optical s e n s o r s
were not originally designed around polar applications, their response is
frequently saturated by bright ice and snow returns, inhibiting detailed
surface discrimination.
Few Arctic images from these platforms h a v e
been archived and some of the early data tapes which did exist are n o w
deteriorating, further reducing the availability of historical data for use i n
long term climate monitoring studies (Massom, 1991).
3.2. MICROWAVE SENSORS
The planetary and atmospheric restrictions imposed on optical a n d
thermal sensors in the polar regions are removed when radiation is
detected in the microwave portion of the electromagnetic spectrum (i.e.,
at wavelengths between 1 mm and 1 m). Microwave sensors d e t e c t
thermal microwave radiation reflected and emitted from the e a r t h ' s
surface. Additional radiative contributions or interference introduced b y
clouds, other atmospheric effects, and extraterrestrial sources exist b u t
are relatively minor in the polar regions (Parkinson et al., 1987b). Two
advantageous features characterize microwave energy from a p o l a r
remote sensing standpoint: (i) microwaves are capable of penetrating t h e
atmosphere under virtually all conditions (of particular significance
during the frequent and extended periods of cloudiness in the Arctic); a n d
(ii) microwave remote sensing systems are not dependent on the sun as a
source of illumination (an important consideration during the p o l a r
night) (Lillesand and Kiefer, 1987; Johnson, 1989).
Remote s e n s o r s
operating in the microwave portion of the spectrum are typically divided
into two types: those which provide their own source of illumination,
called "active" systems; and those which only sense naturally o c c u r r i n g
radiation, called "passive" systems.
So called "active" microwave sensors derive their designation from t h e
fact that, unlike most other remote sensing systems, they send out t h e i r
own energy to illuminate the surface for imaging.
The main active
microwave system in use today is called the synthetic aperture radar, o r
SAR. SAR is arguably the most powerful remote sensor available f o r
cryospheric research due to its high resolution and all-weather, all-season
imaging capabilities (Massom, 1991). It is not without its limitations,
however. While SAR data are quite powerful for the discrimination o f
various ice types under ideal conditions (i.e., in the middle of winter), t h e
energy-matter interactions in a melting snow/ice surface are still p o o r l y
understood (Barber et al., 1992a).
In addition, spaceborne SARs ( f o r
example ERS-1, JERS-1, RADARSAT, and EOS) have (or will have) relatively
narrow swath widths (in the range of 100 km) which are unsuitable f o r
hemispheric climate analyses. Continuous, repetitive SAR remote sensing
of the polar regions has only just begun with the launch of the ERS-1
satellite in 1991. While the future of active microwave remote sensing
looks bright (at least three other spaceborne SARs are to be l a u n c h e d
before the end of the century), the lack of an historical image archive
from which climatic variations can be derived is a limiting factor to t h e
present use of SAR data for climate change studies.
Another type of microwave remote sensing system which exhibits t h e
favourable
characteristics
of all-weather
and day-night
imaging
capabilities, and for which a reasonable historical archive of Arctic d a t a
exists, is that of passive microwave.
As its name suggests, a passive
microwave radiometer detects microwavelength radiation incident u p o n
its antenna without transmitting any illuminating signals by itself. This
microwave energy typically originates as naturally emitted radiation f r o m
the earth's surface and to a lesser extent, the atmosphere.
Three passive microwave sensors have been placed into earth orbit since
1972:
the Electrically Scanning Microwave
Radiometer
(ESMR)
(operational lifespan: December 1972 - October 1976, with some d a t a
gaps), the Scanning Multichannel
Microwave Radiometer
(SMMR)
(operational lifespan: October 1978 - August 1987), and the Special
Sensor Microwave / Imager (SSM/I) (operational lifespan: June 1987 present). With varying degrees of success, data from all three can be u s e d
to derive ice extent, type, concentration, motion dynamics, and the o n s e t
of melt.
The ESMR was the first instrument to provide the research c o m m u n i t y
with the repeated, synoptic overviews of the polar ice masses necessary
for monitoring ice extents. It was only a single-channel system, however,
which inhibits the derivation of ice types solely from the ESMR data a l o n e
(Parkinson et al., 1987b).
Ice concentrations calculated from ESMR
imagery are estimated to have an accuracy range of 15%-25% (Parkinson,
1989; Massom; 1991).
The use of single frequency data from the ESMR sensor creates s e r i o u s
ambiguities in data interpretation, particularly in areas of h e t e r o g e n e o u s
ice cover such as the extensive marginal ice zones of the Arctic.
An
attempt was made to address these problems by designing the s e c o n d
generation of polar orbiting passive microwave sensors, SMMR, to h a v e
ten channels. The availability of passive microwave data at a variety o f
frequencies and dual polarizations permit more accurate determination o f
sea ice characteristics.
Based on the knowledge gained from the ESMR and SMMR systems, t h e
third generation of polar orbiting passive microwave radiometers, SSM/I,
operates at frequencies which are further optimized for the m e a s u r e m e n t
of sea ice. Error analyses of ice parameters derived from SSM/I imagery
using an algorithm developed by researchers at NASA show g o o d
agreement (less than 5% difference in autumn through spring; near 1 0 %
in summer) with supplementary data sources (principally optical r e m o t e
sensing imagery). The analysis technique, however, has been found t o
underestimate
ice concentration
in areas of close pack ice a n d
overestimate ice concentrations in areas of open pack ice (Bjerkelund e t
al., 1990; Cavalieri et al., 1991; Steffen and Schweiger, 1991).
An
alternate algorithm has been developed jointly by the Canadian
Atmospheric Environment Service and the Institute for Space a n d
Terrestrial Science ("AES/ISTS algorithm") which incorporates w e a t h e r
and sea state corrections to aid in the retrieval of ice parameters (LeDrew
et al., 1992c).
In a comparative analysis between the two techniques,
Bjerkelund et al. (1990) found the AES/ISTS algorithm provided m o r e
accurate estimates of ice concentration, type, and extent than the NASA
Team algorithm.
Spaceborne passive microwave sensors are not without limitation,
however. Because of the long integration time required at the sensor t o
receive enough microwave energy to create an acceptable signal, passive
microwave images have very low spatial resolutions - commonly in t h e
range 30 km to 70 km.
This inhibits their application to all b u t
hemispheric-scale analyses. In the context of climate change this m i g h t
not be such a bad restriction, however. Since passive microwave images
tend to be reasonably small it is possible to include many more t e m p o r a l
instances in a long term study than it would be if the images were larger.
3.3. PASSIVE MICROWAVE DATA FOR SEA ICE MONITORING
In section 2, six sea ice climatic indicator parameters were evaluated. In
Table 3, the potential of passive microwave remote sensing to p r o v i d e
data for the measurement of these parameters is assessed.
The only
parameter not directly retrievable from multichannel passive microwave
imagery is ice thickness, although broad generalizations can be i n f e r r e d
from ice type (older ice tends to be thicker). Past and present passive
microwave sensors cannot meet the ideal spatial resolutions proposed f o r
each parameter in Table 3.
The sensors' coarse spatial resolutions
produces a noticeable "land contamination" effect where the s e n s o r ' s
field of view contains enough land (which generally has brightness
temperatures closer to those of sea ice than to those of open water) t o
give anomolously high ice concentration calculations (Parkinson, 1 9 9 1 ) .
This can have a particularly large effect in the Canadian Archipelago,
which may limit the usefulness of satellite passive microwave data in t h i s
region. The spatial averaging inherent to passive microwave imagery m a y
well be useful, however, for reducing the effects of localized anomalies
which could obscure any true climatic signals on broader regional a n d
hemispheric scales.
A finer discrimination of ice types than first-year/multiyear is a desirable
parameter not yet retrievable from passive microwave observations.
In
addition, the extraction of accurate ice information during seasonal
transitions is very limited.
Research should be initiated to develop
measurement techniques for microwave observations of snow c o v e r ,
surface albedo, ice production, and atmospheric water during the sea i c e
seasonal cycle (Carsey, 1984).
These limitations do not negate the usefulness of passive microwave
imagery, however, their potential implications must be acknowledged i n
any long-term climatic studies.
4. HYPERTEMPORAL IMAGE ANALYSIS
Thus far in this report, we have reviewed the need for acquiring a c c u r a t e
historical information on Arctic sea ice for climate change studies a n d
have established that remote sensing imagery from passive microwave
sensors is the only data source available to provide us with t h i s
information. We now focus our attention on the tools required to e x t r a c t
the change information from the remote sensing data.
Time is the fourth dimension; it is unlike the first three (spatial)
dimensions in that it is asymmetrical (it flows in only one direction) a n d
is difficult to envision, much less comprehend (we see the effects o f
time's passage, but we do not perceive it directly) (Saab a n d
Haythornthwaite, 1990).
Because global change is directly related t o
time, the ability to perform multitemporal analyses is vital to global
change research. Thus, multitemporal functionality should be c o n s i d e r e d
a fundamental
1991).
enhancement
to all spatial
analysis systems
(Stutheit,
Although satellite remote sensing has long been advocated for m o n i t o r i n g
surface processes through time, there has been little progress to date i n
spatiotemporal analysis of multitemporal imagery largely because of a
lack of repetitive and inexpensive data and because of a lack of a d e q u a t e
processing procedures (Davis et al., 1991).
The basic premise i n
temporal monitoring using remote sensing data is that changes in s u r f a c e
features must result in changes in sensed radiance values and that t h e
changes in radiance due to surface changes must be large with respect t o
radiance changes caused by other factors (e.g., atmospheric conditions
and sun angle) (Singh, 1989). Most temporal analyses reported to d a t e
examine the changes between an image pair; between time t and time t+1.
We distinguish such two or three period comparisons, or "change
detections", from studies of temporal series containing tens or e v e n
hundreds of images, which is the meaning we ascribe to " h y p e r t e m p o r a l
analysis".
Our intent in this section is to examine the existing status o f
hypertemporal image analysis and to suggest new approaches to analyzing
temporal image sequences which may have potential applications i n
climate change studies. In the discussion that follows, we briefly review
the status of methods for change detection before critiquing potential
hypertemporal analysis techniques.
4.1. CHANGE DETECTION
Two excellent reviews of methods for extracting changes from image p a i r s
can be found in Singh (1989) and Eastman and McKendry (1991), and a r e
summarized in Table 4. Although most of these techniques may be o f
limited use for long-term climatic studies, they may be useful in p r e processing temporal data for subsequent analysis by other methods. For
example, image differencing has been shown to effectively remove a n y
geographically fixed variations in sea ice extent and when the data to b e
analyzed are arranged as monthly averages (Cahalan and Chiu, 1 9 8 6 ) .
Differencing can also be used to remove the seasonal cycle from a t i m e
series (Hipel and McLeod, 1992). Both effects can be advantageous f o r
isolating regional and seasonal patterns in long spatial-temporal d a t a
series.
4.2. VISUALIZATION
The human mind is unequalled in its powers of spatial association a n d
pattern
recognition.
Computers
have unprecedented
powers o f
manipulating large masses of digital data.
Computers, then, a r e
effectively used as display tools for the presentation of unique views o f
data for the mind to analyze. This is the fundamental premise behind all
aspects of image processing, but most importantly that of the descriptive
and exploratory techniques known as computer visualization.
Underlying the concept of visualization is the idea that an observer c a n
build a mental model, the visual attributes of which represent d a t a
attributes in a definable manner. Computer graphics has an i m p o r t a n t
role in scientific visualization: it is not in generating elaborate, realistic
images of ideas; rather it is by creating abstractions of the data. T h e
abstraction process, if successful, helps to distinguish pattern from noise.
Choosing the appropriate graphics display representation can be the k e y
to critical and comprehensive application of the data, thus benefiting
subsequent analysis, processing, or decision making. The choice of t h e
representation ultimately depends on the research context (MacEachren
and Ganter, 1990; Wills et al., 1990; Robertson, 1991).
The most basic method of presenting the temporal elements of a
hypertemporal data set is to display each time slice as an i n d e p e n d e n t
view, ideally juxtaposed so that the viewer can simultaneously c o m p a r e
the patterns at all times in the series. This is a technique we call "static
image sequencing".
The focus here is on the status of the m a p p e d
phenomena at each discrete instant; the search for any patterns of c h a n g e
is left up to the viewer (Monmonier, 1990). This approach is a c o m m o n
method for displaying hypertemporal data because of its c o m p u t a t i o n a l
simplicity. The ice atlases compiled by Zwally et al. (1983a), Parkinson e t
al. (1987b), LeDrew et al. (1992c), and Gloersen et al. (1993) all p r o v i d e
excellent examples of static image sequencing.
The technique benefits
from the arrangement of as many temporal data slices as possible in a
single view (e.g., on a single page). The human mind has superior p a t t e r n
recognition powers within a single image, however, its capabilities f o r
image comparisons is somewhat more limited and reduced further still
when the image arrangement does not promote instantaneous visual
associations (Calkins, 1984).
The next logical step beyond visual examination of static image s e q u e n c e s
is the visualization of dynamic image sequences, or animation.
In a n
animation, temporally sequenced views of the data are displayed in q u i c k
succession so as to give the impression of motion. Animation is based o n
the principle that the eye-brain mechanism retains, momentarily, images
of objects it has seen after the objects have been removed. Therefore, if
the eye is shown a series of static views of objects at a rapid rate, with t h e
objects changing positions only slightly from frame to frame, the illusion
of fluid life-like motion is created by the brain (Campbell and Egbert,
1990).
The technique can provide new insights into the data to b e
analyzed. Gloersen (1990) constructed a time-lapse study of a sea i c e
feature of the Greenland Sea known as the "Odden" from 2-day i c e
concentration estimates over 9 years from SMMR image data.
The
animation allowed the extraction of previously unknown i n f o r m a t i o n
about the formation, growth and movement of this ice feature. In a n o t h e r
study, Assel and Ratkos (1991) created an interactive animation for t h e
seasonal ice cycle on the Great Lakes. Spatial and temporal variations o f
the ice characteristics quickly emerged and lead to a review of potential
forcing mechanisms.
4.3. CLASSIFICATION
If multispectral classification is based on the premise that image pixels
can be grouped according to unique spectral characteristics, then it w o u l d
appear that an image should also be classifiable according to u n i q u e
temporal
characteristics.
Two
requirements
of
multispectral
classification are: (i) a good understanding of the relationship b e t w e e n
the spectral data and the surface features; and (ii) the choice o f
appropriate statistical tools to discriminate relevant categories from t h e
data (Marceau, 1989). With corresponding changes from "spectral" t o
"temporal" this becomes an appropriate creed for h y p e r t e m p o r a l
classification.
Much work has already been done in relating passive
microwave brightness temperatures to ice characteristics and scientists
now have a reasonable understanding of some of the more general
relationships. In this section then, some appropriate statistical tools a r e
reviewed. We focus on unsupervised classification methods because t h e y
provide a more accurate description of the natural patterns in the d a t a
and because of the scarcity of adequate training data for supervised
techniques.
The de facto standard approach to multispectral image segmentation is
the maximum likelihood technique where each pixel is assigned to t h e
class in which it has the highest statistical probability of being a m e m b e r .
For hypertemporal applications, this method assumes a Gaussian t e m p o r a l
distribution of ice parameters. This assumption may hold for Arctic s e a
ice data where the seasonal cycle of growth and decay is approximately
symmetrical (Comiso and Zwally, 1984).
There are other approaches
to multispectral
image classification,
including some nonparametric
techniques which could also h a v e
applicability in the hypertemporal domain. For example, Key et al. ( 1 9 8 9 )
and Maslanik et al. (1990) describe the use of neural networks and SMMR
imagery to examine "second order" ice information (e.g., the onset a n d
duration of melt, ice pack metamorphosis, and timing of new ice g r o w t h )
and relate these to changes in ice condition over an annual cycle. They
conclude that the neural network approach to classification s h o w s
promise for sea ice applications because of its flexible pattern recognition
skills, fuzzy tolerances, and ability to use non-numeric information.
When reviewing classification methods for hypertemporal
imagery,
perhaps we can glean insight from the attempts which have been made t o
analyze hyperspectral data (remote sensing images acquired by imaging
spectrometers potentially including over 200 spectral channels per pixel).
The classification of such high spectral resolution data is clearly a
challenging task due to the large number of channels involved. Analysis
methods are required to extract the most information from each c h a n n e l
while minimizing
computation
time.
Two possible
classification
methodologies include full spectrum classification and feature selection
followed by classification.
Full spectrum classification c o m b i n e s
techniques for a parametric representation of the spectrum derived f o r
each pixel with similarity measures used for classification purposes, while
the second approach uses techniques and transformations for d a t a
reduction prior to classification (Staenz and Goodenough, 1 9 9 0 ) .
Techniques for full spectrum classification do exist, but they are n o t
generally found in standard image analysis software packages (Mazer e t
al., 1988; Cetin and Levandowski, 1991; Edwards et al., 1991).
The main criticism of full spectrum classifiers is that their c o m p u t a t i o n
time increases exponentially with an increase in the number of channels.
It is evident, however, that not every time slice or spectral band c o n t a i n s
unique information, in fact pixel values are often highly c o r r e l a t e d
between channels. Thus, a more efficient approach may be to employ a
data reduction strategy before classification.
Existing methods f o r
channel reduction include band averaging (Staenz and Goodenough,
1990), band-moment analysis (Rundquist and Di, 1989), statistical
channel separability measures (e.g., Sheffield, 1985; Labovitz, 1986; Gong
and Howarth, 1990; Mausel et al., 1990), and principal c o m p o n e n t s
analysis (Chavez and Kwarteng, 1989).
4.4. PRINCIPAL COMPONENTS ANALYSIS
Principal Components Analysis (PCA) is a commonly used technique i n
remote sensing image analysis. Traditionally it has been applied for image
enhancement and channel reduction, however, it has also been effectively
used in change detection studies (Table 5). An appealing aspect of PCA i n
the present context is that the technique is also widely accepted b y
climatologists (who frequently refer to it as eigenvector analysis o r
empirical orthogonal function analysis) as an effective data analysis t o o l
(e.g., Stidd, 1967; LeDrew, 1976; Kelly et al., 1982). This overlap b e t w e e n
disciplines has not been overlooked by scientists looking for t e m p o r a l
changes in both polar and non-polar environments.
In a principal components analysis, the measurement space is rotated i n
such a way as to maximize the variance between the variables. In r e m o t e
sensing, these
variables
can be multispectral
channels
and/or
hypertemporal images. The objectives in applying PCA are to u n d e r s t a n d
how the variables are interrelated and to assess the level of statistical
redundancy present.
In hypertemporal studies, the technique offers a
useful exploratory tool to analyze the interrelationships between regularly
sampled hypertemporal remote sensing data sets (Townshend et al.,
1985).
When images of different dates are compared with a view to mo n i t o r i n g
change, there will be a high correlation between them: those parts of t h e
scene which show an absence of correlation are of interest as t h e y
represent areas of change (Byrne and Crapper, 1980). By the very n a t u r e
of the method, the first component will always contain the m a x i m u m
amount of variation derived from the set of variables. Thus, the first
component is commonly referenced as "integrated brightness"; it provides
a good index of long-term variations since it is not affected by strong, b u t
localized, extremes which can influence large-scale averages and o b s c u r e
underlying trends (Kelly et al., 1982). Experiments using monthly d a t a
over an annual cycle have found that the second component tracks t h e
seasonal evolution quite well (Townshend et al., 1985; Tateishi a n d
Kajiwara, 1990).
The higher components represent increasingly local
variations and anomalies; they are often the ones selected for m o r e
detailed analysis (Walsh and Johnson, 1979b; Crane, 1983).
It is evident that PCA can be a powerful tool for the monitoring of p o l a r
climate with sequential remote sensing observations.
It allows for t h e
abstraction of change from a time series as a whole, not just a segment o f
it.
4.5. TREND SURFACE ANALYSIS
In climate change analyses we are interested in determining how much o f
a variable's response over time is real (i.e., a climatic signal) and h o w
much of it is local variation (i.e., climatic noise). If a regression line ( o r
curve) is fit to a time series plot of sea ice extent, the parameters of t h e
regression equation would give us some indication of a climatic signal,
while the residuals could be interpreted as a measure of local climatic
variability. This technique has been used extensively in climate r e s e a r c h
to extract trends from time series plots (e.g., Chapman and Walsh, 1992).
Image regression is a technique which has been applied to image pairs t o
differentiate between actual change and predicted change (Singh, 1 9 8 9 ;
Eastman and McKendry, 1991). The procedure is initiated by considering
pixel values from the image at time t to be samples of the i n d e p e n d e n t
variable and pixels from the time t+1 image as samples of the d e p e n d e n t
variable. A regression line is then fit to the data points. The c o m p u t e d
slope and intercept values are applied on a per-pixel basis to the first
image to produce a predicted image for time t+1. The actual a n d
predicted time t+1 images are then differenced or ratioed to isolate t h e
changed areas. The image regression technique is limited to pairwise
comparisons, however.
Trend surface analysis extends this concept into two dimensions. Instead
of fitting a regression line to a time series plot, a regression ( " t r e n d " )
surface is fit to a spatial-temporal data set (an image in our case). Ideal
trend surface images provide a smooth characterization of their i n p u t
data. When a trend surface has time as one of its independent variables,
it can be thought of as an isochronic or time contour map ( M o n m o n i e r ,
1990). Some studies have found residual images to be more useful f o r
pattern exploration than the trend image (e.g., Chorley and Haggett,
1965). Trend surface analyses have not been widely applied in climate
change studies although they have the potential for revealing significant
trends and regionalizations.
4.6. TIME SERIES ANALYSIS
In time series analysis, the deterministic components to a series (e.g.,
seasonality and trends) are identified and removed in a process called
"pre-whitening". The residual stochastic component in the series is t h e n
modelled using autoregressive and/or moving average functions (Hipel
and McLeod, 1992).
Although time series analyses have not b e e n
generally used in studies of Arctic sea ice or incorporated into image
analysis programs, they have been extensively applied by climatologists
and have a certain appeal to the problem at hand. This attraction c o m e s
not only from the modelling capabilities of time series equations, but a l s o
from their demonstrated proficiency at generating forecasts beyond t h e
measured data space.
In the 1970s there was a flurry of activity to extend the two-dimensional
nature of conventional time series analysis techniques into t h r e e
dimensions, i.e., spatial time series (Bennett, 1975; Martin and O e p p e n ,
1975; Bennett and Haining, 1976; Bennett, 1979; Pfeifer and Deutsch,
1980). A spatial time series is a multivariate model which attempts t o
simultaneously describe and forecast a collection of time series
representing spatially diverse regions, taking into account the spatial
autocorrelation inherent between neighbouring regions (Pfeifer a n d
Deutsch, 1980; Adamowski et al., 1985).
Unfortunately not m u c h
progress has been made in developing spatial time series since t h e n ,
largely because of the complexity of the model and the difficulty o f
parameter estimation (Curry, 1970; Parkes and Thrift, 1980).
5. DISCUSSION
Mysak (1990) proposed six working questions which should b e
considered when assessing the suitability of sea ice (or any o t h e r
geophysical feature) as a proxy indicator of climate change. Based on t h e
review presented here, we are now in a position to suggest some answers.
1 . Through what mechanisms
climate change?
•
can sea ice be used as an indicator
Sea ice has intimate bi-directional feedback mechanisms with t h e
earth's atmospheric and oceanic circulation systems. The climatic
role that sea ice plays is generally expected to be based on i t s
surface albedo, insulative properties, thermal capacitance, and b r i n e
capacitance.
2 . What specific geophysical parameters are required to characterize
behaviour of these indicators?
•
of
the
The potential sea ice indicators for climate change identified i n
section 2 are extent, type, concentration, thickness, dynamics, a n d
seasonality.
These were shown to have distinct
regional
distributions.
3 . With what accuracy and precision do these indicators
measured? Over what spatial and temporal scales?
•
need
to b e
From the analyses in section 2 and Barber et al. (1992b), we find t h e
following ideals:
Indicator
ice extent
ice type
ice concentration
ice thickness
ice dynamics
onset of melt
Ideal Spatial
Requirements
Ideal Temporal
Requirements
>1 km
> 1 km
> 1 km
10 km
10 km
10 km
3 days
3 days
3 days
3 days
1 day
3 days
4 . What data sources are currently available for the measurement of t h e s e
indicators?
•
An evaluation of remote sensing systems indicated that passive
microwave imagery is the only data source with the necessary
historical record and all weather/day-night sensing capabilities.
5 . Where are the discrepancies
available?
•
between
what is required
and what is
Of all the parameters listed in Question 2, the only one passive
microwave imagers cannot measure ice thickness. Past and p r e s e n t
microwave sensors do not meet the required spatial resolutions o f
any indicator, however, this may not be critical in h e m i s p h e r i c
analyses.
6 . What areas of product development/research are needed to use sea i c e
as an indicator of climate change?
•
If effective use is to be made of the historical passive microwave
data record, new hypertemporal image analysis techniques must b e
developed.
Table 6 lists the potential hypertemporal image analysis f u n c t i o n s
applicable to examining sea ice parameters on remote sensing imagery. It
is clear that there are many options for the analysis of h y p e r t e m p o r a l
image sequences.
How do we proceed when faced with analyzing a
hypertemporal data set? Which analysis methods have the potential t o
help us extract meaningful information from these data? We propose a
hierarchy of analysis functions in Table 7 which may provide s o m e
guidance.
The primary techniques were selected based on t h e i r
demonstrated capabilities of scene processing, information extraction a n d
process description. Primary functions are likely to be the most useful f o r
general explorations of the temporal nature of the data.
Where
interesting patterns of change are found the secondary techniques may b e
helpful in explaining their nature. Re-application of the primary m e t h o d s
using a different parameter set or to different subsets of the data m a y
also be useful. The tertiary functions can be considered to be the " w o r k e r
routines". They have limited applicability, but are frequently useful w h e n
applied to pre-processing or post-processing the data.
During the course of an entire hypertemporal analysis exercise, it is
probable that techniques from each category will be applied in a variety
of sequences until the analyst is satisfied with the meanings a n d
definitions which have been derived from the imagery. Much depends o n
the nature of both the data and the problem.
The key is in effective
combination of the computer's data processing capabilities with t h e
human mind's pattern recognition strengths.
6. CONCLUSIONS
In this review, we have shown the critical role that polar sea ice plays i n
the world's climatic systems; it is clear that the Arctic is a region o f
sensitive indicators of climatic change. An analysis of potential r e m o t e
sensing techniques indicated that sea ice extent, type, c o n c e n t r a t i o n ,
dynamics, and melt onset have been monitored with sufficient ( a l t h o u g h
not ideal) spatial and temporal resolution by orbital passive microwave
sensors to make these variables potential climatic indicators. The passive
microwave data record has revealed large interannual variability in e a c h
of these parameters and few indications of long term trends. Examination
of these images also show, in addition to the large temporal variability o f
these data, that they exhibit strong regionalisms and seasonalities. This
information can be used to focus attention in those seasons and places
when changes may be most apparent.
Hypertemporal image analysis of remote sensing data is still in its infancy,
largely because of a lack of repetitive and inexpensive data, and a d a p t a b l e
processing algorithms. We are on the brink of temporal data explosion a s
a longer historical record is accumulated for existing sensors, as r e m o t e
sensing scientists begin to explore the utility of "less conventional"
sensors for which the data are relatively inexpensive (e.g., AVHRR a n d
passive microwave), and as new, high data-rate instruments are b r o u g h t
on-line towards the end of this century (e.g., EOS and RADARSAT).
Existing image processing tools need to be adapted, and new t e c h n i q u e s
developed, to handle these data.
Application of the hypertemporal image processing techniques reviewed
here is not necessarily restricted to studies of sea ice. Indeed, the a d v e n t
of satellite remote sensing has presented us with intriguing o p p o r t u n i t i e s
to investigate, in good spatial and temporal detail, even the most r e m o t e
or rarely visited areas of the earth's surface.
However, most existing
climate change studies which are derived from remotely sensed data h a v e
been temporal analyses based on aggregation of the spatial i n f o r m a t i o n
from each image, generally into a single parameter (e.g., ice extent). With
the exception of a few recent efforts, not many published results a r e
based on detailed spatial analyses of the type which are possible t h r o u g h
image processing methods. This is partly attributable to the inability o f
standard image analysis software to work in both a spatial and a t e m p o r a l
mode. We have examined the issue of hypertemporal image analysis a n d
conclude that many of the techniques used in multispectral image
processing, such as classification and principal components analysis, a r e
also applicable in the time domain. We call for their increased use in t r u e
spatial-temporal analyses and for the development of new image
processing algorithms specifically for modelling and forecasting change.
ACKNOWLEDGMENTS
We would like to thank Dave Barber, Doug Dudycha, Barry Goodison, Phil
Howarth, and Ed Jernigan for their valuable comments on an earlier
version of this paper.
REFERENCES
Adamowski, Kazimierz, Fadil B. Mohamed, Nicholas R. Dalezios and Louis
G. Birta, 1985. "Space-Time ARIMA Modelling of Precipitation Time
Series", in Multivariate Analysis of Hydrologic Processes, (Shen, Hsieh
Wen, et al., eds.). Fort Collins, Colorado: Engineering Research Center,
Colorado State University. pp. 217-227.
Alfultis, Michael A. and Seelye Martin, 1987. "Satellite Passive Microwave
Studies of the Sea of Okhotsk Ice Cover and its Relation to Oceanic
Processes, 1978-1982", J. Geophysical Research, Vol. 92, No. C12, pp.
13013-13028.
Anderson, Mark R., 1987. "The Onset of Spring Melt in First-Year Ice
Regions of the Arctic as Determined from Scanning Multichannel
Microwave Radiometer Data for 1979 and 1980", J. Geophysical Research,
Vol. 92, No. C12, pp. 13153-13163.
Assel, R.A. and J.M. Ratkos, 1991. "Animation of the Normal Ice Cycle of
the Laurentian Great Lakes of North America", Proceedings, Seventh
International Conference on Interactive Information and Processing
Systems for Meteorology, Hydrology, and Oceanography, January 1991,
New Orleans LA., pp. 331-335.
Barber, D.G., E.F. LeDrew, D.G. Flett, M. Shokr and J. Falkingham, 1 9 9 2 a .
"Seasonal and Diurnal Variations in SAR Signatures of Landfast Sea Ice",
IEEE Transactions on Geoscience and Remote Sensing, Vol. 30, No. 3, p p .
638-642.
Barber, David G., Michael J. Manore, Thomas A. Agnew, Harold Welch,
Eric D. Soulis and Ellsworth F. LeDrew, 1992b. "Science Issues Relating to
Marine Aspects of the Cryosphere: Implications for Remote Sensing", C a n .
J. Remote Sensing, Vol. 18, No. 1, pp. 46-55.
Barry, R.G., 1983. "Arctic Ocean Ice and Climate: Perspectives on a
Century of Polar Research", Annals of the Association of American
Geographers, Vol. 73, No. 4, pp. 485-501.
Bennett, R.J., 1975 "The Representation and Identification of SpatioTemporal Systems: An Example of Population Diffusion in North-West
England", Transactions, Institute of British Geographers, Vol. 66, pp. 7394.
Bennett, R.J., 1979.
Spatial Time Series. London: Pion. 674 pp.
Bennett, R.J. and R.P. Haining, 1976. Space-Time Models: An Introduction
to Concepts. London: University College Department of Geography,
Occasional Paper No. 28. 35 pp.
Bjerkelund, C.A., R.O. Ramseier and I.G. Rubinstein, 1990. "Validation of
the SSM/I and AES/York Algorithms for Sea Ice Parameters", in Ackley, S.F.
and W.F. Weeks (eds.), Sea Ice Properties and Processes, Proceedings:
W.F. Weeks Sea Ice Symposium, San Francisco CA, December 1988. U.S.
Army Cold Regions Research and Engineering Lab, Monograph M90-1, pp.
206-208.
Brown, Ross D. and Phil Cote, 1992. "Interannual Variability of Landfast
Ice Thickness in the Canadian High Arctic, 1950-89", Arctic, Vol. 45, No.
3, pp. 273-284.
Burns, Barbara A. and Thomas Viehoff, 1991. "Merging AVHRR and SSM/I
Data: An Example for Ice Concentration Analyses in the Weddell Sea",
Proceedings, 14th Canadian Symposium on Remote Sensing, Calgary
Alberta, May 1991. Ottawa: Canadian Remote Sensing Society, pp. 371374.
Byrne, G.F. and P.F. Crapper, 1980. "Land Cover Change Detection by
Principal Components Analysis of Multitemporal MSS Data: The Presence
of Clouds", Proceedings, 14th International Symposium on Remote Sensing
of Environment, April 1980, San Jose Costa Rica, pp. 1175-1382.
Byrne, G.F., P.F. Crapper and K.K. Mayo, 1980. "Monitoring Land-Cover
Change by Principal Component Analysis of Multitemporal Landsat
Imagery", Remote Sensing of Environment, Vol. 10, pp. 175-184.
Cahalan, Robert F. and Long S. Chiu, 1986. "Large-Scale Short-Period Sea
Ice Atmosphere Interaction", J. Geophysical Research, Vol. 91, No. C9, pp.
10709-10717.
Calkins, Hugh W., 1984. "Space-Time Data Display Techniques",
Proceedings, First International Symposium on Spatial Data Handling,
Zürich, August 1984. Zürich: Univsersität Zürich-Irchel, Vol. 2, pp. 324331.
Campbell, Craig S. and Stephen L. Egbert, 1990. "Animated Cartography:
Thirty Years of Scratching the Surface", Cartographica, Vol. 27, No. 2, pp.
24-46.
Carsey, F.D., 1984. "Prospects for Describing and Monitoring From Space
the Elements of the Seasonal Cycle of Sea Ice", Annals of Glaciology, Vol.
5, pp. 37-42.
Carsey, F.D., 1985. "Summer Arctic Sea Ice Character from Satellite
Microwave Data", J. Geophysical Research, Vol. 90, No. C3, pp. 50155034.
Cavalieri, D.J. and C.L. Parkinson, 1987. "On the Relationship Between
Atmospheric Circulation and the Fluctuations in the Sea Ice Extents of the
Bering and Okhotsk Seas", J. Geophysical Research, Vol. 92, No. C7, pp.
7141-7162.
Cavalieri, D.J., J.P. Crawford, M.R. Drinkwater, D.T. Eppler, L.D. Farmer,
R.R. Jentz and C.C. Wackerman, 1991. "Aircraft Active and Passive
Microwave Validation of Sea Ice Concentration from the Defense
Meteorological Satellite Program Special Sensor Microwave Imager", J .
Geophysical Research, Vol. 96, No. C12, pp. 21989-22008.
Cetin, Haluk and Donald W. Levandowski, 1991. "Interactive Classification
and Mapping of Multi-Dimensional Remotely Sensed Data Using nDimensional Probability Density Functions (nPDF)", Photogrammetric
Engineering & Remote Sensing, Vol. 57, No. 12, pp. 1579-1587.
CGCP (Canadian Global Change Program), 1989. Contribution of Satellite
Observations to the Canadian Global Change Program, Ottawa: The Royal
Society of Canada and the Canada Centre for Remote Sensing, Canadian
Global Change Program Report No. 3, 48 pp.
Chapman, William L. and John E. Walsh, 1992. "Recent Variations of Sea
Ice and Air Temperature in High Latitudes", Bull. American Meteorological
Society, December 1992, in press.
Chavez, Pat S., Jr. and Andrew Yaw Kwarteng, 1989 "Extracting Spectral
Contrast in Landsat Thematic Mapper Image Data Using Selective Principal
Components Analysis", Photogrammetric Engineering & Remote Sensing,
Vol. 55, No. 3, pp. 339-348.
Chorley, R.J. and P. Haggett, 1965. "Trend-Surface Mapping in
Geographical Research", in B.J.L. Berry, Spatial Analysis . Englewood
Cliffs: Prentice Hall. pp. 195-217.
Clarke, R. Allyn and E. Peter Jones, 1989. "The Arctic Ocean: Its Role in
the Global Climate Engine", Proceedings, Symposium on the Arctic and
Global Change, Ottawa Canada, October 25-27, 1989. Washington D.C.:
The Climate Institute. pp. 42-57.
Comiso, J.C., 1986. "Characteristics of Arctic Winter Sea Ice From
Satellite Multispectral Microwave Observations", J. Geophysical Research,
Vol. 91, No. C1, pp. 975-994.
Comiso, J.C. and C.W. Sullivan, 1986. "Satellite Microwave and In Situ
Observations of the Weddell Sea Ice Cover and its Marginal Ice Zone",
Journal of Geophysical Research, Vol. 91, No. C8, pp. 9663-9681.
Comiso, J.C. and H.J. Zwally, 1982. "Antarctic Sea Ice Concentrations
Inferred from Nimbus 5 ESMR and Landsat Imagery", J. Geophysical
Research, Vol. 87, No. C8, pp. 5836-5844.
Comiso, J.C. and H.J. Zwally, 1984. "Concentration Gradients and
Growth/Decay Characteristics of the Seasonal Sea Ice Cover", J .
Geophysical Research, Vol. 89, No. C5, pp. 8081-8103.
Crane, R.G., 1978. "Seasonal Variations in Sea Ice Extent in the Davis
Strait-Labrador Sea Area and Relationships with Synoptic Scale
Atmospheric Circulation", Arctic, Vol. 31, No. 4, pp. 434-447.
Crane, R.G., 1983. "Atmosphere - Sea Ice Interactions in the
Beaufort/Chukchi Sea and in the European Sector of the Arctic", J .
Geophysical Research, Vol. 88, No. C7, pp. 4505-4523.
Crane, Robert G. and Mark R. Anderson, 1989. "Spring Melt Patterns in
the Kara/Barents Sea: 1984", GeoJournal, Vol. 18, No. 1, pp. 25-33.
Crane, R.G., R.G. Barry and H.J. Zwally, 1982. "Analysis of AtmosphereSea Ice Interactions in the Arctic Basin using ESMR Microwave Data", Int.
J. Remote Sensing, Vol. 3, No. 3, pp. 259-276.
Curry, Leslie, 1970. "Applicability of Space-Time Moving Average
Forecasting", Regional Forecasting (Chisholm, et al., eds.), Hamden CT:
Archon, 470 pp.
Daley, R., 1989 "General Circulation Models", Proceedings, Symposium on
the Arctic and Global Change, Ottawa Canada, October 25-27, 1989.
Washington D.C.: The Climate Institute. pp. 32-33.
Davis, Frank W., Dale A. Quattrochi, Merrill K. Ridd, Nina S-N Lam,
Stephen J. Walsh, Joel C. Michaelsen, Janet Franklin, Douglas A. Stow,
Chris J. Johannsen and Carol A. Johnston, 1991. "Environmental Analysis
Using Integrated GIS and Remotely Sensed Data: Some Research Needs
and Priorities", Photogrammetric Engineering & Remote Sensing, Vol. 57,
No. 6, pp. 689-697.
Dey, B., 1981. "Monitoring Winter Sea Ice Dynamics in the Canadian
Arctic with NOAA-TIR Images", J. Geophysical Research, Vol. 86, No. C4,
pp. 3223-3235.
Eastman, J. Ronald and Jean E. McKendry, 1991. Change and Time Series
Analysis, Volume 1 of Explorations in Geographic Information Systems
Technology, Geneva Switzerland: United Nations Institute for Training
and Research, 85 p.
Edwards, Geoffrey, Annick Jaton and Keith P.B. Thomson, 1991.
"Manipulating Hyperspectral Data", Proceedings, 14th Canadian
Symposium on Remote Sensing, Calgary Alberta, May 1991.
Canadian Remote Sensing Society, pp. 356-359.
Ottawa:
Emery, W.J., M. Radebaugh, C.W. Fowler, D. Cavalieri and K. Steffen, 1991.
"A Comparison of Sea Ice Parameters Computed from Advanced Very High
Resolution Radiometer and Landsat Satellite Imagery and From Airborne
Passive Microwave Radiometry", J. Geophysical Research, Vol. 96, No. C12,
pp. 22075-22085.
Eyton, J. Ronald, 1991. "Rate-of-Change Maps", Cartography and
Geographic Information Systems, Vol. 18, No. 2, pp. 87-103.
Fung, Tung and Ellsworth LeDrew, 1987. "Application of Principal
Components Analysis to Change Detection", Photogrammetric Engineering
and Remote Sensing, Vol. 53, No. 12, pp. 1649-1658.
Gloersen, P. 1990. "A Time-Lapse Motion Study of the 'Odden' During
1978-1987 as Observed with the Nimbus-7 Scanning Multichannel
Microwave Radiometer", in Ackley, S.F. and W.F. Weeks (eds.), Sea Ice
Properties and Processes, Proceedings: W.F. Weeks Sea Ice Symposium,
San Francisco CA, December 1988. U.S. Army Cold Regions Research and
Engineering Lab, Monograph M90-1, pp. 177-179.
Gloersen, Per and William J. Campbell, 1988. "Variations in the Arctic,
Antarctic, and Global Sea Ice Covers During 1978-1987 as Observed with
the Nimbus-7 Scanning Multichannel Microwave Radiometer", J .
Geophysical Research, Vol. 93, No. C9, pp. 10666-10674.
Gloersen, Per, William J. Campbell, Donald J. Cavalieri, Josefino C. Comiso,
Claire L. Parkinson and H. Jay Zwally, 1993. Arctic and Antarctic Sea Ice,
1978-1987: Satellite Passive-Microwave Observations and Analysis,
Washington DC: National Aeronautics and Space Administration, NASA SP511, 290 pp.
Gong, Peng and Philip J. Howarth, 1990. "A Comparison of Two FeatureSelection Procedures for Supervised Land-Cover Classification",
Proceedings, 13th Canadian Symposium on Remote Sensing, Fredericton,
N.B., 16-19 July, pp. 190-202.
Hare, F. Kenneth, 1982. "Confusion About Climate:
Canadian Geographic, Vol. 101, No. 6, pp. 48-53.
What is Happening?",
Hipel, Keith W. and A. Ian McLeod, 1992. Time Series Modelling of Water
Resources and Environmental Systems, Amsterdam: Elsevier.
Houghton, John T., 1990. "Scientific Assessment of Climate Change:
Summary of the IPCC Working Group I Report", Climate Change: Science,
Impacts and Policy, Proceedings of the Second World Climate Conference,
29 October 1990. Cambridge: Cambridge University Press, pp. 23-45.
Howarth, Philip J. and Gregory M. Wickware, 1981. "Procedures for
Change Detection Using Landsat Digital Data", Int. J. Remote Sensing, Vol.
2, No. 3, pp. 277-291.
Ikeda, M., 1990a. "Feedback Mechanism Among Decadal Oscillations in
Northern Hemisphere Atmospheric Circulation, Sea Ice, and Ocean
Circulation", Annals of Glaciology, Vol. 14, pp. 120-123.
Ikeda, M., 1990b. "Decadal Oscillations of the Air-Ice-Ocean System in
the Northern Hemisphere", Atmosphere-Ocean, Vol. 28, No. 1, pp. 106139.
Ingebritsen, S.E. and R.J.P. Lyon, 1985. "Principal Components Analysis of
Multitemporal Image Pairs", Int. J. Remote Sensing, Vol. 6, No. 5, pp. 687696.
Jacobs, John D. and John P. Newell, 1979. "Recent-Year-to-Year Variations
in Seasonal Temperatures and Sea Ice Conditions in the Eastern Canadian
Arctic", Arctic, Vol. 32, No. 4, pp. 345-354.
Johnson, Douglas
its Development
Processes. Earth
002, Department
D., 1989. Remote Sensing of Polar Regions: A Review of
and Application to Atmosphere-Ice-Ocean Interaction
Observations Laboratory Technical Report ISTS-EOL-TR89of Geography, University of Waterloo, 72 pp.
Kelly, P.M., P.D. Jones, C.B. Sear, B.S.G. Cherry and R.K. Tavakol, 1982.
"Variations in Surface Air Temperatures: Part 2. Arctic Regions, 18811980", Monthly Weather Review, Vol. 110, pp. 71-83.
Kelly, P.M., C.M. Goodess and B.S.G. Cherry, 1987. "The Interpretation of
the Icelandic Sea Ice Record", J. Geophysical Research, Vol. 92, No. C10,
pp. 10835-10843.
Key, J., J. A. Maslanik, and A. J. Schweiger, 1989. "Classification of
Merged AVHRR and SMMR Arctic Data with Neural Networks",
Photogrammetric Engineering & Remote Sensing, Vol. 55, No. 9, pp. 13311338.
Kukla, G., 1993. "Glacial Start and Global Warming: What to Watch",
Proceedings, Snow Watch '92: Detection Strategies for Snow and Ice,
Niagara-on-the-Lake Canada, March 1993. Waterloo Canada: Earth
Observations Laboratory, Institute for Space and Terrestrial Science, ISTSEOL-TR93-005, pp. 93-110.
Labovitz, M.L., 1986 "Issues Arising from Sampling Designs and Band
Selection in Discriminating Ground Reference Attributes Using Remotely
Sensed Data", Photogrammetric Engineering & Remote Sensing, Vol. 52,
No. 2, pp. 201
Ledley, Tamara Shapiro, 1990. "The Impact of Milankovitch Solar
Radiation Variations on Sea Ice and Air Temperature in a Coupled EnergyBalance Climate-Sea-Ice Model", Annals of Glaciology, Vol. 14, pp. 144147.
LeDrew, Ellsworth Frank, 1976. Physical Mechanisms Responsible for the
Major Synoptic Systems in the Eastern Canadian Arctic in the Winter and
Summer of 1973. National Center for Atmospheric Research, University of
Colorado, NCAR Cooperative Thesis No. 38, 205 pp.
LeDrew, Ellsworth F., Douglas Johnson and James A. Maslanik, 1991. "An
Examination of Atmospheric Mechanisms that may be Responsible for the
Annual Reversal of the Beaufort Sea Ice Field", Int. J. Climatology, Vol. 11,
pp. 841-859
LeDrew, Ellsworth F., 1992a. "Polar Regions, Influence on Climate
Variability and Change", in Encyclopedia of Earth System Science, Vol. 3
(Nierenberg, William A., ed.), Academic Press.
LeDrew, Ellsworth, 1992b. "The Role of Remote Sensing in the Study of
Atmosphere-Cryosphere Interactions in the Polar Basin", Canadian
Geographer, Vol. 36, No. 4, pp. 337-350.
Lemke, P. E.W. Trinkl and K. Hasselmann, 1980. "Stochastic Dynamic
Analysis of Polar Ice Variability", J. Physical Oceanography, Vol. 10, pp.
2100-2120.
Lillesand, Thomas M. and Ralph W. Kiefer, 1987. Remote Sensing and
Image Interpretation (2nd ed.). New York: Wiley, 721 pp.
Lodwick, G.D., 1979. "Measuring Ecological Changes in Multitemporal
Landsat Data Using Principal Components", Proceedings, 13th
International Symposium on Remote Sensing of Environment, April 1979,
Ann Arbor MI, pp. 1131-1141.
MacEachren, Alan M. and John H. Ganter, 1990. "A Pattern Identification
Approach to Cartographic Visualization", Cartographica, Vol. 27, No. 2,
pp. 64-81.
Marceau, Danielle, 1989. A Review of Image Classification Procedures
with Special Emphasis on the Grey-Level Cooccurrence Matrix Method for
Texture Analysis, ISTS-EOL-TR89-007, 59 pp.
Martin, R.L. and J.E. Oeppen, 1975. "The Identification of Regional
Forecasting Models Using Space-Time Correlation Functions",
Transactions, Institute of British Geographers, Vol. 66, pp. 95-118.
Maslanik, James, Jeff Key and Axel Schweiger, 1990. "Neural Network
Identification of Sea-Ice Seasons in Passive Microwave Data", Proceedings,
IGARSS'90, Washington D.C., 20 May 1990. New York: IEEE, pp. 12811284.
Massom, Robert A., 1991. Satellite Remote Sensing of Polar Regions.
London: Belhaven, 307 pp.
Mausel, P.W., W.J. Kramber, and J.K. Lee, 1990. "Optimum Band Selection
For Supervised Classification of Multispectral Data", Photogrammetric
Engineering & Remote Sensing, Vol. 56, No. 1, pp. 55 - 60.
Maykut, Gary A. and Norbert Untersteiner, 1971. "Some Results from a
Time-Dependent Thermodynamic Model of Sea Ice", J. Geophysical
Research, Vol. 76, No. 6, pp. 1550-1575.
Mazer, Alan S., Miki Martin, Meemong Lee and Jerry S. Solomon, 1988.
"Image Processing Software for Imaging Spectrometry Data Analysis",
Remote Sensing of Environment, Vol. 24, pp. 201-210.
Monmonier, Mark, 1990. "Strategies for the Visualization of Geographic
Time-Series Data", Cartographica, Vol. 27, No. 1, pp. 30 - 45.
Mysak, L., 1990. "Current and Future Trends in Arctic Climate Research:
Can Changes of the Arctic Sea-Ice be Used as an Early Indicator of Global
Warming?", personal notes from a Canadian Ice Working Group workshop,
McGill University, November 1990.
Niebauer, H.J., 1983. "Multiyear Sea Ice Variability in the Eastern Bering
Sea: An Update", J. Geophysical Research, Vol. 88, No. C5, pp. 27332742.
Orheim, Olav and Baerbel Lucchitta, 1990. "Investigating Climate Change
by Digital Analysis of Blue Ice Extent on Satellite Images of Antarctica",
Annals of Glaciology, Vol. 14, pp. 211-215.
Parkes, Don and Nigel Thrift, 1980.
Wiley. 527 pp.
Times, Spaces, and Places. Chichester:
Parkinson, Claire L., 1989. "On the Value of Long-Term Satellite Passive
Microwave Data Sets for Sea Ice/Climate Studies", GeoJournal, Vol. 18,
No. 1, pp. 9-20.
Parkinson, Claire L., 1991. "Interannual Variability of the Spatial
Distribution of Sea Ice in the North Polar Region", J. Geophysical Research,
Vol. 96, No. C3, pp. 4791-4801.
Parkinson, C.L., J.C. Comiso, H.J. Zwally, D.J. Cavalieri, P. Gloersen and
W.J. Campbell, 1987b. Arctic Sea Ice, 1973-1976: Satellite Passive
Microwave Observations, Washington DC: National Aeronautics and Space
Administration, NASA SP-489, 296 pp.
Parkinson, Claire L. and Donald J. Cavalieri, 1989. "Arctic Sea Ice 19731987: Seasonal, Regional, and Interannual Variability", J. Geophysical
Research, Vol. 94, No. C10, pp. 14499-14523.
Pfeifer, Phillip E. and Stuart Jay Deutsch, 1980. "A Three-Stage Iterative
Procedure for Space-Time Modelling", Technometrics, Vol. 22, No. 1, pp.
35-47.
Robertson, Philip K., 1991. "A Methodology for Choosing Data
Representations", IEEE Trans. Computer Graphics & Applications, May
1991, pp. 56-67.
Roots, Fred, 1989. "Environmental Issues Related to Climate Change in
Northern High Latitudes", Proceedings, Symposium on the Arctic and
Global Change, Ottawa Canada, October 25-27, 1989. Washington D.C.:
The Climate Institute. pp. 6-31.
Ropelewski, Chester F., 1983. "Spatial and Temporal Variations in
Antarctic Sea-Ice (1973-82)", J. Climate and Applied Meteorology, Vol.
22, pp. 470-473.
Rothrock, D.A., Donald R. Thomas and Alan S. Thorndike, 1988.
"Principal Component Analysis of Satellite Passive Microwave Data Over
Sea Ice", J. Geophysical Research, Vol. 93, No. C3, pp. 2321-2332.
Rundquist, Donald C. and Liping Di, 1989. "Band-Moment Analysis of
Imaging-Spectrometer Data", Photogrammetric Engineering & Remote
Sensing, Vol. 55, No. 2, pp. 203-208.
Saab, David A. and Thomas W. Haythornthwaite, 1990. "Application of
Spatio-Temporal Database Models to Street Network Files", Proceedings:
GIS for the 1990s; Ottawa Canada. Ottawa: Canadian Institute of
Surveying and Mapping. pp. 1072-1085.
Serreze, Mark C., Roger G. Barry and Alfred S. McLaren, 1989. "Seasonal
Variations in Sea Ice Motion and Effects on Sea Ice Concentration in the
Canada Basin", J. Geophysical Research, Vol. 94, No. C8, pp. 10955-10970.
Sheffield, Charles, 1985 "Selecting Band Combinations from Multispectral
Data", Photogrammetric Engineering & Remote Sensing, Vol. 51, No.12,
AP6, pp. 681
Singh, Ashbindu, 1989. "Digital Change Detection Techniques Using
Remotely-Sensed Data", Int. J. Remote Sensing, Vol. 10, No. 6, pp. 9891003.
Staenz, K. and D.G. Goodenough, 1990. "Airborne Imaging Spectrometer
Data Analysis Applied to an Agricultural Data Set", Global and
Environmental Monitoring Techniques and Impacts, Proceedings: ISPRS
Commission VII Mid-Term Symposium: September 1990, Victoria Canada.
International Society for Photogrammetry and Remote Sensing, pp. 552559.
Steffen, Konrad and Axel Schweiger, 1991. "NASA Team Algorithm for Sea
Ice Concentration Retrieval from Defense Meteorological Satellite Program
Special Sensor Microwave Imager: Comparison with Landsat Satellite
Imagery", J. Geophysical Research, Vol. 96, No. C12, pp. 21971-21987.
Stidd, Charles K., 1967. "The Use of Eigenvectors for Climatic Estimates",
J. Applied Meteorology, Vol. 6, pp. 255-264.
Stutheit, Juliann, 1991. "Temporal GIS Investigates Global Change", G I S
World, Vol. 4, No. 9 (Dec. 1991), pp. 68-72.
Tateishi, Ryutaro and Koji Kajiwara, 1990. "Global Land Cover Monitoring
by NOAA GVI Data", Global and Environmental Monitoring Techniques and
Impacts, Proceedings: ISPRS Commission VII Mid-Term Symposium,
September 1990, Victoria Canada. International Society for
Photogrammetry and Remote Sensing, pp. 34-41.
Thomas, D.R. and D.A. Rothrock, 1989. "Blending Sequential Scanning
Multichannel Microwave Radiometer and Buoy Data Into a Sea Ice Model",
J. Geophysical Research, Vol. 94, No. C8, pp. 10907-10920.
Townshend, John R.G., Thomas E. Goff and Compton J. Tucker, 1985.
"Multitemporal Dimensionality of Images of Normalized Difference
Vegetation Index at Continental Scales", IEEE Transactions on Geoscience
and Remote Sensing, Vol. GE-23, No. 6, pp. 888-895.
Vowinckel, E. and S. Orvig, 1970. "The Climate of the North Polar Basin",
in Orvig, S. (ed.) Climates of the Polar Regions. Amsterdam: Elsevier, pp.
129-252.
Walsh, John E., 1983. "The Role of Sea Ice in Climatic Variability:
Theories and Evidence", Atmosphere-Ocean, Vol. 21, No. 3, pp. 229-242.
Walsh, John E. and Claudia M. Johnson, 1979a. "Interannual Atmospheric
Variability and Associated Fluctuations in Arctic Sea Ice Extent", J .
Geophysical Research, Vol. 84, No. C11, pp. 6915-6928.
Walsh, John E. and Claudia M. Johnson, 1979b. "An Analysis of Arctic Sea
Ice Fluctuations, 1953-77", J. Physical Oceanography, Vol. 9, pp. 580-591.
Washington, W.M. and G.A. Meehl, 1989. "Climate Sensitivity Due to
Increased CO2: Experiments with a Coupled Atmosphere and Ocean
General Circulation Model", Climate Dyn., Vol. 4, pp. 1-38.
Weeks, W.F., 1976. "Sea Ice Conditions in the Arctic", AIDJEX Bulletin, No.
34, Dec. 1976, pp. 173-205.
Wills, G., R. Bradley, J. Haslett and A. Unwin, 1990. "Statistical
Exploration of Spatial Data", Proceedings, 4th International Symposium on
Spatial Data Handling, Zurich Switzerland, July 23-27, pp. 491-500.
Zwally, H.J., J.C. Comiso, C.L. Parkinson, W.J. Campbell, F.D. Carsey and P.
Gloersen, 1983a. Antarctic Sea Ice, 1973-1976: Passive Microwave
Observations, Washington DC: National Aeronautics and Space
Administration, NASA SP-459, 206 pp.
Zwally, H. Jay, C.L. Parkinson, J.C. Comiso, 1983b. "Variability of
Antarctic Sea Ice and Changes in Carbon Dioxide", Science, Vol. 220, No.
4601, pp. 1005-1012.
Figure 1:
Minimum and Maximum Arctic Ice Extents
Averages of February
between 1978-1987.
Figure 2:
Table 1:
and September
SMMR Ice C o n c e n t r a t i o n s
Principal Arctic Ice Drift Patterns
Climatic Influence of Sea Ice
Climatic Property
Physical Properties
Climatic Influence
surface albedo
extent, type,
concentration
sea ice increases the surface albedo from 0.10-0.15
over open water to 0.4-0.6 in the case of snow-free ice
or 0.6-0.9 in the case of snow-covered ice
insulative properties
concentration,
thickness, extent,
motion dynamics
sea ice inhibits exchanges of heat, mass and
momentum between the atmosphere and ocean
thermal capacitance
type (age), seasonality, sea ice acts as a thermal reservoir which delays the
thickness
seasonal temperature cycle
brine capacitance
type (age), motion
dynamics, thickness
sea ice alters the ocean's salinity distribution, both
vertically by the expulsion of brine during freezing and
horizontally by the large-scale transport of low-salinity
ice
Table 2:
Sea Ice as a Climatic Indicator
Sea Ice
Parameter
Observations
Potential*
Indicators
extent
• has been shown to
influence weather and to
be influenced by weather
• extreme variability;
oscillations on a 10-12
year period noted; no
consistent trends found
+
+
+
–
–
–
–
• composition of multiyear
ice pack is a potential
repository for accumulated
climatic changes
• it has not been studied for
trends or variability
concentration • areas of low concentration
are significant areas for
atmosphere-ocean
interaction
• high variability in the
marginal ice zone; no
trends found
thickness
• relative thickness of sea
ice reflects average winter
and early spring
temperatures as well as
the magnitude of midsummer solar radiation
dynamics
• motion changes can lead
to changes in ice
concentration; long-term
motion is a potential
repository for accumulated
climatic changes
• high variability; sometimes
reverses; no trends found
sesonality:
• closely related to winter
melt onset
and spring air
temperatures; long-term
variation is a potential
indicator of climatic
changes
• high variability; no long
term trend studies made
Baffin Bay2,3
Labrador Sea2,4
Greenland Sea5,6
Hudson Bay6
Bering Sea6,7,8
Sea of Okhotsk6,7,9
Kara Sea3,6
type
+
+
+
+
+
Baffin Bay6
Bering Sea6
Labrador Sea6
Sea of Okhotsk6
Kara Sea6,10
– shorefast ice11
+
+
+
–
Chukchi Sea12
Kara Sea12,13
Barents Sea12,13
Labrador Sea6,12
Ideal1
Spatial
Requirements
Ideal1
Temporal
Requirements
>1 km
3 days
> 1 km
3 days
> 1 km
3 days
10 km
3 days
10 km
1 day
10 km
3 days
*potential regions for which consistent deviation from an established norm for that ice parameter might indicate a
change of climate; '+': likely candidate; '-': region to avoid
1Barber et al., 1992b
2Chapman and Walsh, 1992
3Barry, 1993
4Ikeda, 1990a,b
5Kelly et al., 1987
6Parkinson, 1991
7Cavalieri and Parkinson, 1987
8Niebauer, 1983
9Parkinson, 1989
10Crane and Anderson, 1989
11Brown and Cote, 1992
12Anderson, 1987
13Kelly et al., 1982
Table 3: Significant Sea Ice Parameters Retrievable from Passive
Microwave Data
A √ indicates the parameter is retrievable from the data or t h e
data meet the specified requirement; an × indicates the p a r a m e t e r
is not retrievable or requirements are not met.
Spatial a n d
temporal requirements are derived from Barber et al., 1992b.
Sea Ice Parameter
√
√
√
×
√
√
ice extent
ice type
ice concentration
ice thickness
ice dynamics
seasonality
Ideal Spatial Requirements
×
×
×
×
×
×
>1 km
> 1 km
> 1 km
10 km
10 km
10 km
Ideal Temporal Requirements
√
√
√
√
√
√
3 days
3 days
3 days
3 days
1 day
3 days
Table 4:
Change Detection Techniques
Adapted from Howarth and Wickware (1981), Lillesand and K i e f e r
(1987), Singh (1989), Eastman and McKendry (1991), and E y t o n
(1991).
Technique
Synopsis
Comments
colour mapping
temporal composite display images are
created by mapping a different time
slice into a unique colour
transform standard red-green-blue
display structure into another system,
e.g., intensity-hue-saturation
simple; can simultaneously view up to
3 time slices; limited to 3 time periods
colour transformations
image differencing
image ratioing
image regression
subtract time 1 image from time 2
image
divide time 1 image by time 2 image
principal components
analysis
a linear regression is fit between all
pixels at time 1 with those at time 2; a
'predicted' image is created by
projecting time 1 pixels through the
regression; the projected and time 2
images are differenced and examined
for change
calculated as the first derivative of the
difference image
2 images are created: the first showing
the Euclidean distance between pixels
("change vector"); the other showing
the angle of this change vector
low-pass filtered image is subtracted
from original data
comparison of independently
produced classifications;
crossclassification maps the logical
AND of all possible class combinations
in the form of a confusion matrix
measurement space is rotated to
maximize the inter-channel variance
hypertemporal
classification
all data from both dates are classified
concurrently
acceleration mapping
change vector analysis
background
subtraction
post-classification
analysis
produces an effective and controlled
visual presentation which may be more
closely matched to the human visual
system; limited to 3 time slices
simple; indicates absolute change
magnitude, not significance
simple; allows relative image
differences to be examined
accounts for temporal differences in
image means and variances due to
uneven atmospheric or illumination
effects
provides information on the rate of
change between images
the change vector image gives an
indication of the degree of change;
the direction image can be used to
examine the nature of the change
removes slowly varying ("no-change")
background
statistical measures of change are
easily derived from the
crossclassification matrix; difficult to
produce comparable classifications
between dates
allows for the abstraction of change
from a time series as a whole; can be
used to reduce the number of
channels input into other techniques
classification can be complex and
produce too many classes
Table 5: Selected Global Change
Components Analysis
of Remote Sensing Imagery
Studies
Based
on
Principal
Temperate Applications
Variable(s)
Synopsis
Reference
natural vegetation
components from a sequence of 7 Landsat images
show seasonal variations
different components generated from multispectral
and multitemporal imagery reveal local and global
changes
PCA of monthly NDVI images within an annual cycle
highlights seasonal and regional patterns
PCA of monthly GVI images within an annual cycle
highlights seasonal, regional, and other patterns
higher order components provide information on
albedo differences and vegetative changes
standardized components provide more accurate
information for change detection
Lodwick, 1979
Variable(s)
Synopsis
Reference
interannual ice extent
dominant hemispheric and regional variations are
found
regional trends are examined
the leading spatial component was used to divide the
Antarctic region into regions where the magnitudes of
this component were approximately the same
PCA of TM and MLA Antarctic images show a decrease
in blue ice area was observed over a 2 year time span
statistically significant correlations between the
dominant modes of atmospheric and sea ice variability
are found
air temperature, rather than atmospheric pressure,
shows the strongest relationship to ice conditions
using principal components in an ice classification
algorithm would not improve its accuracy; the first 2
components generated from the 10 SMMR channels
are an efficient data set
"extended" PCA is used to categorize the spatial and
temporal patterns of surface change that occur in the
seasonal sea ice zone during the spring/early summer
Walsh & Johnson,
1979a
Lemke et al., 1980
Ropelewski, 1983
geology & natural
vegetation
agricultural & urban
land cover
Byrne et al., 1980
Townshend et al.,
1985
Tateishi & Kajiwara,
1990
Ingebritsen & Lyon,
1985
Fung & LeDrew, 1987
Polar Applications
atmosphere - ice
interactions
seasonal variations in
sea ice type
Orheim & Lucchitta,
1990
Walsh & Johnson,
1979b
Crane et al., 1982;
Crane, 1983
Rothrock et al., 1988;
Thomas & Rothrock,
1989
Crane & Anderson,
1989