Geology Reference
In-Depth Information
10
Retrieval of Ice and Snow Geophysical Parameters
Retrieval of sea ice parameters from remote sensing
data was originally driven by requirements for operational
ice monitoring. The main parameters needed for this pur-
pose were ice types and total ice concentration. The latter
was easier to retrieve and the first attempts to retrieve it
from optical and passive microwave observations started
in the 1970s. The use of optical sensors for this purpose
(started with NOAA satellite series with VHRR and later
AVHRR) led to developing techniques to account for
atmospheric influences and cloud masking. Only limited
results were obtained due to extensive cloud cover espe-
cially in polar regions. With the advent of passive micro-
wave sensors several algorithms for retrieval of total ice
concentration were developed. A few algorithms were
later developed with a capability to identify one or two
ice types (e.g., young ice and MY ice). The momentum
continued after the introduction of space‐borne SAR in
early 1990s. However, because of the remarkably finer
resolution of SAR research was directed to retrieve ice
types rather than ice concentration. The launch of the vis-
ible and infrared sensors onboard the Terra and Aqua sat-
ellites (1999 and 2001, respectively) boosted the research
on ice parameter retrieval. That was because of the need
for development and validation of ice operational prod-
ucts from optical and infrared channels of the Moderate‐
Resolution Imaging Spectroradiometer (MODIS) and
Multi‐angle Imaging SpectroRadioter (MISR).
With the increased capabilities of satellite sensors and
the increasing evidence of ice reduction in the Arctic,
demands for more climate‐related ice parameters have
been identified. Key parameters that can be obtained
from remote sensing includes ice surface temperature and
albedo, ice thickness, ice drift, and ice extent and area.
Most of these parameters can be retrieved from more than
one category of sensors (optical, TIR, and microwave).
Even for albedo, attempts have been undertaken to retrieve
it using correlations with SAR data. Retrieval of some
parameters such as ice surface temperature and ice motion
tracking have matured to satisfy the accuracy, robustness,
and turnaround time required by operational ice monitor-
ing systems as well as the spatial and temporal intervals
required for climate monitoring and prediction tasks.
The Ocean and Sea Ice Satellite Application Facility
(OSI‐SAF), operated jointly by the Norwegian and Danish
Meteorological Institutes, produces daily maps of sea
ice concentration, ice types (only FY and MY ice), and
global sea ice edge delineation for the Arctic and Antarctic
regions. The maps are produced at 10 km resolution in
polar stereographic projection. Low‐resolution sea ice drift
maps (on 62.5 km polar stereographic grid) are also pro-
duced every other day [
Andersen et al
., 2007]. The Radarsat
Geophysical Processor System (RGPS) at the Alaska SAR
Facility (ASF) produces maps of ice motion and deforma-
tion regularly at 3 and 6 day intervals using successive
observations from the high‐resolution SAR data. The U.S.
National Snow and Ice Data Center (NSIDC) produces a
wide variety of sea ice parameters including daily maps of
ice concentration from passive microwave observations, ice
extent from visible and infrared observations, and melt
onset from combination of observations.
Recently, a European study called Developing Arctic
Modeling and Observing Capabilities for Long‐term
Environmental Studies (DAMOCLES) was conducted
from 2006 to 2010 with a goal to identify and understand
the changes occurring in the sea ice, atmosphere, and ocean
of the Arctic and sub‐Arctic domain. Sea ice parameter
retrieval from satellite remote sensing was a major tool in
this project. The progress in sea ice remote sensing capa-
bilities to retrieve a few key parameters is summarized in
Heygster et al
. [2012]. It should be mentioned, however,
