204
C. BERTOIA, J. FALKINGHAM, F. FETTERER
radiometer currently in use at the NIC is the Special Sensor Microwave Imager (SSM/I).
Data from the SSM/I sensor are routinely processed for the NIC at NOAA's National
Centers for Environmental Prediction (Grumbine 1996) using the National Aeronautic
and Space Administration (NASA) Team ice concentration algorithm (Cavalieri 1992)
and incorporating a weather filter (Gloersen and Cavalieri 1986).
Although iliey provide an all-weather sensing capability, the passive microwave sensors are of very coarse resolution; sea ice products are routinely produced wiili 25-km
pixels. Despite iliis relatively poor resolution, NI C analysts have historically relied heavily on microwave data due to ilie extensive cloud cover in the Arctic, which is particularly prevalent in ilie marginal ice zones (MIZ). NIC analyses were based on microwave
data for approximately 30% of the ice analysis chart information content in the summer and 65% in the winter for the period 1972-1983 (Knight 1984). This reliance on passive microwave was reduced to approximately 25% in winter and 35% in summer in the
1990S, when the NIC began to receive digital visible and infrared Operational Linescan
System (OLS) data.
The NIC began its investigation into improved capabilities in ilie microwave band with
the initial receipt of SAR imagery from ilie European Space Agency's (ESA) Earili Remote
Sensing satellite (ERS-1) in March of 1992. The NIC routinely receives imagery from iliree
station masks to provide nearly complete Arctic coverage. Images are processed at each
of ilie three Arctic ground stations wiiliin 6 h of receipt and transferred to the NIC via
the Internet. The Alaska SAR Facility (ASF) processes images at 30-m full resolution and
240-m low resolution; the low resolution products are preferred at the NIC and are simple 8x8 averages of ilie full-resolution images. The quick-turnaround products from the
Troms0, Norway, and Gatineau, Canada station masks are processed to ESA fast -delivery
product standards and received at ilie NIC at 100-m resolution.
The original SAR workstation hosted at the NIC was derived from the system
designed by the Jet Propulsion Laboratory for the Alaska SAR Facility (Kwok et al.1992).
Adaptations to the system, including interactive graphics capabilities required to produce ice products, were developed by the Naval Research Laboratory (NRL), Remote
Sensing Applications Branch, at the Stennis Space Center, Mississippi. The ice classification functions allowed operator-supervised or automated labeling of ice types for
winter SAR images in the high Arctic. The automated classification algorithm used
backscatter-based lookup tables to label four ice types: multiyear ice, two categories of
first-year ice, and new ice/open water.
The NIC currently hosts all satellite imagery and digitized ice reconnaissance information on SUN SPARC and Ultra workstations using the NRL Satellite Image Processing System (NSIPS). NSIPS is a PV-WAVE-based image processing package that enables
display and manipulation of the suite of satellite data sources required to produce an
ice analysis. The NIC uses NSIPS to warp all images to a common polar stereographic
projection; images and associated graphics retain latitude and longitude information
(Fetterer et al.1993). DMSP OLS images and SSM/I-derived ice products are analyzed
in conjunction with ERS-1/-2 SAR and NOAA Advanced Very High Resolution
Radiometer (AVHRR) visible and infrared images. RADARSAT images from Gatineau,
Prince Albert, Troms0, and the ASF are ingested and geocoded into NSIPS. Lines drawn
on these images are transferred to a base map where GIF fIles are created for distribution via the World Wide Web. Analysis graphics fIles are reformatted into various standard ASCII message fIles and are sent into ArcInfo, a Geographic Information System
C. BERTOIA, J. FALKINGHAM, F. FETTERER
radiometer currently in use at the NIC is the Special Sensor Microwave Imager (SSM/I).
Data from the SSM/I sensor are routinely processed for the NIC at NOAA's National
Centers for Environmental Prediction (Grumbine 1996) using the National Aeronautic
and Space Administration (NASA) Team ice concentration algorithm (Cavalieri 1992)
and incorporating a weather filter (Gloersen and Cavalieri 1986).
Although iliey provide an all-weather sensing capability, the passive microwave sensors are of very coarse resolution; sea ice products are routinely produced wiili 25-km
pixels. Despite iliis relatively poor resolution, NI C analysts have historically relied heavily on microwave data due to ilie extensive cloud cover in the Arctic, which is particularly prevalent in ilie marginal ice zones (MIZ). NIC analyses were based on microwave
data for approximately 30% of the ice analysis chart information content in the summer and 65% in the winter for the period 1972-1983 (Knight 1984). This reliance on passive microwave was reduced to approximately 25% in winter and 35% in summer in the
1990S, when the NIC began to receive digital visible and infrared Operational Linescan
System (OLS) data.
The NIC began its investigation into improved capabilities in ilie microwave band with
the initial receipt of SAR imagery from ilie European Space Agency's (ESA) Earili Remote
Sensing satellite (ERS-1) in March of 1992. The NIC routinely receives imagery from iliree
station masks to provide nearly complete Arctic coverage. Images are processed at each
of ilie three Arctic ground stations wiiliin 6 h of receipt and transferred to the NIC via
the Internet. The Alaska SAR Facility (ASF) processes images at 30-m full resolution and
240-m low resolution; the low resolution products are preferred at the NIC and are simple 8x8 averages of ilie full-resolution images. The quick-turnaround products from the
Troms0, Norway, and Gatineau, Canada station masks are processed to ESA fast -delivery
product standards and received at ilie NIC at 100-m resolution.
The original SAR workstation hosted at the NIC was derived from the system
designed by the Jet Propulsion Laboratory for the Alaska SAR Facility (Kwok et al.1992).
Adaptations to the system, including interactive graphics capabilities required to produce ice products, were developed by the Naval Research Laboratory (NRL), Remote
Sensing Applications Branch, at the Stennis Space Center, Mississippi. The ice classification functions allowed operator-supervised or automated labeling of ice types for
winter SAR images in the high Arctic. The automated classification algorithm used
backscatter-based lookup tables to label four ice types: multiyear ice, two categories of
first-year ice, and new ice/open water.
The NIC currently hosts all satellite imagery and digitized ice reconnaissance information on SUN SPARC and Ultra workstations using the NRL Satellite Image Processing System (NSIPS). NSIPS is a PV-WAVE-based image processing package that enables
display and manipulation of the suite of satellite data sources required to produce an
ice analysis. The NIC uses NSIPS to warp all images to a common polar stereographic
projection; images and associated graphics retain latitude and longitude information
(Fetterer et al.1993). DMSP OLS images and SSM/I-derived ice products are analyzed
in conjunction with ERS-1/-2 SAR and NOAA Advanced Very High Resolution
Radiometer (AVHRR) visible and infrared images. RADARSAT images from Gatineau,
Prince Albert, Troms0, and the ASF are ingested and geocoded into NSIPS. Lines drawn
on these images are transferred to a base map where GIF fIles are created for distribution via the World Wide Web. Analysis graphics fIles are reformatted into various standard ASCII message fIles and are sent into ArcInfo, a Geographic Information System