During March 1988 a series of coordinated special sensor microwave imager (SSM/I) underflights were carried out with NASA and Navy aircraft over portions of the Bering, Beaufort, and Chukchi seas as part of the NASA Defense Meteorological Satellite Program SSM/I Sea Ice Validation Program. The two Navy research aircraft, a Naval Research Laboratory P‐3 with the NOARL Ka band radiometric mapping system operating at 33.6 GHz and a Naval Air Development Center (NADC) P‐3 with the NADC‐Environmental Research Institute of Michigan (ERIM) C band synthetic aperture radar (SAR), provided wide‐swath, high‐resolution microwave imagery for direct comparison with sea ice concentrations calculated from SSM/I radiances using the NASA sea ice algorithm. Coincident measurements made with the Jet Propulsion Laboratory (JPL) C band SAR and the Goddard Space Flight Center (GSFC) aircraft multifrequency microwave radiometers (AMMR) on the NASA DC‐8 airborne laboratory provided additional verification of the algorithm. NASA DC‐8 AMMR data from Bering Sea ice edge crossings were used to verify that the ice edge location, defined as the position of the initial ice bands encountered by the aircraft, corresponds to an SSM/I ice concentration of 15%. Direct comparison of SSM/I and aircraft ice concentrations for regions having at least 80% aircraft coverage reveals that the SSM/I total ice concentration is lower on average by 2.4%±2.4%. For multiyear ice, NASA and Navy flights across the Beaufort and Chukchi seas show that the SSM/I algorithm correctly maps the large‐scale distribution of multiyear ice: the zone of first‐year ice off the Alaskan coast, the large areas of mixed first‐year and multiyear ice, and the region of predominantly multiyear ice north of the Canadian archipelago. Quantitative comparisons show that the SSM/I algorithm overestimates multiyear ice concentration by 12%±11% on average in the Chukchi and Beaufort seas. Excluding data for a day which gave anomalously large positive biases, the multiyear ice concentration difference reduces to 5%±4%, also indicating a positive SSM/I bias. Anomalously low SSM/I concentrations were found in the coastal zone north of Ellesmere Island. Differences between multiyear ice concentrations estimated from the JPL C band SAR imagery and from the GSFC AMMR radiances using an SSM/I type algorithm show that the AMMR concentrations are smaller on average by 6%±14%. Sea ice conditions are described, and possible causes of the observed differences are discussed.

No abstract

Remote sensing data collected during the March 1988 Alaska survey is used to compare multiyear ice concentration estimates derived from a satellite mounted passive microwave radiometer to those generated by an aircraft mounted Synthetic Aperture Radar (SAR) system. The passive microwave radiometer data used is produced by the Special Sensor Microwave Imager (SSMII) which is a multichannel system. The SSMII ice concentration estimates are produced by a multichannel algorithm which utilizes both the polarization and spectral gradient radiance ratios to determine the percent of multiyear ice. These estimate are compared to concentration estimates produced by a manual interpretation of mosaicked SAR imagery. A 30 percent discrepancy was found in the multiyear ice concentration estimates for coincident imagery which crosses the first-yearlmultiyear ice edge. KEYWORDSSARI SSM/I, Sea Ice Concentration, Sea Ice Comparison, Passive Microwave Radiometer, Multichannel Concentration Algorithm 1.0 INTRODUCTIONThe SSM/I system is a multichannel coarse resolution passive microwave radiometer mounted on the Defense Meteorological Satellite Program (DMSP-8) satellite. It has been providing nearly full coverage of the arctic region every day since the beginning of 1988, and specifically has been generating large scale ice concentration maps giving the percent concentration of open water, first-year, and multiyear sea ice. The daily production of these sea ice concentration maps is of great interest to both ship navigation and offshore drilling platforms since they can be used to track the motion of the ice edge (the boundary between ice-covered and ice-free water) in the polar regions.However, verifying the SSMII ice concentrations is often difficult due to the lack of available ground truth. One solution is to compare SSMII ice concentration estimates to those derived by other sensors. High resolution aircraft SAR systems can provide such a verification since manual generation of ice concentrations can reliably by done [1,2]. These studies show that the high spatial resolution associated with SAR imagery provides the ability to delineate individual floes which makes the determination o f ice concentration easier. both active and passive sensors have been performed [1,3]. However, the current analysis will use sea ice concentration estimates generated from high resolution SAR imagery to validate ice concentration estimates produced specifically from the multichannel SSM/I data. The SAR derived ice concentration estimates are produced by a manual analysis of mosaicked imagery while the SSMII estimates are generated by a multichannel concentration algorithm which utilizes both the polarization and spectral gradient ratios to determine the percentage of multiyear sea ice [4,5,6]. These ice concentration estimates were produced from imagery gathered during the March 1988 Alaska collection which is coincident with SSM/I overflights. Section 2 of this paper summarizes the sensors and gives a brief description of the Alaskan collection. ...

Previous research s t u d i e s have focused on producing a l g o r i t h m s f o r e x t r a c t i n g geophysical i n f o r m a t i o n from passive microwave d a t a regarding i c e f l o e size, sea i c e concentration, open water lead l o c a t i o n s , and sea i c e extent. These s t u d i e s have r e s u l t e d i n f o u r separate algorithms f o r e x t r a c t i n g these geophysical parameters. Sea i c e concentration estimates generated from each o f these a l g o r i t h m s (i.e., NASA/Team, NASA/Comiso, AES/York, and Navy) a r e compared t o i c e concentration estimates produced from c o i n c i d e n t high r e s o l u t i o n S y n t h e t i c Aperture Radar (SAR) Concentration, Sea I c e A l g o r i t h m Comparison 1.0 INTRODUCTION The p o l a r research community has been i n t e r e s t e d i n t h e determination o f sea i c e products from t h e a r c t i c r e g i o n since t h e launch o f the Nimbus 5 E l e c t r i c a l l y Scanning Microwave Radiometer (ESMR) i n 1972, and continued through 1987 w i t h t h e Nimbus 7 Scanning Multichannel Microwave Radiometer (SMMR). Presently, t h e p o l a r research community has focused i t s a t t e n t i o n towards the SSM/I which was launched i n June o f 1987 aboard t h e Defense Meteorological S a t e l l i t e Program (DMSP) Spacecraft F8. The SSM/I i s the f i r s t o f seven planned SSM/Is scheduled f o r launch over t h e n e x t two decades which w i l l make a v a i l a b l e passive microwave imagery o f t h e a r c t i c r e g i o n w e l l i n t o t h e twenty f i r s t century 113. Since 1972 the p o l a r research community has been developing sea i c e product algorithms designed t o e x t r a c t geophysical i n f o r m a t i o n about the a r c t i c . The focus o f these algorithms has been the generation o f i c e f l o e s i z e d i s t r i b u t i o n s , open water l e a d l o c a t i o n s , sea i c e concentration maps, and sea i c e e x t e n t (the l o c a t i o n of t h e boundary between open water and the i c e pack) which may a s s i s t i n t h e g e n e r a t i o n o f global c l i m a t e models h e l p i n g us t o f u r t h e r understand our biosphere. During t h i s p e r i o d several research teams have developed algorithms James Hol 1 i nger Space Sensing Branch Naval Research Laboratory Washington, D. C. 20375-5000 USA which produce b o t h t h e t o t a l and f r a c t i o n a l sea i c e percentages from passive microwave data. V e r i f i c a t i o n o f t h e r e s u l t s generated f r o m these sea i c e c o n c e n t r a t i o n algorithms was performed s e p a r a t e l y by v a r i o u s research teams. T h e i r approach (when ground t r u t h data was n o t a v a i l a b l e ) has been t o compare t h e r e s u l t s produced from c o i n c i d e n t d a t a sets c o l l e c t e d by m u l t i p l e sensors, then t r y and e x p l a i n any discrepancies found. High r e s o l u t i o n a i r c r a f t SAR systems can p r o v i d e such a v e r i f i c a t i o n since t h e generation o f r ...