5 Fusion of Satellite SAR with Passive Microwave Data for Sea Ice Remote Sensing
107
approach capitalizes on the fact that C-band backscatter from Arctic multiyear ice is
relatively invariant under cold conditions, allowing reliable estimates of multiyear ice
concentration from SAR (Kwok et al.1992; Kwok and Cunningham 1994). Improved estimates of ice type concentration are obtained from a combination of satellite active and
passive microwave data. First, we derived the multiyear ice concentration from ERS-1
SAR using a local-area threshold method. We then used the information derived from
SAR as a constraint in a modification of an existing satellite passive microwave algorithm. The estimates of first-year ice concentration obtained from the fused result for
the region in which we had ship observations were 18.7 and 14.5% for Julian day 249
and 255, respectively. Ship-based observations of ice type were obtained during Julian
days 240-242 from the same region as the satellite measurements. These observations
showed thatthe first -year ice coverage was approximately 10-15% and the multiyear ice
coverage was approximately 70% along the ship's track in this region. The estimated
concentration from the satellite passive microwave data alone yields 27-40% first-year
ice concentrations. These results demonstrate that an improved estimate of first-year
ice concentration during the freeze-up season is obtained by fusing of active and passive satellite data. The estimate of multiyear ice concentration shows similar improvement.
Combining SAR and passive microwave data has been demonstrated here and
shows great potential. Further research is needed to determine the usefulness of this
method for different seasons and, potentially, to extend the technique to subdivide the
first-year ice category into thick and thin first-year ice. RADARSAT now provides
wide-swath data so that the coverage of the Arctic will be greatly increased. This
increased coverage affords an opportunity for improvements in this approach since
the SAR coverage will be more commensurate with the SSM!I coverage than that from
the ERS-l SAR.
References
Beaven SG (1995) Sea ice radar backscatter modeling, measurements, and the fusion of
active and passive microwave data. PhD Dissertation, The University of Kansas,
Lawrence, Kansas (also: RSL Tech Rep 8243-4, March 1995)
Beaven SG, Gogineni SP (1994) Shipborne radar backscatter measurements from Arctic sea ice during the fall freeze-up. Remote Sens Rev 9, nos 1-2: 3-25
Beaven SG, Gogineni SP, ED. Carsey FD (1996) Fusion of satellite active and passive
microwave data for sea ice type concentration estimates. IEEE Trans Geosci Remote
Sens 34, no 5: 1172-1183
Burns BA, Cavalieri DJ, Keller MR, Campbell WJ, Grenfell TC, Maykut GA, Gloersen P
(1987) Multisensor comparison of ice concentration estimates in the marginal ice
zone. J Geophys Res 92, no C7: 6843-6856
Carsey FD (1982) Arctic sea ice distribution at end of summer 1973-1976 from satellite
microwave data. J Geophys Res 87, no C8: 5809-5835
Carsey FD (1985) Summer Arctic sea ice character from satellite microwave data. J Geophys Res 90, no C3: 5015-5034
Cavalieri DJ, Gloersen P, Campbell WJ (1984) Determination of sea ice parameters with
the Nimbus 7 SMMR. J Geophys Res 89, no D4: 5355-5369
107
approach capitalizes on the fact that C-band backscatter from Arctic multiyear ice is
relatively invariant under cold conditions, allowing reliable estimates of multiyear ice
concentration from SAR (Kwok et al.1992; Kwok and Cunningham 1994). Improved estimates of ice type concentration are obtained from a combination of satellite active and
passive microwave data. First, we derived the multiyear ice concentration from ERS-1
SAR using a local-area threshold method. We then used the information derived from
SAR as a constraint in a modification of an existing satellite passive microwave algorithm. The estimates of first-year ice concentration obtained from the fused result for
the region in which we had ship observations were 18.7 and 14.5% for Julian day 249
and 255, respectively. Ship-based observations of ice type were obtained during Julian
days 240-242 from the same region as the satellite measurements. These observations
showed thatthe first -year ice coverage was approximately 10-15% and the multiyear ice
coverage was approximately 70% along the ship's track in this region. The estimated
concentration from the satellite passive microwave data alone yields 27-40% first-year
ice concentrations. These results demonstrate that an improved estimate of first-year
ice concentration during the freeze-up season is obtained by fusing of active and passive satellite data. The estimate of multiyear ice concentration shows similar improvement.
Combining SAR and passive microwave data has been demonstrated here and
shows great potential. Further research is needed to determine the usefulness of this
method for different seasons and, potentially, to extend the technique to subdivide the
first-year ice category into thick and thin first-year ice. RADARSAT now provides
wide-swath data so that the coverage of the Arctic will be greatly increased. This
increased coverage affords an opportunity for improvements in this approach since
the SAR coverage will be more commensurate with the SSM!I coverage than that from
the ERS-l SAR.
References
Beaven SG (1995) Sea ice radar backscatter modeling, measurements, and the fusion of
active and passive microwave data. PhD Dissertation, The University of Kansas,
Lawrence, Kansas (also: RSL Tech Rep 8243-4, March 1995)
Beaven SG, Gogineni SP (1994) Shipborne radar backscatter measurements from Arctic sea ice during the fall freeze-up. Remote Sens Rev 9, nos 1-2: 3-25
Beaven SG, Gogineni SP, ED. Carsey FD (1996) Fusion of satellite active and passive
microwave data for sea ice type concentration estimates. IEEE Trans Geosci Remote
Sens 34, no 5: 1172-1183
Burns BA, Cavalieri DJ, Keller MR, Campbell WJ, Grenfell TC, Maykut GA, Gloersen P
(1987) Multisensor comparison of ice concentration estimates in the marginal ice
zone. J Geophys Res 92, no C7: 6843-6856
Carsey FD (1982) Arctic sea ice distribution at end of summer 1973-1976 from satellite
microwave data. J Geophys Res 87, no C8: 5809-5835
Carsey FD (1985) Summer Arctic sea ice character from satellite microwave data. J Geophys Res 90, no C3: 5015-5034
Cavalieri DJ, Gloersen P, Campbell WJ (1984) Determination of sea ice parameters with
the Nimbus 7 SMMR. J Geophys Res 89, no D4: 5355-5369