Investigation of the effects of summer melt on the calculation of sea ice concentration using active and passive microwave data

Abstract

Active and passive microwave data sets acquired during the 1984 Marginal Ice Zone (MIZ) Experiment aircraft flights in the Fram Strait region are used to examine the effects of ice surface melt on microwave signatures and their resulting error in the calculation of sea ice concentration. Conditions examined with the active‐passive data set include ice floes with moist and dry snow cover and both heavily ponded and ridged surfaces. Passive sensors on the NASA CV‐990 aircraft included the aircraft electrically scanning microwave radiometer (ESMR) operating at 19.4 GHz and aircraft multichannel microwave radiometer (AMMR) operating at 10.7, 18.0, 21.0, and 37.0 GHz. Active microwave sensors flown on the Canadian Centre for Remote Sensing CV‐580 aircraft included the Environmental Research Institute of Michigan synthetic aperture radar (SAR) operating at 1.2 and 9.4 GHz. Coincident AMMR and SAR measurements of individual floes identified in aerial photography are used to describe the effects of surface melt on the calculation of sea ice concentration, and in particular, the response of the passive microwave polarization and spectral gradient characteristics to different stages of surface melt. Although the onset and progression of summer melt are not uniform throughout the Arctic, the stages of summer melt observed in the MIZ are also observed on a large‐scale in the central Arctic. This is demonstrated using Nimbus 7 SMMR data and Arctic Ocean buoy temperature data over one annual cycle. Finally, the potential of optimally combining both active and passive microwave data in an effort to ameliorate these surface melt effects during the summer months is also explored.