Logo Universitat Bremen
Page path:

HB-1-1

Climatological time series of Arctic multiyear sea ice area

PhD student :Raul Scarlat
Project supervisors :J. Notholt (Germany)
B. Tremblay (Canada)
Key hypothesis
A forward model that includes both atmosphere and surface will be used in order to retrieve surface emissivities for all sensor footprints at the required frequencies. Having these surface emissivities allows for a physically meaningful retrieval of brightness temperatures and from there on to retrieving simultaneously both surface parameters (sea ice concentration and type) and atmospheric parameters (surface temperature, atmospheric water vapor, cloud liquid water, surface wind speed). This approach would allow a retrieval of these parameters while taking into account the interconnections that exist between them, as opposed to the classical case when the retrieval of a single parameter implies an artificial parametrization of all others.
 
Current climate projections suggest that some of the most dramatic climatic changes of the century will take place in the Arctic. The repercussions these changes will have for the mid-latitudes and the importance of Arctic influences on oceanic and atmospheric circulation give cause for the intense efforts to monitor the physical parameters of the Arctic atmosphere and ocean.
The difficulty of doing direct observations in this region means that remote sensing methods represent the only viable option for a meaningful monitoring of the atmospheric and surface physical parameters .
Observations from satellite passive microwave sensors are ideal for such remote sensing applications because they are independent of the polar night and mostly independent of cloud cover. This method of retrieval is however highly dependent on an accurate modeling of the surface microwave emissivity. In order to achieve this, the contribution of the atmosphere above the ice has to be properly taken into consideration. Until now, sea ice concentration and sea ice type retrieval methods use a simplified model of the atmosphere and do not take into account the local atmospheric particularities of each sensor footprint.
Because sea ice and atmospheric parameters are interdependent in the context of climatic monitoring and passive microwave retrieval, this project aims to achieve an integrated retrieval of sea ice surface parameters and atmospheric data that would provide a more physically realistic picture of the Arctic.
The goal of this work is to establish a consistent and reliable time series of the Arctic MYI area based on sea ice concentrations and drift information, both derived from passive microwave satellite sensors, best from the start of continuous sea ice observations with passive microwave sensors in 1978 and onwards. The baseline will be existing MYI concentration retrieval algorithms like the bootstrap and NASA Team algorithms. The selected algorithms will be modified by using dynamical tie-points adjusted to the typical signatures of 100% first-year ice, MYI, and open water for each involved channel on a monthly basis. The seasonal evolution of the emissivity of sea at the frequencies of various microwave sensors has been already been investigated. Sea ice drift is a standard product derived from passive microwave sensors with many different algorithms. Here the OSI SAF product, relying on the Continuous Maximum Cross Correlation method is suggested to be used because of its improved directional accuracy. Passive sensors are preferred over scatterometers (from which time series are available over similar periods) because the latter typically only observe one parameter (backscatter coefficient) from which only one geophysical quantity can be derived, typically the sea ice type, but not concentration of one or even more ice types. The results will be compared with runs of sea ice and ocean circulation models; among them those run at the McGill University, Montreal, Canada (Prof. B. Tremblay), and suggestions for improved model parameterizations of the sea ice dynamic processes will be derived in collaboration with projects CA-1 and CA-2 and under consideration of the results of project HB-2.