Adjustment and filtering of seasat altimetry with the least squares response technique

The least squares response technique has been applied to SEASAT altimetry data for the adjustment and filtering of individual profiles without the use of standard crossover analyses. Based on a multichannel input-single output model, the technique allows a straightforward multi-correlation of the altimetry information with different types of forcing inputs. In addition, it can account for other known constituents such as datum biases, linear trends, periodic constituents of known periods and arbitrarily defined user-specified functions. In this paper, results of this approach are compared with geoid profiles computed from standard geopotential models such as GEM-10B which show a consistently good agreement with rms differences found at the level of the errors at long wavelengths of the geoid model used.