Abstract: The cryosat-2 satellite recently published its first altimeter dataincluding periodically drifting orbit and dense ground tracks. These data can be used to improve the accuracy and resolution of the marine gravity field. In this paper, we constructs auto-regressive models of the altimeter data series by considering two aspects of signal and noise, and propose a new method called the signal-to-noise ratio analytical method for estimating the resolution capability of Cryosat-2 data. in five test cases, this new method offers a consistent alternative to the traditional spectral correlation method. The resolution capability of Cryostat-2 altimeter data from single cycle is in the range of 23 km~33 km, with an average value of 27 km, and better than 38 km, 43 km and 37 km of the Geosat/ERM, ERS-1/ERM and T/P data. This result shows that the Cryosat-2 altimeter data has a lower noise level and a higher geoid signal recognition. Therefore, it is reasonable to believe that the Cryostat-2 could provide shorter wavelength information for the construction of marine gravity field with higher resolution.