A minimal, statistical model for the surface albedo of Vestfonna ice cap, Svalbard

Abstract. The ice cap Vestfonna is located in northeastern Svalbard and forms one of the largest ice bodies of the Eurasian Arctic. Its surface albedo plays a key role in the understanding and modelling of its energy and mass balance. The principle governing factors for albedo evolution, i.e. precipitation and air temperature and therewith snow depth and melt duration, were found to vary almost exclusively with terrain elevation throughout the ice cap. Hence, surface albedo can be expected to develop a comparable pattern. A new statistical model is presented that estimates this mean altitudinal albedo profile of the ice cap on the basis of a minimal set of meteorological variables on a monthly resolution. Model calculations are based on a sigmoid function of the artificial quantity rain-snow ratio and a linear function of cumulative snowfall and cumulative positive degree days. Surface albedo fields of the MODIS snow product MOD10A1 from the period March to October in the years 2001–2008 serve as a basis for both calibration and cross-validation of the model. The meteorological model input covers the period September 2000 until October 2008 and is based on ERA-Interim data of a grid point located close to the ice cap. The albedo model shows a good performance. The root mean square error between observed and modelled albedo values along the altitudinal profile is 0.057±0.028 (mean ± one standard deviation). The area weighted mean even reduces to a value of 0.054. Distinctly higher deviations (0.07–0.09) are only present throughout the very lowest and uppermost parts of the ice cap that are either small in area or hardly affected by surface melt. Thus, the new, minimal, statistical albedo model presented in this study is found to reproduce the albedo evolution on Vestfonna ice cap on a high level of accuracy and is thus suggested to be fully suitable for further application in broader energy or mass-balance studies of the ice cap.