A first estimate of permafrost distribution from BTS measurements in the Romanian Carpathians (Retezat Mountains)

A first estimate of permafrost distribution in the central part of the Retezat Mountains (Southern Carpathians, Romania) was computed using basal temperature measurements of winter snow cover as input data for an empirical-statistical modelling approach. During the 2012-2013 winter season, a total of 170 bottom temperature of snow cover measurements (Haeberli, 1973) were made. The permafrost distribution within the investigated area was modelled based on multiple linear regression analysis of the relationship between bottom temperature of snow cover measurements and five predictor variables: elevation, solar radiation, vegetation cover, slope and profile curvature. The final model, having an overall accuracy of 0.48, included only the first three variables as the main factors controlling permafrost occurrence within the investigated area, while slope and profile curvature proved to be statistically insignificant. Permafrost occurrence, as predicted by the model, covers 31 km(2) (52% of the investigated area), of which 14 km(2) is predicted as probable permafrost and the remaining 17 km(2) is possible permafrost. The obtained results indicate the importance of solar radiation, elevation and vegetation cover in controlling alpine permafrost occurrence in the central part of the Retezat Mountains.