The mountain permafrost distribution in the Abisko region in northern Sweden has been assessed using a combination of empirical and statistical analysis. The empirical data was obtained using the bottom temperature of snow cover (BTS) method, supported by continuous ground temperature measurements. The statistical analysis was based on 148 data points in total and used logistic regression to model the probability of permafrost occurrence. Further, Geographically Weighted Regression (GWR) was introduced as an exploratory tool for detecting non-stationarity in the relationships between permafrost and the independent variables and showed to be a useful tool in the statistical analysis. As a result of the GWR analysis the region was divided into two subregions. The models show probabilities 0.8 for permafrost at elevations above 1300 m a.s.l. in the western part of the region. In the eastern part, the probabilities are likely to be influenced by the potential incoming shortwave summer radiation, indicating a probability 0.8 above 850 m a.s.l. on the north-east and east-facing slopes, above 1000 m on the west-facing slopes and above 1100 m a.s.l. on the south-facing slopes. Permafrost conditions throughout the region were found to be marginal and sensitive to current warming trends.

The occurrence of turf‐banked solifluction landforms in the Abisko region was analysed using a grid‐based approach and statistical modelling through logistic regression. Significant parameters in the model were the vegetation index NDVI, annual incoming potential radiation, wetness index, slope gradient and elevation. The model had an acceptable discrimination capacity and rather low model‐fit values, but clearly showed the importance of vegetation patterns for the occurrence of solifluction at a regional scale. Solifluction movement rates measured at eight sites were combined with model parameters and the annual duration of sun hours to regionalise solifluction movement rates through an unsupervised terrain classification. For comparison, the linear relationship between the probability of solifluction occurrence and variations in movement rates was also used to regionalise movement rates. Potential geomorphic work was calculated for six different areas within the region, with the greatest being for Kärkevagge, the area with the highest precipitation. The combination of a logistic regression model of mapped landforms and field measurements of solifluction rates represents a promising methodology to assess the occurrence and activity of the process at a regional scale. Copyright © 2010 John Wiley & Sons, Ltd.

ABSTRACT. Solifluction movement rates from 1952 to 2008 for the Abisko region, northern Sweden, have been compiled and analysed through correlation tests and multiple regression. The temporal analysis is based on two datasets (Lobe11 & gridAB and Line B) from Kärkevagge. The dataset Lobe11 & gridAB show a strong correlation between movement rates and mean annual air temperature (MAAT) and MAAT is also identified as one of the significant contributing parameters in the multiple regression model. No significant correlations were found for the Line B dataset. The spatial analysis indicates generally higher movement rates in the western part of the region and at lower altitudes mainly between 700 and 900 m a.s.l., but the spatial variability is high. To reduce the influence of the temporal variation the data for the correlation tests of the spatial variations were divided into two parts: 1957 to 1980 and 1981 to 2008. The correlation analysis of the dataset 1957 to 1980 shows a significant negative correlation between annual average movement rates and permafrost probability and altitude. The dataset 1981 to 2008 shows a positive correlation between movement rates and wetness index. It is concluded that movement rates may increase with higher MAAT in the western part of the region (Kärkevagge), the spatial variability of movement rates within the region is very high and that altitude (and/or permafrost) together with wetness index are the main controls on the regional spatial variation. The study highlights the limitations in establishing statistical relationships between movement rates and climate using data from different field empirical studies.

Turf-banked lobes and terraces on the valley slopes and stone-banked or non-sorted lobes at summits show a distinct regional trend of decreasing size from west to east in the Abisko region, northern Sweden. Significant correlations with environmental parameters exist between morphometry and elevation, soil moisture and soil texture. At a regional scale, high soil moisture environments can be associated with larger dimensions. The turf-banked forms in the valleys are associated with deep seasonal snow cover and therefore saturated conditions are believed to enhance gelifluction at these locations. The smaller forms on the vegetation-and snow-free summit sites are believed to be dominated by frost creep. Figure 4 Comparison between mean monthly temperatures from Latnjajaure (1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001) and Abisko (1979Abisko ( -2004 weather stations and the corresponding mean monthly temperatures calculated from the Meteonorm 1 V4.0 program.

A new method is presented to assess decadal solifluction lobe movement and volumetric changes at the catchment scale. The method is based on photogrammetrical analysis of orthorectified repeat aerial photography in a geographic information system (GIS). High resolution aerial photographs from two valleys in northern Sweden, Kärkevagge and Låktatjåkka valleys, are used to quantify changes of the lobe fronts, as an indication of solifluction rates over a period of 41 years. Two methods are explored, the flow direction method and the front line method, in order to quantify the movement rates. Results show that the maximum combined root mean square error (RMSE) and resolution error amount to c. 1·7 m; no clear preference of one measurement method over the other was found. Application potential of the method is promising but reliability depends on the movement rates of the landforms and sufficient time span between the images. In the pilot study the measured yearly movement over the period 1959-2000 ranges from not-detectable to 63 mm/yr. Not-detectable means that movement cannot be detected within the resolution used in the study. Lobe advance rates are similar to measured surface movement rates and volumetric sediment flux by solifluction is found to be considerably lower than a previous estimate by Rapp (1960). Implications for lobe advance models are discussed. Photogrammetrical analysis is recommended as an important tool for detecting decadal solifluction movement at the catchment scale.