3.6 Correlation analysis
We investigated links between snow cover variability, atmospheric circulation and temperatures through correlation analysis, assessing the strength of relationships using Pearson’s correlation coefficient (r), and evaluating significance (p) using a two-tailed student test. We compared winter/spring indices with DSCD and DPSCD at separate elevation ranges and basins, and per pixel in the case of DSCD. To assess the influence of temperatures, we also compared winter/spring indices with temperature anomalies and temperature anomalies with snow cover variables. Temperatures from stations and ERA-Interim reanalysis between 2000 and 2017 were used for this test. Before this operation, data were detrended to isolate the inter-annual fluctuations (Panagiotopoulos et al., 2005). This analysis was also performed for each basin as a whole and at separate elevation ranges, with temperatures from each weather station and ERA-Interim grid cells in the study area.