Short-scale variations in high-resolution crystal-preferred orientation data in an alpine ice core - do we need a new statistical approach?