Question about AKDE vs occurrence for wintering individuals with non-stationary variograms

[Lara Moreno Zárate]

Hello,

I am working with tracking data from a migratory bird species during the wintering period. I have identified wintering areas for each individual and I would like to estimate their space use (AKDE 95% and 50%).

However, after inspecting the variograms, I see two different situations: some individuals show a clear asymptote, consistent with range residency, but others show increasing variograms without a clear plateau, or apparent shifts between spatial patches within the wintering period.

Ecologically, these individuals are still within their wintering grounds, but they may be expanding or shifting their space use rather than remaining stationary. I am unsure how best to proceed in these cases. Would it be appropriate to estimate AKDE for individuals showing clear residency, and use occurrence distributions for non-stationary ones Alternatively, is there a preferred approach in ctmm for describing space use during such partially non-stationary periods? My main goal is to quantify the area used during the wintering period in a way that is both ecologically meaningful and methodologically sound.

Any advice or references would be greatly appreciated.

Many thanks in advance.

Lara