I wonder if there is some resampling exercise we could do on some of the other tracks (not MPT) to understand the variation in the KDE and the separation we detected. I am just coming up with this idea so I have not fully thought it through as to what it would entail and if it would actually help us understand a) if our sample size is big enough (tracks per year) or b) if by choosing other possible tracks we would get vastly different results?
For a) in the past, Rachael has done resampling to look at how many birds are needed to stabilize the home range area (see Orben et al. 2018). I wonder if you could do a similar approach but instead of calculating area, calculate overlap with the MPT KDE?
For b) Sample a random track from each bird, build a KDE, and compare overlap with the MPT KDE, repeat 1000 times? High average overlap would suggest that our result holds true, low average overlap would not?
I wonder if there is some resampling exercise we could do on some of the other tracks (not MPT) to understand the variation in the KDE and the separation we detected. I am just coming up with this idea so I have not fully thought it through as to what it would entail and if it would actually help us understand a) if our sample size is big enough (tracks per year) or b) if by choosing other possible tracks we would get vastly different results?
For a) in the past, Rachael has done resampling to look at how many birds are needed to stabilize the home range area (see Orben et al. 2018). I wonder if you could do a similar approach but instead of calculating area, calculate overlap with the MPT KDE?
For b) Sample a random track from each bird, build a KDE, and compare overlap with the MPT KDE, repeat 1000 times? High average overlap would suggest that our result holds true, low average overlap would not?