OUF vs OU describing movement patterns

[Zoe Bonerbo]

Hi there,

I've been reading up on the differences between OUF and OU models as we have quite a bit of variation between which model is selected among individuals we're tracking. 

My PI keeps saying that OUF describes individuals with multiple centers of activity that aren't connected while OU is meant for those with continuous ranges. However, none of the resources I have found mention that to be the case and that doesn't match the data. I'm not sure where she got that idea from and she hasn't been able to provide me with a citation for it, so I wanted to make sure I wasn't overlooking something.

Cheers,

Zoe

[Christen Fleming]

Hi Zoe,

We had some odd descriptions of the models in the original paper on variograms.

I think the more accurate description is that the OUF model describes a behavioral state with random motion within a fuzzy range, and where the motion has a measurable degree of persistence to it, rather than being fractal and admitting infinite velocities.

The OU model is then one where the persistence of motion is not measurable,  because of the coarseness of the data, and you can't resolve speeds or velocity autocorrelation. The true process will have finite velocities, but in coarse data the OU model with its fractal motion and infinite velocities is all that you can estimate from the data alone.

The models are nested and are equivalent in all other respects.

Best,

Chris