unrealistically small density estimates

[Asia Murphy]

Hello!

I'm attempting to model deer density and I've been receiving really small density estimates (on the order of *10-4 per ha) that I don't believe are realistic based on what I know of the area and published estimates. I've tried changing buffer sizes and I've collapsed trap locations depending on which ones are within a certain distance of each other; neither attempt helped anything.

Here's the summary of my current capt object:

Object class capthist Detector type proximity Detector number 793 Average spacing 9.433981 m x-range 561677 761627 m y-range 4248142 4652789 m Usage range by occasion 1 2 3 4 min 1 0 0 0 max 1 1 1 1 Counts by occasion 1 2 3 4 Total n 192 103 77 64 436 u 192 60 32 22 306 f 210 67 24 5 306 M(t+1) 192 252 284 306 306 losses 0 0 0 0 0 detections 331 132 104 79 646 detectors visited 328 132 104 79 643 detectors used 793 787 773 773 3126 Individual covariates Sex F:205 M:101

Is there anything else I could do? I'm assuming the issue is the lack of spatial recaptures and failure to correct estimate sigma. Thank you!

[Murray Efford]

I can't read the summary. How many re-detections are there? What is the distribution of unlist(moves(ch))? Have you plotted the data, including tracks=TRUE? What does RPSV(ch, CC=TRUE) give you?

[Asia Murphy]

Sorry about that! I've attached screenshots.

So it looks like I have few movers and a pretty significant RPSV value.

--
You received this message because you are subscribed to a topic in the Google Groups "secr" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/secrgroup/WqB6f2RV9lg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to secrgroup+...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/secrgroup/a1042485-106f-444e-be3b-6beeca3f095dn%40googlegroups.com.

--

Asia Murphy, Ph.D. (she/her)

www.asiajmurphy.com

Support Palestinian families: https://chuffed.org/project/hope-giving-circle

[Murray Efford]

Thanks for additional detail. The non-spatial counts look healthy. The one long movement (>80 km) may be the source of the problem. Even if this is not a data error I would remove it and see what happens. All other moves seem to be within your tight detector clusters? The long move presumably forces the HN detection function to fit a large sigma, which intuitively attributes the detected individuals to a large area around the detectors (with correspondingly low density). Exponential detection functions are less sensitive to extreme movements.

Murray