Interpreting the u summary from a count model with one occasion

[Greg Distiller]

Hi
I have used secrdesign to simulate a bunch of different datasets, all for a count model that uses one occasion. All the data summaries have n = u but with a higher number of total detections, for example:

Counts by occasion 
                    1 Total
n                 113   113
u                 113   113
f                 113   113
M(t+1)            113   113
losses              0     0
detections        212   212
detectors visited  39    39
detectors used     40    40

If I manually extract the number of individuals only caught once (singletons) I also get a smaller number than u:

> individual_totals <- apply(ch, 1, sum, na.rm = TRUE)
> singletons <- sum(individual_totals == 1)
> singletons
[1] 57

Is u interpreted differently for a count model with one occasion (something about being only detected in a single occasion)?

Many thanks

Greg

[Greg Distiller]

Sorry Murray but one more follow up question: 

I have noticed that the session name for the data that are simulated by secrdesign (for two groups) is labelled as: `session = 1+2, g = S1` or `session = 1+2, g = S2`. I assume the 'session = 1+2' is nothing alarming but thought I should just double check since I did not specify more than one session.

Thanks!

[Greg Distiller]

Hi again

I have realised that my code above to calculate singletons will exclude individuals caught more than once at the same detector. Is u meant to be the number of individuals that are detected at only one detector rather than individuals with a single detection? 

Thanks

Greg

[Murray Efford]

I'm catching up - sorry for slowness.

The counts are defined (e.g., ?counts) as

"n number of individuals detected on each occasion
u number of individuals detected for the first time on each occasion
f number of individuals detected exactly f times
M(t+1) cumulative number of individuals detected

..."

So 'u' will always equal n and M(t+1) for a single sampling occasion. The definition of 'f' is imprecise; it should read "number of individuals detected on exactly f occasions".

Murray