My name is Charlotte Searle, and I’m a PhD student at the University of Oxford. I’m currently using secr to estimate population density for leopards at sites in the Ruaha-Rungwa landscape in southern Tanzania, using camera trap data I collected in 2018 and 2019.
I originally ran some of these analyses back in February 2019 in secr 3.2.0. This week I have been re-running the same analyses in secr 4.1.0, as I wanted to check the buffer widths and finalise my results, but noticed that the density estimates were coming out slightly different from those produced last year using the exact same input data.
My first thought was that there is a problem with the input files – so I checked and re-checked these – but on closer inspection, it seems that there is a problem arising when fitting the models. When I fit a model and print the output, where previously it would say "N occasions : [correct number of sampling occasions]", it now always says "N occasions : 1". I’ve tried this for four different data sets, with different numbers of sampling occasions, and the same problem is affecting all of them. However, when I print a summary of the capthist it does show the correct number of occasions, so the input files seem to be behaving as expected.
The other change I’ve noticed is that the detector type in the model output from secr 4.1.0 is listed as count, rather than proximity, despite including "detector = “proximity”" in the read.capthist function, as per my previous analyses. As above, when I print the capthist summary, it does list the detector type as proximity, so the problem seems to be arising later on.
Output of model fitted in February 2019 (secr 3.2.0):
secr.fit(capthist = RNP.Lonly.capthist, model = list(g0 ~ 1,
sigma ~ 1, D ~ 1), buffer = buffer, detectfn = "HN", binomN = 1,
hcov = "V5", details = list(minprob = 1e-200), trace = FALSE)
secr 3.2.0, 16:41:01 06 Feb 2019
Detector type proximity
Detector number 45
Average spacing 1957.079 m
x-range 680198 706553 m
y-range 9146385 9159243 m
Usage range by occasion
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
min 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
max 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
min 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
max 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
min 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0
max 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
N animals : 37
N detections : 201
N occasions : 83
Mask area : 144114.7 ha
...
Output of same model fitted to the same data in January 2020 (secr 4.1.0):
secr.fit(capthist = RNP.Lonly.capthist, model = list(g0 ~ 1,
sigma ~ 1, D ~ 1), buffer = buffer, detectfn = "HN",
binomN = 1, hcov = "V5", details = list(minprob = 1e-200),
trace = FALSE)
secr 4.1.0, 15:11:10 05 Jan 2020
Detector type count
Detector number 45
Average spacing 1957.079 m
x-range 680198 706553 m
y-range 9146385 9159243 m
Usage range by occasion
1
min 63
max 83
N animals : 37
N detections : 201
N occasions : 1
Count model : Binomial, size from usage
Mask area : 144114.7 ha
...
If anyone has any advice on how to fix this problem I would really appreciate it!
Best,
Charlotte