1
$\begingroup$

I'm a student in ecology and I'm currently doing an internship to validate the first year of my master degree in France. I have to calculate roe deer density for the Mavrovo National Park in Macedonia which is the most important stronghold of Balkan lynk. For that, I'm using Spatially Explicit Capture Recapture models on camera trap data with secr Rpackage. The habitat mask was constructed with a discrete 120-meters state-space over a 800-meters buffer width and those potential roe deer activity centres restricted to the forest habitat.

When I try to compare AICc of several models with predefined (t, T, B, b, bk, Bk, k, K, h2) and two user defined detector variables (slope, elevation), I have a warning message:

AIC(SCR0, SCR.sig.b, SCR.sig.B, SCR.sig.bk, SCR.sig.Bk, SCR.sig.k, SCR.sig.K, SCR.sig.h2, SCR.sig.T, SCR.sig.t, SCR.sig.slope, SCR.sig.ele)
                                  model    detectfn npar    logLik      AIC     AICc dAICc AICcwt
SCR0                   D~1 g0~1 sigma~1 exponential    3 -172.9945  351.989  356.789    NA     NA
SCR.sig.slope     D~1 g0~1 sigma~Kslope exponential    4 -169.9071  347.814  357.814    NA     NA
SCR.sig.h2    D~1 g0~1 sigma~h2 pmix~h2 exponential    5 -165.6934  341.387  361.387    NA     NA
SCR.sig.K              D~1 g0~1 sigma~K exponential    4 -172.9945  353.989  363.989    NA     NA
SCR.sig.ele   D~1 g0~1 sigma~Kelevation exponential    4 -175.1276  358.255  368.255    NA     NA
SCR.sig.bk            D~1 g0~1 sigma~bk exponential    4 -538.9850 1085.970 1095.970    NA     NA
SCR.sig.B              D~1 g0~1 sigma~B exponential    4 -541.6033 1091.207 1101.207    NA     NA
SCR.sig.b              D~1 g0~1 sigma~b exponential    4 -542.8917 1093.783 1103.783    NA     NA
SCR.sig.T              D~1 g0~1 sigma~T exponential    4 -543.4159 1094.832 1104.832    NA     NA
SCR.sig.Bk            D~1 g0~1 sigma~Bk exponential    4 -544.6537 1097.307 1107.307    NA     NA
SCR.sig.k              D~1 g0~1 sigma~k exponential    4 -544.6679 1097.336 1107.336    NA     NA
SCR.sig.t              D~1 g0~1 sigma~t exponential   63 -494.5765 1115.153       NA    NA     NA
Warning message:
In AIC.secr(SCR0, SCR.sig.b, SCR.sig.B, SCR.sig.bk, SCR.sig.Bk,  :
  models not compatible for AIC

I use the function AICcompatible to see where is the problem :

AICcompatible(SCR0, SCR.sig.b, SCR.sig.B, SCR.sig.bk, SCR.sig.Bk, SCR.sig.k, SCR.sig.K, SCR.sig.h2, SCR.sig.T, SCR.sig.t)
  data     CL groups   hcov binomN 
 FALSE   TRUE   TRUE   TRUE   TRUE 

I don't understand why and how the data changed ... After more research, I was able to form two groups with no problem of AIC compatibility.

AICcompatible(SCR0, SCR.sig.h2, SCR.sig.K, SCR.sig.slope, SCR.sig.ele)
  data     CL groups   hcov binomN 
  TRUE   TRUE   TRUE   TRUE   TRUE 
> AIC(SCR0, SCR.sig.h2, SCR.sig.K, SCR.sig.slope, SCR.sig.ele)
                                  model    detectfn npar    logLik     AIC    AICc  dAICc AICcwt
SCR0                   D~1 g0~1 sigma~1 exponential    3 -172.9945 351.989 356.789  0.000 0.5791
SCR.sig.slope     D~1 g0~1 sigma~Kslope exponential    4 -169.9071 347.814 357.814  1.025 0.3469
SCR.sig.h2    D~1 g0~1 sigma~h2 pmix~h2 exponential    5 -165.6934 341.387 361.387  4.598 0.0581
SCR.sig.K              D~1 g0~1 sigma~K exponential    4 -172.9945 353.989 363.989  7.200 0.0158
SCR.sig.ele   D~1 g0~1 sigma~Kelevation exponential    4 -175.1276 358.255 368.255 11.466 0.0000

> AICcompatible(SCR.sig.b, SCR.sig.B, SCR.sig.bk, SCR.sig.Bk, SCR.sig.k, SCR.sig.t, SCR.sig.T)
  data     CL groups   hcov binomN 
  TRUE   TRUE   TRUE   TRUE   TRUE 
> AIC(SCR.sig.b, SCR.sig.B, SCR.sig.bk, SCR.sig.Bk, SCR.sig.k, SCR.sig.t, SCR.sig.T)
                       model    detectfn npar    logLik      AIC     AICc dAICc AICcwt
SCR.sig.bk D~1 g0~1 sigma~bk exponential    4 -538.9850 1085.970 1095.970    NA     NA
SCR.sig.B   D~1 g0~1 sigma~B exponential    4 -541.6033 1091.207 1101.207    NA     NA
SCR.sig.b   D~1 g0~1 sigma~b exponential    4 -542.8917 1093.783 1103.783    NA     NA
SCR.sig.T   D~1 g0~1 sigma~T exponential    4 -543.4159 1094.832 1104.832    NA     NA
SCR.sig.Bk D~1 g0~1 sigma~Bk exponential    4 -544.6537 1097.307 1107.307    NA     NA
SCR.sig.k   D~1 g0~1 sigma~k exponential    4 -544.6679 1097.336 1107.336    NA     NA
SCR.sig.t   D~1 g0~1 sigma~t exponential   63 -494.5765 1115.153       NA    NA     NA

But again I don't understand why data are not the same for these two groups.

Does anyone have a lead? Tell me if you want more details or rather what details you want ;)

Thanks

Yon

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.