Heterogeneity models to estimate species richness in MARK

[mspinola]

Dear list members,

I am running mixture closed population models (closed captures with heterogeneity) to estimate bird species richness in 6 different cover types.
I have 8 occasions and my best model is a model that includes 2 groups (mixture) with p for each cover type and an overall N.
These are the estimates:


Real Function Parameters
95% Confidence Interval
Parameter Estimate Standard Error Lower Upper
------------------------- -------------- -------------- --------------
1:pi 0.1520450 0.0509366 0.0763498 0.2800339
2:pi 0.1404894 0.0565588 0.0612681 0.2904519
3:pi 0.1137196 0.0475850 0.0483985 0.2445455
4:pi 0.7210107 0.1070836 0.4765381 0.8800472
5:pi 0.1673584 0.0695076 0.0702984 0.3482326
6:pi 0.3411857 0.1201921 0.1536760 0.5962889
7:p 0.5469759 0.0762680 0.3977601 0.6882017
8:p 0.0899726 0.0250098 0.0515313 0.1524802
9:p 0.4399541 0.0788610 0.2955320 0.5953120
10:p 0.0839005 0.0228605 0.0486399 0.1409359
11:p 0.6099638 0.1041373 0.3987267 0.7866909
12:p 0.0921614 0.0263457 0.0519217 0.1583774
13:p 0.0613161 0.0305395 0.0225676 0.1559785
14:p 0.3621260 0.0725660 0.2347134 0.5123944
15:p 0.4158665 0.0813875 0.2696427 0.5785700
16:p 0.0732242 0.0249726 0.0369890 0.1398030
17:p 0.3177394 0.0600720 0.2129367 0.4449624
18:p 0.0514507 0.0289204 0.0166954 0.1476898
19:N 97.204479 12.918187 79.046728 131.80900



Estimates of Derived Parameters

95% Confidence Interval
Grp. Sess. N-hat Standard Error Lower Upper
---- ----- -------------- -------------- -------------- --------------
1 1 97.204479 12.918187 79.046728 131.80900
2 1 95.204479 12.918187 77.046728 129.80900
3 1 90.204479 12.918187 72.046728 124.80900
4 1 85.204479 12.918187 67.046728 119.80900
5 1 85.204479 12.918187 67.046728 119.80900
6 1 85.204479 12.918187 67.046728 119.80900

Questions:
Is it ok to use this type of model with 2 mixture (my idea is to model easy and difficult species to detect)?
Is it ok to have 6 different parameters, even when I set up the model for only 1 N?
Does the derived parameters make sense? They appear to have a suspicious pattern (same numbers after the dot).
Thank you very much in advance.
Best,

Manuel



--
Manuel Spínola, Ph.D.
Instituto Internacional en Conservación y Manejo de Vida Silvestre
Universidad Nacional
Apartado 1350-3000
Heredia
COSTA RICA
mspinola@una.ac.cr
Teléfono: 277-3598
Fax: 237-7036