POPAN gross/net births and the right N

[nlsanto2]

Hi everybody,

I am using POPAN (as implemented in program MARK 7.1) to estimate the population size of striped dolphins in the Gulf of Corinth, Greece.

I have a very specific question to ask to anyone who has some experience with POPAN.

The MARK manual (Chapter 12, pp 20) says that the estimate of the super population size is given under the REAL parameter estimates (and in my case is 720 dolphins).
Under the DERIVED parameters estimates though, there is another parameter (Gross Population Estimates N*-hat), that is in my case 881… quite larger!

I tried to understand the difference between the two parameters by reading the Schwarz & Arnason (1996) and the Schwarz et al. (1993) papers and a very interesting previous conversation on this forum, but I am still left with a doubt.

Given that I understood the difference, which one should I pick as my population estimate?
I think the latter one is the right one because in my study I have monthly capture occasions separated by one or two weeks, so it is possible that some dolphins entered the population between two occasions and then left before the next one.

Since the manual does not suggest this choice I would be very grateful if anybody wants to give me his/her opinion about this problem.

Moreover, my best model is phi(t)p(t)pent(t). If I cannot assume that the recruitment and survival probability are constant, as reccomended in Schwartz et al. 1993, should I consider that I cannot estimate the gross number of entrants and then the gross population size is not valid? In this case I would pick the N in the real parameter estimates, knowing that I am overlooking the presence of animals that may have entered the population and died before being sampled.

Nina

[murray.efford]

I suspect you want neither the superpopulation nor the gross superpopulation. The POPAN 'superpopulation' N consists of all individuals available for detection in the sampled area at some time during sampling. N* is adjusted for animals that come and go between sampling times. The POPAN superpopulation will usually exceed the local population size at any one time, but it bears no particular relationship to the biological superpopulation, the number of individuals in the wider geographical area.
Murray

[nlsanto2]

Hi Murray!

Thank you so much for taking the time to reply to my post.

I think my question was not clear though.

I think am interested in a super-population estimate, since I want to have an idea about the number of animals that were in the study area during the time of the study (and I have some problems assuming a closed population), but I was confused by the choice between the gross and the net population estimate.

Thank you again,

Nina

[jlaake]

Nina-

I believe that gross N includes animals that could arrive and depart between occasions. I suggest looking at some of the original papers by Neil Arnason and Carl Schwarz. Carl can correct me if I'm wrong here.

regards --jeff

[murray.efford]

Nina
My point was that the 'superpopulation' from POPAN is not the geographical superpopulation that you want. The term 'superpopulation' is used for two quite different concepts. I'm aware that some cetacean workers have used the POPAN superpopulation as if it was a geographical superpopulation, but this seems to me a fundamental mistake. A POPAN superpopulation always increases as you add more samples if there is any recruitment, even if that is balanced by mortality. It is a mathematical convenience rather than a biological reality.
Murray

[cooch]

Murray's point is very well-taken. It is also generally established that 'biological reality' and 'open population abundance estimates' are often oxymorons. Everyone wants to estimate the geographical superpopulation. In general, you simply can't get there from here with open population models.

[nlsanto2]

Dear all, thank you again for you replies. I am very interested in this discussion, even if I may be a little ignorant in statistic to understand it completely.

I would like to ask to any of you to explain it to me why the POPAN superpopulation cannot be used in my case (estimating the number of dolphins that were in the study area during the duration of the study), and in what case the use of POPAN superpopulation approach would be theoretically correct.

Also, should I decide to use a different estimation method (not CMR) to estimate the population size in my study area, then?

Any thought on this would be extremely helpful to me, thank you.
Nina