separating sources of mortality

[Mark Trinder]

Hi all

apologies for the repeat posting of the following message, but i had no replies last time! just like to clarify the reasons - noone knows the answer or noone cares...

I'm working on a heavily hunted population of greylag geese, for which some comprehensive analyses have been undertaken (Frederiksen et al, 2004) using Burnham combined live-dead models. I am trying to use the results of this analysis as well as data from a variety of other sources (population counts, partial hunting bag values) to provide parameters for population modelling. Currently the best estimate for survival (as modelled by Frederiksen above) includes both hunting and natural mortality, but in order to explore the impact of changes in the level of hunting it would be nice to be able to partition the mortality out. So, my question is, if we have a combined survival estimate already (Burnham S), is it possible (and permissable) to derive a second survival estimate based on recovery data alone, and thereby find the natural survival rate as:

total mortality/recovery mortality ?

Or would this approach, a- not work, and/or; b-be statistically invalid?

cheers
Mark

[gwhite]

Mark:

Your question was:

If we have a combined survival estimate already (Burnham S), is it possible (and permissable) to derive a second survival estimate based on recovery data alone, and thereby find the natural survival rate as:

total mortality/recovery mortality ?

The band recovery models, as well as the joint live/dead models, estimate total mortality rates, not cause specific mortality rates. Even if you only use dead recoveries from hunting, the estimate of survival is overall survival, not just mortality from hunting. I know that the idea that only using dead recoveries from hunting gives you an estimate of only hunting mortality floats around, but the idea is wrong. The survival rate from the band recovery models is overall survival no matter what the source of recoveries.

Gary

[Mark Trinder]

gary

many thanks for the prompt reply. shame you had to rain on our parade though!

m

[dglover]

Mark,

Would you happen to know the reporting rate from hunters? I'm sure you already know that you could separate total mortality into natural and hunting mortality if you have this information.

Dave

[cooch]

Mark,

Would you happen to know the reporting rate from hunters? I'm sure you already know that you could separate total mortality into natural and hunting mortality if you have this information.

Dave

If not, I'd suggest looking at 3 references:

1. the 'big book' by Williams, Nichols & Conroy (which everyone should own anyway)

2. the relevant sections of chapter 10 (especially p. 19, which points out separating natural from hunter mortality can't be done using the non-Brownie approach, but can be if you use the Brownie approach, and have estimates of reporting and retrieval rates).

3. also, have a look at

Schaub,M. & Pradel,R. (2004) Assessing the relative importance of different sources of mortality from recoveries of marked animals. Ecology, 85, 930-938.

A nice recent paper that talks about a number of issues related to the original question.