[Ipmpack-users] modelling a gender-switching iteroparous herb

[Christopher Heckel]

Hi Eelke,

Thank you.  I am interested in the population structure of 
non-flowering, male, and female individuals because the multiple 
populations that I have studied have very different sex-ratios.  I don't 
think at this time I need to include marriage functions because seed 
production in this species can be pollen limited even when there are 
high densities of males.

I've begun writing code with the series of conditional probabilities and 
there certainly are many parameters.  I may be posting more specific 
questions soon.

Cheers,

Christpher
On 6/9/2014 5:37 PM, Eelke Jongejans wrote:
> Dear Christopher,
>
> there are probably multiple ways to do what you're describing, but I 
> guess your modelling choices also depend on what you want to achieve. 
> Is it the population trends you are interested in, and/or the 
> population structure of non-flowering, male and female flowering 
> individuals? If you are just interested in seed production and assume 
> that pollen is not limiting, a size-dependent female-flowering 
> probability in the F matrix might be enough. You would then have a 
> size-structured model and no need for additional states. I guess you 
> could even translate the stable size distribution to proportions males 
> and females. However, if you do need males as a state, the options you 
> describe should work: either model a series of conditional 
> probabilities as functions of the previous state and size of 
> individuals, or use multinomial models. The number of needed 
> parameters will increase rapidly if you want both size and state 
> transitions to depend on both previous size and previous stage. Or are 
> you interested to include 'marriage functions' as well? Not sure if 
> you could include those with just size as a continuous state variable 
> and size-dependent male and female flowering probabilities. Anyone?
>
> As for IPMpack: we now do accomodate simple age x size IPMs, but I 
> don't think you can make such a complicated size x flowering stage IPM 
> with IPMpack functions only. Probably (as often when you want to 
> include very complex life histories) you will have to develop code to 
> construct your own IPM. However, it is often very useful to use bits 
> and pieces of IPMpack to achieve this (e.g. by using IPMpack functions 
> to create survival, growth and fecundity objects for subsets of 
> individuals).
>
> best wishes,
> Eelke
>
>
> On 09-Jun-14 8:03 PM, Christopher Heckel wrote:
>>
>> Hello all,
>>
>>
>> I would like to use IPMs to model the population dynamics of a plant 
>> that is not only an iteroparous perennial, but also exhibits 
>> size-dependent gender switching.  Generally for this species 
>> (/Arisaema triphyllum/) plants progress from non-flowering to 
>> male-flowering to female-flowering over the course of several years 
>> depending mainly on biomass accumulation.  However, plants also can 
>> retrogress to previous states; for example, a plant that was female 
>> in year t could return as male in year t+1.
>>
>>
>> My questions:
>>
>>
>> 1. I think I need to set this up as a stage-size-dependent IPM but I 
>> am unsure how to allow individuals to move back and forth between stages.
>>
>>
>> 2. I can use polytomous logistic regression to find size-based 
>> transition probabilities among the three stages with greater power 
>> compared to a series of binomial logistic regressions.
>>
>>     a. Is polytomous log-reg appropriate for this?
>>
>>     b. Could IPMpack tools handle this?
>>
>>
>> Thanks to all for any suggestions to help get me moving forward.
>>
>>
>> Christopher Heckel
>>
>>
>> PhD candidate
>>
>> University of Pittsburgh
>>
>>
>>
>>
>>
>> _______________________________________________
>> IPMpack-users mailing list
>> IPMpack-users at lists.r-forge.r-project.org
>> http://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/ipmpack-users
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/ipmpack-users/attachments/20140611/e505056b/attachment.html>