Ne/Nc ratio

[Viktoria Köppä]

Hi everyone

Given the discussion Monday we are wondering how others have approached the question of known ratio. 

This question is prompted in section 4 when you tick that there is census data for populations. 

I wonder what sources you have used to answer this question, and for approx how many species? I also wonder if you have used closely related species to answer this question, and in that case how closely related? 

I have used the supplementary table from Hoban et al 2020 Biol. consv. However some of these values are very high and some species have more than one estimate. 

Best regards

Viktoria (Swedish team)

[Joachim Mergeay]

Hi Viktoria,

good question. The table in Hoban et al. 2020 is a compilation of data from very different sources, and I wouldn't bet my money on more than 30% of these to actually represent reliable Ne/Nc, because both are notoriously hard to estimate at the same spatial scale. Quite a few estimates are from populations kept in captivity, which often reduces the variance in reproductive success, thereby increasing the Ne/Nc artificially.

Also, some species show Ne/Nc >1, which is, although mathematically not entirely impossible, extremely unlikely. 

There are, however, general trends, which I think relate to the possible range of variance in reproductive success Vk.

Species which have over their lifetime a very limited number of offspring will tend to have higher Ne/Nc, simply because the variance is mathematically bound to be smaller. 
A black rhinoceros, for example, will have a max of 8 offspring across its lifetime. So the variance in the population can never be as large as with a species capable of having millions of offspring. 

In general, we see larger Ne/Nc in large mammals. In wolves, it's about 1/5, and they have pretty large Vk for a large mammal. Bears will likely have larger Ne/Nc, because Vk will be smaller than in wolves. 

Also in birds, Vk is never very large, because they never produce massive amounts of eggs as they need to be able to fly.

In reptiles I see Ne/Nc close to 0.10 (own data, which I trust pretty well). 

Amphibians: depending on the species you have a few to 1000s of offspring. Ne/Nc I expect to be close to 1/10 often as well. and with bony fish, trees, corals, bivalves, you get huge ranges of survival, and hence high Vk, and low Ne/Nc

Mark also that when Nc is very small, Vk is by definition smaller as well. Take the extreme example of Nc=2. Since both parents have the same reproductive output, Vk=0, and Ne=(4Nc-2)/(Vk+2)=3, which is 1.5 times Nc. So when population sizes are very (very!) small Ne/Nc will be larger than expected. 

At the extreme end of the range, we have species with extreme fertility, such as marine fishes, trees, orchids, corals, ... 
You can wonder to what extent Ne is still a relevant metric in such populations. We silently assume that the most important force affecting Ne is genetic drift (Vk among individuals caused by chance alone), but when you produce 10 million offspring, and on average only 2 survive, natural selection is going to be hugely important, and much more important than genetic drift in defining the genetic makeup of those that survive to the next generation. The two survivors will thus not be a random sample of the genetic makeup of the parents. Ne may be small due to selection, but that's OK. What we don't want is genetic drift affecting Ne. This happens inevitably when the number of offspring, the number of tickets in the lottery of life, is small. When you're putting millions in the raffle, there's bound to be a bunch of decent genotypes for a myriad of adaptive traits. 

In such species, the guidelines for Nc>5000 still hold, as that number will reduce different sources of stochasticity in the population to acceptable levels. I would never advocate, however, a marine fish stock to be in safe bounds if Nc=5000. But that's for other reasons than genetic ones. 

I would love to make a statistical analysis of what traits affect Ne/Nc, but for this you first need reliable Ne and Nc estimates, plus life history trait values. But this brings us back to square one... 

Joachim  

--
You received this message because you are subscribed to the Google Groups "Genetic indicators project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/7741ddc2-e495-4aca-8097-febf6296a367n%40googlegroups.com.

--

Joachim Mergeay (he/him)

Research Institute for Nature and Forest - Belgium                                  

Associate Professor KULeuven Ecology Evolution and Biodiversity Conservation

Tel: +32 499 942 942

Meldpunt wolven Vlaamse overheid : wo...@inbo.be

///////////////////////////////////////////

[vcmi]

Hi Joachim

Thank you for this very thorough answer. That is the same discussion me and Linda had, that some of the values were unlikely, and that we should therefore be careful with applying other values than 0.10. This reminds me of the discussions from Brasov, concerning traits to take into consideration. 

I assume from your answer that you have mainly relied on your own data for the question in KOBO? 

Best // Viktoria

To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/CAFP9MUsR92%3Dz-5%3DcUsOTefQ-L9cE%3D98v3MpdgOaULnG9FxmSTA%40mail.gmail.com.

[Alicia Mastretta]

I haven't estimated it, but I would say only around 5% of the entries have data in this field. Considering all countries.

To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/HE1PR0902MB2089E78C36CC7DADEB8E5518812DA%40HE1PR0902MB2089.eurprd09.prod.outlook.com.