Question about unit of migration rate

[Sakawrat]

Dear Christian,

Thank you very much for your help during the rmetasim simulation in skelesim. We have one more question to confirm about the unit of migration rate. When we prepared an input file for 3 scenarios with m = 0.02, 0.04, and 0.08. We assume that the unit is per year. Our first scenario (m=0.02) is based on the situation we observed in the field: 2 fragmented populations, a single migrant detected in pop1 (N of pop1 = 50) in the study period of 1 year. Therefore, we set m=0.02 in one of the total 3 scenarios, assuming that it's migration rate per year. Could you please verify if we could use this set up of migration rate (0.02, 0.04, 0.08) with the unit per year, to forecast the level of Fst and He in the next 100 years? Thank you very much.

Gif

[stranda]

Migration in rmetasim (parameterized by skelesim) is specified as the mean of a Poisson distribution giving number of offspring moving to the specified population. Therefore there could easily be no movement in a particular simulation time-click (usually thought of as years in sims) with small m and smaller populations (smaller populations mean less draws from the Poisson)

cheers allan

On Wed, Sep 12, 2018 at 2:59 AM Sakawrat notifications@github.com wrote:

Dear Christian,

Thank you very much for your help during the rmetasim simulation in skelesim. We have a one more question to confirm about the unit of migration rate. When we prepared an input file for 3 scenarios with m = 0.02, 0.04, and 0.08. We assume that the unit is per year. Our first scenario (m=0.02) is based on the situation we observed in the field: 2 fragmented populations, a single migrant detected in pop1 (N of pop1 = 50) in the study period of 1 year. Therefore, we set m=0.02 in one of the total 3 scenarios, assuming that it's migration rate per year. Could you please verify if we could use this set up of migration rate (0.02, 0.04, 0.08) with the unit per year, to forecast the level of Fst and He in the next 100 years? Thank you very much.

Gif

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/christianparobek/skeleSim/issues/74, or mute the thread https://github.com/notifications/unsubscribe-auth/ABoaS9YrNZfZG_WadvZLrljo84KHZH9iks5uaLDlgaJpZM4Wkwsc .

-- Allan Strand Department of Biology/Grice Marine Lab College of Charleston Charleston, SC 29424 ph: 843-953-9189 http://linum.cofc.edu

[Sakawrat]

Thank you Allan for your prompt reply. Could you please clarify one issue for me. In your R package manual of skelesim (skelesim.pdf), you mentioed the parameters "mig.rates" and "num.gen", so I assume that the unit of mig.rates, for example mig.rates=0.01 is the average number of migrant drew from the poisson distribution "per generation", but the current input of time to simulate now change to "year". Therefore, does it mean that I need to specify mig.rates by assume that it's the "per year" rate?

Thank you very much Gif

[stranda]

I think this represents some sloppiness in the description of time in rmetasim. All on me, I'm afraid.

Time is arbitrary in rmetasim, I have used generation and year interchangeably; one could easily use the software to simulate multiple time points within a year, for example an organism with multiple generations per year. So it might be best to think of each "turn of the crank" in rmetasim simulations (survive-reproduce-carry_capacity_imposition) as a 'time-click' until you set up a specific model for an organism.

To answer your question though, yes these dispersal means are per time-click so in most simulations that translates into per-year

On Thu, Sep 13, 2018 at 8:49 AM Sakawrat notifications@github.com wrote:

Thank you Allan for your prompt reply. Could you please clarify one issue for me. In your R package manual of skelesim (skelesim.pdf), you mentioed the parameters "mig.rates" and "num.gen", so I assume that the unit of mig.rates, for example mig.rates=0.01 is the average number of migrant drew from the poisson distribution "per generation", but the current input of time to simulate now change to "year". Therefore, does it mean that I need to specify mig.rates by assume that it's the "per year" rate?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/christianparobek/skeleSim/issues/74#issuecomment-420994640, or mute the thread https://github.com/notifications/unsubscribe-auth/ABoaSxMlnkOKO2APqiCgKXxZYSHSx4GBks5ualRlgaJpZM4Wkwsc .