Scale digestion rate in DEB

[drinky fan]

Hi Mike,

I am running a standard DEB model for a viper species. It is gape-limted species so the feeding is discrete. We measured the time from predation to excretion in the laboratory condition. I assumed the individuals had assimilated food in gut into reserve completely during this period and I want to use it to calibrate digestion-related parameters of DEB (currently the DEB parameters refers to a related species in add-my-pet). Based on digestion rate of delta E_s = {PAm} * surface area * kap_X^-1 (given a constant body temperature and the energy amount of the prey), I calculated a digestion time, which was much lower than the observation. I tried to scale maximum assimilation power ({Pam}) and digestion efficiency (kap_X), but (1) {PAm} is controled internally by somatic maintain power (line 1485 P_AM_REF=P_MREF*ZFACT/KAP in ECTOTHREM.f); (2) too low kap_X is biologically meaningless. My questions are (1) how can I scale the digestion rate to match the observation better and (2) what is the idea behind the link between {P_Am} and [P_M]. Is it due to the gut model?

Thanks for you help.

Best regards,

Junqi

[drinky fan]

correct: I calculated a digestion time, which was much longer than the observation.

[NicheMapR]

Hi Junqi,

Most of the DEB parameters need to be estimated simultaneously from life history data in one hit because there's a one-to-many relationship between the core parameters and the life history observations. You can't change those parameters once you've estimated them because you'll end up with a model that doesn't fit the original data used to estimate the parameters as well.  The maximum specific assimilation rate {p_Am} is a primary/core parameter like the specific maintenance rate [p_M]. In the estimation procedure you get an estimate from the data of [p_M] and the "zoom factor", z where {p_Am} = z * [p_M] / kappa. So you shouldn't tweak z or {p_Am} or [p_M] if you want to maintain the fit of your model to the original data used to estimate the parameters.

The parameters you don't usually estimate (because of lack of data on the feeding process), and that link to your problem, are the structure-specific energy content of the stomach, [E_sm], the maximum surface area-specific food intake rate {p_Xm}, the digestion efficiency kappa_X and the half saturation constant K. You can play around with these and try and find sensible values if you are trying to model the feeding process. They will just alter the amount of food needed to be eaten and how big the stomach can get, and the assimilation process and everything downstream of that remains unchanged. If you are finding the digestion rates are too slow or fast it has to do with your original DEB parameter estimates for the core parameters.

For gape-limited predators I have developed some code to simulate the feeding process that I'll add to the package at some point this year but it's not quite ready yet. It's a very simple stomach model so don't expect too much from it :).

All the best,

Mike